Initial commit
This commit is contained in:
Zhongwei Li
129
skills/clinical-decision-support/README.md
Normal file
129
skills/clinical-decision-support/README.md
Normal file
@@ -0,0 +1,129 @@
|
||||
# Clinical Decision Support Skill
|
||||
|
||||
Professional clinical decision support documents for medical professionals in pharmaceutical and clinical research settings.
|
||||
|
||||
## Quick Start
|
||||
|
||||
This skill enables generation of three types of clinical documents:
|
||||
|
||||
1. **Individual Patient Treatment Plans** - Personalized protocols for specific patients
|
||||
2. **Patient Cohort Analysis** - Biomarker-stratified group analyses with outcomes
|
||||
3. **Treatment Recommendation Reports** - Evidence-based clinical guidelines
|
||||
|
||||
All documents are generated as compact, professional LaTeX/PDF files.
|
||||
|
||||
## Directory Structure
|
||||
|
||||
```
|
||||
clinical-decision-support/
|
||||
├── SKILL.md # Main skill definition
|
||||
├── README.md # This file
|
||||
│
|
||||
├── references/ # Clinical guidance documents
|
||||
│ ├── patient_cohort_analysis.md
|
||||
│ ├── treatment_recommendations.md
|
||||
│ ├── clinical_decision_algorithms.md
|
||||
│ ├── biomarker_classification.md
|
||||
│ ├── outcome_analysis.md
|
||||
│ └── evidence_synthesis.md
|
||||
│
|
||||
├── assets/ # Templates and examples
|
||||
│ ├── cohort_analysis_template.tex
|
||||
│ ├── treatment_recommendation_template.tex
|
||||
│ ├── clinical_pathway_template.tex
|
||||
│ ├── biomarker_report_template.tex
|
||||
│ ├── example_gbm_cohort.md
|
||||
│ ├── recommendation_strength_guide.md
|
||||
│ └── color_schemes.tex
|
||||
│
|
||||
└── scripts/ # Analysis and generation tools
|
||||
├── generate_survival_analysis.py
|
||||
├── create_cohort_tables.py
|
||||
├── build_decision_tree.py
|
||||
├── biomarker_classifier.py
|
||||
└── validate_cds_document.py
|
||||
```
|
||||
|
||||
## Example Use Cases
|
||||
|
||||
### Create a Patient Cohort Analysis
|
||||
```
|
||||
> Analyze a cohort of 45 NSCLC patients stratified by PD-L1 expression
|
||||
(<1%, 1-49%, ≥50%) including ORR, PFS, and OS outcomes
|
||||
```
|
||||
|
||||
### Generate Treatment Recommendations
|
||||
```
|
||||
> Create evidence-based treatment recommendations for HER2-positive
|
||||
metastatic breast cancer with GRADE methodology
|
||||
```
|
||||
|
||||
### Build Clinical Pathway
|
||||
```
|
||||
> Generate a clinical decision algorithm for acute chest pain
|
||||
management with TIMI risk score
|
||||
```
|
||||
|
||||
## Key Features
|
||||
|
||||
- **GRADE Methodology**: Evidence quality grading (High/Moderate/Low/Very Low)
|
||||
- **Recommendation Strength**: Strong (Grade 1) vs Conditional (Grade 2)
|
||||
- **Biomarker Integration**: Genomic, expression, and molecular subtype classification
|
||||
- **Statistical Analysis**: Kaplan-Meier, Cox regression, log-rank tests
|
||||
- **Guideline Concordance**: NCCN, ASCO, ESMO, AHA/ACC integration
|
||||
- **Professional Output**: 0.5in margins, color-coded boxes, publication-ready
|
||||
|
||||
## Dependencies
|
||||
|
||||
Python scripts require:
|
||||
- `pandas`, `numpy`, `scipy`: Data analysis and statistics
|
||||
- `lifelines`: Survival analysis (Kaplan-Meier, Cox regression)
|
||||
- `matplotlib`: Visualization
|
||||
- `pyyaml` (optional): YAML input for decision trees
|
||||
|
||||
Install with:
|
||||
```bash
|
||||
pip install pandas numpy scipy lifelines matplotlib pyyaml
|
||||
```
|
||||
|
||||
## References Included
|
||||
|
||||
1. **Patient Cohort Analysis**: Stratification methods, biomarker correlations, statistical comparisons
|
||||
2. **Treatment Recommendations**: Evidence grading, treatment sequencing, special populations
|
||||
3. **Clinical Decision Algorithms**: Risk scores, decision trees, TikZ flowcharts
|
||||
4. **Biomarker Classification**: Genomic alterations, molecular subtypes, companion diagnostics
|
||||
5. **Outcome Analysis**: Survival methods, response criteria (RECIST), effect sizes
|
||||
6. **Evidence Synthesis**: Guideline integration, systematic reviews, meta-analysis
|
||||
|
||||
## Templates Provided
|
||||
|
||||
1. **Cohort Analysis**: Demographics table, biomarker profile, outcomes, statistics, recommendations
|
||||
2. **Treatment Recommendations**: Evidence review, GRADE-graded options, monitoring, decision algorithm
|
||||
3. **Clinical Pathway**: TikZ flowchart with risk stratification and urgency-coded actions
|
||||
4. **Biomarker Report**: Genomic profiling with tier-based actionability and therapy matching
|
||||
|
||||
## Scripts Included
|
||||
|
||||
1. **`generate_survival_analysis.py`**: Create Kaplan-Meier curves with hazard ratios
|
||||
2. **`create_cohort_tables.py`**: Generate baseline, efficacy, and safety tables
|
||||
3. **`build_decision_tree.py`**: Convert text/JSON to TikZ flowcharts
|
||||
4. **`biomarker_classifier.py`**: Stratify patients by PD-L1, HER2, molecular subtypes
|
||||
5. **`validate_cds_document.py`**: Quality checks for completeness and compliance
|
||||
|
||||
## Integration
|
||||
|
||||
Integrates with existing skills:
|
||||
- **scientific-writing**: Citation management, statistical reporting
|
||||
- **clinical-reports**: Medical terminology, HIPAA compliance
|
||||
- **scientific-schematics**: TikZ flowcharts
|
||||
|
||||
## Version
|
||||
|
||||
Version 1.0 - Initial release
|
||||
Created: November 2024
|
||||
Last Updated: November 5, 2024
|
||||
|
||||
## Questions or Feedback
|
||||
|
||||
This skill was designed for pharmaceutical and clinical research professionals creating clinical decision support documents. For questions about usage or suggestions for improvements, contact the Scientific Writer development team.
|
||||
|
||||
501
skills/clinical-decision-support/SKILL.md
Normal file
501
skills/clinical-decision-support/SKILL.md
Normal file
@@ -0,0 +1,501 @@
|
||||
---
|
||||
name: clinical-decision-support
|
||||
description: "Generate professional clinical decision support (CDS) documents for pharmaceutical and clinical research settings, including patient cohort analyses (biomarker-stratified with outcomes) and treatment recommendation reports (evidence-based guidelines with decision algorithms). Supports GRADE evidence grading, statistical analysis (hazard ratios, survival curves, waterfall plots), biomarker integration, and regulatory compliance. Outputs publication-ready LaTeX/PDF format optimized for drug development, clinical research, and evidence synthesis."
|
||||
allowed-tools: [Read, Write, Edit, Bash]
|
||||
---
|
||||
|
||||
# Clinical Decision Support Documents
|
||||
|
||||
## Description
|
||||
|
||||
Generate professional clinical decision support (CDS) documents for pharmaceutical companies, clinical researchers, and medical decision-makers. This skill specializes in analytical, evidence-based documents that inform treatment strategies and drug development:
|
||||
|
||||
1. **Patient Cohort Analysis** - Biomarker-stratified group analyses with statistical outcome comparisons
|
||||
2. **Treatment Recommendation Reports** - Evidence-based clinical guidelines with GRADE grading and decision algorithms
|
||||
|
||||
All documents are generated as publication-ready LaTeX/PDF files optimized for pharmaceutical research, regulatory submissions, and clinical guideline development.
|
||||
|
||||
**Note:** For individual patient treatment plans at the bedside, use the `treatment-plans` skill instead. This skill focuses on group-level analyses and evidence synthesis for pharmaceutical/research settings.
|
||||
|
||||
## Capabilities
|
||||
|
||||
### Document Types
|
||||
|
||||
**Patient Cohort Analysis**
|
||||
- Biomarker-based patient stratification (molecular subtypes, gene expression, IHC)
|
||||
- Molecular subtype classification (e.g., GBM mesenchymal-immune-active vs proneural, breast cancer subtypes)
|
||||
- Outcome metrics with statistical analysis (OS, PFS, ORR, DOR, DCR)
|
||||
- Statistical comparisons between subgroups (hazard ratios, p-values, 95% CI)
|
||||
- Survival analysis with Kaplan-Meier curves and log-rank tests
|
||||
- Efficacy tables and waterfall plots
|
||||
- Comparative effectiveness analyses
|
||||
- Pharmaceutical cohort reporting (trial subgroups, real-world evidence)
|
||||
|
||||
**Treatment Recommendation Reports**
|
||||
- Evidence-based treatment guidelines for specific disease states
|
||||
- Strength of recommendation grading (GRADE system: 1A, 1B, 2A, 2B, 2C)
|
||||
- Quality of evidence assessment (high, moderate, low, very low)
|
||||
- Treatment algorithm flowcharts with TikZ diagrams
|
||||
- Line-of-therapy sequencing based on biomarkers
|
||||
- Decision pathways with clinical and molecular criteria
|
||||
- Pharmaceutical strategy documents
|
||||
- Clinical guideline development for medical societies
|
||||
|
||||
### Clinical Features
|
||||
|
||||
- **Biomarker Integration**: Genomic alterations (mutations, CNV, fusions), gene expression signatures, IHC markers, PD-L1 scoring
|
||||
- **Statistical Analysis**: Hazard ratios, p-values, confidence intervals, survival curves, Cox regression, log-rank tests
|
||||
- **Evidence Grading**: GRADE system (1A/1B/2A/2B/2C), Oxford CEBM levels, quality of evidence assessment
|
||||
- **Clinical Terminology**: SNOMED-CT, LOINC, proper medical nomenclature, trial nomenclature
|
||||
- **Regulatory Compliance**: HIPAA de-identification, confidentiality headers, ICH-GCP alignment
|
||||
- **Professional Formatting**: Compact 0.5in margins, color-coded recommendations, publication-ready, suitable for regulatory submissions
|
||||
|
||||
## Pharmaceutical and Research Use Cases
|
||||
|
||||
This skill is specifically designed for pharmaceutical and clinical research applications:
|
||||
|
||||
**Drug Development**
|
||||
- **Phase 2/3 Trial Analyses**: Biomarker-stratified efficacy and safety analyses
|
||||
- **Subgroup Analyses**: Forest plots showing treatment effects across patient subgroups
|
||||
- **Companion Diagnostic Development**: Linking biomarkers to drug response
|
||||
- **Regulatory Submissions**: IND/NDA documentation with evidence summaries
|
||||
|
||||
**Medical Affairs**
|
||||
- **KOL Education Materials**: Evidence-based treatment algorithms for thought leaders
|
||||
- **Medical Strategy Documents**: Competitive landscape and positioning strategies
|
||||
- **Advisory Board Materials**: Cohort analyses and treatment recommendation frameworks
|
||||
- **Publication Planning**: Manuscript-ready analyses for peer-reviewed journals
|
||||
|
||||
**Clinical Guidelines**
|
||||
- **Guideline Development**: Evidence synthesis with GRADE methodology for specialty societies
|
||||
- **Consensus Recommendations**: Multi-stakeholder treatment algorithm development
|
||||
- **Practice Standards**: Biomarker-based treatment selection criteria
|
||||
- **Quality Measures**: Evidence-based performance metrics
|
||||
|
||||
**Real-World Evidence**
|
||||
- **RWE Cohort Studies**: Retrospective analyses of patient cohorts from EMR data
|
||||
- **Comparative Effectiveness**: Head-to-head treatment comparisons in real-world settings
|
||||
- **Outcomes Research**: Long-term survival and safety in clinical practice
|
||||
- **Health Economics**: Cost-effectiveness analyses by biomarker subgroup
|
||||
|
||||
## When to Use
|
||||
|
||||
Use this skill when you need to:
|
||||
|
||||
- **Analyze patient cohorts** stratified by biomarkers, molecular subtypes, or clinical characteristics
|
||||
- **Generate treatment recommendation reports** with evidence grading for clinical guidelines or pharmaceutical strategies
|
||||
- **Compare outcomes** between patient subgroups with statistical analysis (survival, response rates, hazard ratios)
|
||||
- **Produce pharmaceutical research documents** for drug development, clinical trials, or regulatory submissions
|
||||
- **Develop clinical practice guidelines** with GRADE evidence grading and decision algorithms
|
||||
- **Document biomarker-guided therapy selection** at the population level (not individual patients)
|
||||
- **Synthesize evidence** from multiple trials or real-world data sources
|
||||
- **Create clinical decision algorithms** with flowcharts for treatment sequencing
|
||||
|
||||
**Do NOT use this skill for:**
|
||||
- Individual patient treatment plans (use `treatment-plans` skill)
|
||||
- Bedside clinical care documentation (use `treatment-plans` skill)
|
||||
- Simple patient-specific treatment protocols (use `treatment-plans` skill)
|
||||
|
||||
## Visual Enhancement with Scientific Schematics
|
||||
|
||||
**When creating documents with this skill, always consider adding scientific diagrams and schematics to enhance visual communication.**
|
||||
|
||||
If your document does not already contain schematics or diagrams:
|
||||
- Use the **scientific-schematics** skill to generate AI-powered publication-quality diagrams
|
||||
- Simply describe your desired diagram in natural language
|
||||
- Nano Banana Pro will automatically generate, review, and refine the schematic
|
||||
|
||||
**For new documents:** Scientific schematics should be generated by default to visually represent key concepts, workflows, architectures, or relationships described in the text.
|
||||
|
||||
**How to generate schematics:**
|
||||
```bash
|
||||
python scripts/generate_schematic.py "your diagram description" -o figures/output.png
|
||||
```
|
||||
|
||||
The AI will automatically:
|
||||
- Create publication-quality images with proper formatting
|
||||
- Review and refine through multiple iterations
|
||||
- Ensure accessibility (colorblind-friendly, high contrast)
|
||||
- Save outputs in the figures/ directory
|
||||
|
||||
**When to add schematics:**
|
||||
- Clinical decision algorithm flowcharts
|
||||
- Treatment pathway diagrams
|
||||
- Biomarker stratification trees
|
||||
- Patient cohort flow diagrams (CONSORT-style)
|
||||
- Survival curve visualizations
|
||||
- Molecular mechanism diagrams
|
||||
- Any complex concept that benefits from visualization
|
||||
|
||||
For detailed guidance on creating schematics, refer to the scientific-schematics skill documentation.
|
||||
|
||||
---
|
||||
|
||||
## Document Structure
|
||||
|
||||
**CRITICAL REQUIREMENT: All clinical decision support documents MUST begin with a complete executive summary on page 1 that spans the entire first page before any table of contents or detailed sections.**
|
||||
|
||||
### Page 1 Executive Summary Structure
|
||||
|
||||
The first page of every CDS document should contain ONLY the executive summary with the following components:
|
||||
|
||||
**Required Elements (all on page 1):**
|
||||
1. **Document Title and Type**
|
||||
- Main title (e.g., "Biomarker-Stratified Cohort Analysis" or "Evidence-Based Treatment Recommendations")
|
||||
- Subtitle with disease state and focus
|
||||
|
||||
2. **Report Information Box** (using colored tcolorbox)
|
||||
- Document type and purpose
|
||||
- Date of analysis/report
|
||||
- Disease state and patient population
|
||||
- Author/institution (if applicable)
|
||||
- Analysis framework or methodology
|
||||
|
||||
3. **Key Findings Boxes** (3-5 colored boxes using tcolorbox)
|
||||
- **Primary Results** (blue box): Main efficacy/outcome findings
|
||||
- **Biomarker Insights** (green box): Key molecular subtype findings
|
||||
- **Clinical Implications** (yellow/orange box): Actionable treatment implications
|
||||
- **Statistical Summary** (gray box): Hazard ratios, p-values, key statistics
|
||||
- **Safety Highlights** (red box, if applicable): Critical adverse events or warnings
|
||||
|
||||
**Visual Requirements:**
|
||||
- Use `\thispagestyle{empty}` to remove page numbers from page 1
|
||||
- All content must fit on page 1 (before `\newpage`)
|
||||
- Use colored tcolorbox environments with different colors for visual hierarchy
|
||||
- Boxes should be scannable and highlight most critical information
|
||||
- Use bullet points, not narrative paragraphs
|
||||
- End page 1 with `\newpage` before table of contents or detailed sections
|
||||
|
||||
**Example First Page LaTeX Structure:**
|
||||
```latex
|
||||
\maketitle
|
||||
\thispagestyle{empty}
|
||||
|
||||
% Report Information Box
|
||||
\begin{tcolorbox}[colback=blue!5!white, colframe=blue!75!black, title=Report Information]
|
||||
\textbf{Document Type:} Patient Cohort Analysis\\
|
||||
\textbf{Disease State:} HER2-Positive Metastatic Breast Cancer\\
|
||||
\textbf{Analysis Date:} \today\\
|
||||
\textbf{Population:} 60 patients, biomarker-stratified by HR status
|
||||
\end{tcolorbox}
|
||||
|
||||
\vspace{0.3cm}
|
||||
|
||||
% Key Finding #1: Primary Results
|
||||
\begin{tcolorbox}[colback=blue!5!white, colframe=blue!75!black, title=Primary Efficacy Results]
|
||||
\begin{itemize}
|
||||
\item Overall ORR: 72\% (95\% CI: 59-83\%)
|
||||
\item Median PFS: 18.5 months (95\% CI: 14.2-22.8)
|
||||
\item Median OS: 35.2 months (95\% CI: 28.1-NR)
|
||||
\end{itemize}
|
||||
\end{tcolorbox}
|
||||
|
||||
\vspace{0.3cm}
|
||||
|
||||
% Key Finding #2: Biomarker Insights
|
||||
\begin{tcolorbox}[colback=green!5!white, colframe=green!75!black, title=Biomarker Stratification Findings]
|
||||
\begin{itemize}
|
||||
\item HR+/HER2+: ORR 68\%, median PFS 16.2 months
|
||||
\item HR-/HER2+: ORR 78\%, median PFS 22.1 months
|
||||
\item HR status significantly associated with outcomes (p=0.041)
|
||||
\end{itemize}
|
||||
\end{tcolorbox}
|
||||
|
||||
\vspace{0.3cm}
|
||||
|
||||
% Key Finding #3: Clinical Implications
|
||||
\begin{tcolorbox}[colback=orange!5!white, colframe=orange!75!black, title=Clinical Recommendations]
|
||||
\begin{itemize}
|
||||
\item Strong efficacy observed regardless of HR status (Grade 1A)
|
||||
\item HR-/HER2+ patients showed numerically superior outcomes
|
||||
\item Treatment recommended for all HER2+ MBC patients
|
||||
\end{itemize}
|
||||
\end{tcolorbox}
|
||||
|
||||
\newpage
|
||||
\tableofcontents % TOC on page 2
|
||||
\newpage % Detailed content starts page 3
|
||||
```
|
||||
|
||||
### Patient Cohort Analysis (Detailed Sections - Page 3+)
|
||||
- **Cohort Characteristics**: Demographics, baseline features, patient selection criteria
|
||||
- **Biomarker Stratification**: Molecular subtypes, genomic alterations, IHC profiles
|
||||
- **Treatment Exposure**: Therapies received, dosing, treatment duration by subgroup
|
||||
- **Outcome Analysis**: Response rates (ORR, DCR), survival data (OS, PFS), DOR
|
||||
- **Statistical Methods**: Kaplan-Meier survival curves, hazard ratios, log-rank tests, Cox regression
|
||||
- **Subgroup Comparisons**: Biomarker-stratified efficacy, forest plots, statistical significance
|
||||
- **Safety Profile**: Adverse events by subgroup, dose modifications, discontinuations
|
||||
- **Clinical Recommendations**: Treatment implications based on biomarker profiles
|
||||
- **Figures**: Waterfall plots, swimmer plots, survival curves, forest plots
|
||||
- **Tables**: Demographics table, biomarker frequency, outcomes by subgroup
|
||||
|
||||
### Treatment Recommendation Reports (Detailed Sections - Page 3+)
|
||||
|
||||
**Page 1 Executive Summary for Treatment Recommendations should include:**
|
||||
1. **Report Information Box**: Disease state, guideline version/date, target population
|
||||
2. **Key Recommendations Box** (green): Top 3-5 GRADE-graded recommendations by line of therapy
|
||||
3. **Biomarker Decision Criteria Box** (blue): Key molecular markers influencing treatment selection
|
||||
4. **Evidence Summary Box** (gray): Major trials supporting recommendations (e.g., KEYNOTE-189, FLAURA)
|
||||
5. **Critical Monitoring Box** (orange/red): Essential safety monitoring requirements
|
||||
|
||||
**Detailed Sections (Page 3+):**
|
||||
- **Clinical Context**: Disease state, epidemiology, current treatment landscape
|
||||
- **Target Population**: Patient characteristics, biomarker criteria, staging
|
||||
- **Evidence Review**: Systematic literature synthesis, guideline summary, trial data
|
||||
- **Treatment Options**: Available therapies with mechanism of action
|
||||
- **Evidence Grading**: GRADE assessment for each recommendation (1A, 1B, 2A, 2B, 2C)
|
||||
- **Recommendations by Line**: First-line, second-line, subsequent therapies
|
||||
- **Biomarker-Guided Selection**: Decision criteria based on molecular profiles
|
||||
- **Treatment Algorithms**: TikZ flowcharts showing decision pathways
|
||||
- **Monitoring Protocol**: Safety assessments, efficacy monitoring, dose modifications
|
||||
- **Special Populations**: Elderly, renal/hepatic impairment, comorbidities
|
||||
- **References**: Full bibliography with trial names and citations
|
||||
|
||||
## Output Format
|
||||
|
||||
**MANDATORY FIRST PAGE REQUIREMENT:**
|
||||
- **Page 1**: Full-page executive summary with 3-5 colored tcolorbox elements
|
||||
- **Page 2**: Table of contents (optional)
|
||||
- **Page 3+**: Detailed sections with methods, results, figures, tables
|
||||
|
||||
**Document Specifications:**
|
||||
- **Primary**: LaTeX/PDF with 0.5in margins for compact, data-dense presentation
|
||||
- **Length**: Typically 5-15 pages (1 page executive summary + 4-14 pages detailed content)
|
||||
- **Style**: Publication-ready, pharmaceutical-grade, suitable for regulatory submissions
|
||||
- **First Page**: Always a complete executive summary spanning entire page 1 (see Document Structure section)
|
||||
|
||||
**Visual Elements:**
|
||||
- **Colors**:
|
||||
- Page 1 boxes: blue=data/information, green=biomarkers/recommendations, yellow/orange=clinical implications, red=warnings
|
||||
- Recommendation boxes (green=strong recommendation, yellow=conditional, blue=research needed)
|
||||
- Biomarker stratification (color-coded molecular subtypes)
|
||||
- Statistical significance (color-coded p-values, hazard ratios)
|
||||
- **Tables**:
|
||||
- Demographics with baseline characteristics
|
||||
- Biomarker frequency by subgroup
|
||||
- Outcomes table (ORR, PFS, OS, DOR by molecular subtype)
|
||||
- Adverse events by cohort
|
||||
- Evidence summary tables with GRADE ratings
|
||||
- **Figures**:
|
||||
- Kaplan-Meier survival curves with log-rank p-values and number at risk tables
|
||||
- Waterfall plots showing best response by patient
|
||||
- Forest plots for subgroup analyses with confidence intervals
|
||||
- TikZ decision algorithm flowcharts
|
||||
- Swimmer plots for individual patient timelines
|
||||
- **Statistics**: Hazard ratios with 95% CI, p-values, median survival times, landmark survival rates
|
||||
- **Compliance**: De-identification per HIPAA Safe Harbor, confidentiality notices for proprietary data
|
||||
|
||||
## Integration
|
||||
|
||||
This skill integrates with:
|
||||
- **scientific-writing**: Citation management, statistical reporting, evidence synthesis
|
||||
- **clinical-reports**: Medical terminology, HIPAA compliance, regulatory documentation
|
||||
- **scientific-schematics**: TikZ flowcharts for decision algorithms and treatment pathways
|
||||
- **treatment-plans**: Individual patient applications of cohort-derived insights (bidirectional)
|
||||
|
||||
## Key Differentiators from Treatment-Plans Skill
|
||||
|
||||
**Clinical Decision Support (this skill):**
|
||||
- **Audience**: Pharmaceutical companies, clinical researchers, guideline committees, medical affairs
|
||||
- **Scope**: Population-level analyses, evidence synthesis, guideline development
|
||||
- **Focus**: Biomarker stratification, statistical comparisons, evidence grading
|
||||
- **Output**: Multi-page analytical documents (5-15 pages typical) with extensive figures and tables
|
||||
- **Use Cases**: Drug development, regulatory submissions, clinical practice guidelines, medical strategy
|
||||
- **Example**: "Analyze 60 HER2+ breast cancer patients by hormone receptor status with survival outcomes"
|
||||
|
||||
**Treatment-Plans Skill:**
|
||||
- **Audience**: Clinicians, patients, care teams
|
||||
- **Scope**: Individual patient care planning
|
||||
- **Focus**: SMART goals, patient-specific interventions, monitoring plans
|
||||
- **Output**: Concise 1-4 page actionable care plans
|
||||
- **Use Cases**: Bedside clinical care, EMR documentation, patient-centered planning
|
||||
- **Example**: "Create treatment plan for a 55-year-old patient with newly diagnosed type 2 diabetes"
|
||||
|
||||
**When to use each:**
|
||||
- Use **clinical-decision-support** for: cohort analyses, biomarker stratification studies, treatment guideline development, pharmaceutical strategy documents
|
||||
- Use **treatment-plans** for: individual patient care plans, treatment protocols for specific patients, bedside clinical documentation
|
||||
|
||||
## Example Usage
|
||||
|
||||
### Patient Cohort Analysis
|
||||
|
||||
**Example 1: NSCLC Biomarker Stratification**
|
||||
```
|
||||
> Analyze a cohort of 45 NSCLC patients stratified by PD-L1 expression (<1%, 1-49%, ≥50%)
|
||||
> receiving pembrolizumab. Include outcomes: ORR, median PFS, median OS with hazard ratios
|
||||
> comparing PD-L1 ≥50% vs <50%. Generate Kaplan-Meier curves and waterfall plot.
|
||||
```
|
||||
|
||||
**Example 2: GBM Molecular Subtype Analysis**
|
||||
```
|
||||
> Generate cohort analysis for 30 GBM patients classified into Cluster 1 (Mesenchymal-Immune-Active)
|
||||
> and Cluster 2 (Proneural) molecular subtypes. Compare outcomes including median OS, 6-month PFS rate,
|
||||
> and response to TMZ+bevacizumab. Include biomarker profile table and statistical comparison.
|
||||
```
|
||||
|
||||
**Example 3: Breast Cancer HER2 Cohort**
|
||||
```
|
||||
> Analyze 60 HER2-positive metastatic breast cancer patients treated with trastuzumab-deruxtecan,
|
||||
> stratified by prior trastuzumab exposure (yes/no). Include ORR, DOR, median PFS with forest plot
|
||||
> showing subgroup analyses by hormone receptor status, brain metastases, and number of prior lines.
|
||||
```
|
||||
|
||||
### Treatment Recommendation Report
|
||||
|
||||
**Example 1: HER2+ Metastatic Breast Cancer Guidelines**
|
||||
```
|
||||
> Create evidence-based treatment recommendations for HER2-positive metastatic breast cancer including
|
||||
> biomarker-guided therapy selection. Use GRADE system to grade recommendations for first-line
|
||||
> (trastuzumab+pertuzumab+taxane), second-line (trastuzumab-deruxtecan), and third-line options.
|
||||
> Include decision algorithm flowchart based on brain metastases, hormone receptor status, and prior therapies.
|
||||
```
|
||||
|
||||
**Example 2: Advanced NSCLC Treatment Algorithm**
|
||||
```
|
||||
> Generate treatment recommendation report for advanced NSCLC based on PD-L1 expression, EGFR mutation,
|
||||
> ALK rearrangement, and performance status. Include GRADE-graded recommendations for each molecular subtype,
|
||||
> TikZ flowchart for biomarker-directed therapy selection, and evidence tables from KEYNOTE-189, FLAURA,
|
||||
> and CheckMate-227 trials.
|
||||
```
|
||||
|
||||
**Example 3: Multiple Myeloma Line-of-Therapy Sequencing**
|
||||
```
|
||||
> Create treatment algorithm for newly diagnosed multiple myeloma through relapsed/refractory setting.
|
||||
> Include GRADE recommendations for transplant-eligible vs ineligible, high-risk cytogenetics considerations,
|
||||
> and sequencing of daratumumab, carfilzomib, and CAR-T therapy. Provide flowchart showing decision points
|
||||
> at each line of therapy.
|
||||
```
|
||||
|
||||
## Key Features
|
||||
|
||||
### Biomarker Classification
|
||||
- Genomic: Mutations, CNV, gene fusions
|
||||
- Expression: RNA-seq, IHC scores
|
||||
- Molecular subtypes: Disease-specific classifications
|
||||
- Clinical actionability: Therapy selection guidance
|
||||
|
||||
### Outcome Metrics
|
||||
- Survival: OS (overall survival), PFS (progression-free survival)
|
||||
- Response: ORR (objective response rate), DOR (duration of response), DCR (disease control rate)
|
||||
- Quality: ECOG performance status, symptom burden
|
||||
- Safety: Adverse events, dose modifications
|
||||
|
||||
### Statistical Methods
|
||||
- Survival analysis: Kaplan-Meier curves, log-rank tests
|
||||
- Group comparisons: t-tests, chi-square, Fisher's exact
|
||||
- Effect sizes: Hazard ratios, odds ratios with 95% CI
|
||||
- Significance: p-values, multiple testing corrections
|
||||
|
||||
### Evidence Grading
|
||||
|
||||
**GRADE System**
|
||||
- **1A**: Strong recommendation, high-quality evidence
|
||||
- **1B**: Strong recommendation, moderate-quality evidence
|
||||
- **2A**: Weak recommendation, high-quality evidence
|
||||
- **2B**: Weak recommendation, moderate-quality evidence
|
||||
- **2C**: Weak recommendation, low-quality evidence
|
||||
|
||||
**Recommendation Strength**
|
||||
- **Strong**: Benefits clearly outweigh risks
|
||||
- **Conditional**: Trade-offs exist, patient values important
|
||||
- **Research**: Insufficient evidence, clinical trials needed
|
||||
|
||||
## Best Practices
|
||||
|
||||
### For Cohort Analyses
|
||||
|
||||
1. **Patient Selection Transparency**: Clearly document inclusion/exclusion criteria, patient flow, and reasons for exclusions
|
||||
2. **Biomarker Clarity**: Specify assay methods, platforms (e.g., FoundationOne, Caris), cut-points, and validation status
|
||||
3. **Statistical Rigor**:
|
||||
- Report hazard ratios with 95% confidence intervals, not just p-values
|
||||
- Include median follow-up time for survival analyses
|
||||
- Specify statistical tests used (log-rank, Cox regression, Fisher's exact)
|
||||
- Account for multiple comparisons when appropriate
|
||||
4. **Outcome Definitions**: Use standard criteria:
|
||||
- Response: RECIST 1.1, iRECIST for immunotherapy
|
||||
- Adverse events: CTCAE version 5.0
|
||||
- Performance status: ECOG or Karnofsky
|
||||
5. **Survival Data Presentation**:
|
||||
- Median OS/PFS with 95% CI
|
||||
- Landmark survival rates (6-month, 12-month, 24-month)
|
||||
- Number at risk tables below Kaplan-Meier curves
|
||||
- Censoring clearly indicated
|
||||
6. **Subgroup Analyses**: Pre-specify subgroups; clearly label exploratory vs pre-planned analyses
|
||||
7. **Data Completeness**: Report missing data and how it was handled
|
||||
|
||||
### For Treatment Recommendation Reports
|
||||
|
||||
1. **Evidence Grading Transparency**:
|
||||
- Use GRADE system consistently (1A, 1B, 2A, 2B, 2C)
|
||||
- Document rationale for each grade
|
||||
- Clearly state quality of evidence (high, moderate, low, very low)
|
||||
2. **Comprehensive Evidence Review**:
|
||||
- Include phase 3 randomized trials as primary evidence
|
||||
- Supplement with phase 2 data for emerging therapies
|
||||
- Note real-world evidence and meta-analyses
|
||||
- Cite trial names (e.g., KEYNOTE-189, CheckMate-227)
|
||||
3. **Biomarker-Guided Recommendations**:
|
||||
- Link specific biomarkers to therapy recommendations
|
||||
- Specify testing methods and validated assays
|
||||
- Include FDA/EMA approval status for companion diagnostics
|
||||
4. **Clinical Actionability**: Every recommendation should have clear implementation guidance
|
||||
5. **Decision Algorithm Clarity**: TikZ flowcharts should be unambiguous with clear yes/no decision points
|
||||
6. **Special Populations**: Address elderly, renal/hepatic impairment, pregnancy, drug interactions
|
||||
7. **Monitoring Guidance**: Specify safety labs, imaging, and frequency
|
||||
8. **Update Frequency**: Date recommendations and plan for periodic updates
|
||||
|
||||
### General Best Practices
|
||||
|
||||
1. **First Page Executive Summary (MANDATORY)**:
|
||||
- ALWAYS create a complete executive summary on page 1 that spans the entire first page
|
||||
- Use 3-5 colored tcolorbox elements to highlight key findings
|
||||
- No table of contents or detailed sections on page 1
|
||||
- Use `\thispagestyle{empty}` and end with `\newpage`
|
||||
- This is the single most important page - it should be scannable in 60 seconds
|
||||
2. **De-identification**: Remove all 18 HIPAA identifiers before document generation (Safe Harbor method)
|
||||
3. **Regulatory Compliance**: Include confidentiality notices for proprietary pharmaceutical data
|
||||
4. **Publication-Ready Formatting**: Use 0.5in margins, professional fonts, color-coded sections
|
||||
5. **Reproducibility**: Document all statistical methods to enable replication
|
||||
6. **Conflict of Interest**: Disclose pharmaceutical funding or relationships when applicable
|
||||
7. **Visual Hierarchy**: Use colored boxes consistently (blue=data, green=biomarkers, yellow/orange=recommendations, red=warnings)
|
||||
|
||||
## References
|
||||
|
||||
See the `references/` directory for detailed guidance on:
|
||||
- Patient cohort analysis and stratification methods
|
||||
- Treatment recommendation development
|
||||
- Clinical decision algorithms
|
||||
- Biomarker classification and interpretation
|
||||
- Outcome analysis and statistical methods
|
||||
- Evidence synthesis and grading systems
|
||||
|
||||
## Templates
|
||||
|
||||
See the `assets/` directory for LaTeX templates:
|
||||
- `cohort_analysis_template.tex` - Biomarker-stratified patient cohort analysis with statistical comparisons
|
||||
- `treatment_recommendation_template.tex` - Evidence-based clinical practice guidelines with GRADE grading
|
||||
- `clinical_pathway_template.tex` - TikZ decision algorithm flowcharts for treatment sequencing
|
||||
- `biomarker_report_template.tex` - Molecular subtype classification and genomic profile reports
|
||||
- `evidence_synthesis_template.tex` - Systematic evidence review and meta-analysis summaries
|
||||
|
||||
**Template Features:**
|
||||
- 0.5in margins for compact presentation
|
||||
- Color-coded recommendation boxes
|
||||
- Professional tables for demographics, biomarkers, outcomes
|
||||
- Built-in support for Kaplan-Meier curves, waterfall plots, forest plots
|
||||
- GRADE evidence grading tables
|
||||
- Confidentiality headers for pharmaceutical documents
|
||||
|
||||
## Scripts
|
||||
|
||||
See the `scripts/` directory for analysis and visualization tools:
|
||||
- `generate_survival_analysis.py` - Kaplan-Meier curve generation with log-rank tests, hazard ratios, 95% CI
|
||||
- `create_waterfall_plot.py` - Best response visualization for cohort analyses
|
||||
- `create_forest_plot.py` - Subgroup analysis visualization with confidence intervals
|
||||
- `create_cohort_tables.py` - Demographics, biomarker frequency, and outcomes tables
|
||||
- `build_decision_tree.py` - TikZ flowchart generation for treatment algorithms
|
||||
- `biomarker_classifier.py` - Patient stratification algorithms by molecular subtype
|
||||
- `calculate_statistics.py` - Hazard ratios, Cox regression, log-rank tests, Fisher's exact
|
||||
- `validate_cds_document.py` - Quality and compliance checks (HIPAA, statistical reporting standards)
|
||||
- `grade_evidence.py` - Automated GRADE assessment helper for treatment recommendations
|
||||
|
||||
@@ -0,0 +1,380 @@
|
||||
\documentclass[10pt,letterpaper]{article}
|
||||
|
||||
% Packages
|
||||
\usepackage[margin=0.5in]{geometry}
|
||||
\usepackage[utf8]{inputenc}
|
||||
\usepackage[T1]{fontenc}
|
||||
\usepackage{helvet}
|
||||
\renewcommand{\familydefault}{\sfdefault}
|
||||
\usepackage{xcolor}
|
||||
\usepackage{tcolorbox}
|
||||
\usepackage{array}
|
||||
\usepackage{tabularx}
|
||||
\usepackage{booktabs}
|
||||
\usepackage{enumitem}
|
||||
\usepackage{titlesec}
|
||||
\usepackage{fancyhdr}
|
||||
\usepackage{graphicx}
|
||||
|
||||
% Color definitions
|
||||
\definecolor{headerblue}{RGB}{0,102,204}
|
||||
\definecolor{tier1green}{RGB}{0,153,76}
|
||||
\definecolor{tier2orange}{RGB}{255,152,0}
|
||||
\definecolor{tier3gray}{RGB}{158,158,158}
|
||||
\definecolor{mutationred}{RGB}{244,67,54}
|
||||
\definecolor{amplificationblue}{RGB}{33,150,243}
|
||||
\definecolor{fusionpurple}{RGB}{156,39,176}
|
||||
\definecolor{highlightgray}{RGB}{240,240,240}
|
||||
|
||||
% Section formatting
|
||||
\titleformat{\section}{\normalfont\fontsize{11}{12}\bfseries\color{headerblue}}{\thesection}{0.5em}{}
|
||||
\titlespacing*{\section}{0pt}{4pt}{2pt}
|
||||
|
||||
\titleformat{\subsection}{\normalfont\fontsize{10}{11}\bfseries}{\thesubsection}{0.5em}{}
|
||||
\titlespacing*{\subsection}{0pt}{3pt}{1pt}
|
||||
|
||||
% List formatting
|
||||
\setlist[itemize]{leftmargin=*,itemsep=0pt,parsep=0pt,topsep=1pt}
|
||||
\setlist[enumerate]{leftmargin=*,itemsep=0pt,parsep=0pt,topsep=1pt}
|
||||
|
||||
\setlength{\parindent}{0pt}
|
||||
\setlength{\parskip}{2pt}
|
||||
|
||||
% Header/footer
|
||||
\pagestyle{fancy}
|
||||
\fancyhf{}
|
||||
\fancyhead[L]{\footnotesize \textbf{Genomic Profile Report: [PATIENT ID]}}
|
||||
\fancyhead[R]{\footnotesize Page \thepage}
|
||||
\renewcommand{\headrulewidth}{0.5pt}
|
||||
\fancyfoot[C]{\footnotesize Confidential Laboratory Report - CLIA/CAP Certified}
|
||||
|
||||
\begin{document}
|
||||
|
||||
% Title block
|
||||
\begin{center}
|
||||
{\fontsize{14}{16}\selectfont\bfseries\color{headerblue} COMPREHENSIVE GENOMIC PROFILING REPORT}\\[2pt]
|
||||
{\fontsize{10}{12}\selectfont [Laboratory Name] | CLIA \#: [Number] | CAP \#: [Number]}
|
||||
\end{center}
|
||||
|
||||
\vspace{2pt}
|
||||
|
||||
% Patient/Specimen Information
|
||||
\begin{tcolorbox}[colback=highlightgray,colframe=black]
|
||||
\begin{minipage}{0.48\textwidth}
|
||||
{\small
|
||||
\textbf{Patient Information}\\
|
||||
Patient ID: [De-identified ID]\\
|
||||
Date of Birth: [De-identified/Age only]\\
|
||||
Sex: [M/F]\\
|
||||
Ordering Physician: [Name, MD]
|
||||
}
|
||||
\end{minipage}
|
||||
\hfill
|
||||
\begin{minipage}{0.48\textwidth}
|
||||
{\small
|
||||
\textbf{Specimen Information}\\
|
||||
Specimen Type: [Tissue/Blood/Other]\\
|
||||
Collection Date: [Date]\\
|
||||
Received Date: [Date]\\
|
||||
Report Date: [Date]
|
||||
}
|
||||
\end{minipage}
|
||||
\end{tcolorbox}
|
||||
|
||||
\vspace{2pt}
|
||||
|
||||
% Diagnosis
|
||||
\textbf{Diagnosis}: [Cancer type, stage, histology]
|
||||
|
||||
\textbf{Testing Performed}: [Assay name - e.g., FoundationOne CDx, NGS Panel]
|
||||
|
||||
\vspace{2pt}
|
||||
|
||||
% Results Summary Box
|
||||
\begin{tcolorbox}[enhanced,colback=tier1green!10,colframe=tier1green,
|
||||
title=\textbf{RESULTS SUMMARY},fonttitle=\bfseries,coltitle=black]
|
||||
{\small
|
||||
\textbf{Actionable Findings}: [X] alteration(s) detected
|
||||
\begin{itemize}
|
||||
\item \textbf{Tier 1}: [Number] FDA-approved therapy target(s)
|
||||
\item \textbf{Tier 2}: [Number] clinical trial or off-label option(s)
|
||||
\item \textbf{Tier 3}: [Number] variant(s) of uncertain significance
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Additional Biomarkers}:
|
||||
\begin{itemize}
|
||||
\item Tumor Mutational Burden (TMB): [X.X] mutations/Mb - [High/Intermediate/Low]
|
||||
\item Microsatellite Status: [MSI-H / MSS / Not assessed]
|
||||
\item PD-L1 Expression: [X\% TPS / Not assessed]
|
||||
\end{itemize}
|
||||
}
|
||||
\end{tcolorbox}
|
||||
|
||||
\section{Tier 1: FDA-Approved Targeted Therapies}
|
||||
|
||||
\begin{tcolorbox}[enhanced,colback=tier1green!5,colframe=tier1green,
|
||||
title={\colorbox{mutationred!60}{\textcolor{white}{\textbf{MUTATION}}} \textbf{[Gene Name] [Alteration]} \hfill \textbf{TIER 1 - ACTIONABLE}},
|
||||
fonttitle=\bfseries\small,coltitle=black]
|
||||
{\small
|
||||
\textbf{Alteration}: [Gene] [Specific variant - e.g., EGFR p.L858R (c.2573T>G)]\\
|
||||
\textbf{Variant Allele Frequency (VAF)}: XX\% (suggests [clonal/subclonal] mutation)\\
|
||||
\textbf{Classification}: [Pathogenic / Likely Pathogenic] (ClinVar, OncoKB)
|
||||
|
||||
\textbf{Clinical Significance}: \textcolor{tier1green}{\textbf{ACTIONABLE - FDA-APPROVED THERAPY AVAILABLE}}
|
||||
|
||||
\textbf{FDA-Approved Therapy}:
|
||||
\begin{itemize}
|
||||
\item \textbf{Drug}: [Drug name (brand name)] XX mg [PO/IV] [schedule]
|
||||
\item \textbf{Indication}: [Specific disease, line of therapy]
|
||||
\item \textbf{Evidence}: [Pivotal trial] - [Key results with HR, ORR, median survival]
|
||||
\item \textbf{Guideline}: NCCN Category [1/2A], [ESMO/ASCO recommendation]
|
||||
\item \textbf{Expected Outcomes}: ORR XX\%, median PFS XX months
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Alternative Therapies}:
|
||||
\begin{itemize}
|
||||
\item [Alternative drug] - [Indication, evidence level]
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Recommendation}: \textbf{STRONG} - Consider [drug name] as [first-line/second-line] therapy (GRADE 1A)
|
||||
}
|
||||
\end{tcolorbox}
|
||||
|
||||
\vspace{3pt}
|
||||
|
||||
\begin{tcolorbox}[enhanced,colback=tier1green!5,colframe=tier1green,
|
||||
title={\colorbox{amplificationblue!60}{\textcolor{white}{\textbf{AMPLIFICATION}}} \textbf{[Gene] Amplification} \hfill \textbf{TIER 1}},
|
||||
fonttitle=\bfseries\small,coltitle=black]
|
||||
{\small
|
||||
\textbf{Alteration}: [Gene name] amplification\\
|
||||
\textbf{Copy Number}: [X.X] copies per cell (threshold for positivity: ≥[Y])\\
|
||||
\textbf{Method}: [NGS copy number analysis / FISH]
|
||||
|
||||
\textbf{Clinical Significance}: \textcolor{tier1green}{\textbf{ACTIONABLE - COMPANION DIAGNOSTIC}}
|
||||
|
||||
\textbf{Therapy Options}: [Similar structure as mutation section]
|
||||
}
|
||||
\end{tcolorbox}
|
||||
|
||||
\section{Tier 2: Clinical Trial or Guideline-Recommended Off-Label}
|
||||
|
||||
\begin{tcolorbox}[enhanced,colback=tier2orange!5,colframe=tier2orange,
|
||||
title={\colorbox{fusionpurple!60}{\textcolor{white}{\textbf{FUSION}}} \textbf{[Gene] Rearrangement} \hfill \textbf{TIER 2 - INVESTIGATIONAL}},
|
||||
fonttitle=\bfseries\small,coltitle=black]
|
||||
{\small
|
||||
\textbf{Alteration}: [Gene A]-[Gene B] fusion detected\\
|
||||
\textbf{Method}: [RNA-seq / DNA NGS / FISH]
|
||||
|
||||
\textbf{Clinical Significance}: \textcolor{tier2orange}{\textbf{INVESTIGATIONAL - CLINICAL TRIAL PREFERRED}}
|
||||
|
||||
\textbf{Treatment Options}:
|
||||
\begin{itemize}
|
||||
\item \textbf{Clinical Trial}: [Specific trial or trial search guidance]
|
||||
\item \textbf{Off-Label Option}: [Drug] - NCCN Category 2A recommendation
|
||||
\item \textbf{Evidence}: [Phase 2 data, basket trial results, case series]
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Recommendation}: \textbf{CONDITIONAL} - Consider clinical trial enrollment or off-label use after standard therapy (GRADE 2B)
|
||||
}
|
||||
\end{tcolorbox}
|
||||
|
||||
\section{Tier 3: Variants of Uncertain Significance (VUS)}
|
||||
|
||||
\begin{tcolorbox}[colback=tier3gray!10,colframe=tier3gray]
|
||||
{\small
|
||||
\textbf{[Gene] [Variant]}: [Description]\\
|
||||
\textbf{Classification}: Variant of Uncertain Significance (VUS)\\
|
||||
\textbf{Clinical Actionability}: None currently - insufficient evidence\\
|
||||
\textbf{Recommendation}: No treatment change based on this finding; may be reclassified as evidence emerges
|
||||
}
|
||||
\end{tcolorbox}
|
||||
|
||||
\section{Biomarkers Assessed - Negative}
|
||||
|
||||
\textbf{No Alterations Detected in}:
|
||||
\begin{multicols}{3}
|
||||
\begin{itemize}
|
||||
\item [Gene 1]
|
||||
\item [Gene 2]
|
||||
\item [Gene 3]
|
||||
\item [Gene 4]
|
||||
\item [Gene 5]
|
||||
\item [Gene 6]
|
||||
\end{itemize}
|
||||
\end{multicols}
|
||||
|
||||
\section{Additional Biomarkers}
|
||||
|
||||
\subsection{Tumor Mutational Burden (TMB)}
|
||||
|
||||
\textbf{TMB}: [X.X] mutations per megabase
|
||||
|
||||
\textbf{Classification}:
|
||||
\begin{itemize}
|
||||
\item $\geq$10 mut/Mb: TMB-high (potential immunotherapy benefit)
|
||||
\item 6-9 mut/Mb: TMB-intermediate
|
||||
\item <6 mut/Mb: TMB-low
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Result}: [TMB-high / TMB-intermediate / TMB-low]
|
||||
|
||||
\textbf{Clinical Implication}:
|
||||
\begin{itemize}
|
||||
\item TMB-high: Consider immunotherapy; pembrolizumab FDA-approved for TMB-H ($\geq$10) solid tumors
|
||||
\item TMB-intermediate/low: Standard chemotherapy or biomarker-directed therapy
|
||||
\end{itemize}
|
||||
|
||||
\subsection{Microsatellite Instability (MSI)}
|
||||
|
||||
\textbf{MSI Status}: [MSI-H / MSI-L / MSS]
|
||||
|
||||
\textbf{Method}: [NGS-based MSI calling / PCR-based assay]
|
||||
|
||||
\textbf{Clinical Implication}:
|
||||
\begin{itemize}
|
||||
\item MSI-H: Immunotherapy highly effective (ORR 30-60\%); pembrolizumab, nivolumab approved
|
||||
\item MSS: Standard therapy; MSI-H-specific therapies not indicated
|
||||
\item If MSI-H + [relevant cancer] + young age: Consider germline Lynch syndrome testing
|
||||
\end{itemize}
|
||||
|
||||
\section{Integrated Treatment Recommendations}
|
||||
|
||||
\begin{tcolorbox}[enhanced,colback=stronggreen!10,colframe=tier1green,
|
||||
title=\textbf{PERSONALIZED TREATMENT PLAN},fonttitle=\bfseries,coltitle=black]
|
||||
{\small
|
||||
Based on the genomic profile, the following treatment approach is recommended:
|
||||
|
||||
\textbf{Primary Recommendation (GRADE 1A)}:
|
||||
\begin{itemize}
|
||||
\item \textbf{[Drug targeting identified alteration]}
|
||||
\item Dosing: [Specific dose and schedule]
|
||||
\item Evidence: [Supporting data]
|
||||
\item Expected outcomes: ORR XX\%, median PFS XX months
|
||||
\end{itemize}
|
||||
|
||||
\textbf{If Primary Recommendation Contraindicated}:
|
||||
\begin{itemize}
|
||||
\item Alternative 1: [Second-line biomarker-directed option]
|
||||
\item Alternative 2: [Standard therapy if targeted therapy ineligible]
|
||||
\end{itemize}
|
||||
|
||||
\textbf{At Progression}:
|
||||
\begin{itemize}
|
||||
\item Repeat molecular profiling (liquid biopsy or tissue) for resistance mechanisms
|
||||
\item Expected resistance alterations: [e.g., EGFR T790M, MET amplification]
|
||||
\item Sequential targeted therapy if secondary actionable alteration identified
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Clinical Trial Matching}:
|
||||
\begin{itemize}
|
||||
\item [List relevant trials based on identified alterations]
|
||||
\item ClinicalTrials.gov search terms: [Suggested keywords]
|
||||
\end{itemize}
|
||||
}
|
||||
\end{tcolorbox}
|
||||
|
||||
\section{Clinical Trial Matching}
|
||||
|
||||
\begin{table}[H]
|
||||
\centering
|
||||
\small
|
||||
\begin{tabular}{llll}
|
||||
\toprule
|
||||
\textbf{Trial} & \textbf{Intervention} & \textbf{Biomarker} & \textbf{Phase} \\
|
||||
\midrule
|
||||
[NCT Number] & [Drug/regimen] & [Matching biomarker] & Phase [1/2/3] \\
|
||||
[NCT Number] & [Drug/regimen] & [Matching biomarker] & Phase [1/2/3] \\
|
||||
\bottomrule
|
||||
\end{tabular}
|
||||
\caption{Potential clinical trials based on molecular profile (as of [date])}
|
||||
\end{table}
|
||||
|
||||
\textit{Note: Trial availability changes frequently. Search ClinicalTrials.gov for current options.}
|
||||
|
||||
\section{Methodology}
|
||||
|
||||
\subsection{Assay Information}
|
||||
|
||||
\textbf{Test Name}: [FoundationOne CDx / Custom NGS Panel / Other]\\
|
||||
\textbf{Methodology}: Next-generation sequencing (NGS)\\
|
||||
\textbf{Genes Analyzed}: [Number] genes for SNVs, indels, CNVs, and rearrangements\\
|
||||
\textbf{Coverage Depth}: [XXX]x median coverage\\
|
||||
\textbf{Limit of Detection}: [X\%] variant allele frequency
|
||||
|
||||
\textbf{Specimen Details}:
|
||||
\begin{itemize}
|
||||
\item Specimen type: [FFPE tissue block / Blood (ctDNA)]
|
||||
\item Tumor content: [XX\%] (minimum 20\% required for optimal sensitivity)
|
||||
\item DNA quality: [Adequate / Suboptimal]
|
||||
\item DNA quantity: [XX ng] (minimum [Y ng] required)
|
||||
\end{itemize}
|
||||
|
||||
\subsection{Interpretation}
|
||||
|
||||
\textbf{Variant Classification}:
|
||||
\begin{itemize}
|
||||
\item Pathogenic: Disease-causing, clinically significant
|
||||
\item Likely Pathogenic: Probably disease-causing based on available evidence
|
||||
\item VUS: Uncertain significance, insufficient evidence for classification
|
||||
\item Likely Benign: Probably not disease-causing
|
||||
\item Benign: Not disease-causing
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Databases Referenced}:
|
||||
\begin{itemize}
|
||||
\item OncoKB (Memorial Sloan Kettering)
|
||||
\item CIViC (Clinical Interpretations of Variants in Cancer)
|
||||
\item ClinVar (NCBI)
|
||||
\item COSMIC (Catalogue of Somatic Mutations in Cancer)
|
||||
\item [Others - PMKB, CGI, etc.]
|
||||
\end{itemize}
|
||||
|
||||
\section{Limitations}
|
||||
|
||||
\begin{itemize}
|
||||
\item This test analyzes [somatic/germline] alterations in tumor tissue. [If somatic: Results not informative for inherited cancer risk]
|
||||
\item Negative result does not exclude presence of alterations in genes not covered by this panel
|
||||
\item Low VAF alterations (<5\%) may not be detected due to assay sensitivity limits
|
||||
\item Copy number analysis limited for small amplifications or deletions
|
||||
\item Structural variants detection depends on breakpoint location within sequenced regions
|
||||
\item TMB and MSI calculations are estimate-based; consider orthogonal testing if borderline
|
||||
\end{itemize}
|
||||
|
||||
\section{Recommendations for Referring Clinician}
|
||||
|
||||
\begin{enumerate}
|
||||
\item \textbf{[Action 1]}: [e.g., Initiate targeted therapy with drug X based on detected alteration]
|
||||
\item \textbf{[Action 2]}: [e.g., Consider clinical trial enrollment for Tier 2 alteration]
|
||||
\item \textbf{[Action 3]}: [e.g., Repeat molecular profiling at progression to identify resistance mechanisms]
|
||||
\item \textbf{[Action 4]}: [e.g., If MSI-H detected and patient <50 years, refer for genetic counseling for Lynch syndrome]
|
||||
\item \textbf{[Action 5]}: [e.g., Share report with molecular tumor board for complex decision-making]
|
||||
\end{enumerate}
|
||||
|
||||
\section{References}
|
||||
|
||||
\begin{enumerate}
|
||||
\item [FDA Label for companion diagnostic]
|
||||
\item [Key clinical trial supporting biomarker-therapy association]
|
||||
\item [NCCN Guideline reference]
|
||||
\item [OncoKB database version]
|
||||
\item [Assay validation publication]
|
||||
\end{enumerate}
|
||||
|
||||
\vspace{10pt}
|
||||
|
||||
\hrule
|
||||
\vspace{4pt}
|
||||
{\footnotesize
|
||||
\textbf{Laboratory Director}: [Name, MD, PhD] | [Board certifications]\\
|
||||
\textbf{Report Authorized By}: [Name, credentials] | Date: [Date]\\
|
||||
\textbf{Laboratory}: [Name, address]\\
|
||||
\textbf{CLIA \#}: [Number] | \textbf{CAP \#}: [Number]\\
|
||||
\textbf{Questions}: Contact [Name] at [Phone] or [Email]
|
||||
|
||||
\vspace{2pt}
|
||||
|
||||
\textit{This report is intended for use by qualified healthcare professionals. The information provided is based on current scientific literature and databases. Interpretation and treatment decisions should be made by qualified physicians in consultation with the patient. This test was performed in a CLIA-certified, CAP-accredited laboratory.}
|
||||
}
|
||||
|
||||
\end{document}
|
||||
|
||||
@@ -0,0 +1,222 @@
|
||||
\documentclass[10pt,letterpaper,landscape]{article}
|
||||
|
||||
% Landscape for wider flowcharts
|
||||
\usepackage[margin=0.4in]{geometry}
|
||||
\usepackage[utf8]{inputenc}
|
||||
\usepackage[T1]{fontenc}
|
||||
\usepackage{helvet}
|
||||
\renewcommand{\familydefault}{\sfdefault}
|
||||
\usepackage{xcolor}
|
||||
\usepackage{tcolorbox}
|
||||
\usepackage{tikz}
|
||||
\usetikzlibrary{shapes,arrows,positioning,fit,calc}
|
||||
\usepackage{fancyhdr}
|
||||
|
||||
% Color definitions
|
||||
\definecolor{headerblue}{RGB}{0,102,204}
|
||||
\definecolor{actiongreen}{RGB}{0,153,76}
|
||||
\definecolor{decisionyellow}{RGB}{255,193,7}
|
||||
\definecolor{urgentred}{RGB}{220,20,60}
|
||||
\definecolor{infobox}{RGB}{33,150,243}
|
||||
\definecolor{routineblue}{RGB}{100,181,246}
|
||||
|
||||
% Header/footer
|
||||
\pagestyle{fancy}
|
||||
\fancyhf{}
|
||||
\fancyhead[L]{\footnotesize \textbf{Clinical Pathway: [CONDITION/DISEASE]}}
|
||||
\fancyhead[R]{\footnotesize Version X.X | [Date]}
|
||||
\renewcommand{\headrulewidth}{0.5pt}
|
||||
\fancyfoot[C]{\footnotesize Evidence-Based Clinical Decision Pathway | For Professional Use Only | Page \thepage}
|
||||
|
||||
% TikZ styles
|
||||
\tikzstyle{startstop} = [rectangle, rounded corners=8pt, minimum width=3cm, minimum height=1cm, text centered, draw=black, fill=headerblue!20, font=\small\bfseries]
|
||||
\tikzstyle{decision} = [diamond, minimum width=3cm, minimum height=1.2cm, text centered, draw=black, fill=decisionyellow!40, font=\small, aspect=2, inner sep=0pt]
|
||||
\tikzstyle{process} = [rectangle, rounded corners=4pt, minimum width=3.5cm, minimum height=0.9cm, text centered, draw=black, fill=actiongreen!20, font=\small]
|
||||
\tikzstyle{urgent} = [rectangle, rounded corners=4pt, minimum width=3.5cm, minimum height=0.9cm, text centered, draw=urgentred, line width=1.5pt, fill=urgentred!15, font=\small\bfseries]
|
||||
\tikzstyle{routine} = [rectangle, rounded corners=4pt, minimum width=3.5cm, minimum height=0.9cm, text centered, draw=black, fill=routineblue!20, font=\small]
|
||||
\tikzstyle{info} = [rectangle, rounded corners=2pt, minimum width=2.5cm, minimum height=0.7cm, text centered, draw=infobox, fill=infobox!10, font=\footnotesize]
|
||||
\tikzstyle{arrow} = [thick,->,>=stealth]
|
||||
\tikzstyle{urgentarrow} = [ultra thick,->,>=stealth,color=urgentred]
|
||||
|
||||
\setlength{\parindent}{0pt}
|
||||
|
||||
\begin{document}
|
||||
|
||||
\begin{center}
|
||||
{\fontsize{16}{18}\selectfont\bfseries\color{headerblue} CLINICAL DECISION PATHWAY}\\[2pt]
|
||||
{\fontsize{13}{15}\selectfont\bfseries [Disease/Condition - e.g., Acute Chest Pain Management]}\\[2pt]
|
||||
{\fontsize{10}{12}\selectfont [Institution Name] | Version X.X | Effective Date: [Date]}
|
||||
\end{center}
|
||||
|
||||
\vspace{6pt}
|
||||
|
||||
% Legend box
|
||||
\begin{tcolorbox}[colback=white,colframe=black,width=\textwidth]
|
||||
\begin{minipage}{0.48\textwidth}
|
||||
\textbf{Pathway Symbols:}\\[2pt]
|
||||
\begin{tikzpicture}[node distance=0.5cm]
|
||||
\node[startstop, scale=0.7] (start) {Start/End};
|
||||
\node[decision, right=1cm of start, scale=0.7] (dec) {Decision\\Point};
|
||||
\node[process, right=1cm of dec, scale=0.7] (proc) {Action/Process};
|
||||
\end{tikzpicture}
|
||||
\end{minipage}
|
||||
\begin{minipage}{0.48\textwidth}
|
||||
\textbf{Urgency Color Coding:}\\[2pt]
|
||||
\begin{tikzpicture}[node distance=0.5cm]
|
||||
\node[urgent, scale=0.7] (urg) {URGENT\\<1 hour};
|
||||
\node[process, right=1cm of urg, scale=0.7] (sem) {Semi-Urgent\\<24 hours};
|
||||
\node[routine, right=1cm of sem, scale=0.7] (rout) {Routine\\>24 hours};
|
||||
\end{tikzpicture}
|
||||
\end{minipage}
|
||||
\end{tcolorbox}
|
||||
|
||||
\vspace{4pt}
|
||||
|
||||
% Main flowchart
|
||||
\begin{center}
|
||||
\begin{tikzpicture}[node distance=2.2cm and 3cm, auto]
|
||||
|
||||
% Start
|
||||
\node [startstop] (start) {Patient Presentation:\\[2pt] [Chief Complaint]};
|
||||
|
||||
% First decision
|
||||
\node [decision, below=of start] (decision1) {[Critical\\Criteria\\Present?]};
|
||||
|
||||
% Urgent pathway (left branch)
|
||||
\node [urgent, left=of decision1, below=1.8cm] (urgent1) {IMMEDIATE ACTION:\\[2pt] [Specific intervention]\\[2pt] Call Code/Transfer};
|
||||
|
||||
% Continue evaluation (right branch)
|
||||
\node [process, right=of decision1, below=1.8cm] (eval1) {Continue\\Evaluation:\\[2pt][Tests/Assessment]};
|
||||
|
||||
% Second decision
|
||||
\node [decision, below=of eval1] (decision2) {[Risk\\Score\\$\geq$X?]};
|
||||
|
||||
% High risk pathway
|
||||
\node [urgent, left=of decision2, below=1.8cm] (high) {HIGH RISK:\\[2pt] Admit ICU/Telemetry\\[2pt] [Specific management]};
|
||||
|
||||
% Moderate risk
|
||||
\node [process, below=of decision2] (moderate) {MODERATE RISK:\\[2pt] Admit for observation\\[2pt] Serial testing};
|
||||
|
||||
% Low risk pathway
|
||||
\node [routine, right=of decision2, below=1.8cm] (low) {LOW RISK:\\[2pt] Outpatient management\\[2pt] Follow-up in X days};
|
||||
|
||||
% Final outcome node
|
||||
\node [startstop, below=of moderate, node distance=2.5cm] (outcome) {Definitive Management\\Based on Results};
|
||||
|
||||
% Arrows
|
||||
\draw [urgentarrow] (start) -- (decision1);
|
||||
\draw [urgentarrow] (decision1) -| node[near start,left] {YES} (urgent1);
|
||||
\draw [arrow] (decision1) -| node[near start,right] {NO} (eval1);
|
||||
\draw [arrow] (eval1) -- (decision2);
|
||||
\draw [arrow] (decision2) -| node[near start,left] {HIGH} (high);
|
||||
\draw [arrow] (decision2) -- node[right] {MODERATE} (moderate);
|
||||
\draw [arrow] (decision2) -| node[near start,right] {LOW} (low);
|
||||
\draw [arrow] (urgent1) |- (outcome);
|
||||
\draw [arrow] (high) |- (outcome);
|
||||
\draw [arrow] (moderate) -- (outcome);
|
||||
\draw [arrow] (low) |- (outcome);
|
||||
|
||||
% Information boxes
|
||||
\node [info, right=1.5cm of eval1] (info1) {[Criteria]:\\[1pt] \footnotesize • Item 1\\• Item 2\\• Item 3};
|
||||
\node [info, right=1.5cm of decision2] (info2) {[Score]:\\[1pt] \footnotesize Calculate:\\risk score};
|
||||
|
||||
\end{tikzpicture}
|
||||
\end{center}
|
||||
|
||||
\vspace{8pt}
|
||||
|
||||
% Detailed pathway steps
|
||||
\begin{tcolorbox}[colback=highlightgray!30,colframe=headerblue,title=\textbf{Detailed Pathway Steps},fonttitle=\bfseries]
|
||||
|
||||
\textbf{STEP 1: Initial Assessment}
|
||||
\begin{itemize}
|
||||
\item Vital signs: BP, HR, RR, temp, O₂ saturation
|
||||
\item Focused history: [Key elements]
|
||||
\item Physical examination: [Key findings]
|
||||
\item Initial labs: [Specify tests]
|
||||
\item ECG (if applicable)
|
||||
\end{itemize}
|
||||
|
||||
\textbf{STEP 2: Risk Stratification}
|
||||
\begin{itemize}
|
||||
\item Calculate [Risk Score Name] (see scoring table below)
|
||||
\item Identify high-risk features requiring immediate intervention
|
||||
\item Document risk category in medical record
|
||||
\end{itemize}
|
||||
|
||||
\textbf{STEP 3: Treatment Initiation}
|
||||
\begin{itemize}
|
||||
\item Urgent: [Specific interventions within 1 hour]
|
||||
\item Semi-urgent: [Interventions within 24 hours]
|
||||
\item Routine: [Standard management approach]
|
||||
\end{itemize}
|
||||
|
||||
\textbf{STEP 4: Monitoring and Reassessment}
|
||||
\begin{itemize}
|
||||
\item Frequency: [Based on risk category]
|
||||
\item Parameters: [What to monitor]
|
||||
\item Escalation criteria: [When to intensify treatment]
|
||||
\item De-escalation criteria: [When to transition to lower intensity]
|
||||
\end{itemize}
|
||||
|
||||
\end{tcolorbox}
|
||||
|
||||
\vspace{4pt}
|
||||
|
||||
% Risk scoring table
|
||||
\begin{tcolorbox}[colback=white,colframe=headerblue,title=\textbf{[Risk Score Name] Calculation},fonttitle=\bfseries]
|
||||
{\small
|
||||
\begin{tabular}{lc}
|
||||
\toprule
|
||||
\textbf{Clinical Feature} & \textbf{Points} \\
|
||||
\midrule
|
||||
[Feature 1 - e.g., Age $\geq$65 years] & +1 \\
|
||||
[Feature 2 - e.g., Prior history] & +1 \\
|
||||
[Feature 3 - e.g., Abnormal lab value] & +2 \\
|
||||
[Feature 4 - e.g., Specific symptom] & +1 \\
|
||||
[Feature 5 - e.g., Imaging finding] & +2 \\
|
||||
\midrule
|
||||
\textbf{Total Score} & \textbf{0-X points} \\
|
||||
\bottomrule
|
||||
\end{tabular}
|
||||
|
||||
\vspace{4pt}
|
||||
|
||||
\textbf{Risk Categories}:
|
||||
\begin{itemize}
|
||||
\item \textbf{Low Risk}: 0-1 points → [Management approach, predicted outcome]
|
||||
\item \textbf{Moderate Risk}: 2-3 points → [Management approach, predicted outcome]
|
||||
\item \textbf{High Risk}: $\geq$4 points → [Management approach, predicted outcome]
|
||||
\end{itemize}
|
||||
}
|
||||
\end{tcolorbox}
|
||||
|
||||
\vspace{4pt}
|
||||
|
||||
% Evidence basis
|
||||
\begin{tcolorbox}[colback=actiongreen!5,colframe=actiongreen,title=\textbf{Evidence Basis for Pathway},fonttitle=\bfseries]
|
||||
{\small
|
||||
\textbf{Key Supporting Evidence}:
|
||||
\begin{enumerate}
|
||||
\item \textbf{[Clinical Trial/Study]}: [Key finding supporting pathway decision]
|
||||
\item \textbf{Guidelines}: NCCN/ASCO/AHA/ACC/[Relevant society] [Year] - [Recommendation level]
|
||||
\item \textbf{Meta-Analysis}: [If applicable - pooled results supporting approach]
|
||||
\end{enumerate}
|
||||
|
||||
\textbf{Validation}: Pathway validated at [institution] with [X\%] adherence rate and [outcome metrics].
|
||||
|
||||
\textbf{Last Updated}: [Date] based on [new trial, guideline update, or scheduled review]
|
||||
}
|
||||
\end{tcolorbox}
|
||||
|
||||
\vspace{8pt}
|
||||
|
||||
\hrule
|
||||
\vspace{4pt}
|
||||
{\footnotesize
|
||||
\textbf{Pathway Committee}: [Names, titles] | \textbf{Approved}: [Date] | \textbf{Next Review}: [Date]\\
|
||||
\textbf{Contact for Questions}: [Name, email, phone]
|
||||
}
|
||||
|
||||
\end{document}
|
||||
|
||||
@@ -0,0 +1,359 @@
|
||||
\documentclass[10pt,letterpaper]{article}
|
||||
|
||||
% Packages
|
||||
\usepackage[margin=0.5in]{geometry}
|
||||
\usepackage[utf8]{inputenc}
|
||||
\usepackage[T1]{fontenc}
|
||||
\usepackage{helvet}
|
||||
\renewcommand{\familydefault}{\sfdefault}
|
||||
\usepackage{xcolor}
|
||||
\usepackage{tcolorbox}
|
||||
\usepackage{array}
|
||||
\usepackage{tabularx}
|
||||
\usepackage{booktabs}
|
||||
\usepackage{enumitem}
|
||||
\usepackage{titlesec}
|
||||
\usepackage{fancyhdr}
|
||||
\usepackage{multicol}
|
||||
\usepackage{graphicx}
|
||||
\usepackage{float}
|
||||
|
||||
% Color definitions
|
||||
\definecolor{headerblue}{RGB}{0,102,204}
|
||||
\definecolor{highlightgreen}{RGB}{0,153,76}
|
||||
\definecolor{warningred}{RGB}{204,0,0}
|
||||
\definecolor{highlightgray}{RGB}{240,240,240}
|
||||
\definecolor{biomarkerblue}{RGB}{51,102,204}
|
||||
|
||||
% Section formatting - compact
|
||||
\titleformat{\section}{\normalfont\fontsize{11}{12}\bfseries\color{headerblue}}{\thesection}{0.5em}{}
|
||||
\titlespacing*{\section}{0pt}{4pt}{2pt}
|
||||
|
||||
\titleformat{\subsection}{\normalfont\fontsize{10}{11}\bfseries}{\thesubsection}{0.5em}{}
|
||||
\titlespacing*{\subsection}{0pt}{3pt}{1pt}
|
||||
|
||||
% List formatting - ultra compact
|
||||
\setlist[itemize]{leftmargin=*,itemsep=0pt,parsep=0pt,topsep=1pt}
|
||||
\setlist[enumerate]{leftmargin=*,itemsep=0pt,parsep=0pt,topsep=1pt}
|
||||
|
||||
% Remove paragraph indentation
|
||||
\setlength{\parindent}{0pt}
|
||||
\setlength{\parskip}{2pt}
|
||||
|
||||
% Header/footer
|
||||
\pagestyle{fancy}
|
||||
\fancyhf{}
|
||||
\fancyhead[L]{\footnotesize \textbf{Clinical Decision Support: [COHORT NAME]}}
|
||||
\fancyhead[R]{\footnotesize Page \thepage}
|
||||
\renewcommand{\headrulewidth}{0.5pt}
|
||||
\fancyfoot[C]{\footnotesize Confidential Medical Document - For Professional Use Only}
|
||||
|
||||
\begin{document}
|
||||
|
||||
% Title block - compact
|
||||
\begin{center}
|
||||
{\fontsize{14}{16}\selectfont\bfseries\color{headerblue} PATIENT COHORT ANALYSIS REPORT}\\[2pt]
|
||||
{\fontsize{12}{14}\selectfont\bfseries [Cohort Description - e.g., NSCLC Patients Stratified by PD-L1 Expression]}\\[2pt]
|
||||
{\fontsize{10}{12}\selectfont [Institution/Study Name]}\\[1pt]
|
||||
{\fontsize{9}{11}\selectfont Report Date: [Date]}
|
||||
\end{center}
|
||||
|
||||
\vspace{4pt}
|
||||
|
||||
% Executive Summary Box
|
||||
\begin{tcolorbox}[colback=highlightgray,colframe=headerblue,title=\textbf{Executive Summary},fonttitle=\bfseries\small,coltitle=black]
|
||||
{\small
|
||||
\textbf{Cohort}: [n=XX] patients with [disease] stratified by [biomarker/characteristic]
|
||||
|
||||
\textbf{Key Findings}:
|
||||
\begin{itemize}
|
||||
\item [Primary finding - e.g., Biomarker+ patients had significantly longer PFS]
|
||||
\item [Secondary finding - e.g., ORR 45\% vs 30\%, p=0.023]
|
||||
\item [Safety finding - e.g., Similar toxicity profiles between groups]
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Clinical Implications}: [Treatment recommendations based on findings]
|
||||
}
|
||||
\end{tcolorbox}
|
||||
|
||||
\vspace{2pt}
|
||||
|
||||
\section{Cohort Characteristics}
|
||||
|
||||
\subsection{Patient Demographics}
|
||||
|
||||
[Narrative description of cohort composition, inclusion/exclusion criteria, time period]
|
||||
|
||||
\begin{table}[H]
|
||||
\centering
|
||||
\small
|
||||
\begin{tabular}{lccc}
|
||||
\toprule
|
||||
\textbf{Characteristic} & \textbf{Group A (n=XX)} & \textbf{Group B (n=XX)} & \textbf{p-value} \\
|
||||
\midrule
|
||||
Age, years (median [IQR]) & XX [XX-XX] & XX [XX-XX] & X.XX \\
|
||||
Sex, n (\%) & & & \\
|
||||
\quad Male & XX (XX\%) & XX (XX\%) & X.XX \\
|
||||
\quad Female & XX (XX\%) & XX (XX\%) & \\
|
||||
ECOG PS, n (\%) & & & \\
|
||||
\quad 0-1 & XX (XX\%) & XX (XX\%) & X.XX \\
|
||||
\quad 2 & XX (XX\%) & XX (XX\%) & \\
|
||||
Disease Stage, n (\%) & & & \\
|
||||
\quad III & XX (XX\%) & XX (XX\%) & X.XX \\
|
||||
\quad IV & XX (XX\%) & XX (XX\%) & \\
|
||||
Prior Lines of Therapy & & & \\
|
||||
\quad 0 (treatment-naïve) & XX (XX\%) & XX (XX\%) & X.XX \\
|
||||
\quad 1-2 & XX (XX\%) & XX (XX\%) & \\
|
||||
\quad $\geq$3 & XX (XX\%) & XX (XX\%) & \\
|
||||
\bottomrule
|
||||
\end{tabular}
|
||||
\caption{Baseline patient demographics and clinical characteristics}
|
||||
\end{table}
|
||||
|
||||
\subsection{Biomarker Profile}
|
||||
|
||||
\begin{tcolorbox}[colback=biomarkerblue!10,colframe=biomarkerblue,title=\textbf{Biomarker Stratification},fonttitle=\bfseries\small]
|
||||
{\small
|
||||
\textbf{Classification Method}: [e.g., IHC for PD-L1 expression, NGS for mutations, gene expression clustering]
|
||||
|
||||
\textbf{Group Definitions}:
|
||||
\begin{itemize}
|
||||
\item \textbf{Group A (Biomarker+)}: [n=XX] - [Definition, e.g., PD-L1 TPS $\geq$50\%, or Mesenchymal-Immune-Active subtype]
|
||||
\item \textbf{Group B (Biomarker-)}: [n=XX] - [Definition, e.g., PD-L1 TPS <50\%]
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Molecular Features of Group A}:
|
||||
\begin{itemize}
|
||||
\item [Feature 1]: XX\% (n=XX) - [Clinical significance]
|
||||
\item [Feature 2]: XX\% (n=XX) - [Clinical significance]
|
||||
\item [Feature 3]: Elevated/decreased [marker] (median [value])
|
||||
\end{itemize}
|
||||
}
|
||||
\end{tcolorbox}
|
||||
|
||||
\section{Treatment Exposures}
|
||||
|
||||
\begin{table}[H]
|
||||
\centering
|
||||
\small
|
||||
\begin{tabular}{lcc}
|
||||
\toprule
|
||||
\textbf{Treatment Received} & \textbf{Group A, n (\%)} & \textbf{Group B, n (\%)} \\
|
||||
\midrule
|
||||
[Treatment regimen 1] & XX (XX\%) & XX (XX\%) \\
|
||||
[Treatment regimen 2] & XX (XX\%) & XX (XX\%) \\
|
||||
[Treatment regimen 3] & XX (XX\%) & XX (XX\%) \\
|
||||
Median cycles received (range) & X (X-X) & X (X-X) \\
|
||||
\bottomrule
|
||||
\end{tabular}
|
||||
\caption{Treatment exposures by biomarker group}
|
||||
\end{table}
|
||||
|
||||
\section{Treatment Outcomes}
|
||||
|
||||
\subsection{Response Rates}
|
||||
|
||||
\begin{table}[H]
|
||||
\centering
|
||||
\small
|
||||
\begin{tabular}{lccc}
|
||||
\toprule
|
||||
\textbf{Response Category} & \textbf{Group A (n=XX)} & \textbf{Group B (n=XX)} & \textbf{p-value} \\
|
||||
\midrule
|
||||
Objective Response Rate (ORR) & XX\% [95\% CI] & XX\% [95\% CI] & X.XXX \\
|
||||
\quad Complete Response (CR) & XX (XX\%) & XX (XX\%) & \\
|
||||
\quad Partial Response (PR) & XX (XX\%) & XX (XX\%) & \\
|
||||
Disease Control Rate (DCR) & XX\% [95\% CI] & XX\% [95\% CI] & X.XXX \\
|
||||
\quad Stable Disease (SD) & XX (XX\%) & XX (XX\%) & \\
|
||||
Progressive Disease (PD) & XX (XX\%) & XX (XX\%) & \\
|
||||
\midrule
|
||||
Median Duration of Response (months) & X.X (95\% CI X.X-X.X) & X.X (95\% CI X.X-X.X) & X.XXX \\
|
||||
\bottomrule
|
||||
\end{tabular}
|
||||
\caption{Best overall response by biomarker group (RECIST v1.1 criteria)}
|
||||
\end{table}
|
||||
|
||||
\subsection{Survival Outcomes}
|
||||
|
||||
\textbf{Progression-Free Survival (PFS)}:
|
||||
\begin{itemize}
|
||||
\item Group A: Median X.X months (95\% CI X.X-X.X), 12-month PFS rate: XX\%
|
||||
\item Group B: Median X.X months (95\% CI X.X-X.X), 12-month PFS rate: XX\%
|
||||
\item Hazard Ratio: X.XX (95\% CI X.XX-X.XX), log-rank p = X.XXX
|
||||
\item \textit{[Interpretation: Group A had XX\% reduction in risk of progression compared to Group B]}
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Overall Survival (OS)}:
|
||||
\begin{itemize}
|
||||
\item Group A: Median XX.X months (95\% CI XX.X-XX.X), 12-month OS rate: XX\%
|
||||
\item Group B: Median XX.X months (95\% CI XX.X-XX.X), 12-month OS rate: XX\%
|
||||
\item Hazard Ratio: X.XX (95\% CI X.XX-X.XX), log-rank p = X.XXX
|
||||
\item \textit{[Interpretation: XX\% reduction in risk of death for Group A]}
|
||||
\end{itemize}
|
||||
|
||||
% Note: Include Kaplan-Meier curves as figures if available
|
||||
% \begin{figure}[H]
|
||||
% \centering
|
||||
% \includegraphics[width=0.9\textwidth]{figures/pfs_by_biomarker.pdf}
|
||||
% \caption{Progression-free survival by biomarker status}
|
||||
% \end{figure}
|
||||
|
||||
\section{Safety and Tolerability}
|
||||
|
||||
\begin{table}[H]
|
||||
\centering
|
||||
\small
|
||||
\begin{tabular}{lcccc}
|
||||
\toprule
|
||||
\multirow{2}{*}{\textbf{Adverse Event}} & \multicolumn{2}{c}{\textbf{Any Grade, n (\%)}} & \multicolumn{2}{c}{\textbf{Grade 3-4, n (\%)}} \\
|
||||
\cmidrule(lr){2-3} \cmidrule(lr){4-5}
|
||||
& Group A & Group B & Group A & Group B \\
|
||||
\midrule
|
||||
[AE 1 - e.g., Fatigue] & XX (XX\%) & XX (XX\%) & X (X\%) & X (X\%) \\
|
||||
[AE 2 - e.g., Nausea] & XX (XX\%) & XX (XX\%) & X (X\%) & X (X\%) \\
|
||||
[AE 3 - e.g., Neutropenia] & XX (XX\%) & XX (XX\%) & X (X\%) & X (X\%) \\
|
||||
[AE 4 - e.g., Diarrhea] & XX (XX\%) & XX (XX\%) & X (X\%) & X (X\%) \\
|
||||
[AE 5 - immune-related] & XX (XX\%) & XX (XX\%) & X (X\%) & X (X\%) \\
|
||||
\midrule
|
||||
Treatment discontinuation & XX (XX\%) & XX (XX\%) & \multicolumn{2}{c}{-} \\
|
||||
Dose reductions & XX (XX\%) & XX (XX\%) & \multicolumn{2}{c}{-} \\
|
||||
\bottomrule
|
||||
\end{tabular}
|
||||
\caption{Treatment-emergent adverse events by biomarker group (CTCAE v5.0)}
|
||||
\end{table}
|
||||
|
||||
\section{Statistical Analysis}
|
||||
|
||||
\subsection{Methods}
|
||||
|
||||
\textbf{Study Design}: [Retrospective cohort analysis / Prospective cohort / Post-hoc analysis of clinical trial]
|
||||
|
||||
\textbf{Statistical Tests}:
|
||||
\begin{itemize}
|
||||
\item Continuous variables: [t-test / Mann-Whitney U test], reported as [mean $\pm$ SD / median [IQR]]
|
||||
\item Categorical variables: Chi-square test or Fisher's exact test (if expected count <5)
|
||||
\item Survival analysis: Kaplan-Meier method, log-rank test, Cox proportional hazards regression
|
||||
\item Significance level: Two-sided p<0.05 considered statistically significant
|
||||
\item Software: [R version X.X.X, survival package / SAS / Stata / Python lifelines]
|
||||
\end{itemize}
|
||||
|
||||
\subsection{Multivariable Analysis}
|
||||
|
||||
Cox regression model adjusting for baseline prognostic factors:
|
||||
|
||||
\begin{table}[H]
|
||||
\centering
|
||||
\small
|
||||
\begin{tabular}{lccc}
|
||||
\toprule
|
||||
\textbf{Variable} & \textbf{Hazard Ratio} & \textbf{95\% CI} & \textbf{p-value} \\
|
||||
\midrule
|
||||
Biomarker+ (vs Biomarker-) & X.XX & X.XX-X.XX & X.XXX \\
|
||||
Age (per 10 years) & X.XX & X.XX-X.XX & X.XXX \\
|
||||
ECOG PS 2 (vs 0-1) & X.XX & X.XX-X.XX & X.XXX \\
|
||||
Stage IV (vs III) & X.XX & X.XX-X.XX & X.XXX \\
|
||||
[Additional variable] & X.XX & X.XX-X.XX & X.XXX \\
|
||||
\bottomrule
|
||||
\end{tabular}
|
||||
\caption{Multivariable Cox regression for progression-free survival}
|
||||
\end{table}
|
||||
|
||||
\textbf{Interpretation}: After adjusting for age, performance status, and disease stage, [biomarker status] remained an independent predictor of [PFS/OS] (HR X.XX, 95\% CI X.XX-X.XX, p=X.XXX).
|
||||
|
||||
\section{Clinical Implications}
|
||||
|
||||
\begin{tcolorbox}[colback=highlightgreen!10,colframe=highlightgreen,title=\textbf{Treatment Recommendations},fonttitle=\bfseries\small]
|
||||
{\small
|
||||
\textbf{For Biomarker-Positive Patients (Group A)}:
|
||||
|
||||
\textbf{Preferred Regimen} (GRADE 1A):
|
||||
\begin{itemize}
|
||||
\item [Specific treatment based on biomarker]
|
||||
\item Evidence: [Trial name/data showing benefit in biomarker+ population]
|
||||
\item Expected outcomes: ORR XX\%, median PFS XX months
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Monitoring}:
|
||||
\begin{itemize}
|
||||
\item Imaging every [X weeks] for response assessment
|
||||
\item [Specific lab monitoring for biomarker+ patients]
|
||||
\item Watch for [specific toxicities more common in this group]
|
||||
\end{itemize}
|
||||
|
||||
\textbf{For Biomarker-Negative Patients (Group B)}:
|
||||
|
||||
\textbf{Standard Regimen} (GRADE 1B):
|
||||
\begin{itemize}
|
||||
\item [Standard therapy for biomarker- population]
|
||||
\item Expected outcomes: ORR XX\%, median PFS XX months
|
||||
\item Consider [alternative approaches or clinical trial enrollment]
|
||||
\end{itemize}
|
||||
}
|
||||
\end{tcolorbox}
|
||||
|
||||
\section{Subgroup Analyses}
|
||||
|
||||
\textbf{Interaction Testing}: Treatment effect by biomarker subgroup (p-interaction = X.XXX)
|
||||
|
||||
[Describe whether treatment benefit differs by biomarker status - i.e., predictive biomarker]
|
||||
|
||||
Additional exploratory subgroups:
|
||||
\begin{itemize}
|
||||
\item Age <65 vs $\geq$65 years
|
||||
\item Sex (male vs female)
|
||||
\item Prior lines of therapy (0 vs 1+ prior treatments)
|
||||
\item Disease burden (high vs low tumor burden)
|
||||
\end{itemize}
|
||||
|
||||
\section{Strengths and Limitations}
|
||||
|
||||
\subsection{Strengths}
|
||||
\begin{itemize}
|
||||
\item [e.g., Biomarker-stratified analysis with prospectively defined groups]
|
||||
\item [e.g., Adequate sample size for statistical power]
|
||||
\item [e.g., Standardized response assessment using RECIST v1.1]
|
||||
\item [e.g., Multivariable analysis adjusting for confounders]
|
||||
\end{itemize}
|
||||
|
||||
\subsection{Limitations}
|
||||
\begin{itemize}
|
||||
\item [e.g., Retrospective design with potential selection bias]
|
||||
\item [e.g., Single-institution cohort may limit generalizability]
|
||||
\item [e.g., Biomarker testing not available for all patients (XX\% tested)]
|
||||
\item [e.g., Limited follow-up for OS (median X months)]
|
||||
\item [e.g., Heterogeneous treatment regimens across cohort]
|
||||
\end{itemize}
|
||||
|
||||
\section{Conclusions}
|
||||
|
||||
[Paragraph summarizing key findings]
|
||||
|
||||
[Biomarker-positive patients demonstrated [significantly better/worse] outcomes compared to biomarker-negative patients, with [outcome metric] of [values] (HR X.XX, p=X.XXX). These findings support [biomarker-guided therapy selection / routine biomarker testing / specific treatment approach].]
|
||||
|
||||
[Future directions: Prospective validation in independent cohort, investigation of mechanisms, clinical trial design implications]
|
||||
|
||||
\section{References}
|
||||
|
||||
\begin{enumerate}
|
||||
\item [Reference 1 - Key clinical trial]
|
||||
\item [Reference 2 - Biomarker validation study]
|
||||
\item [Reference 3 - Guideline reference (NCCN, ASCO, ESMO)]
|
||||
\item [Reference 4 - Statistical methods reference]
|
||||
\item [Reference 5 - Additional supporting evidence]
|
||||
\end{enumerate}
|
||||
|
||||
\vspace{10pt}
|
||||
|
||||
\hrule
|
||||
\vspace{4pt}
|
||||
{\footnotesize
|
||||
\textbf{Report Prepared By}: [Name, Title]\\
|
||||
\textbf{Date}: [Date]\\
|
||||
\textbf{Contact}: [Email/Phone]\\
|
||||
\textbf{Institutional Review}: [IRB approval number if applicable]\\
|
||||
\textbf{Data Cut-Off Date}: [Date]\\
|
||||
\textbf{Confidentiality}: This document contains proprietary clinical data. Distribution restricted to authorized personnel only.
|
||||
}
|
||||
|
||||
\end{document}
|
||||
|
||||
149
skills/clinical-decision-support/assets/color_schemes.tex
Normal file
149
skills/clinical-decision-support/assets/color_schemes.tex
Normal file
@@ -0,0 +1,149 @@
|
||||
% Clinical Decision Support Color Schemes
|
||||
% For use in LaTeX documents
|
||||
|
||||
% ============================================================================
|
||||
% PRIMARY THEME COLORS
|
||||
% ============================================================================
|
||||
|
||||
% Header and structural elements
|
||||
\definecolor{headerblue}{RGB}{0,102,204} % Section headers, titles
|
||||
\definecolor{highlightgray}{RGB}{240,240,240} % Background boxes
|
||||
|
||||
% ============================================================================
|
||||
% RECOMMENDATION STRENGTH COLORS
|
||||
% ============================================================================
|
||||
|
||||
% Strong recommendations (benefits clearly outweigh risks)
|
||||
\definecolor{stronggreen}{RGB}{0,153,76} % Grade 1A, 1B
|
||||
\definecolor{strongdark}{RGB}{0,120,60} % Darker variant for emphasis
|
||||
|
||||
% Conditional recommendations (trade-offs exist)
|
||||
\definecolor{conditionalyellow}{RGB}{255,193,7} % Grade 2A, 2B, 2C
|
||||
\definecolor{conditionalamber}{RGB}{255,160,0} % Darker variant
|
||||
|
||||
% Research/Investigational (insufficient evidence)
|
||||
\definecolor{researchblue}{RGB}{33,150,243} % Clinical trials
|
||||
\definecolor{researchdark}{RGB}{25,118,210} % Darker variant
|
||||
|
||||
% Not recommended / Contraindicated
|
||||
\definecolor{warningred}{RGB}{204,0,0} % Strong recommendation against
|
||||
\definecolor{dangerred}{RGB}{220,20,60} % Critical warnings, urgent actions
|
||||
|
||||
% ============================================================================
|
||||
% URGENCY LEVELS (Clinical Pathways)
|
||||
% ============================================================================
|
||||
|
||||
\definecolor{urgentred}{RGB}{220,20,60} % Immediate action (<1 hour)
|
||||
\definecolor{semiurgent}{RGB}{255,152,0} % Action within 24 hours
|
||||
\definecolor{routineblue}{RGB}{100,181,246} % Routine care (>24 hours)
|
||||
\definecolor{actiongreen}{RGB}{0,153,76} % Standard interventions
|
||||
|
||||
% ============================================================================
|
||||
% BIOMARKER CATEGORIES
|
||||
% ============================================================================
|
||||
|
||||
% Alteration types
|
||||
\definecolor{mutationred}{RGB}{244,67,54} % Point mutations, SNVs
|
||||
\definecolor{amplificationblue}{RGB}{33,150,243} % Copy number gains
|
||||
\definecolor{deletionpurple}{RGB}{156,39,176} % Copy number losses
|
||||
\definecolor{fusionpurple}{RGB}{156,39,176} % Gene fusions/rearrangements
|
||||
\definecolor{expressionorange}{RGB}{255,152,0} % Expression alterations
|
||||
|
||||
% Actionability tiers
|
||||
\definecolor{tier1green}{RGB}{0,153,76} % FDA-approved therapy
|
||||
\definecolor{tier2orange}{RGB}{255,152,0} % Clinical trial/off-label
|
||||
\definecolor{tier3gray}{RGB}{158,158,158} % VUS, no action
|
||||
|
||||
% ============================================================================
|
||||
% STATISTICAL SIGNIFICANCE
|
||||
% ============================================================================
|
||||
|
||||
\definecolor{significant}{RGB}{0,153,76} % p < 0.05, statistically significant
|
||||
\definecolor{trending}{RGB}{255,193,7} % p = 0.05-0.10, trending
|
||||
\definecolor{nonsignificant}{RGB}{158,158,158} % p > 0.10, not significant
|
||||
|
||||
% ============================================================================
|
||||
% OUTCOME CATEGORIES
|
||||
% ============================================================================
|
||||
|
||||
% Response assessment (RECIST)
|
||||
\definecolor{completeresponse}{RGB}{0,153,76} % CR (complete response)
|
||||
\definecolor{partialresponse}{RGB}{76,175,80} % PR (partial response)
|
||||
\definecolor{stabledisease}{RGB}{255,193,7} % SD (stable disease)
|
||||
\definecolor{progressivedisease}{RGB}{244,67,54} % PD (progressive disease)
|
||||
|
||||
% Survival outcomes
|
||||
\definecolor{survivedgreen}{RGB}{0,153,76} % Patient alive
|
||||
\definecolor{eventred}{RGB}{244,67,54} % Event occurred (death, progression)
|
||||
\definecolor{censoredgray}{RGB}{158,158,158} % Censored observation
|
||||
|
||||
% ============================================================================
|
||||
% ADVERSE EVENT SEVERITY (CTCAE)
|
||||
% ============================================================================
|
||||
|
||||
\definecolor{grade1}{RGB}{255,235,59} % Mild
|
||||
\definecolor{grade2}{RGB}{255,193,7} % Moderate
|
||||
\definecolor{grade3}{RGB}{255,152,0} % Severe
|
||||
\definecolor{grade4}{RGB}{244,67,54} % Life-threatening
|
||||
\definecolor{grade5}{RGB}{198,40,40} % Fatal
|
||||
|
||||
% ============================================================================
|
||||
% COLORBLIND-SAFE PALETTE (Okabe-Ito)
|
||||
% ============================================================================
|
||||
% Use these for graphs/figures to ensure accessibility
|
||||
|
||||
\definecolor{okabe1}{RGB}{230,159,0} % Orange
|
||||
\definecolor{okabe2}{RGB}{86,180,233} % Sky blue
|
||||
\definecolor{okabe3}{RGB}{0,158,115} % Bluish green
|
||||
\definecolor{okabe4}{RGB}{240,228,66} % Yellow
|
||||
\definecolor{okabe5}{RGB}{0,114,178} % Blue
|
||||
\definecolor{okabe6}{RGB}{213,94,0} % Vermillion
|
||||
\definecolor{okabe7}{RGB}{204,121,167} % Reddish purple
|
||||
|
||||
% ============================================================================
|
||||
% USAGE EXAMPLES
|
||||
% ============================================================================
|
||||
|
||||
% Example 1: Strong recommendation box
|
||||
% \begin{tcolorbox}[enhanced,colback=stronggreen!10,colframe=stronggreen,
|
||||
% title={\textbf{STRONG RECOMMENDATION} \hfill \textbf{GRADE: 1A}}]
|
||||
% We recommend osimertinib for EGFR-mutated NSCLC...
|
||||
% \end{tcolorbox}
|
||||
|
||||
% Example 2: Conditional recommendation box
|
||||
% \begin{tcolorbox}[enhanced,colback=conditionalyellow!10,colframe=conditionalyellow,
|
||||
% title={\textbf{CONDITIONAL RECOMMENDATION} \hfill \textbf{GRADE: 2B}}]
|
||||
% We suggest considering maintenance therapy...
|
||||
% \end{tcolorbox}
|
||||
|
||||
% Example 3: Biomarker alteration
|
||||
% \colorbox{mutationred!60}{\textcolor{white}{\textbf{MUTATION}}}
|
||||
|
||||
% Example 4: Statistical significance in table
|
||||
% \cellcolor{significant!20} p < 0.001
|
||||
|
||||
% Example 5: Adverse event severity
|
||||
% \textcolor{grade3}{Grade 3} or \colorbox{grade3!30}{Grade 3}
|
||||
|
||||
% ============================================================================
|
||||
% ACCESSIBILITY NOTES
|
||||
% ============================================================================
|
||||
|
||||
% 1. Always use sufficient color contrast (4.5:1 ratio for normal text)
|
||||
% 2. Do not rely on color alone - use symbols/text as well
|
||||
% 3. Test in grayscale to ensure readability
|
||||
% 4. Use Okabe-Ito palette for colorblind accessibility in figures
|
||||
% 5. Add text labels to colored boxes ("STRONG", "CONDITIONAL", etc.)
|
||||
|
||||
% ============================================================================
|
||||
% STYLE CONSISTENCY
|
||||
% ============================================================================
|
||||
|
||||
% Font: Helvetica (sans-serif) for clinical documents
|
||||
% Margins: 0.5 inches for compact professional appearance
|
||||
% Font sizes: 10pt body, 11pt subsections, 12-14pt headers
|
||||
% Line spacing: Compact (minimal whitespace for dense information)
|
||||
% Boxes: tcolorbox with rounded corners, colored backgrounds at 10-20% opacity
|
||||
|
||||
% End of color scheme definitions
|
||||
|
||||
208
skills/clinical-decision-support/assets/example_gbm_cohort.md
Normal file
208
skills/clinical-decision-support/assets/example_gbm_cohort.md
Normal file
@@ -0,0 +1,208 @@
|
||||
# Example: GBM Molecular Subtype Cohort Analysis
|
||||
|
||||
## Clinical Context
|
||||
|
||||
This example demonstrates a patient cohort analysis stratified by molecular biomarkers, similar to the GBM Mesenchymal-Immune-Active cluster analysis provided as reference.
|
||||
|
||||
## Cohort Overview
|
||||
|
||||
**Disease**: Glioblastoma (GBM), IDH-wild-type
|
||||
|
||||
**Study Population**: n=60 patients with newly diagnosed GBM treated with standard Stupp protocol (temozolomide + radiation → adjuvant temozolomide)
|
||||
|
||||
**Molecular Classification**: Verhaak 2010 subtypes with immune signature refinement
|
||||
- **Group A**: Mesenchymal-Immune-Active subtype (n=18, 30%)
|
||||
- **Group B**: Other molecular subtypes (Proneural, Classical, Neural) (n=42, 70%)
|
||||
|
||||
**Study Period**: January 2019 - December 2022
|
||||
|
||||
**Data Source**: Single academic medical center, retrospective cohort analysis
|
||||
|
||||
## Biomarker Classification
|
||||
|
||||
### Mesenchymal-Immune-Active Subtype Characteristics
|
||||
|
||||
**Molecular Features**:
|
||||
- NF1 alterations (mutations or deletions): 72% (13/18)
|
||||
- High YKL-40 (CHI3L1) expression: 100% (18/18, median z-score +2.8)
|
||||
- Immune gene signature: Elevated (median ESTIMATE immune score +1250)
|
||||
- CD163+ macrophage infiltration: High density (median 195 cells/mm², range 120-340)
|
||||
- MES (mesenchymal) signature score: >0.5 (all patients)
|
||||
|
||||
**Clinical Characteristics**:
|
||||
- Median age: 64 years (range 42-76)
|
||||
- Male: 61% (11/18)
|
||||
- Tumor location: Temporal lobe predominant (55%)
|
||||
- Multifocal disease: 33% (6/18) - higher than overall cohort
|
||||
|
||||
### Comparison Groups (Other Subtypes)
|
||||
|
||||
**Molecular Features**:
|
||||
- Proneural: n=15 (25%) - PDGFRA amplification, younger age
|
||||
- Classical: n=18 (30%) - EGFR amplification, chromosome 7+/10-
|
||||
- Neural: n=9 (15%) - neuronal markers, may include normal tissue
|
||||
|
||||
## Treatment Outcomes
|
||||
|
||||
### Response Assessment (RANO Criteria)
|
||||
|
||||
**Objective Response Rate** (after chemoradiation, ~3 months):
|
||||
- Mesenchymal-Immune-Active: 6/18 (33%) - CR 0, PR 6
|
||||
- Other subtypes: 18/42 (43%) - CR 1, PR 17
|
||||
- p = 0.48 (Fisher's exact)
|
||||
|
||||
**Interpretation**: No significant difference in initial response rates
|
||||
|
||||
### Survival Outcomes
|
||||
|
||||
**Progression-Free Survival (PFS)**:
|
||||
- Mesenchymal-Immune-Active: Median 7.2 months (95% CI 5.8-9.1)
|
||||
- Other subtypes: Median 9.5 months (95% CI 8.1-11.3)
|
||||
- Hazard Ratio: 1.58 (95% CI 0.89-2.81), p = 0.12
|
||||
- 6-month PFS rate: 61% vs 74%
|
||||
|
||||
**Overall Survival (OS)**:
|
||||
- Mesenchymal-Immune-Active: Median 12.8 months (95% CI 10.2-15.4)
|
||||
- Other subtypes: Median 16.3 months (95% CI 14.7-18.9)
|
||||
- Hazard Ratio: 1.72 (95% CI 0.95-3.11), p = 0.073
|
||||
- 12-month OS rate: 55% vs 68%
|
||||
- 24-month OS rate: 17% vs 31%
|
||||
|
||||
**Interpretation**: Trend toward worse survival in mesenchymal-immune-active subtype, not reaching statistical significance in this cohort size
|
||||
|
||||
### Response to Bevacizumab at Recurrence
|
||||
|
||||
**Subset Analysis** (patients receiving bevacizumab at first recurrence, n=35):
|
||||
- Mesenchymal-Immune-Active: n=12
|
||||
- ORR: 58% (7/12)
|
||||
- Median PFS2 (from bevacizumab start): 6.8 months
|
||||
- Other subtypes: n=23
|
||||
- ORR: 35% (8/23)
|
||||
- Median PFS2: 4.2 months
|
||||
- p = 0.19 (Fisher's exact for ORR)
|
||||
- HR for PFS2: 0.62 (95% CI 0.29-1.32), p = 0.21
|
||||
|
||||
**Interpretation**: Exploratory finding suggesting enhanced benefit from bevacizumab in mesenchymal-immune-active subtype (not statistically significant with small sample)
|
||||
|
||||
## Safety Profile
|
||||
|
||||
**Treatment-Related Adverse Events** (Temozolomide):
|
||||
|
||||
No significant differences in toxicity between molecular subtypes:
|
||||
- Lymphopenia (any grade): 89% vs 86%, p = 0.77
|
||||
- Thrombocytopenia (grade 3-4): 22% vs 19%, p = 0.79
|
||||
- Fatigue (any grade): 94% vs 90%, p = 0.60
|
||||
- Treatment discontinuation: 17% vs 14%, p = 0.77
|
||||
|
||||
## Clinical Implications
|
||||
|
||||
### Treatment Recommendations
|
||||
|
||||
**For Mesenchymal-Immune-Active GBM**:
|
||||
|
||||
1. **First-Line**: Standard Stupp protocol (no change based on subtype)
|
||||
- Evidence: No proven benefit for alternative first-line strategies
|
||||
- GRADE: 1A (strong recommendation, high-quality evidence)
|
||||
|
||||
2. **At Recurrence - Consider Bevacizumab Earlier**:
|
||||
- Rationale: Exploratory data suggesting enhanced anti-angiogenic response
|
||||
- Evidence: Mesenchymal GBM has high VEGF expression, angiogenic phenotype
|
||||
- GRADE: 2C (conditional recommendation, low-quality evidence from subset)
|
||||
|
||||
3. **Clinical Trial Enrollment - Immunotherapy Combinations**:
|
||||
- Rationale: High immune cell infiltration may predict immunotherapy benefit
|
||||
- Targets: PD-1/PD-L1 blockade ± anti-CTLA-4 or anti-angiogenic agents
|
||||
- Evidence: Ongoing trials (CheckMate-498, CheckMate-548 showed negative results, but did not select for immune-active)
|
||||
- GRADE: R (research recommendation)
|
||||
|
||||
**For Other GBM Subtypes**:
|
||||
- Standard treatment per NCCN guidelines
|
||||
- Consider tumor treating fields (Optune) after radiation completion
|
||||
- Clinical trials based on specific molecular features (EGFR amplification → EGFR inhibitor trials)
|
||||
|
||||
### Prognostic Information
|
||||
|
||||
**Counseling Patients**:
|
||||
- Mesenchymal-immune-active subtype associated with trend toward shorter survival (12.8 vs 16.3 months)
|
||||
- Not definitive due to small sample size and confidence intervals overlapping
|
||||
- Prospective validation needed
|
||||
- Should not alter standard first-line treatment
|
||||
|
||||
## Study Limitations
|
||||
|
||||
1. **Small Sample Size**: n=18 in mesenchymal-immune-active group limits statistical power
|
||||
2. **Retrospective Design**: Potential selection bias, unmeasured confounders
|
||||
3. **Single Institution**: May not generalize to other populations
|
||||
4. **Heterogeneous Recurrence Treatment**: Not all patients received bevacizumab; treatment selection bias
|
||||
5. **Molecular Classification**: Based on bulk tumor RNA-seq; intratumoral heterogeneity not captured
|
||||
6. **No Central Pathology Review**: Molecular classification performed locally
|
||||
|
||||
## Future Directions
|
||||
|
||||
1. **Prospective Validation**: Confirm survival differences in independent cohort (n>100 per group for adequate power)
|
||||
2. **Biomarker Testing**: Develop clinically feasible assay for mesenchymal-immune subtype identification
|
||||
3. **Clinical Trial Design**: Immunotherapy combinations targeting mesenchymal-immune-active GBM specifically
|
||||
4. **Mechanistic Studies**: Investigate why mesenchymal-immune GBM may respond better to bevacizumab
|
||||
5. **Longitudinal Analysis**: Track molecular subtype evolution over treatment course
|
||||
|
||||
## Data Presentation Example
|
||||
|
||||
### Baseline Characteristics Table
|
||||
|
||||
```
|
||||
Characteristic Mesenchymal-IA (n=18) Other (n=42) p-value
|
||||
Age, years (median [IQR]) 64 [56-71] 61 [53-68] 0.42
|
||||
Sex, n (%)
|
||||
Male 11 (61%) 24 (57%) 0.78
|
||||
Female 7 (39%) 18 (43%)
|
||||
ECOG PS, n (%)
|
||||
0-1 15 (83%) 37 (88%) 0.63
|
||||
2 3 (17%) 5 (12%)
|
||||
Tumor location
|
||||
Frontal 4 (22%) 15 (36%) 0.35
|
||||
Temporal 10 (56%) 16 (38%)
|
||||
Parietal/Occipital 4 (22%) 11 (26%)
|
||||
Extent of resection
|
||||
Gross total 8 (44%) 22 (52%) 0.58
|
||||
Subtotal 10 (56%) 20 (48%)
|
||||
MGMT promoter methylated 5 (28%) 18 (43%) 0.27
|
||||
```
|
||||
|
||||
### Survival Outcomes Summary
|
||||
|
||||
```
|
||||
Endpoint Mesenchymal-IA Other HR (95% CI) p-value
|
||||
Median PFS, months (95% CI) 7.2 (5.8-9.1) 9.5 (8.1-11.3) 1.58 (0.89-2.81) 0.12
|
||||
6-month PFS rate 61% 74%
|
||||
Median OS, months (95% CI) 12.8 (10.2-15.4) 16.3 (14.7-18.9) 1.72 (0.95-3.11) 0.073
|
||||
12-month OS rate 55% 68%
|
||||
24-month OS rate 17% 31%
|
||||
```
|
||||
|
||||
## Key Takeaways
|
||||
|
||||
1. **Molecular heterogeneity exists** in GBM with distinct subtypes
|
||||
2. **Mesenchymal-immune-active subtype** characterized by NF1 alterations, immune infiltration
|
||||
3. **Trend toward worse prognosis** but not statistically significant (power limitations)
|
||||
4. **Potential bevacizumab benefit** hypothesis-generating, requires prospective validation
|
||||
5. **Immunotherapy target**: High immune infiltration rational for checkpoint inhibitor trials
|
||||
6. **Clinical implementation pending**: Need prospective validation before routine subtyping
|
||||
|
||||
## References
|
||||
|
||||
1. Verhaak RG, et al. Integrated genomic analysis identifies clinically relevant subtypes of glioblastoma characterized by abnormalities in PDGFRA, IDH1, EGFR, and NF1. Cancer Cell. 2010;17(1):98-110.
|
||||
2. Wang Q, et al. Tumor Evolution of Glioma-Intrinsic Gene Expression Subtypes Associates with Immunological Changes in the Microenvironment. Cancer Cell. 2017;32(1):42-56.
|
||||
3. Stupp R, et al. Radiotherapy plus Concomitant and Adjuvant Temozolomide for Glioblastoma. NEJM. 2005;352(10):987-996.
|
||||
4. Gilbert MR, et al. Bevacizumab for Newly Diagnosed Glioblastoma. NEJM. 2014;370(8):699-708.
|
||||
5. NCCN Clinical Practice Guidelines in Oncology: Central Nervous System Cancers. Version 1.2024.
|
||||
|
||||
---
|
||||
|
||||
**This example demonstrates**:
|
||||
- Biomarker-based stratification methodology
|
||||
- Outcome reporting with appropriate statistics
|
||||
- Clinical contextualization of findings
|
||||
- Evidence-based recommendations with grading
|
||||
- Transparent limitation discussion
|
||||
- Structure suitable for pharmaceutical/clinical research documentation
|
||||
|
||||
@@ -0,0 +1,328 @@
|
||||
# Recommendation Strength Guide
|
||||
|
||||
## GRADE Framework for Clinical Recommendations
|
||||
|
||||
### Components of a Recommendation
|
||||
|
||||
Every clinical recommendation should address:
|
||||
|
||||
1. **Population**: Who should receive the intervention?
|
||||
2. **Intervention**: What specific treatment/action?
|
||||
3. **Comparator**: Compared to what alternative?
|
||||
4. **Outcome**: What are the expected results?
|
||||
5. **Strength**: How strong is the recommendation?
|
||||
6. **Quality of Evidence**: How confident are we in the evidence?
|
||||
|
||||
### Recommendation Strength (Grade 1 vs Grade 2)
|
||||
|
||||
#### Strong Recommendation (Grade 1)
|
||||
|
||||
**When to Use**:
|
||||
- Desirable effects clearly outweigh undesirable effects (or vice versa)
|
||||
- High or moderate quality evidence
|
||||
- Values and preferences: Little variability expected
|
||||
- Resource implications: Cost-effective or cost considerations minor
|
||||
|
||||
**Wording**: "We recommend..." or "Clinicians should..."
|
||||
|
||||
**Implications**:
|
||||
- Most patients should receive the recommended intervention
|
||||
- Adherence to recommendation could be a quality indicator
|
||||
- Policy-makers can adapt as performance measure
|
||||
|
||||
**Examples**:
|
||||
|
||||
```
|
||||
STRONG RECOMMENDATION FOR (Grade 1):
|
||||
|
||||
"We recommend osimertinib 80 mg daily as first-line therapy for adults with
|
||||
advanced NSCLC harboring EGFR exon 19 deletion or L858R mutation (Strong
|
||||
recommendation, High-quality evidence - GRADE 1A)."
|
||||
|
||||
Rationale:
|
||||
- Large PFS benefit: 18.9 vs 10.2 months (HR 0.46, p<0.001)
|
||||
- OS benefit: 38.6 vs 31.8 months (HR 0.80, p=0.046)
|
||||
- Better tolerability: Lower grade 3-4 AEs
|
||||
- Evidence: High-quality (large RCT, low risk of bias)
|
||||
- Benefits clearly outweigh harms
|
||||
```
|
||||
|
||||
```
|
||||
STRONG RECOMMENDATION AGAINST (Grade 1):
|
||||
|
||||
"We recommend against using bevacizumab in the first-line treatment of newly
|
||||
diagnosed glioblastoma to improve overall survival (Strong recommendation against,
|
||||
High-quality evidence - GRADE 1A)."
|
||||
|
||||
Rationale:
|
||||
- No OS benefit: HR 0.88 (0.76-1.02), p=0.10 (AVAglio trial)
|
||||
- Toxicity: Increased grade ≥3 AEs (66% vs 52%)
|
||||
- Evidence: High-quality (two large phase 3 RCTs)
|
||||
- Harms outweigh lack of survival benefit
|
||||
```
|
||||
|
||||
#### Conditional/Weak Recommendation (Grade 2)
|
||||
|
||||
**When to Use**:
|
||||
- Desirable and undesirable effects closely balanced
|
||||
- Low or very low quality evidence
|
||||
- Values and preferences: Substantial variability
|
||||
- Resource implications: High cost or limited access
|
||||
|
||||
**Wording**: "We suggest..." or "Clinicians might..."
|
||||
|
||||
**Implications**:
|
||||
- Different choices will be appropriate for different patients
|
||||
- Shared decision-making essential
|
||||
- Policy-making requires substantial debate and stakeholder involvement
|
||||
|
||||
**Examples**:
|
||||
|
||||
```
|
||||
CONDITIONAL RECOMMENDATION FOR (Grade 2):
|
||||
|
||||
"We suggest considering maintenance pemetrexed after first-line platinum-pemetrexed
|
||||
chemotherapy for advanced non-squamous NSCLC in patients without disease progression
|
||||
(Conditional recommendation, Moderate-quality evidence - GRADE 2B)."
|
||||
|
||||
Rationale:
|
||||
- Modest PFS benefit: 4.0 vs 2.0 months (HR 0.62)
|
||||
- No OS benefit: 13.9 vs 11.0 months (HR 0.79, p=0.23)
|
||||
- Toxicity: Continued chemotherapy burden
|
||||
- Quality of life: Trade-off between symptom control and treatment side effects
|
||||
- Patient values: Some prioritize time off treatment, others prioritize disease control
|
||||
- Shared decision-making essential
|
||||
```
|
||||
|
||||
```
|
||||
CONDITIONAL RECOMMENDATION - EITHER OPTION ACCEPTABLE (Grade 2):
|
||||
|
||||
"We suggest either pembrolizumab monotherapy OR pembrolizumab plus platinum-doublet
|
||||
chemotherapy as first-line treatment for PD-L1 ≥50% NSCLC, based on patient
|
||||
preferences and clinical factors (Conditional recommendation, High-quality evidence -
|
||||
GRADE 2A)."
|
||||
|
||||
Rationale:
|
||||
- Both regimens NCCN Category 1 preferred
|
||||
- Monotherapy: Less toxicity, oral vs IV, better quality of life
|
||||
- Combination: Higher ORR (48% vs 39%), numerically longer PFS
|
||||
- OS: Similar between strategies
|
||||
- Patient values: Varies widely (tolerability vs response rate priority)
|
||||
```
|
||||
|
||||
### Evidence Quality (⊕⊕⊕⊕ to ⊕○○○)
|
||||
|
||||
#### High Quality (⊕⊕⊕⊕)
|
||||
|
||||
- Further research very unlikely to change confidence in effect estimate
|
||||
- Consistent results from well-designed RCTs
|
||||
- No serious limitations
|
||||
- Direct evidence (target population, intervention, outcomes)
|
||||
- Precise estimate (narrow CI)
|
||||
|
||||
**Example**: FLAURA trial for osimertinib in EGFR+ NSCLC - Large RCT, consistent results, low risk of bias, direct outcomes
|
||||
|
||||
#### Moderate Quality (⊕⊕⊕○)
|
||||
|
||||
- Further research likely to impact confidence and may change estimate
|
||||
- RCTs with some limitations OR very strong evidence from observational studies
|
||||
- Some inconsistency, indirectness, imprecision, or publication bias
|
||||
|
||||
**Example**: Single RCT with some limitations, or multiple RCTs with moderate heterogeneity
|
||||
|
||||
#### Low Quality (⊕⊕○○)
|
||||
|
||||
- Further research very likely to have important impact on confidence in estimate
|
||||
- Observational studies OR RCTs with serious limitations
|
||||
- Serious issues with consistency, directness, precision, or bias
|
||||
|
||||
**Example**: Well-conducted cohort study, or RCT with high attrition and unclear allocation concealment
|
||||
|
||||
#### Very Low Quality (⊕○○○)
|
||||
|
||||
- Estimate of effect very uncertain
|
||||
- Case series, expert opinion, mechanistic reasoning
|
||||
- Very serious limitations
|
||||
|
||||
**Example**: Retrospective case series, expert consensus without systematic review
|
||||
|
||||
## Combining Strength and Quality
|
||||
|
||||
### All Nine Possible Combinations
|
||||
|
||||
| Evidence Quality | Strong For (↑↑) | Weak For (↑) | Strong Against (↓↓) | Weak Against (↓) |
|
||||
|-----------------|----------------|--------------|---------------------|------------------|
|
||||
| **High (⊕⊕⊕⊕)** | Grade 1A | Grade 2A | Grade 1A (against) | Grade 2A (against) |
|
||||
| **Moderate (⊕⊕⊕○)** | Grade 1B | Grade 2B | Grade 1B (against) | Grade 2B (against) |
|
||||
| **Low (⊕⊕○○)** | Grade 1C* | Grade 2C | Grade 1C (against)* | Grade 2C (against) |
|
||||
| **Very Low (⊕○○○)** | Grade 1D* | Grade 2D | Grade 1D (against)* | Grade 2D (against) |
|
||||
|
||||
*Rare: Strong recommendations usually require at least moderate-quality evidence
|
||||
|
||||
### Unusual Combinations (When They Occur)
|
||||
|
||||
**Strong Recommendation with Low Quality Evidence (Grade 1C)**
|
||||
|
||||
Rare, but can occur when:
|
||||
- Large magnitude of effect from observational data (RR >5 or <0.2)
|
||||
- Low quality evidence, but clear benefit-harm balance
|
||||
- Example: Anticoagulation for atrial fibrillation (before RCTs, strong observational data)
|
||||
|
||||
**Weak Recommendation with High Quality Evidence (Grade 2A)**
|
||||
|
||||
Occurs when:
|
||||
- Benefits and harms closely balanced
|
||||
- Patient values highly variable
|
||||
- Example: Aspirin for primary prevention in low-risk individuals (benefits small, bleeding risk present, patient values vary)
|
||||
|
||||
## Wording Templates
|
||||
|
||||
### Strong Recommendations
|
||||
|
||||
**FOR (↑↑)**:
|
||||
- "We recommend [intervention] for [population]."
|
||||
- "Clinicians should [action]."
|
||||
- "[Intervention] is recommended."
|
||||
|
||||
**AGAINST (↓↓)**:
|
||||
- "We recommend against [intervention] for [population]."
|
||||
- "Clinicians should not [action]."
|
||||
- "[Intervention] is not recommended."
|
||||
|
||||
### Conditional/Weak Recommendations
|
||||
|
||||
**FOR (↑)**:
|
||||
- "We suggest [intervention] for [population]."
|
||||
- "Clinicians might consider [action]."
|
||||
- "[Intervention] may be considered for selected patients."
|
||||
|
||||
**AGAINST (↓)**:
|
||||
- "We suggest not using [intervention] for [population]."
|
||||
- "Clinicians might avoid [action]."
|
||||
- "[Intervention] is generally not recommended."
|
||||
|
||||
**EITHER ACCEPTABLE**:
|
||||
- "We suggest either [option A] or [option B] based on patient preferences."
|
||||
- "Either approach is reasonable."
|
||||
|
||||
## Color Coding for Visual Documents
|
||||
|
||||
**Strong Recommendations (Green Background)**:
|
||||
- RGB(0, 153, 76) or #009954
|
||||
- Clear visual priority
|
||||
- Use for Grade 1A, 1B
|
||||
|
||||
**Conditional Recommendations (Yellow Background)**:
|
||||
- RGB(255, 193, 7) or #FFC107
|
||||
- Indicates discussion needed
|
||||
- Use for Grade 2A, 2B, 2C
|
||||
|
||||
**Research/Investigational (Blue Background)**:
|
||||
- RGB(33, 150, 243) or #2196F3
|
||||
- Clinical trial consideration
|
||||
- Insufficient evidence for standard care
|
||||
|
||||
**Not Recommended (Red Border/Background)**:
|
||||
- RGB(220, 20, 60) or #DC143C
|
||||
- Strong recommendation against
|
||||
- Evidence of harm or no benefit
|
||||
|
||||
## Common Scenarios
|
||||
|
||||
### Scenario 1: Strong Evidence, Clear Benefit-Harm Balance
|
||||
|
||||
**Example**: Pembrolizumab for PD-L1 ≥50% NSCLC
|
||||
|
||||
- Evidence: Large phase 3 RCT (KEYNOTE-024), n=305, well-designed
|
||||
- Results: PFS HR 0.50 (0.37-0.68), OS HR 0.60 (0.41-0.89)
|
||||
- Toxicity: Lower grade 3-5 AEs than chemotherapy (27% vs 53%)
|
||||
- Patient values: Most prioritize efficacy and tolerability
|
||||
|
||||
**Recommendation**: STRONG FOR (Grade 1A)
|
||||
|
||||
### Scenario 2: Moderate Evidence, Balanced Trade-Offs
|
||||
|
||||
**Example**: Adjuvant immunotherapy for resected melanoma
|
||||
|
||||
- Evidence: RCT showing relapse-free survival benefit, OS data immature
|
||||
- Results: Recurrence risk reduced but ongoing toxicity
|
||||
- Toxicity: Immune-related AEs requiring steroids (some severe)
|
||||
- Cost: High annual cost for 12 months treatment
|
||||
- Patient values: Variable (some prioritize recurrence prevention, others avoid toxicity)
|
||||
|
||||
**Recommendation**: CONDITIONAL FOR (Grade 2B)
|
||||
|
||||
### Scenario 3: Low Evidence, but Severe Consequence
|
||||
|
||||
**Example**: Anticoagulation for prosthetic heart valve
|
||||
|
||||
- Evidence: No RCTs (would be unethical), observational data and mechanistic reasoning
|
||||
- Consequence: Very high thromboembolic risk without anticoagulation
|
||||
- Benefit-harm: Clear despite low quality evidence
|
||||
|
||||
**Recommendation**: STRONG FOR (Grade 1C)
|
||||
|
||||
### Scenario 4: High Evidence, but Patient Preferences Vary
|
||||
|
||||
**Example**: Breast reconstruction after mastectomy
|
||||
|
||||
- Evidence: High-quality data on outcomes and satisfaction
|
||||
- Trade-offs: Cosmetic benefit vs additional surgery, recovery time
|
||||
- Values: Highly personal decision, wide preference variability
|
||||
|
||||
**Recommendation**: CONDITIONAL (Grade 2A) - discuss options, patient decides
|
||||
|
||||
## Documentation Template
|
||||
|
||||
```
|
||||
RECOMMENDATION: [State recommendation clearly]
|
||||
|
||||
Strength: [STRONG / CONDITIONAL]
|
||||
Quality of Evidence: [HIGH / MODERATE / LOW / VERY LOW]
|
||||
GRADE: [1A / 1B / 2A / 2B / 2C]
|
||||
|
||||
Evidence Summary:
|
||||
- Primary study: [Citation]
|
||||
- Design: [RCT / Observational / Meta-analysis]
|
||||
- Sample size: n = [X]
|
||||
- Results: [Primary outcome with effect size, CI, p-value]
|
||||
- Quality assessment: [Strengths and limitations]
|
||||
|
||||
Benefits:
|
||||
- [Quantified benefit 1]
|
||||
- [Quantified benefit 2]
|
||||
|
||||
Harms:
|
||||
- [Quantified harm 1]
|
||||
- [Quantified harm 2]
|
||||
|
||||
Balance: [Benefits clearly outweigh harms / Close balance requiring discussion / etc.]
|
||||
|
||||
Values and Preferences: [Little variability / Substantial variability]
|
||||
|
||||
Cost Considerations: [If relevant]
|
||||
|
||||
Guideline Concordance:
|
||||
- NCCN: [Category and recommendation]
|
||||
- ASCO: [Recommendation]
|
||||
- ESMO: [Grade and recommendation]
|
||||
```
|
||||
|
||||
## Quality Checklist
|
||||
|
||||
Before finalizing recommendations, verify:
|
||||
|
||||
- [ ] Recommendation statement is clear and actionable
|
||||
- [ ] Strength is explicitly stated (strong vs conditional)
|
||||
- [ ] Quality of evidence is graded (high/moderate/low/very low)
|
||||
- [ ] GRADE notation provided (1A, 1B, 2A, 2B, 2C)
|
||||
- [ ] Evidence is cited with specific study results
|
||||
- [ ] Benefits are quantified (effect sizes with CIs)
|
||||
- [ ] Harms are quantified (AE rates)
|
||||
- [ ] Balance of benefits/harms is explained
|
||||
- [ ] Patient values consideration is addressed (if conditional)
|
||||
- [ ] Alternative options are mentioned
|
||||
- [ ] Guideline concordance is documented
|
||||
- [ ] Special populations are addressed (elderly, renal/hepatic impairment)
|
||||
- [ ] Monitoring requirements are specified
|
||||
|
||||
@@ -0,0 +1,529 @@
|
||||
\documentclass[10pt,letterpaper]{article}
|
||||
|
||||
% Packages
|
||||
\usepackage[margin=0.5in]{geometry}
|
||||
\usepackage[utf8]{inputenc}
|
||||
\usepackage[T1]{fontenc}
|
||||
\usepackage{helvet}
|
||||
\renewcommand{\familydefault}{\sfdefault}
|
||||
\usepackage{xcolor}
|
||||
\usepackage{tcolorbox}
|
||||
\usepackage{array}
|
||||
\usepackage{tabularx}
|
||||
\usepackage{booktabs}
|
||||
\usepackage{enumitem}
|
||||
\usepackage{titlesec}
|
||||
\usepackage{fancyhdr}
|
||||
\usepackage{multicol}
|
||||
\usepackage{graphicx}
|
||||
\usepackage{tikz}
|
||||
\usetikzlibrary{shapes,arrows,positioning}
|
||||
|
||||
% Color definitions
|
||||
\definecolor{headerblue}{RGB}{0,102,204}
|
||||
\definecolor{stronggreen}{RGB}{0,153,76}
|
||||
\definecolor{conditionalyellow}{RGB}{255,193,7}
|
||||
\definecolor{researchblue}{RGB}{33,150,243}
|
||||
\definecolor{warningred}{RGB}{204,0,0}
|
||||
\definecolor{highlightgray}{RGB}{240,240,240}
|
||||
|
||||
% Section formatting - compact
|
||||
\titleformat{\section}{\normalfont\fontsize{11}{12}\bfseries\color{headerblue}}{\thesection}{0.5em}{}
|
||||
\titlespacing*{\section}{0pt}{4pt}{2pt}
|
||||
|
||||
\titleformat{\subsection}{\normalfont\fontsize{10}{11}\bfseries}{\thesubsection}{0.5em}{}
|
||||
\titlespacing*{\subsection}{0pt}{3pt}{1pt}
|
||||
|
||||
% List formatting - ultra compact
|
||||
\setlist[itemize]{leftmargin=*,itemsep=0pt,parsep=0pt,topsep=1pt}
|
||||
\setlist[enumerate]{leftmargin=*,itemsep=0pt,parsep=0pt,topsep=1pt}
|
||||
|
||||
% Remove paragraph indentation
|
||||
\setlength{\parindent}{0pt}
|
||||
\setlength{\parskip}{2pt}
|
||||
|
||||
% Header/footer
|
||||
\pagestyle{fancy}
|
||||
\fancyhf{}
|
||||
\fancyhead[L]{\footnotesize \textbf{Treatment Recommendations: [CONDITION]}}
|
||||
\fancyhead[R]{\footnotesize Page \thepage}
|
||||
\renewcommand{\headrulewidth}{0.5pt}
|
||||
\fancyfoot[C]{\footnotesize Evidence-Based Clinical Guideline - For Professional Use Only}
|
||||
|
||||
\begin{document}
|
||||
|
||||
% Title block
|
||||
\begin{center}
|
||||
{\fontsize{14}{16}\selectfont\bfseries\color{headerblue} EVIDENCE-BASED TREATMENT RECOMMENDATIONS}\\[2pt]
|
||||
{\fontsize{12}{14}\selectfont\bfseries [Disease/Condition - e.g., HER2+ Metastatic Breast Cancer]}\\[2pt]
|
||||
{\fontsize{10}{12}\selectfont [Institution/Organization]}\\[1pt]
|
||||
{\fontsize{9}{11}\selectfont Version X.X | Effective Date: [Date] | Next Review: [Date]}
|
||||
\end{center}
|
||||
|
||||
\vspace{4pt}
|
||||
|
||||
% Recommendation Strength Legend
|
||||
\begin{tcolorbox}[colback=highlightgray,colframe=black,title=\textbf{Recommendation Strength Key},fonttitle=\bfseries\small,coltitle=black]
|
||||
{\small
|
||||
\begin{itemize}
|
||||
\item \colorbox{stronggreen!30}{\textbf{STRONG (Grade 1)}} - Benefits clearly outweigh risks; most patients should receive intervention
|
||||
\item \colorbox{conditionalyellow!30}{\textbf{CONDITIONAL (Grade 2)}} - Trade-offs exist; shared decision-making essential
|
||||
\item \colorbox{researchblue!30}{\textbf{RESEARCH (Grade R)}} - Insufficient evidence; clinical trial enrollment preferred
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Evidence Quality}: \textbf{A} = High (RCTs), \textbf{B} = Moderate (RCTs with limitations), \textbf{C} = Low (observational), \textbf{D} = Very low (expert opinion)
|
||||
}
|
||||
\end{tcolorbox}
|
||||
|
||||
\vspace{2pt}
|
||||
|
||||
\section{Clinical Context}
|
||||
|
||||
\subsection{Disease Overview}
|
||||
|
||||
[Brief description of disease state, epidemiology, natural history]
|
||||
|
||||
\subsection{Patient Population}
|
||||
|
||||
\textbf{Target Population}:
|
||||
\begin{itemize}
|
||||
\item [Demographic characteristics - e.g., Adults $\geq$18 years]
|
||||
\item [Disease stage/severity - e.g., Metastatic disease, Stage IV]
|
||||
\item [Biomarker status - e.g., HER2-positive (IHC 3+ or FISH+)]
|
||||
\item [Performance status - e.g., ECOG 0-2]
|
||||
\item [Line of therapy - e.g., First-line, previously untreated]
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Exclusions}:
|
||||
\begin{itemize}
|
||||
\item [Contraindications to recommended therapies]
|
||||
\item [Comorbidities affecting eligibility]
|
||||
\end{itemize}
|
||||
|
||||
\section{Evidence Review}
|
||||
|
||||
\subsection{Key Clinical Trials}
|
||||
|
||||
\textbf{[Trial Name 1]} (Author, Journal Year):
|
||||
\begin{itemize}
|
||||
\item \textbf{Design}: Phase 3 RCT, n=XXX, [Treatment A] vs [Treatment B]
|
||||
\item \textbf{Population}: [Key eligibility criteria]
|
||||
\item \textbf{Primary Endpoint}: [Outcome] - XX vs XX months (HR X.XX, 95\% CI X.XX-X.XX, p<X.XXX)
|
||||
\item \textbf{Secondary Endpoints}: [Additional outcomes]
|
||||
\item \textbf{Safety}: Grade 3-4 AEs XX\% vs XX\%
|
||||
\item \textbf{Quality}: \textbf{High} (low risk of bias, adequate power, intention-to-treat analysis)
|
||||
\end{itemize}
|
||||
|
||||
\textbf{[Trial Name 2]} (Author, Journal Year):
|
||||
\begin{itemize}
|
||||
\item \textbf{Design}: Phase 3 RCT, n=XXX, [Treatment C] vs [Standard of care]
|
||||
\item \textbf{Primary Endpoint}: [Outcome and results]
|
||||
\item \textbf{Quality}: \textbf{Moderate} (some limitations)
|
||||
\end{itemize}
|
||||
|
||||
\subsection{Guideline Concordance}
|
||||
|
||||
\begin{table}[H]
|
||||
\centering
|
||||
\small
|
||||
\begin{tabular}{lll}
|
||||
\toprule
|
||||
\textbf{Guideline} & \textbf{Recommendation} & \textbf{Evidence Level} \\
|
||||
\midrule
|
||||
NCCN vX.XXXX & [Specific recommendation] & Category 1 (preferred) \\
|
||||
ASCO Year & [Recommendation] & Strong, Evidence A \\
|
||||
ESMO Year & [Recommendation] & Grade I, A \\
|
||||
\bottomrule
|
||||
\end{tabular}
|
||||
\caption{Major guideline recommendations}
|
||||
\end{table}
|
||||
|
||||
\section{Treatment Options}
|
||||
|
||||
\subsection{First-Line Therapy}
|
||||
|
||||
\begin{tcolorbox}[enhanced,colback=stronggreen!10,colframe=stronggreen,
|
||||
title={\textbf{Option 1: [Regimen Name]} \hfill \colorbox{white}{\textbf{STRONG (1A)}}},
|
||||
fonttitle=\bfseries\small,coltitle=black]
|
||||
{\small
|
||||
\textbf{Regimen}:
|
||||
\begin{itemize}
|
||||
\item [Drug A]: XX mg [IV/PO] [schedule]
|
||||
\item [Drug B]: XX mg [IV/PO] [schedule]
|
||||
\item Cycle length: XX days
|
||||
\item Duration: Until progression or unacceptable toxicity
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Evidence Basis}:
|
||||
\begin{itemize}
|
||||
\item Primary study: [Trial name], n=XXX
|
||||
\item Primary outcome: [Endpoint] XX vs XX months (HR X.XX, p<X.XXX)
|
||||
\item ORR: XX\% vs XX\% (control)
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Indications}:
|
||||
\begin{itemize}
|
||||
\item [Biomarker-defined population or all patients]
|
||||
\item [Performance status requirement]
|
||||
\item [Organ function requirements]
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Key Toxicities}:
|
||||
\begin{itemize}
|
||||
\item Grade 3-4 AEs: XX\%
|
||||
\item Common: [List 3-5 most common AEs with incidence]
|
||||
\item Serious: [SAEs, discontinuation rate]
|
||||
\item Management: [Key mitigation strategies]
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Monitoring}:
|
||||
\begin{itemize}
|
||||
\item Labs: [Specific tests, frequency]
|
||||
\item Imaging: Every [X weeks] (RECIST v1.1)
|
||||
\item Clinical assessment: Every cycle
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Recommendation Strength}: \textbf{STRONG} - Benefits clearly outweigh risks\\
|
||||
\textbf{Evidence Quality}: \textbf{HIGH} - Well-designed RCT with consistent results
|
||||
}
|
||||
\end{tcolorbox}
|
||||
|
||||
\vspace{3pt}
|
||||
|
||||
\begin{tcolorbox}[enhanced,colback=conditionalyellow!10,colframe=conditionalyellow,
|
||||
title={\textbf{Option 2: [Alternative Regimen]} \hfill \colorbox{white}{\textbf{CONDITIONAL (2B)}}},
|
||||
fonttitle=\bfseries\small,coltitle=black]
|
||||
{\small
|
||||
\textbf{Regimen}: [Dosing details]
|
||||
|
||||
\textbf{Evidence Basis}: [Moderate-quality evidence or specific population subset]
|
||||
|
||||
\textbf{Indications}: [When to consider this option - e.g., patient preference for oral therapy, specific contraindication to Option 1]
|
||||
|
||||
\textbf{Trade-offs}:
|
||||
\begin{itemize}
|
||||
\item Advantages: [e.g., Oral administration, better tolerability]
|
||||
\item Disadvantages: [e.g., Lower response rate, less survival benefit]
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Recommendation Strength}: \textbf{CONDITIONAL} - Patient values important in decision\\
|
||||
\textbf{Evidence Quality}: \textbf{MODERATE} - Some limitations in evidence base
|
||||
}
|
||||
\end{tcolorbox}
|
||||
|
||||
\vspace{3pt}
|
||||
|
||||
\begin{tcolorbox}[enhanced,colback=researchblue!10,colframe=researchblue,
|
||||
title={\textbf{Option 3: Clinical Trial} \hfill \colorbox{white}{\textbf{RESEARCH (R)}}},
|
||||
fonttitle=\bfseries\small,coltitle=black]
|
||||
{\small
|
||||
\textbf{Recommendation}: Consider clinical trial enrollment for [specific scenario - e.g., biomarker-selected patients, refractory disease]
|
||||
|
||||
\textbf{Available Trials}: [List relevant trials if known, or state "ClinicalTrials.gov search"]
|
||||
|
||||
\textbf{Rationale}: [Why clinical trial appropriate - e.g., novel mechanism, unmet medical need, investigational biomarker]
|
||||
}
|
||||
\end{tcolorbox}
|
||||
|
||||
\subsection{Second-Line and Beyond}
|
||||
|
||||
\textbf{At Progression on First-Line Therapy}:
|
||||
|
||||
\begin{itemize}
|
||||
\item \textbf{Biomarker Re-Testing}: [If applicable - e.g., liquid biopsy for resistance mutations]
|
||||
\item \textbf{Second-Line Options}:
|
||||
\begin{itemize}
|
||||
\item Preferred: [Regimen] (Evidence level)
|
||||
\item Alternative: [Regimen] (Evidence level)
|
||||
\end{itemize}
|
||||
\item \textbf{Third-Line Options}: [Subsequent therapy options]
|
||||
\end{itemize}
|
||||
|
||||
\section{Special Populations}
|
||||
|
||||
\subsection{Elderly Patients ($\geq$70 years)}
|
||||
|
||||
\textbf{Considerations}:
|
||||
\begin{itemize}
|
||||
\item Geriatric assessment recommended (G8 screening tool)
|
||||
\item Dose reductions: [Specific adjustments for frail patients]
|
||||
\item Monitoring: More frequent assessments for toxicity
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Regimen Modifications}:
|
||||
\begin{itemize}
|
||||
\item [Reduced-intensity regimens if appropriate]
|
||||
\item [Single-agent vs combination considerations]
|
||||
\end{itemize}
|
||||
|
||||
\subsection{Renal Impairment}
|
||||
|
||||
\begin{table}[H]
|
||||
\centering
|
||||
\footnotesize
|
||||
\begin{tabular}{lll}
|
||||
\toprule
|
||||
\textbf{eGFR (mL/min/1.73m²)} & \textbf{Category} & \textbf{Dose Adjustment} \\
|
||||
\midrule
|
||||
$\geq$60 & Normal/Mild & Standard dosing \\
|
||||
30-59 & Moderate & [Specific adjustment - e.g., Reduce 25\%] \\
|
||||
15-29 & Severe & [Specific adjustment - e.g., Reduce 50\% or avoid] \\
|
||||
<15 or dialysis & ESRD & [Use with caution or contraindicated] \\
|
||||
\bottomrule
|
||||
\end{tabular}
|
||||
\caption{Dose adjustments for renal impairment}
|
||||
\end{table}
|
||||
|
||||
\subsection{Hepatic Impairment}
|
||||
|
||||
[Similar table for hepatic dose adjustments using Child-Pugh class or bilirubin/transaminases]
|
||||
|
||||
\section{Clinical Decision Algorithm}
|
||||
|
||||
% Simple flowchart example - can be expanded with more complex TikZ
|
||||
\begin{center}
|
||||
\begin{tikzpicture}[node distance=1.8cm, auto,
|
||||
decision/.style={diamond, draw, fill=conditionalyellow!30, text width=4.5em, text centered, inner sep=1pt, font=\tiny},
|
||||
process/.style={rectangle, draw, fill=stronggreen!20, text width=5.5em, text centered, rounded corners, minimum height=2em, font=\tiny},
|
||||
terminal/.style={rectangle, draw, fill=highlightgray, text width=5.5em, text centered, rounded corners=6pt, minimum height=2em, font=\tiny},
|
||||
alert/.style={rectangle, draw=warningred, line width=1pt, fill=warningred!10, text width=5.5em, text centered, rounded corners, minimum height=2em, font=\tiny\bfseries},
|
||||
arrow/.style={thick,->,>=stealth}]
|
||||
|
||||
\node [terminal] (start) {[Disease] Diagnosis Confirmed};
|
||||
\node [decision, below of=start, node distance=1.8cm] (biomarker) {Biomarker\\ Positive?};
|
||||
\node [process, left of=biomarker, node distance=3.5cm] (optionA) {Targeted\\ Therapy};
|
||||
\node [process, right of=biomarker, node distance=3.5cm] (optionB) {Standard\\ Therapy};
|
||||
\node [terminal, below of=biomarker, node distance=2.5cm] (monitor) {Monitor Response\\ Every X weeks};
|
||||
|
||||
\draw [arrow] (start) -- (biomarker);
|
||||
\draw [arrow] (biomarker) -- node[above] {Yes} (optionA);
|
||||
\draw [arrow] (biomarker) -- node[above] {No} (optionB);
|
||||
\draw [arrow] (optionA) |- (monitor);
|
||||
\draw [arrow] (optionB) |- (monitor);
|
||||
\end{tikzpicture}
|
||||
\end{center}
|
||||
|
||||
{\footnotesize \textit{Figure 1: Simplified treatment selection algorithm. See detailed algorithm in references for complete decision pathway.}}
|
||||
|
||||
\section{Monitoring Protocol}
|
||||
|
||||
\subsection{On-Treatment Monitoring}
|
||||
|
||||
\begin{table}[H]
|
||||
\centering
|
||||
\footnotesize
|
||||
\begin{tabular}{lccl}
|
||||
\toprule
|
||||
\textbf{Assessment} & \textbf{Baseline} & \textbf{Frequency} & \textbf{Rationale} \\
|
||||
\midrule
|
||||
CBC with differential & $\checkmark$ & Before each cycle & Myelosuppression \\
|
||||
Comprehensive metabolic panel & $\checkmark$ & Before each cycle & Organ function \\
|
||||
[Specific biomarker] & $\checkmark$ & Every X cycles & [Reason] \\
|
||||
Imaging (CT chest/abd/pelvis) & $\checkmark$ & Every X weeks & Response assessment \\
|
||||
ECOG performance status & $\checkmark$ & Every visit & Functional status \\
|
||||
Toxicity assessment (CTCAE) & - & Every visit & Safety monitoring \\
|
||||
\bottomrule
|
||||
\end{tabular}
|
||||
\caption{Recommended monitoring schedule}
|
||||
\end{table}
|
||||
|
||||
\subsection{Dose Modification Guidelines}
|
||||
|
||||
\textbf{Hematologic Toxicity}:
|
||||
\begin{itemize}
|
||||
\item \textbf{ANC <1.0 or Platelets <75k}: Delay treatment, recheck weekly, dose reduce 20\% when recovered
|
||||
\item \textbf{ANC <0.5 or Platelets <50k}: Hold treatment, G-CSF support, dose reduce 25-40\%
|
||||
\item \textbf{Febrile neutropenia}: Hold, hospitalize, antibiotics, dose reduce 25\% when recovered
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Non-Hematologic Toxicity}:
|
||||
\begin{itemize}
|
||||
\item \textbf{Grade 2}: Continue with supportive care, consider dose modification if persistent
|
||||
\item \textbf{Grade 3}: Hold until $\leq$Grade 1, resume at reduced dose (20-25\% reduction)
|
||||
\item \textbf{Grade 4}: Discontinue treatment or hold pending recovery (case-by-case)
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Specific Toxicity Management}:
|
||||
\begin{itemize}
|
||||
\item \textbf{[Specific AE]}: [Management approach - e.g., Diarrhea Grade 3: Hold treatment, loperamide, hydration, resume at reduced dose when $\leq$Grade 1]
|
||||
\item \textbf{[Immune-related AE]}: [Management - e.g., Pneumonitis Grade 2+: Hold immunotherapy, corticosteroids, pulmonology consultation]
|
||||
\end{itemize}
|
||||
|
||||
\section{Treatment Recommendations by Clinical Scenario}
|
||||
|
||||
\subsection{Scenario 1: [Specific Clinical Situation]}
|
||||
|
||||
\begin{tcolorbox}[enhanced,colback=stronggreen!10,colframe=stronggreen,
|
||||
title={\textbf{RECOMMENDATION} \hfill \textbf{GRADE: 1A}},
|
||||
fonttitle=\bfseries\small,coltitle=black]
|
||||
{\small
|
||||
\textbf{We recommend} [specific intervention] for [patient population].
|
||||
|
||||
\textbf{Evidence}:
|
||||
\begin{itemize}
|
||||
\item [Primary supporting evidence with results]
|
||||
\item [Guideline concordance - NCCN, ASCO, ESMO]
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Benefits}: [Quantified improvements - e.g., 8.7-month PFS benefit, HR 0.46]
|
||||
|
||||
\textbf{Harms}: [Quantified risks - e.g., 15\% grade 3-4 immune-related AEs]
|
||||
|
||||
\textbf{Balance}: Benefits clearly outweigh harms for most patients
|
||||
}
|
||||
\end{tcolorbox}
|
||||
|
||||
\subsection{Scenario 2: [Alternative Clinical Situation]}
|
||||
|
||||
\begin{tcolorbox}[enhanced,colback=conditionalyellow!10,colframe=conditionalyellow,
|
||||
title={\textbf{RECOMMENDATION} \hfill \textbf{GRADE: 2B}},
|
||||
fonttitle=\bfseries\small,coltitle=black]
|
||||
{\small
|
||||
\textbf{We suggest} [intervention] for [patient population] who value [specific outcome].
|
||||
|
||||
\textbf{Evidence}: [Moderate-quality evidence summary]
|
||||
|
||||
\textbf{Trade-offs}:
|
||||
\begin{itemize}
|
||||
\item \textbf{Advantages}: [e.g., Oral administration, less frequent monitoring]
|
||||
\item \textbf{Disadvantages}: [e.g., Lower response rate, more out-of-pocket cost]
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Patient Values}: Substantial variability in how patients value outcomes; shared decision-making essential
|
||||
}
|
||||
\end{tcolorbox}
|
||||
|
||||
\section{Alternative Approaches}
|
||||
|
||||
\subsection{Non-Recommended Options}
|
||||
|
||||
\begin{tcolorbox}[enhanced,colback=warningred!10,colframe=warningred,
|
||||
title={\textbf{NOT RECOMMENDED}},
|
||||
fonttitle=\bfseries\small,coltitle=white,colbacktitle=warningred]
|
||||
{\small
|
||||
\textbf{[Intervention X]} is \textbf{not recommended} for [population].
|
||||
|
||||
\textbf{Reason}: [Evidence of harm, lack of benefit, or superior alternatives available]
|
||||
|
||||
\textbf{Evidence}: [Supporting data showing no benefit or harm]
|
||||
}
|
||||
\end{tcolorbox}
|
||||
|
||||
\section{Supportive Care}
|
||||
|
||||
\subsection{Symptom Management}
|
||||
|
||||
\begin{itemize}
|
||||
\item \textbf{Pain Control}: [Analgesic recommendations, WHO ladder]
|
||||
\item \textbf{Nausea Prevention}: [Antiemetics - e.g., 5-HT3 antagonists, NK1 antagonists for highly emetogenic]
|
||||
\item \textbf{Bone Health}: [e.g., Bisphosphonates or denosumab if bone metastases]
|
||||
\item \textbf{Nutritional Support}: [Consult if weight loss >5\%, cachexia management]
|
||||
\item \textbf{Psychosocial Support}: [Depression screening, support groups, palliative care early integration]
|
||||
\end{itemize}
|
||||
|
||||
\subsection{Growth Factor Support}
|
||||
|
||||
\textbf{G-CSF Prophylaxis}:
|
||||
\begin{itemize}
|
||||
\item \textbf{Primary prophylaxis}: If febrile neutropenia risk $\geq$20\%
|
||||
\item \textbf{Secondary prophylaxis}: After prior febrile neutropenia episode
|
||||
\item Agent: [Pegfilgrastim 6 mg SC day 2 or filgrastim 5 mcg/kg SC daily days 3-10]
|
||||
\end{itemize}
|
||||
|
||||
\section{Follow-Up and Surveillance}
|
||||
|
||||
\subsection{During Active Treatment}
|
||||
|
||||
[Schedule outlined in Monitoring Protocol section above]
|
||||
|
||||
\subsection{Post-Treatment Surveillance}
|
||||
|
||||
\begin{table}[H]
|
||||
\centering
|
||||
\footnotesize
|
||||
\begin{tabular}{lccc}
|
||||
\toprule
|
||||
\textbf{Time Period} & \textbf{Imaging} & \textbf{Labs} & \textbf{Clinical Visits} \\
|
||||
\midrule
|
||||
Year 1 & Every 3 months & Every 3 months & Every 3 months \\
|
||||
Year 2 & Every 3-4 months & Every 3-4 months & Every 3-4 months \\
|
||||
Years 3-5 & Every 6 months & Every 6 months & Every 6 months \\
|
||||
Year 5+ & Annually & Annually & Annually \\
|
||||
\bottomrule
|
||||
\end{tabular}
|
||||
\caption{Post-treatment surveillance schedule (adjust based on risk of recurrence)}
|
||||
\end{table}
|
||||
|
||||
\section{Clinical Trial Opportunities}
|
||||
|
||||
\textbf{When to Consider Clinical Trials}:
|
||||
\begin{itemize}
|
||||
\item After progression on standard therapies
|
||||
\item High-risk disease with poor prognosis on standard therapy
|
||||
\item Novel biomarker potentially predictive of response
|
||||
\item Patient preference for investigational approach
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Resources}:
|
||||
\begin{itemize}
|
||||
\item ClinicalTrials.gov search: [Specific keywords]
|
||||
\item [Institution] clinical trials office: [Contact information]
|
||||
\end{itemize}
|
||||
|
||||
\section{Shared Decision-Making}
|
||||
|
||||
\subsection{Key Discussion Points}
|
||||
|
||||
\textbf{Goals of Care}:
|
||||
\begin{itemize}
|
||||
\item Curative intent vs prolonged disease control vs palliation
|
||||
\item Quality of life vs quantity of life trade-offs
|
||||
\item Functional independence goals
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Treatment Options Counseling}:
|
||||
\begin{itemize}
|
||||
\item Expected benefits (median survival, response rates)
|
||||
\item Potential harms (toxicity profile, quality of life impact)
|
||||
\item Treatment schedule and logistics (frequency of visits, IV vs oral)
|
||||
\item Financial considerations (out-of-pocket costs, time off work)
|
||||
\end{itemize}
|
||||
|
||||
\textbf{Decision Aids}:
|
||||
\begin{itemize}
|
||||
\item Number Needed to Treat: [e.g., Treat X patients to prevent 1 progression event]
|
||||
\item Survival benefit visualization: [X-month improvement in median survival]
|
||||
\end{itemize}
|
||||
|
||||
\section{References}
|
||||
|
||||
\begin{enumerate}
|
||||
\item [Primary clinical trial reference]
|
||||
\item [Secondary supporting trial]
|
||||
\item [NCCN Guidelines, version]
|
||||
\item [ASCO/ESMO Guideline reference]
|
||||
\item [Meta-analysis or systematic review if applicable]
|
||||
\item [Biomarker validation reference]
|
||||
\end{enumerate}
|
||||
|
||||
\vspace{10pt}
|
||||
|
||||
\hrule
|
||||
\vspace{4pt}
|
||||
{\footnotesize
|
||||
\textbf{Guideline Development Committee}:\\
|
||||
[Names and titles of committee members, affiliations]
|
||||
|
||||
\textbf{Evidence Review Date}: [Date]\\
|
||||
\textbf{Guideline Effective Date}: [Date]\\
|
||||
\textbf{Next Scheduled Review}: [Date] (or earlier if practice-changing evidence published)
|
||||
|
||||
\textbf{Conflicts of Interest}: [None / See disclosure statements]
|
||||
|
||||
\textbf{Methodology}: GRADE framework for evidence evaluation and recommendation development. Systematic literature review conducted [date range]. Guidelines concordance checked with NCCN, ASCO, ESMO current versions.
|
||||
|
||||
\textbf{For Questions}: Contact [Name], [Title] at [Email/Phone]
|
||||
}
|
||||
|
||||
\end{document}
|
||||
|
||||
@@ -0,0 +1,719 @@
|
||||
# Biomarker Classification and Interpretation Guide
|
||||
|
||||
## Overview
|
||||
|
||||
Biomarkers are measurable indicators of biological state or condition. In clinical decision support, biomarkers guide diagnosis, prognosis, treatment selection, and monitoring. This guide covers genomic, proteomic, and molecular biomarkers with emphasis on clinical actionability.
|
||||
|
||||
## Biomarker Categories
|
||||
|
||||
### Prognostic Biomarkers
|
||||
|
||||
**Definition**: Predict clinical outcome (survival, recurrence) regardless of treatment received
|
||||
|
||||
**Examples by Disease**
|
||||
|
||||
**Cancer**
|
||||
- **Ki-67 index**: High proliferation (>20%) predicts worse outcome in breast cancer
|
||||
- **TP53 mutation**: Poor prognosis across many cancer types
|
||||
- **Tumor stage/grade**: TNM staging, histologic grade
|
||||
- **LDH elevation**: Poor prognosis in melanoma, lymphoma
|
||||
- **AFP elevation**: Poor prognosis in hepatocellular carcinoma
|
||||
|
||||
**Cardiovascular**
|
||||
- **NT-proBNP/BNP**: Elevated levels predict mortality in heart failure
|
||||
- **Troponin**: Predicts adverse events in ACS
|
||||
- **CRP**: Inflammation marker, predicts cardiovascular events
|
||||
|
||||
**Infectious Disease**
|
||||
- **HIV viral load**: Predicts disease progression if untreated
|
||||
- **HCV genotype**: Predicts treatment duration needed
|
||||
|
||||
**Application**: Risk stratification, treatment intensity selection, clinical trial enrollment
|
||||
|
||||
### Predictive Biomarkers
|
||||
|
||||
**Definition**: Identify patients likely to benefit (or not benefit) from specific therapy
|
||||
|
||||
**Positive Predictive Biomarkers (Treatment Benefit)**
|
||||
|
||||
**Oncology - Targeted Therapy**
|
||||
- **EGFR exon 19 del/L858R → EGFR TKIs**: Response rate 60-70%, PFS 10-14 months
|
||||
- **ALK rearrangement → ALK inhibitors**: ORR 70-90%, PFS 25-34 months
|
||||
- **HER2 amplification → Trastuzumab**: Benefit only in HER2+ (IHC 3+ or FISH+)
|
||||
- **BRAF V600E → BRAF inhibitors**: ORR 50%, PFS 6-7 months (melanoma)
|
||||
- **PD-L1 ≥50% → Pembrolizumab**: ORR 45%, PFS 10 months vs 6 months (chemo)
|
||||
|
||||
**Oncology - Immunotherapy**
|
||||
- **MSI-H/dMMR → Anti-PD-1**: ORR 40-60% across tumor types
|
||||
- **TMB-high → Immunotherapy**: Investigational, some benefit signals
|
||||
- **PD-L1 expression → Anti-PD-1/PD-L1**: Higher expression correlates with better response
|
||||
|
||||
**Hematology**
|
||||
- **BCR-ABL → Imatinib (CML)**: Complete cytogenetic response 80%
|
||||
- **CD20+ → Rituximab (lymphoma)**: Benefit only if CD20-expressing cells
|
||||
- **CD33+ → Gemtuzumab ozogamicin (AML)**: Benefit in CD33+ subset
|
||||
|
||||
**Negative Predictive Biomarkers (Resistance/No Benefit)**
|
||||
- **KRAS mutation → Anti-EGFR mAbs (CRC)**: No benefit, contraindicated
|
||||
- **EGFR T790M → 1st/2nd-gen TKIs**: Resistance mechanism, use osimertinib
|
||||
- **RAS/RAF wild-type required → BRAF inhibitors (melanoma)**: Paradoxical MAPK activation
|
||||
|
||||
### Diagnostic Biomarkers
|
||||
|
||||
**Definition**: Detect or confirm presence of disease
|
||||
|
||||
**Infectious Disease**
|
||||
- **PCR for pathogen DNA/RNA**: SARS-CoV-2, HIV, HCV viral load
|
||||
- **Antibody titers**: IgM (acute), IgG (prior exposure/immunity)
|
||||
- **Antigen tests**: Rapid detection (strep, flu, COVID)
|
||||
|
||||
**Autoimmune**
|
||||
- **ANA**: Screen for lupus, connective tissue disease
|
||||
- **Anti-CCP**: Specific for rheumatoid arthritis
|
||||
- **Anti-dsDNA**: Lupus, correlates with disease activity
|
||||
- **ANCA**: Vasculitis (c-ANCA for GPA, p-ANCA for MPA)
|
||||
|
||||
**Cancer**
|
||||
- **PSA**: Prostate cancer screening/monitoring
|
||||
- **CA 19-9**: Pancreatic cancer, biliary obstruction
|
||||
- **CEA**: Colorectal cancer monitoring
|
||||
- **AFP**: Hepatocellular carcinoma, germ cell tumors
|
||||
|
||||
### Pharmacodynamic Biomarkers
|
||||
|
||||
**Definition**: Assess treatment response or mechanism of action
|
||||
|
||||
**Examples**
|
||||
- **HbA1c**: Glycemic control in diabetes (target <7% typically)
|
||||
- **LDL cholesterol**: Statin efficacy (target <70 mg/dL in high-risk)
|
||||
- **Blood pressure**: Antihypertensive efficacy (target <130/80 mmHg)
|
||||
- **Viral load suppression**: Antiretroviral efficacy (target <20 copies/mL)
|
||||
- **INR**: Warfarin anticoagulation monitoring (target 2-3 for most indications)
|
||||
|
||||
## Genomic Biomarkers
|
||||
|
||||
### Mutation Analysis
|
||||
|
||||
**Driver Mutations (Oncogenic)**
|
||||
- **Activating mutations**: Constitutive pathway activation (BRAF V600E, EGFR L858R)
|
||||
- **Inactivating mutations**: Tumor suppressor loss (TP53, PTEN)
|
||||
- **Hotspot mutations**: Recurrent positions (KRAS G12/G13, PIK3CA H1047R)
|
||||
- **Variant allele frequency (VAF)**: Clonality (VAF ≈50% clonal, <10% subclonal)
|
||||
|
||||
**Resistance Mutations**
|
||||
- **EGFR T790M**: Resistance to 1st/2nd-gen TKIs (40-60% of cases)
|
||||
- **ALK G1202R, I1171N**: Resistance to early ALK inhibitors
|
||||
- **ESR1 mutations**: Resistance to aromatase inhibitors (breast cancer)
|
||||
- **RAS mutations**: Acquired resistance to anti-EGFR therapy (CRC)
|
||||
|
||||
**Mutation Detection Methods**
|
||||
- **Tissue NGS**: Comprehensive genomic profiling, 300-500 genes
|
||||
- **Liquid biopsy**: ctDNA analysis, non-invasive, serial monitoring
|
||||
- **PCR-based assays**: Targeted hotspot detection, FDA-approved companion diagnostics
|
||||
- **Allele-specific PCR**: High sensitivity for known mutations (cobas EGFR test)
|
||||
|
||||
### Copy Number Variations (CNV)
|
||||
|
||||
**Amplifications**
|
||||
- **HER2 (ERBB2)**: Breast, gastric cancer → trastuzumab, pertuzumab
|
||||
- Testing: IHC (0, 1+, 2+, 3+) → FISH if 2+ (HER2/CEP17 ratio ≥2.0)
|
||||
- **MET amplification**: NSCLC resistance mechanism → crizotinib, capmatinib
|
||||
- Cut-point: Gene copy number ≥5, GCN/CEP7 ratio ≥2.0
|
||||
- **EGFR amplification**: Glioblastoma, some NSCLC
|
||||
- **FGFR2 amplification**: Gastric cancer → investigational FGFR inhibitors
|
||||
|
||||
**Deletions**
|
||||
- **PTEN loss**: Common in many cancers, predicts PI3K pathway activation
|
||||
- **RB1 loss**: Small cell transformation, poor prognosis
|
||||
- **CDKN2A/B deletion**: Cell cycle dysregulation
|
||||
- **Homozygous deletion**: Complete loss of both alleles (more significant)
|
||||
|
||||
**Detection Methods**
|
||||
- **FISH (Fluorescence In Situ Hybridization)**: HER2, ALK rearrangements
|
||||
- **NGS copy number calling**: Depth of coverage analysis
|
||||
- **SNP array**: Genome-wide CNV detection
|
||||
- **ddPCR**: Quantitative copy number measurement
|
||||
|
||||
### Gene Fusions and Rearrangements
|
||||
|
||||
**Oncogenic Fusions**
|
||||
- **ALK fusions** (NSCLC): EML4-ALK most common (60%), 20+ partners
|
||||
- Detection: IHC (D5F3 antibody), FISH (break-apart probe), NGS/RNA-seq
|
||||
- **ROS1 fusions** (NSCLC, glioblastoma): CD74-ROS1, SLC34A2-ROS1, others
|
||||
- **RET fusions** (NSCLC, thyroid): KIF5B-RET, CCDC6-RET
|
||||
- **NTRK fusions** (many tumor types, rare): ETV6-NTRK3, others
|
||||
- Pan-cancer: Larotrectinib, entrectinib approved across tumor types
|
||||
- **BCR-ABL** (CML, ALL): t(9;22), Philadelphia chromosome
|
||||
|
||||
**Fusion Partner Considerations**
|
||||
- Partner influences drug sensitivity (EML4-ALK variant 3 more sensitive)
|
||||
- 5' vs 3' fusion affects detection methods
|
||||
- Intron breakpoints vary (RNA-seq more comprehensive than DNA panels)
|
||||
|
||||
**Detection Methods**
|
||||
- **FISH break-apart probes**: ALK, ROS1, RET
|
||||
- **IHC**: ALK protein overexpression (screening), ROS1
|
||||
- **RT-PCR**: Targeted fusion detection
|
||||
- **RNA-seq**: Comprehensive fusion detection, identifies novel partners
|
||||
|
||||
### Tumor Mutational Burden (TMB)
|
||||
|
||||
**Definition**: Number of somatic mutations per megabase of DNA
|
||||
|
||||
**Classification**
|
||||
- **TMB-high**: ≥10 mutations/Mb (some definitions ≥20 mut/Mb)
|
||||
- **TMB-intermediate**: 6-9 mutations/Mb
|
||||
- **TMB-low**: <6 mutations/Mb
|
||||
|
||||
**Clinical Application**
|
||||
- **Predictive for immunotherapy**: Higher TMB → more neoantigens → better immune response
|
||||
- **FDA approval**: Pembrolizumab for TMB-H (≥10 mut/Mb) solid tumors (2020)
|
||||
- **Limitations**: Not validated in all tumor types, assay variability
|
||||
|
||||
**Tumor Types with Typically High TMB**
|
||||
- Melanoma (median 10-15 mut/Mb)
|
||||
- NSCLC (especially smoking-associated, 8-12 mut/Mb)
|
||||
- Urothelial carcinoma (8-10 mut/Mb)
|
||||
- Microsatellite instable tumors (30-50 mut/Mb)
|
||||
|
||||
### Microsatellite Instability (MSI) and Mismatch Repair (MMR)
|
||||
|
||||
**Classification**
|
||||
- **MSI-high (MSI-H)**: Instability at ≥2 of 5 loci or ≥30% of markers
|
||||
- **MSI-low (MSI-L)**: Instability at <2 of 5 loci
|
||||
- **Microsatellite stable (MSS)**: No instability
|
||||
|
||||
**Mismatch Repair Status**
|
||||
- **dMMR (deficient)**: Loss of MLH1, MSH2, MSH6, or PMS2 by IHC
|
||||
- **pMMR (proficient)**: Intact expression of all four MMR proteins
|
||||
|
||||
**Clinical Significance**
|
||||
- **MSI-H/dMMR Tumors**: 3-5% of most solid tumors, 15% of colorectal cancer
|
||||
- **Immunotherapy Sensitivity**: ORR 30-60% to anti-PD-1 therapy
|
||||
- Pembrolizumab FDA-approved for MSI-H/dMMR solid tumors (2017)
|
||||
- Nivolumab ± ipilimumab approved
|
||||
- **Chemotherapy Resistance**: MSI-H CRC does not benefit from 5-FU adjuvant therapy
|
||||
- **Lynch Syndrome**: Germline MMR mutation if MSI-H + young age + family history
|
||||
|
||||
**Testing Algorithm**
|
||||
```
|
||||
Colorectal Cancer (all newly diagnosed):
|
||||
1. IHC for MMR proteins (MLH1, MSH2, MSH6, PMS2)
|
||||
├─ All intact → pMMR (MSS) → Standard chemotherapy if indicated
|
||||
│
|
||||
└─ Loss of one or more → dMMR (likely MSI-H)
|
||||
└─ Reflex MLH1 promoter hypermethylation test
|
||||
├─ Methylated → Sporadic MSI-H, immunotherapy option
|
||||
└─ Unmethylated → Germline testing for Lynch syndrome
|
||||
```
|
||||
|
||||
## Expression Biomarkers
|
||||
|
||||
### Immunohistochemistry (IHC)
|
||||
|
||||
**PD-L1 Expression (Immune Checkpoint)**
|
||||
- **Assays**: 22C3 (FDA), 28-8, SP263, SP142 (some differences in scoring)
|
||||
- **Scoring**: Tumor Proportion Score (TPS) = % tumor cells with membrane staining
|
||||
- TPS <1%: Low/negative
|
||||
- TPS 1-49%: Intermediate
|
||||
- TPS ≥50%: High
|
||||
- **Combined Positive Score (CPS)**: (PD-L1+ tumor + immune cells) / total tumor cells × 100
|
||||
- Used for some indications (e.g., CPS ≥10 for pembrolizumab in HNSCC)
|
||||
|
||||
**Hormone Receptors (Breast Cancer)**
|
||||
- **ER/PR Positivity**: ≥1% nuclear staining by IHC (ASCO/CAP guidelines)
|
||||
- Allred Score 0-8 (proportion + intensity) - historical
|
||||
- H-score 0-300 (percentage at each intensity) - quantitative
|
||||
- **Clinical Cut-Points**:
|
||||
- ER ≥1%: Endocrine therapy indicated
|
||||
- ER 1-10%: "Low positive," may have lower benefit
|
||||
- PR loss with ER+: Possible endocrine resistance
|
||||
|
||||
**HER2 Testing (Breast/Gastric Cancer)**
|
||||
```
|
||||
IHC Initial Test:
|
||||
├─ 0 or 1+: HER2-negative (no further testing)
|
||||
│
|
||||
├─ 2+: Equivocal → Reflex FISH testing
|
||||
│ ├─ FISH+ (HER2/CEP17 ratio ≥2.0 OR HER2 copies ≥6/cell) → HER2-positive
|
||||
│ └─ FISH- → HER2-negative
|
||||
│
|
||||
└─ 3+: HER2-positive (no FISH needed)
|
||||
└─ Uniform intense complete membrane staining in >10% of tumor cells
|
||||
|
||||
HER2-positive: Trastuzumab-based therapy indicated
|
||||
HER2-low (IHC 1+ or 2+/FISH-): Trastuzumab deruxtecan eligibility (2022)
|
||||
```
|
||||
|
||||
### RNA Expression Analysis
|
||||
|
||||
**Gene Expression Signatures (Breast Cancer)**
|
||||
|
||||
**Oncotype DX (21-gene assay)**
|
||||
- **Recurrence Score (RS)**: 0-100
|
||||
- RS <26: Low risk → Endocrine therapy alone (most patients)
|
||||
- RS 26-100: High risk → Chemotherapy + endocrine therapy
|
||||
- **Population**: ER+/HER2-, node-negative or 1-3 positive nodes
|
||||
- **Evidence**: TAILORx trial (N=10,273) validated RS <26 can omit chemo
|
||||
|
||||
**MammaPrint (70-gene assay)**
|
||||
- **Result**: High risk vs Low risk (binary)
|
||||
- **Population**: Early-stage breast cancer, ER+/HER2-
|
||||
- **Evidence**: MINDACT trial validated low-risk can omit chemo
|
||||
|
||||
**Prosigna (PAM50)**
|
||||
- **Result**: Risk of Recurrence (ROR) score + intrinsic subtype
|
||||
- **Subtypes**: Luminal A, Luminal B, HER2-enriched, Basal-like
|
||||
- **Application**: Post-menopausal, ER+, node-negative or 1-3 nodes
|
||||
|
||||
**RNA-Seq for Fusion Detection**
|
||||
- **Advantage**: Detects novel fusion partners, quantifies expression
|
||||
- **Application**: NTRK fusions (rare, many partners), RET fusions
|
||||
- **Limitation**: Requires fresh/frozen tissue or good-quality FFPE RNA
|
||||
|
||||
## Molecular Subtypes
|
||||
|
||||
### Glioblastoma (GBM) Molecular Classification
|
||||
|
||||
**Verhaak 2010 Classification (4 subtypes)**
|
||||
|
||||
**Proneural Subtype**
|
||||
- **Characteristics**: PDGFRA amplification, IDH1 mutations (secondary GBM), TP53 mutations
|
||||
- **Age**: Younger patients typically
|
||||
- **Prognosis**: Better prognosis (median OS 15-18 months)
|
||||
- **Treatment**: May benefit from bevacizumab less than other subtypes
|
||||
|
||||
**Neural Subtype**
|
||||
- **Characteristics**: Neuron markers (NEFL, GABRA1, SYT1, SLC12A5)
|
||||
- **Controversy**: May represent normal brain contamination
|
||||
- **Prognosis**: Intermediate
|
||||
- **Treatment**: Standard temozolomide-based therapy
|
||||
|
||||
**Classical Subtype**
|
||||
- **Characteristics**: EGFR amplification (97%), chromosome 7 gain, chromosome 10 loss
|
||||
- **Association**: Lacks TP53, PDGFRA, NF1 mutations
|
||||
- **Prognosis**: Intermediate
|
||||
- **Treatment**: May benefit from EGFR inhibitors (investigational)
|
||||
|
||||
**Mesenchymal Subtype**
|
||||
- **Characteristics**: NF1 mutations/deletions, high expression of mesenchymal markers (CHI3L1/YKL-40)
|
||||
- **Immune Features**: Higher macrophage/microglia infiltration
|
||||
- **Subgroup**: Mesenchymal-immune-active (high immune signature)
|
||||
- **Prognosis**: Poor prognosis (median OS 12-13 months)
|
||||
- **Treatment**: May respond better to anti-angiogenic therapy, immunotherapy investigational
|
||||
|
||||
**Clinical Application**
|
||||
```
|
||||
GBM Molecular Subtyping Report:
|
||||
|
||||
Patient Cohort: Mesenchymal-Immune-Active Subtype (n=15)
|
||||
|
||||
Molecular Features:
|
||||
- NF1 alterations: 73% (11/15)
|
||||
- High YKL-40 expression: 100% (15/15)
|
||||
- Immune gene signature: Elevated (median z-score +2.3)
|
||||
- CD163+ macrophages: High density (median 180/mm²)
|
||||
|
||||
Treatment Implications:
|
||||
- Standard therapy: Temozolomide-based (Stupp protocol)
|
||||
- Consider: Bevacizumab for recurrent disease (may have enhanced benefit)
|
||||
- Clinical trial: Immune checkpoint inhibitors ± anti-angiogenic therapy
|
||||
- Prognosis: Median OS 12-14 months (worse than proneural)
|
||||
|
||||
Recommendation:
|
||||
Enroll in combination immunotherapy trial if eligible, otherwise standard therapy
|
||||
with early consideration of bevacizumab at progression.
|
||||
```
|
||||
|
||||
### Breast Cancer Intrinsic Subtypes
|
||||
|
||||
**PAM50-Based Classification**
|
||||
|
||||
**Luminal A**
|
||||
- **Characteristics**: ER+, HER2-, low proliferation (Ki-67 <20%)
|
||||
- **Gene signature**: High ER-related genes, low proliferation genes
|
||||
- **Prognosis**: Best prognosis, low recurrence risk
|
||||
- **Treatment**: Endocrine therapy alone usually sufficient
|
||||
- **Chemotherapy**: Rarely needed unless high-risk features
|
||||
|
||||
**Luminal B**
|
||||
- **Characteristics**: ER+, HER2- or HER2+, high proliferation (Ki-67 ≥20%)
|
||||
- **Subtypes**: Luminal B (HER2-) and Luminal B (HER2+)
|
||||
- **Prognosis**: Intermediate prognosis
|
||||
- **Treatment**: Chemotherapy + endocrine therapy; add trastuzumab if HER2+
|
||||
|
||||
**HER2-Enriched**
|
||||
- **Characteristics**: HER2+, ER-, PR-
|
||||
- **Gene signature**: High HER2 and proliferation genes, low ER genes
|
||||
- **Prognosis**: Poor if untreated, good with HER2-targeted therapy
|
||||
- **Treatment**: Chemotherapy + trastuzumab + pertuzumab
|
||||
|
||||
**Basal-Like**
|
||||
- **Characteristics**: ER-, PR-, HER2- (triple-negative), high proliferation
|
||||
- **Gene signature**: Basal cytokeratins (CK5/6, CK17), EGFR
|
||||
- **Overlap**: 80% concordance with TNBC, but not identical
|
||||
- **Prognosis**: Aggressive, high early recurrence risk
|
||||
- **Treatment**: Chemotherapy (platinum, anthracycline), PARP inhibitors if BRCA-mutated
|
||||
- **Immunotherapy**: PD-L1+ may benefit from pembrolizumab + chemotherapy
|
||||
|
||||
### Colorectal Cancer Consensus Molecular Subtypes (CMS)
|
||||
|
||||
**CMS1 (14%): MSI Immune**
|
||||
- **Features**: MSI-high, BRAF mutations, strong immune activation
|
||||
- **Prognosis**: Poor survival after relapse despite immune infiltration
|
||||
- **Treatment**: Immunotherapy highly effective, 5-FU chemotherapy ineffective
|
||||
|
||||
**CMS2 (37%): Canonical**
|
||||
- **Features**: Epithelial, marked WNT and MYC activation
|
||||
- **Prognosis**: Better survival
|
||||
- **Treatment**: Benefits from adjuvant chemotherapy
|
||||
|
||||
**CMS3 (13%): Metabolic**
|
||||
- **Features**: Metabolic dysregulation, KRAS mutations
|
||||
- **Prognosis**: Intermediate survival
|
||||
- **Treatment**: May benefit from targeted metabolic therapies (investigational)
|
||||
|
||||
**CMS4 (23%): Mesenchymal**
|
||||
- **Features**: Stromal infiltration, TGF-β activation, angiogenesis
|
||||
- **Prognosis**: Worst survival, often diagnosed at advanced stage
|
||||
- **Treatment**: May benefit from anti-angiogenic therapy (bevacizumab)
|
||||
|
||||
## Companion Diagnostics
|
||||
|
||||
### FDA-Approved Biomarker-Drug Pairs
|
||||
|
||||
**Required Testing (Label Indication)**
|
||||
```
|
||||
Biomarker Drug(s) Indication Assay
|
||||
EGFR exon 19 del/L858R Osimertinib NSCLC cobas EGFR v2, NGS
|
||||
ALK rearrangement Alectinib, brigatinib NSCLC Vysis ALK FISH, IHC (D5F3)
|
||||
BRAF V600E Vemurafenib, dabrafenib Melanoma, NSCLC THxID BRAF, cobas BRAF
|
||||
HER2 amplification Trastuzumab, pertuzumab Breast, gastric HercepTest IHC, FISH
|
||||
ROS1 rearrangement Crizotinib, entrectinib NSCLC FISH, NGS
|
||||
PD-L1 ≥50% TPS Pembrolizumab (mono) NSCLC first-line 22C3 pharmDx
|
||||
MSI-H/dMMR Pembrolizumab Any solid tumor IHC (MMR), PCR (MSI)
|
||||
NTRK fusion Larotrectinib, entrectinib Pan-cancer FoundationOne CDx
|
||||
BRCA1/2 mutations Olaparib, talazoparib Breast, ovarian, prostate BRACAnalysis CDx
|
||||
```
|
||||
|
||||
### Complementary Diagnostics (Informative, Not Required)
|
||||
|
||||
- **PD-L1 1-49%**: Informs combination vs monotherapy choice
|
||||
- **TMB-high**: May predict immunotherapy benefit (not FDA-approved indication)
|
||||
- **STK11/KEAP1 mutations**: Associated with immunotherapy resistance
|
||||
- **Homologous recombination deficiency (HRD)**: Predicts PARP inhibitor benefit
|
||||
|
||||
## Clinical Actionability Frameworks
|
||||
|
||||
### OncoKB Levels of Evidence (Memorial Sloan Kettering)
|
||||
|
||||
**Level 1: FDA-Approved**
|
||||
- Biomarker-drug pair with FDA approval in specific tumor type
|
||||
- Example: EGFR L858R → osimertinib in NSCLC
|
||||
|
||||
**Level 2: Standard Care Off-Label**
|
||||
- Biomarker-drug in professional guidelines for specific tumor type (not FDA-approved for biomarker)
|
||||
- Example: BRAF V600E → dabrafenib + trametinib in CRC (NCCN-recommended)
|
||||
|
||||
**Level 3: Clinical Evidence**
|
||||
- Clinical trial evidence supporting biomarker-drug association
|
||||
- 3A: Compelling clinical evidence
|
||||
- 3B: Standard care for different tumor type or investigational
|
||||
|
||||
**Level 4: Biological Evidence**
|
||||
- Preclinical evidence only (cell lines, mouse models)
|
||||
- 4: Biological evidence supporting association
|
||||
|
||||
**Level R1-R2: Resistance**
|
||||
- R1: Standard care associated with resistance
|
||||
- R2: Investigational or preclinical resistance evidence
|
||||
|
||||
### CIViC (Clinical Interpretation of Variants in Cancer)
|
||||
|
||||
**Evidence Levels**
|
||||
- **A**: Validated in clinical practice or validated by regulatory association
|
||||
- **B**: Clinical trial or other primary patient data supporting association
|
||||
- **C**: Case study with molecular analysis
|
||||
- **D**: Preclinical evidence (cell culture, animal models)
|
||||
- **E**: Inferential association (literature review, expert opinion)
|
||||
|
||||
**Clinical Significance Tiers**
|
||||
- **Tier I**: Variants with strong clinical significance (predictive, diagnostic, prognostic in professional guidelines)
|
||||
- **Tier II**: Variants with potential clinical significance (clinical trial or case study evidence)
|
||||
- **Tier III**: Variants with uncertain significance
|
||||
- **Tier IV**: Benign or likely benign variants
|
||||
|
||||
## Multi-Biomarker Panels
|
||||
|
||||
### Comprehensive Genomic Profiling (CGP)
|
||||
|
||||
**FoundationOne CDx**
|
||||
- **Genes**: 324 genes (SNVs, indels, CNVs, rearrangements)
|
||||
- **Additional**: TMB, MSI status
|
||||
- **FDA-Approved**: Companion diagnostic for 18+ targeted therapies
|
||||
- **Turnaround**: 10-14 days
|
||||
- **Tissue**: FFPE, 40 unstained slides or tissue block
|
||||
|
||||
**Guardant360 CDx (Liquid Biopsy)**
|
||||
- **Genes**: 74 genes in cell-free DNA (cfDNA)
|
||||
- **Sample**: 2 tubes of blood (20 mL total)
|
||||
- **FDA-Approved**: Companion diagnostic for osimertinib (EGFR), NSCLC
|
||||
- **Application**: Non-invasive, serial monitoring, when tissue unavailable
|
||||
- **Limitation**: Lower sensitivity than tissue (especially for low tumor burden)
|
||||
|
||||
**Tempus xT**
|
||||
- **Genes**: 648 genes (DNA) + whole transcriptome (RNA)
|
||||
- **Advantage**: RNA detects fusions, expression signatures
|
||||
- **Application**: Research and clinical use
|
||||
- **Not FDA-Approved**: Not a companion diagnostic currently
|
||||
|
||||
### Testing Recommendations by Tumor Type
|
||||
|
||||
**NSCLC (NCCN Guidelines)**
|
||||
```
|
||||
Broad molecular profiling for all advanced NSCLC at diagnosis:
|
||||
|
||||
Required (FDA-approved therapies available):
|
||||
✓ EGFR mutations (exons 18, 19, 20, 21)
|
||||
✓ ALK rearrangement
|
||||
✓ ROS1 rearrangement
|
||||
✓ BRAF V600E
|
||||
✓ MET exon 14 skipping
|
||||
✓ RET rearrangements
|
||||
✓ NTRK fusions
|
||||
✓ KRAS G12C
|
||||
✓ PD-L1 IHC
|
||||
|
||||
Recommended (to inform treatment strategy):
|
||||
✓ Comprehensive NGS panel (captures all above + emerging targets)
|
||||
✓ Consider liquid biopsy if tissue insufficient
|
||||
|
||||
At progression on targeted therapy:
|
||||
✓ Repeat tissue biopsy or liquid biopsy for resistance mechanisms
|
||||
✓ Examples: EGFR T790M, ALK resistance mutations, MET amplification
|
||||
```
|
||||
|
||||
**Metastatic Colorectal Cancer**
|
||||
```
|
||||
Required before anti-EGFR therapy (cetuximab, panitumumab):
|
||||
✓ RAS testing (KRAS exons 2, 3, 4; NRAS exons 2, 3, 4)
|
||||
└─ RAS mutation → Do NOT use anti-EGFR therapy (resistance)
|
||||
✓ BRAF V600E
|
||||
└─ If BRAF V600E+ → Consider encorafenib + cetuximab + binimetinib
|
||||
|
||||
Recommended for all metastatic CRC:
|
||||
✓ MSI/MMR testing (immunotherapy indication)
|
||||
✓ HER2 amplification (investigational trastuzumab-based therapy if RAS/BRAF WT)
|
||||
✓ NTRK fusions (rare, <1%, but actionable)
|
||||
|
||||
Left-sided vs Right-sided:
|
||||
- Left-sided (descending, sigmoid, rectum): Better prognosis, anti-EGFR more effective
|
||||
- Right-sided (cecum, ascending): Worse prognosis, anti-EGFR less effective, consider bevacizumab
|
||||
```
|
||||
|
||||
**Melanoma**
|
||||
```
|
||||
All advanced melanoma:
|
||||
✓ BRAF V600 mutation (30-50% of cutaneous melanoma)
|
||||
└─ If BRAF V600E/K → Dabrafenib + trametinib or vemurafenib + cobimetinib
|
||||
✓ NRAS mutation (20-30%)
|
||||
└─ No targeted therapy approved, consider MEK inhibitor trials
|
||||
✓ KIT mutations (mucosal, acral, chronic sun-damaged melanoma)
|
||||
└─ If KIT exon 11 or 13 mutation → Imatinib (off-label)
|
||||
✓ PD-L1 (optional, not required for immunotherapy eligibility)
|
||||
|
||||
Note: Uveal melanoma has different biology (GNAQ, GNA11 mutations)
|
||||
```
|
||||
|
||||
## Biomarker Cut-Points and Thresholds
|
||||
|
||||
### Establishing Clinical Cut-Points
|
||||
|
||||
**Methods for Cut-Point Determination**
|
||||
|
||||
**Data-Driven Approaches**
|
||||
- **Median split**: Simple but arbitrary, may not be optimal
|
||||
- **Tertiles/quartiles**: Categorizes into 3-4 groups
|
||||
- **ROC curve analysis**: Maximizes sensitivity and specificity
|
||||
- **Maximally selected rank statistics**: Finds optimal prognostic cut-point
|
||||
- **Validation required**: Independent cohort confirmation essential
|
||||
|
||||
**Biologically Informed**
|
||||
- **Detection limit**: Assay lower limit of quantification
|
||||
- **Mechanism-based**: Threshold for pathway activation
|
||||
- **Pharmacodynamic**: Threshold for target engagement
|
||||
- **Normal range**: Comparison to healthy individuals
|
||||
|
||||
**Clinically Defined**
|
||||
- **Guideline-recommended**: Established by professional societies
|
||||
- **Regulatory-approved**: FDA-specified threshold for companion diagnostic
|
||||
- **Trial-defined**: Cut-point used in pivotal clinical trial
|
||||
|
||||
**PD-L1 Example**
|
||||
- **Cut-points**: 1%, 5%, 10%, 50% TPS used in different trials
|
||||
- **Context-dependent**: Varies by drug, disease, line of therapy
|
||||
- **≥50%**: Pembrolizumab monotherapy (KEYNOTE-024)
|
||||
- **≥1%**: Atezolizumab combinations, broader population
|
||||
|
||||
### Continuous vs Categorical
|
||||
|
||||
**Continuous Analysis Advantages**
|
||||
- Preserves information (no dichotomization loss)
|
||||
- Statistical power maintained
|
||||
- Can assess dose-response relationship
|
||||
- HR per unit increase or per standard deviation
|
||||
|
||||
**Categorical Analysis Advantages**
|
||||
- Clinically interpretable (high vs low)
|
||||
- Facilitates treatment decisions (binary: use targeted therapy yes/no)
|
||||
- Aligns with regulatory approvals (biomarker-positive = eligible)
|
||||
|
||||
**Best Practice**: Report both continuous and categorical analyses
|
||||
- Cox model with continuous biomarker
|
||||
- Stratified analysis by clinically relevant cut-point
|
||||
- Subgroup analysis to confirm consistency
|
||||
|
||||
## Germline vs Somatic Testing
|
||||
|
||||
### Germline (Inherited) Mutations
|
||||
|
||||
**Indications for Germline Testing**
|
||||
- **Cancer predisposition syndromes**: BRCA1/2, Lynch syndrome (MLH1, MSH2), Li-Fraumeni (TP53)
|
||||
- **Family history**: Multiple affected relatives, young age at diagnosis
|
||||
- **Tumor features**: MSI-H in young patient, triple-negative breast cancer <60 years
|
||||
- **Treatment implications**: PARP inhibitors for BRCA-mutated (germline or somatic)
|
||||
|
||||
**Common Hereditary Cancer Syndromes**
|
||||
- **BRCA1/2**: Breast, ovarian, pancreatic, prostate cancer
|
||||
- Testing: All ovarian cancer, TNBC <60 years, male breast cancer
|
||||
- Treatment: PARP inhibitors (olaparib, talazoparib)
|
||||
- Prevention: Prophylactic mastectomy, oophorectomy (risk-reducing)
|
||||
- **Lynch syndrome (MLH1, MSH2, MSH6, PMS2)**: Colorectal, endometrial, ovarian, gastric
|
||||
- Testing: MSI-H/dMMR tumors, Amsterdam II criteria families
|
||||
- Surveillance: Colonoscopy every 1-2 years starting age 20-25
|
||||
- **Li-Fraumeni (TP53)**: Diverse cancers at young age
|
||||
- **PTEN (Cowden syndrome)**: Breast, thyroid, endometrial cancer
|
||||
|
||||
**Genetic Counseling**
|
||||
- Pre-test counseling: Implications for patient and family
|
||||
- Post-test counseling: Management, surveillance, family testing
|
||||
- Informed consent: Genetic discrimination concerns (GINA protections)
|
||||
|
||||
### Somatic (Tumor-Only) Testing
|
||||
|
||||
**Tumor Tissue Testing**
|
||||
- Detects mutations present in cancer cells only (not inherited)
|
||||
- Most cancer driver mutations are somatic (KRAS, EGFR in lung cancer)
|
||||
- No implications for family members
|
||||
- Guides therapy selection
|
||||
|
||||
**Distinguishing Germline from Somatic**
|
||||
- **Variant allele frequency**: Germline ~50% (heterozygous) or ~100% (homozygous); somatic variable
|
||||
- **Matched normal**: Paired tumor-normal sequencing definitive
|
||||
- **Databases**: Germline variant databases (gnomAD, ClinVar)
|
||||
- **Reflex germline testing**: Trigger testing if pathogenic germline variant suspected
|
||||
|
||||
## Reporting Biomarker Results
|
||||
|
||||
### Structured Report Template
|
||||
|
||||
```
|
||||
MOLECULAR PROFILING REPORT
|
||||
|
||||
Patient: [De-identified ID]
|
||||
Tumor Type: Non-Small Cell Lung Adenocarcinoma
|
||||
Specimen: Lung biopsy (left upper lobe)
|
||||
Testing Date: [Date]
|
||||
Report Date: [Date]
|
||||
|
||||
METHODOLOGY
|
||||
- Assay: FoundationOne CDx (comprehensive genomic profiling)
|
||||
- Specimen Type: Formalin-fixed paraffin-embedded (FFPE)
|
||||
- Tumor Content: 40% (adequate for testing)
|
||||
|
||||
RESULTS SUMMARY
|
||||
Biomarkers Detected: 4
|
||||
- 1 FDA-approved therapy target
|
||||
- 1 prognostic biomarker
|
||||
- 2 variants of uncertain significance
|
||||
|
||||
ACTIONABLE FINDINGS
|
||||
|
||||
Tier 1: FDA-Approved Targeted Therapy Available
|
||||
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
|
||||
EGFR Exon 19 Deletion (p.E746_A750del)
|
||||
Variant Allele Frequency: 42%
|
||||
Clinical Significance: Sensitizing mutation
|
||||
FDA-Approved Therapy: Osimertinib (Tagrisso) 80 mg daily
|
||||
Evidence: FLAURA trial - median PFS 18.9 vs 10.2 months (HR 0.46, p<0.001)
|
||||
Guideline: NCCN Category 1 preferred first-line
|
||||
Recommendation: Strong recommendation for EGFR TKI therapy (GRADE 1A)
|
||||
|
||||
Tier 2: Prognostic Biomarker
|
||||
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
|
||||
TP53 Mutation (p.R273H)
|
||||
Variant Allele Frequency: 85%
|
||||
Clinical Significance: Poor prognostic marker, no targeted therapy
|
||||
Implication: Associated with worse survival, does not impact first-line treatment selection
|
||||
|
||||
BIOMARKERS ASSESSED - NEGATIVE
|
||||
- ALK rearrangement: Not detected
|
||||
- ROS1 rearrangement: Not detected
|
||||
- BRAF V600E: Not detected
|
||||
- MET exon 14 skipping: Not detected
|
||||
- RET rearrangement: Not detected
|
||||
- KRAS mutation: Not detected
|
||||
- PD-L1 IHC: Separate report (TPS 30%)
|
||||
|
||||
TUMOR MUTATIONAL BURDEN: 8 mutations/Mb (Intermediate)
|
||||
- Interpretation: Below threshold for TMB-high designation (≥10 mut/Mb)
|
||||
- Clinical relevance: May still benefit from immunotherapy combinations
|
||||
|
||||
MICROSATELLITE STATUS: Stable (MSS)
|
||||
|
||||
CLINICAL RECOMMENDATIONS
|
||||
|
||||
Primary Recommendation:
|
||||
First-line therapy with osimertinib 80 mg PO daily until progression or unacceptable toxicity.
|
||||
|
||||
Monitoring:
|
||||
- CT imaging every 6 weeks for first 12 weeks, then every 9 weeks
|
||||
- At progression, repeat tissue or liquid biopsy for resistance mechanisms (T790M, C797S, MET amplification)
|
||||
|
||||
Alternative Options:
|
||||
- Clinical trial enrollment for novel EGFR TKI combinations
|
||||
- Erlotinib or afatinib (second-line for osimertinib if used first-line)
|
||||
|
||||
References:
|
||||
1. Soria JC, et al. Osimertinib in Untreated EGFR-Mutated Advanced NSCLC. NEJM 2018.
|
||||
2. NCCN Guidelines for Non-Small Cell Lung Cancer v4.2024.
|
||||
|
||||
Report Prepared By: [Lab Name]
|
||||
Medical Director: [Name, MD, PhD]
|
||||
CLIA #: [Number] | CAP #: [Number]
|
||||
```
|
||||
|
||||
## Quality Assurance
|
||||
|
||||
### Analytical Validation
|
||||
|
||||
- **Sensitivity**: Minimum 5-10% variant allele frequency detection
|
||||
- **Specificity**: <1% false positive rate
|
||||
- **Reproducibility**: >95% concordance between replicates
|
||||
- **Accuracy**: >99% concordance with validated orthogonal method
|
||||
- **Turnaround time**: Median time from sample receipt to report
|
||||
|
||||
### Clinical Validation
|
||||
|
||||
- **Positive Predictive Value**: % biomarker+ patients who respond to therapy
|
||||
- **Negative Predictive Value**: % biomarker- patients who do not respond
|
||||
- **Clinical Utility**: Does testing improve patient outcomes?
|
||||
- **Cost-Effectiveness**: QALY gained vs cost of testing and treatment
|
||||
|
||||
### Proficiency Testing
|
||||
|
||||
- CAP/CLIA proficiency testing for clinical labs
|
||||
- Participate in external quality assurance schemes
|
||||
- Blinded sample exchange with reference laboratories
|
||||
- Document corrective actions for failures
|
||||
|
||||
@@ -0,0 +1,604 @@
|
||||
# Clinical Decision Algorithms Guide
|
||||
|
||||
## Overview
|
||||
|
||||
Clinical decision algorithms provide systematic, step-by-step guidance for diagnosis, treatment selection, and patient management. This guide covers algorithm development, validation, and visual presentation using decision trees and flowcharts.
|
||||
|
||||
## Algorithm Design Principles
|
||||
|
||||
### Key Components
|
||||
|
||||
**Decision Nodes**
|
||||
- **Question/Criteria**: Clear, measurable clinical parameter
|
||||
- **Binary vs Multi-Way**: Yes/no (simple) vs multiple options (complex)
|
||||
- **Objective**: Lab value, imaging finding vs Subjective: Clinical judgment
|
||||
|
||||
**Action Nodes**
|
||||
- **Treatment**: Specific intervention with dosing
|
||||
- **Test**: Additional diagnostic procedure
|
||||
- **Referral**: Specialist consultation, higher level of care
|
||||
- **Observation**: Watchful waiting with defined follow-up
|
||||
|
||||
**Terminal Nodes**
|
||||
- **Outcome**: Final decision point
|
||||
- **Follow-up**: Schedule for reassessment
|
||||
- **Exit criteria**: When to exit algorithm
|
||||
|
||||
### Design Criteria
|
||||
|
||||
**Clarity**
|
||||
- Unambiguous decision points
|
||||
- Mutually exclusive pathways
|
||||
- No circular loops (unless intentional reassessment cycles)
|
||||
- Clear entry and exit points
|
||||
|
||||
**Clinical Validity**
|
||||
- Evidence-based decision criteria
|
||||
- Validated cut-points for biomarkers
|
||||
- Guideline-concordant recommendations
|
||||
- Expert consensus where evidence limited
|
||||
|
||||
**Usability**
|
||||
- Maximum 7 decision points per pathway (cognitive load)
|
||||
- Visual hierarchy (most common path highlighted)
|
||||
- Printable single-page format preferred
|
||||
- Color coding for urgency/safety
|
||||
|
||||
**Completeness**
|
||||
- All possible scenarios covered
|
||||
- Default pathway for edge cases
|
||||
- Safety-net provisions for unusual presentations
|
||||
- Escalation criteria clearly stated
|
||||
|
||||
## Clinical Decision Trees
|
||||
|
||||
### Diagnostic Algorithms
|
||||
|
||||
**Chest Pain Evaluation Algorithm**
|
||||
|
||||
```
|
||||
Entry: Patient with chest pain
|
||||
|
||||
├─ STEMI Criteria? (ST elevation ≥1mm in ≥2 contiguous leads)
|
||||
│ ├─ YES → Activate cath lab, aspirin 325mg, heparin, clopidogrel 600mg
|
||||
│ │ Transfer for primary PCI (goal door-to-balloon <90 minutes)
|
||||
│ └─ NO → Continue evaluation
|
||||
|
||||
├─ High-Risk Features? (Hemodynamic instability, arrhythmia, troponin elevation)
|
||||
│ ├─ YES → Admit CCU, serial troponins, cardiology consultation
|
||||
│ │ Consider early angiography if NSTEMI
|
||||
│ └─ NO → Calculate TIMI or HEART score
|
||||
|
||||
├─ TIMI Score 0-1 or HEART Score 0-3? (Low risk)
|
||||
│ ├─ YES → Observe 6-12 hours, serial troponins, stress test if negative
|
||||
│ │ Discharge if all negative with cardiology follow-up in 72 hours
|
||||
│ └─ NO → TIMI 2-4 or HEART 4-6 (Intermediate risk)
|
||||
|
||||
├─ TIMI Score 2-4 or HEART Score 4-6? (Intermediate risk)
|
||||
│ ├─ YES → Admit telemetry, serial troponins, stress imaging vs CT angiography
|
||||
│ │ Medical management: Aspirin, statin, beta-blocker
|
||||
│ └─ NO → TIMI ≥5 or HEART ≥7 (High risk) → Treat as NSTEMI
|
||||
|
||||
Decision Endpoint: Risk-stratified pathway with 30-day event rate documented
|
||||
```
|
||||
|
||||
**Pulmonary Embolism Diagnostic Algorithm (Wells Criteria)**
|
||||
|
||||
```
|
||||
Entry: Suspected PE
|
||||
|
||||
Step 1: Calculate Wells Score
|
||||
Clinical features points:
|
||||
- Clinical signs of DVT: 3 points
|
||||
- PE more likely than alternative diagnosis: 3 points
|
||||
- Heart rate >100: 1.5 points
|
||||
- Immobilization/surgery in past 4 weeks: 1.5 points
|
||||
- Previous PE/DVT: 1.5 points
|
||||
- Hemoptysis: 1 point
|
||||
- Malignancy: 1 point
|
||||
|
||||
Step 2: Risk Stratify
|
||||
├─ Wells Score ≤4 (PE unlikely)
|
||||
│ └─ D-dimer test
|
||||
│ ├─ D-dimer negative (<500 ng/mL) → PE excluded, consider alternative diagnosis
|
||||
│ └─ D-dimer positive (≥500 ng/mL) → CTPA
|
||||
│
|
||||
└─ Wells Score >4 (PE likely)
|
||||
└─ CTPA (skip D-dimer)
|
||||
|
||||
Step 3: CTPA Results
|
||||
├─ Positive for PE → Risk stratify severity
|
||||
│ ├─ Massive PE (hypotension, shock) → Thrombolytics vs embolectomy
|
||||
│ ├─ Submassive PE (RV strain, troponin+) → Admit ICU, consider thrombolytics
|
||||
│ └─ Low-risk PE → Anticoagulation, consider outpatient management
|
||||
│
|
||||
└─ Negative for PE → PE excluded, investigate alternative diagnosis
|
||||
|
||||
Step 4: Treatment Decision (if PE confirmed)
|
||||
├─ Absolute contraindication to anticoagulation?
|
||||
│ ├─ YES → IVC filter placement, treat underlying condition
|
||||
│ └─ NO → Anticoagulation therapy
|
||||
│
|
||||
├─ Cancer-associated thrombosis?
|
||||
│ ├─ YES → LMWH preferred (edoxaban alternative)
|
||||
│ └─ NO → DOAC preferred (apixaban, rivaroxaban, edoxaban)
|
||||
│
|
||||
└─ Duration: Minimum 3 months, extended if unprovoked or recurrent
|
||||
```
|
||||
|
||||
### Treatment Selection Algorithms
|
||||
|
||||
**NSCLC First-Line Treatment Algorithm**
|
||||
|
||||
```
|
||||
Entry: Advanced/Metastatic NSCLC, adequate PS (ECOG 0-2)
|
||||
|
||||
Step 1: Biomarker Testing Complete?
|
||||
├─ NO → Reflex testing: EGFR, ALK, ROS1, BRAF, PD-L1, consider NGS
|
||||
│ Hold systemic therapy pending results (unless rapidly progressive)
|
||||
└─ YES → Proceed to Step 2
|
||||
|
||||
Step 2: Actionable Genomic Alteration?
|
||||
├─ EGFR exon 19 deletion or L858R → Osimertinib 80mg daily
|
||||
│ └─ Alternative: Erlotinib, gefitinib, afatinib (less preferred)
|
||||
│
|
||||
├─ ALK rearrangement → Alectinib 600mg BID
|
||||
│ └─ Alternatives: Brigatinib, lorlatinib, crizotinib (less preferred)
|
||||
│
|
||||
├─ ROS1 rearrangement → Crizotinib 250mg BID or entrectinib
|
||||
│
|
||||
├─ BRAF V600E → Dabrafenib + trametinib
|
||||
│
|
||||
├─ MET exon 14 skipping → Capmatinib or tepotinib
|
||||
│
|
||||
├─ RET rearrangement → Selpercatinib or pralsetinib
|
||||
│
|
||||
├─ NTRK fusion → Larotrectinib or entrectinib
|
||||
│
|
||||
├─ KRAS G12C → Sotorasib or adagrasib (if no other options)
|
||||
│
|
||||
└─ NO actionable alteration → Proceed to Step 3
|
||||
|
||||
Step 3: PD-L1 Testing Result?
|
||||
├─ PD-L1 ≥50% (TPS)
|
||||
│ ├─ Option 1: Pembrolizumab 200mg Q3W (monotherapy, NCCN Category 1)
|
||||
│ ├─ Option 2: Pembrolizumab + platinum doublet chemotherapy
|
||||
│ └─ Option 3: Atezolizumab + bevacizumab + carboplatin + paclitaxel
|
||||
│
|
||||
├─ PD-L1 1-49% (TPS)
|
||||
│ ├─ Preferred: Pembrolizumab + platinum doublet chemotherapy
|
||||
│ └─ Alternative: Platinum doublet chemotherapy alone
|
||||
│
|
||||
└─ PD-L1 <1% (TPS)
|
||||
├─ Preferred: Pembrolizumab + platinum doublet chemotherapy
|
||||
└─ Alternative: Platinum doublet chemotherapy ± bevacizumab
|
||||
|
||||
Step 4: Platinum Doublet Selection (if applicable)
|
||||
├─ Squamous histology
|
||||
│ └─ Carboplatin AUC 6 + paclitaxel 200 mg/m² Q3W (4 cycles)
|
||||
│ or Carboplatin AUC 5 + nab-paclitaxel 100 mg/m² D1,8,15 Q4W
|
||||
│
|
||||
└─ Non-squamous histology
|
||||
└─ Carboplatin AUC 6 + pemetrexed 500 mg/m² Q3W (4 cycles)
|
||||
Continue pemetrexed maintenance if responding
|
||||
Add bevacizumab 15 mg/kg if eligible (no hemoptysis, brain mets)
|
||||
|
||||
Step 5: Monitoring and Response Assessment
|
||||
- Imaging every 6 weeks for first 12 weeks, then every 9 weeks
|
||||
- Continue until progression or unacceptable toxicity
|
||||
- At progression, proceed to second-line algorithm
|
||||
```
|
||||
|
||||
**Heart Failure Management Algorithm (AHA/ACC Guidelines)**
|
||||
|
||||
```
|
||||
Entry: Heart Failure Diagnosis Confirmed
|
||||
|
||||
Step 1: Determine HF Type
|
||||
├─ HFrEF (EF ≤40%)
|
||||
│ └─ Proceed to Guideline-Directed Medical Therapy (GDMT)
|
||||
│
|
||||
├─ HFpEF (EF ≥50%)
|
||||
│ └─ Treat comorbidities, diuretics for congestion, consider SGLT2i
|
||||
│
|
||||
└─ HFmrEF (EF 41-49%)
|
||||
└─ Consider HFrEF GDMT, evidence less robust
|
||||
|
||||
Step 2: GDMT for HFrEF (All patients unless contraindicated)
|
||||
|
||||
Quadruple Therapy (Class 1 recommendations):
|
||||
|
||||
1. ACE Inhibitor/ARB/ARNI
|
||||
├─ Preferred: Sacubitril-valsartan 49/51mg BID → titrate to 97/103mg BID
|
||||
│ └─ If ACE-I naïve or taking <10mg enalapril equivalent
|
||||
├─ Alternative: ACE-I (enalapril, lisinopril, ramipril) to target dose
|
||||
└─ Alternative: ARB (losartan, valsartan) if ACE-I intolerant
|
||||
|
||||
2. Beta-Blocker (start low, titrate slowly)
|
||||
├─ Bisoprolol 1.25mg daily → 10mg daily target
|
||||
├─ Metoprolol succinate 12.5mg daily → 200mg daily target
|
||||
└─ Carvedilol 3.125mg BID → 25mg BID target (50mg BID if >85kg)
|
||||
|
||||
3. Mineralocorticoid Receptor Antagonist (MRA)
|
||||
├─ Spironolactone 12.5-25mg daily → 50mg daily target
|
||||
└─ Eplerenone 25mg daily → 50mg daily target
|
||||
└─ Contraindications: K >5.0, CrCl <30 mL/min
|
||||
|
||||
4. SGLT2 Inhibitor (regardless of diabetes status)
|
||||
├─ Dapagliflozin 10mg daily
|
||||
└─ Empagliflozin 10mg daily
|
||||
|
||||
Step 3: Additional Therapies Based on Phenotype
|
||||
|
||||
├─ Sinus rhythm + HR ≥70 despite beta-blocker?
|
||||
│ └─ YES: Add ivabradine 5mg BID → 7.5mg BID target
|
||||
│
|
||||
├─ African American + NYHA III-IV?
|
||||
│ └─ YES: Add hydralazine 37.5mg TID + isosorbide dinitrate 20mg TID
|
||||
│ (Target: hydralazine 75mg TID + ISDN 40mg TID)
|
||||
│
|
||||
├─ Atrial fibrillation?
|
||||
│ ├─ Rate control (target <80 bpm at rest, <110 bpm with activity)
|
||||
│ └─ Anticoagulation (DOAC preferred, warfarin if valvular)
|
||||
│
|
||||
└─ Iron deficiency (ferritin <100 or <300 with TSAT <20%)?
|
||||
└─ YES: IV iron supplementation (ferric carboxymaltose)
|
||||
|
||||
Step 4: Device Therapy Evaluation
|
||||
|
||||
├─ EF ≤35%, NYHA II-III, LBBB with QRS ≥150 ms, sinus rhythm?
|
||||
│ └─ YES: Cardiac resynchronization therapy (CRT-D)
|
||||
│
|
||||
├─ EF ≤35%, NYHA II-III, on GDMT ≥3 months?
|
||||
│ └─ YES: ICD for primary prevention
|
||||
│ (if life expectancy >1 year with good functional status)
|
||||
│
|
||||
└─ EF ≤35%, NYHA IV despite GDMT, or advanced HF?
|
||||
└─ Refer to advanced HF specialist
|
||||
├─ LVAD evaluation
|
||||
├─ Heart transplant evaluation
|
||||
└─ Palliative care consultation
|
||||
|
||||
Step 5: Monitoring and Titration
|
||||
|
||||
Weekly to biweekly visits during titration:
|
||||
- Blood pressure (target SBP ≥90 mmHg)
|
||||
- Heart rate (target 50-60 bpm)
|
||||
- Potassium (target 4.0-5.0 mEq/L, hold MRA if >5.5)
|
||||
- Creatinine (expect 10-20% increase, acceptable if <30% and stable)
|
||||
- Symptoms and congestion status (daily weights, NYHA class)
|
||||
|
||||
Stable on GDMT:
|
||||
- Visits every 3-6 months
|
||||
- Echocardiogram at 3-6 months after GDMT optimization, then annually
|
||||
- NT-proBNP or BNP trending (biomarker-guided therapy investigational)
|
||||
```
|
||||
|
||||
## Risk Stratification Tools
|
||||
|
||||
### Cardiovascular Risk Scores
|
||||
|
||||
**TIMI Risk Score (NSTEMI/Unstable Angina)**
|
||||
|
||||
```
|
||||
Score Calculation (0-7 points):
|
||||
☐ Age ≥65 years (1 point)
|
||||
☐ ≥3 cardiac risk factors (HTN, hyperlipidemia, diabetes, smoking, family history) (1)
|
||||
☐ Known CAD (stenosis ≥50%) (1)
|
||||
☐ ASA use in past 7 days (1)
|
||||
☐ Severe angina (≥2 episodes in 24 hours) (1)
|
||||
☐ ST deviation ≥0.5 mm (1)
|
||||
☐ Elevated cardiac biomarkers (1)
|
||||
|
||||
Risk Stratification:
|
||||
├─ Score 0-1: 5% risk of death/MI/urgent revasc at 14 days (Low)
|
||||
│ └─ Management: Observation, stress test, outpatient follow-up
|
||||
│
|
||||
├─ Score 2: 8% risk (Low-intermediate)
|
||||
│ └─ Management: Admission, medical therapy, stress imaging
|
||||
│
|
||||
├─ Score 3-4: 13-20% risk (Intermediate-high)
|
||||
│ └─ Management: Admission, aggressive medical therapy, early invasive strategy
|
||||
│
|
||||
└─ Score 5-7: 26-41% risk (High)
|
||||
└─ Management: Aggressive treatment, urgent angiography (<24 hours)
|
||||
```
|
||||
|
||||
**CHA2DS2-VASc Score (Stroke Risk in Atrial Fibrillation)**
|
||||
|
||||
```
|
||||
Score Calculation:
|
||||
☐ Congestive heart failure (1 point)
|
||||
☐ Hypertension (1)
|
||||
☐ Age ≥75 years (2)
|
||||
☐ Diabetes mellitus (1)
|
||||
☐ Prior stroke/TIA/thromboembolism (2)
|
||||
☐ Vascular disease (MI, PAD, aortic plaque) (1)
|
||||
☐ Age 65-74 years (1)
|
||||
☐ Sex category (female) (1)
|
||||
|
||||
Maximum score: 9 points
|
||||
|
||||
Treatment Algorithm:
|
||||
├─ Score 0 (male) or 1 (female): 0-1.3% annual stroke risk
|
||||
│ └─ No anticoagulation or aspirin (Class IIb)
|
||||
│
|
||||
├─ Score 1 (male): 1.3% annual stroke risk
|
||||
│ └─ Consider anticoagulation (Class IIa)
|
||||
│ Factors: Patient preference, bleeding risk, comorbidities
|
||||
│
|
||||
└─ Score ≥2 (male) or ≥3 (female): ≥2.2% annual stroke risk
|
||||
└─ Anticoagulation recommended (Class I)
|
||||
├─ Preferred: DOAC (apixaban, rivaroxaban, edoxaban, dabigatran)
|
||||
└─ Alternative: Warfarin (INR 2-3) if DOAC contraindicated
|
||||
|
||||
Bleeding Risk Assessment (HAS-BLED):
|
||||
H - Hypertension (SBP >160)
|
||||
A - Abnormal renal/liver function (1 point each)
|
||||
S - Stroke history
|
||||
B - Bleeding history or predisposition
|
||||
L - Labile INR (if on warfarin)
|
||||
E - Elderly (age >65)
|
||||
D - Drugs (antiplatelet, NSAIDs) or alcohol (1 point each)
|
||||
|
||||
HAS-BLED ≥3: High bleeding risk → Modifiable factors, consider DOAC over warfarin
|
||||
```
|
||||
|
||||
### Oncology Risk Calculators
|
||||
|
||||
**MELD Score (Hepatocellular Carcinoma Eligibility)**
|
||||
|
||||
```
|
||||
MELD = 3.78×ln(bilirubin mg/dL) + 11.2×ln(INR) + 9.57×ln(creatinine mg/dL) + 6.43
|
||||
|
||||
Interpretation:
|
||||
├─ MELD <10: 1.9% 3-month mortality (Low)
|
||||
│ └─ Consider resection or ablation for HCC
|
||||
│
|
||||
├─ MELD 10-19: 6-20% 3-month mortality (Moderate)
|
||||
│ └─ Transplant evaluation if within Milan criteria
|
||||
│ Milan: Single ≤5cm or ≤3 lesions each ≤3cm, no vascular invasion
|
||||
│
|
||||
├─ MELD 20-29: 20-45% 3-month mortality (High)
|
||||
│ └─ Urgent transplant evaluation, bridge therapy (TACE, ablation)
|
||||
│
|
||||
└─ MELD ≥30: 50-70% 3-month mortality (Very high)
|
||||
└─ Transplant vs palliative care discussion
|
||||
Too ill for transplant if MELD >35-40 typically
|
||||
```
|
||||
|
||||
**Adjuvant! Online (Breast Cancer Recurrence Risk)**
|
||||
|
||||
```
|
||||
Input Variables:
|
||||
- Age at diagnosis
|
||||
- Tumor size
|
||||
- Tumor grade (1-3)
|
||||
- ER status
|
||||
- Node status (0, 1-3, 4-9, ≥10)
|
||||
- HER2 status
|
||||
- Comorbidity index
|
||||
|
||||
Output: 10-year risk of:
|
||||
- Recurrence
|
||||
- Breast cancer mortality
|
||||
- Overall mortality
|
||||
|
||||
Treatment Benefit Estimates:
|
||||
- Chemotherapy: Absolute reduction in recurrence
|
||||
- Endocrine therapy: Absolute reduction in recurrence
|
||||
- Trastuzumab: Absolute reduction (if HER2+)
|
||||
|
||||
Clinical Application:
|
||||
├─ Low risk (<10% recurrence): Consider endocrine therapy alone if ER+
|
||||
├─ Intermediate risk (10-20%): Chemotherapy discussion, genomic assay
|
||||
│ └─ Oncotype DX score <26: Endocrine therapy alone
|
||||
│ └─ Oncotype DX score ≥26: Chemotherapy + endocrine therapy
|
||||
└─ High risk (>20%): Chemotherapy + endocrine therapy if ER+
|
||||
```
|
||||
|
||||
## TikZ Flowchart Best Practices
|
||||
|
||||
### Visual Design Principles
|
||||
|
||||
**Node Styling**
|
||||
```latex
|
||||
% Decision nodes (diamond)
|
||||
\tikzstyle{decision} = [diamond, draw, fill=yellow!20, text width=4.5em, text centered, inner sep=0pt]
|
||||
|
||||
% Process nodes (rectangle)
|
||||
\tikzstyle{process} = [rectangle, draw, fill=blue!20, text width=5em, text centered, rounded corners, minimum height=3em]
|
||||
|
||||
% Terminal nodes (rounded rectangle)
|
||||
\tikzstyle{terminal} = [rectangle, draw, fill=green!20, text width=5em, text centered, rounded corners=1em, minimum height=3em]
|
||||
|
||||
% Input/Output (parallelogram)
|
||||
\tikzstyle{io} = [trapezium, draw, fill=purple!20, text width=5em, text centered, minimum height=3em]
|
||||
```
|
||||
|
||||
**Color Coding by Urgency**
|
||||
- **Red**: Life-threatening, immediate action required
|
||||
- **Orange**: Urgent, action within hours
|
||||
- **Yellow**: Semi-urgent, action within 24-48 hours
|
||||
- **Green**: Routine, stable clinical situation
|
||||
- **Blue**: Informational, monitoring only
|
||||
|
||||
**Pathway Emphasis**
|
||||
- Bold arrows for most common pathway
|
||||
- Dashed arrows for rare scenarios
|
||||
- Arrow thickness proportional to pathway frequency
|
||||
- Highlight boxes around critical decision points
|
||||
|
||||
### LaTeX TikZ Template
|
||||
|
||||
```latex
|
||||
\documentclass{article}
|
||||
\usepackage{tikz}
|
||||
\usetikzlibrary{shapes, arrows, positioning}
|
||||
|
||||
\begin{document}
|
||||
|
||||
\tikzstyle{decision} = [diamond, draw, fill=yellow!20, text width=4em, text centered, inner sep=2pt, font=\small]
|
||||
\tikzstyle{process} = [rectangle, draw, fill=blue!20, text width=6em, text centered, rounded corners, minimum height=2.5em, font=\small]
|
||||
\tikzstyle{terminal} = [rectangle, draw, fill=green!20, text width=6em, text centered, rounded corners=8pt, minimum height=2.5em, font=\small]
|
||||
\tikzstyle{alert} = [rectangle, draw=red, line width=1.5pt, fill=red!10, text width=6em, text centered, rounded corners, minimum height=2.5em, font=\small\bfseries]
|
||||
\tikzstyle{arrow} = [thick,->,>=stealth]
|
||||
|
||||
\begin{tikzpicture}[node distance=2cm, auto]
|
||||
% Nodes
|
||||
\node [terminal] (start) {Patient presents with symptom X};
|
||||
\node [decision, below of=start] (decision1) {Criterion A met?};
|
||||
\node [alert, below of=decision1, node distance=2.5cm] (alert1) {Immediate action};
|
||||
\node [process, right of=decision1, node distance=4cm] (process1) {Standard evaluation};
|
||||
\node [terminal, below of=process1, node distance=2.5cm] (end) {Outcome};
|
||||
|
||||
% Arrows
|
||||
\draw [arrow] (start) -- (decision1);
|
||||
\draw [arrow] (decision1) -- node {Yes} (alert1);
|
||||
\draw [arrow] (decision1) -- node {No} (process1);
|
||||
\draw [arrow] (process1) -- (end);
|
||||
\draw [arrow] (alert1) -| (end);
|
||||
\end{tikzpicture}
|
||||
|
||||
\end{document}
|
||||
```
|
||||
|
||||
## Algorithm Validation
|
||||
|
||||
### Development Process
|
||||
|
||||
**Step 1: Literature Review and Evidence Synthesis**
|
||||
- Systematic review of guidelines (NCCN, ASCO, ESMO, AHA/ACC)
|
||||
- Meta-analyses of clinical trials
|
||||
- Expert consensus statements
|
||||
- Local practice patterns and resource availability
|
||||
|
||||
**Step 2: Draft Algorithm Development**
|
||||
- Multidisciplinary team input (physicians, nurses, pharmacists)
|
||||
- Define decision nodes and criteria
|
||||
- Specify actions and outcomes
|
||||
- Identify areas of uncertainty
|
||||
|
||||
**Step 3: Pilot Testing**
|
||||
- Retrospective application to historical cases (n=20-50)
|
||||
- Identify scenarios not covered by algorithm
|
||||
- Refine decision criteria
|
||||
- Usability testing with end-users
|
||||
|
||||
**Step 4: Prospective Validation**
|
||||
- Implement in clinical practice with data collection
|
||||
- Track adherence rate (target >80%)
|
||||
- Monitor outcomes vs historical controls
|
||||
- User satisfaction surveys
|
||||
|
||||
**Step 5: Continuous Quality Improvement**
|
||||
- Quarterly review of algorithm performance
|
||||
- Update based on new evidence
|
||||
- Address deviations and reasons for non-adherence
|
||||
- Version control and change documentation
|
||||
|
||||
### Performance Metrics
|
||||
|
||||
**Process Metrics**
|
||||
- Algorithm adherence rate (% cases following algorithm)
|
||||
- Time to decision (median time from presentation to treatment start)
|
||||
- Completion rate (% cases reaching terminal node)
|
||||
|
||||
**Outcome Metrics**
|
||||
- Appropriateness of care (concordance with guidelines)
|
||||
- Clinical outcomes (mortality, morbidity, readmissions)
|
||||
- Resource utilization (length of stay, unnecessary tests)
|
||||
- Safety (adverse events, errors)
|
||||
|
||||
**User Experience Metrics**
|
||||
- Ease of use (Likert scale survey)
|
||||
- Time to use (median time to navigate algorithm)
|
||||
- Perceived utility (% users reporting algorithm helpful)
|
||||
- Barriers to use (qualitative feedback)
|
||||
|
||||
## Implementation Strategies
|
||||
|
||||
### Integration into Clinical Workflow
|
||||
|
||||
**Electronic Health Record Integration**
|
||||
- Clinical decision support (CDS) alerts at key decision points
|
||||
- Order sets linked to algorithm pathways
|
||||
- Auto-population of risk scores from EHR data
|
||||
- Documentation templates following algorithm structure
|
||||
|
||||
**Point-of-Care Tools**
|
||||
- Pocket cards for quick reference
|
||||
- Mobile apps with interactive algorithms
|
||||
- Wall posters in clinical areas
|
||||
- QR codes linking to full algorithm
|
||||
|
||||
**Education and Training**
|
||||
- Didactic presentation of algorithm rationale
|
||||
- Case-based exercises
|
||||
- Simulation scenarios
|
||||
- Audit and feedback on adherence
|
||||
|
||||
### Overcoming Barriers
|
||||
|
||||
**Common Barriers**
|
||||
- Algorithm complexity (too many decision points)
|
||||
- Lack of awareness (not disseminated effectively)
|
||||
- Disagreement with recommendations (perceived as cookbook medicine)
|
||||
- Competing priorities (time pressure, multiple patients)
|
||||
- Resource limitations (recommended tests/treatments not available)
|
||||
|
||||
**Mitigation Strategies**
|
||||
- Simplify algorithms (≤7 decision points per pathway preferred)
|
||||
- Champion network (local opinion leaders promoting algorithm)
|
||||
- Customize to local context (allow flexibility for clinical judgment)
|
||||
- Measure and report outcomes (demonstrate value)
|
||||
- Provide resources (ensure algorithm-recommended options available)
|
||||
|
||||
## Algorithm Maintenance and Updates
|
||||
|
||||
### Version Control
|
||||
|
||||
**Change Log Documentation**
|
||||
```
|
||||
Algorithm: NSCLC First-Line Treatment
|
||||
Version: 3.2
|
||||
Effective Date: January 1, 2024
|
||||
Previous Version: 3.1 (effective July 1, 2023)
|
||||
|
||||
Changes in Version 3.2:
|
||||
1. Added KRAS G12C-mutated pathway (sotorasib, adagrasib)
|
||||
- Evidence: FDA approval May 2021/2022
|
||||
- Guideline: NCCN v4.2023
|
||||
|
||||
2. Updated PD-L1 ≥50% recommendation to include pembrolizumab monotherapy as Option 1
|
||||
- Evidence: KEYNOTE-024 5-year follow-up
|
||||
- Guideline: NCCN Category 1 preferred
|
||||
|
||||
3. Removed crizotinib as preferred ALK inhibitor, moved to alternative
|
||||
- Evidence: ALEX, CROWN trials showing superiority of alectinib, lorlatinib
|
||||
- Guideline: NCCN/ESMO Category 1 for alectinib as first-line
|
||||
|
||||
Reviewed by: Thoracic Oncology Committee
|
||||
Approved by: Dr. [Name], Medical Director
|
||||
Next Review Date: July 1, 2024
|
||||
```
|
||||
|
||||
### Trigger for Updates
|
||||
|
||||
**Mandatory Updates (Within 3 Months)**
|
||||
- FDA approval of new drug for algorithm indication
|
||||
- Guideline change (NCCN, ASCO, ESMO Category 1 recommendation)
|
||||
- Safety alert or black box warning added to recommended agent
|
||||
- Major clinical trial results changing standard of care
|
||||
|
||||
**Routine Updates (Annually)**
|
||||
- Minor evidence updates
|
||||
- Optimization based on local performance data
|
||||
- Formatting or usability improvements
|
||||
- Addition of new clinical scenarios encountered
|
||||
|
||||
**Emergency Updates (Within 1 Week)**
|
||||
- Drug shortage requiring alternative pathways
|
||||
- Drug recall or safety withdrawal
|
||||
- Outbreak or pandemic requiring modified protocols
|
||||
|
||||
@@ -0,0 +1,840 @@
|
||||
# Evidence Synthesis and Guideline Integration Guide
|
||||
|
||||
## Overview
|
||||
|
||||
Evidence synthesis involves systematically reviewing, analyzing, and integrating research findings to inform clinical recommendations. This guide covers guideline sources, evidence hierarchies, systematic reviews, meta-analyses, and integration of multiple evidence streams for clinical decision support.
|
||||
|
||||
## Major Clinical Practice Guidelines
|
||||
|
||||
### Oncology Guidelines
|
||||
|
||||
**NCCN (National Comprehensive Cancer Network)**
|
||||
- **Scope**: 60+ cancer types, supportive care guidelines
|
||||
- **Update Frequency**: Continuous (online), 1-3 updates per year per guideline
|
||||
- **Evidence Categories**:
|
||||
- **Category 1**: High-level evidence, uniform NCCN consensus
|
||||
- **Category 2A**: Lower-level evidence, uniform consensus (appropriate)
|
||||
- **Category 2B**: Lower-level evidence, non-uniform consensus (appropriate)
|
||||
- **Category 3**: Major disagreement or insufficient evidence
|
||||
- **Access**: Free for patients, subscription for providers (institutional access common)
|
||||
- **Application**: US-focused, most widely used in clinical practice
|
||||
|
||||
**ASCO (American Society of Clinical Oncology)**
|
||||
- **Scope**: Evidence-based clinical practice guidelines
|
||||
- **Methodology**: Systematic review, GRADE-style evidence tables
|
||||
- **Endorsements**: Often endorses NCCN, ESMO, or other guidelines
|
||||
- **Focused Topics**: Specific clinical questions (e.g., biomarker testing, supportive care)
|
||||
- **Guideline Products**: Full guidelines, rapid recommendations, endorsements
|
||||
- **Quality**: Rigorous methodology, peer-reviewed publication
|
||||
|
||||
**ESMO (European Society for Medical Oncology)**
|
||||
- **Scope**: European guidelines for cancer management
|
||||
- **Evidence Levels**:
|
||||
- **I**: Evidence from at least one large RCT or meta-analysis
|
||||
- **II**: Evidence from at least one well-designed non-randomized trial, cohort study
|
||||
- **III**: Evidence from well-designed non-experimental study
|
||||
- **IV**: Evidence from expert committee reports or opinions
|
||||
- **V**: Evidence from case series, case reports
|
||||
- **Recommendation Grades**:
|
||||
- **A**: Strong evidence for efficacy, substantial clinical benefit (strongly recommended)
|
||||
- **B**: Strong or moderate evidence, limited clinical benefit (generally recommended)
|
||||
- **C**: Insufficient evidence, benefit not sufficiently well established
|
||||
- **D**: Moderate evidence against efficacy or for adverse effects (not recommended)
|
||||
- **E**: Strong evidence against efficacy (never recommended)
|
||||
- **ESMO-MCBS**: Magnitude of Clinical Benefit Scale (grades 1-5 for meaningful benefit)
|
||||
|
||||
### Cardiovascular Guidelines
|
||||
|
||||
**AHA/ACC (American Heart Association / American College of Cardiology)**
|
||||
- **Scope**: Cardiovascular disease prevention, diagnosis, management
|
||||
- **Class of Recommendation (COR)**:
|
||||
- **Class I**: Strong recommendation - should be performed/administered
|
||||
- **Class IIa**: Moderate recommendation - is reasonable
|
||||
- **Class IIb**: Weak recommendation - may be considered
|
||||
- **Class III - No Benefit**: Not recommended
|
||||
- **Class III - Harm**: Potentially harmful
|
||||
- **Level of Evidence (LOE)**:
|
||||
- **A**: High-quality evidence from >1 RCT, meta-analyses
|
||||
- **B-R**: Moderate-quality evidence from ≥1 RCT
|
||||
- **B-NR**: Moderate-quality evidence from non-randomized studies
|
||||
- **C-LD**: Limited data from observational studies, registries
|
||||
- **C-EO**: Expert opinion based on clinical experience
|
||||
- **Example**: "Statin therapy is recommended for adults with LDL-C ≥190 mg/dL (Class I, LOE A)"
|
||||
|
||||
**ESC (European Society of Cardiology)**
|
||||
- **Scope**: European cardiovascular guidelines
|
||||
- **Class of Recommendation**:
|
||||
- **I**: Recommended or indicated
|
||||
- **II**: Should be considered
|
||||
- **III**: Not recommended
|
||||
- **Level of Evidence**: A (RCTs), B (single RCT or observational), C (expert opinion)
|
||||
|
||||
### Other Specialties
|
||||
|
||||
**IDSA (Infectious Diseases Society of America)**
|
||||
- Antimicrobial guidelines, infection management
|
||||
- GRADE methodology
|
||||
- Strong vs weak recommendations
|
||||
|
||||
**ATS/ERS (American Thoracic Society / European Respiratory Society)**
|
||||
- Respiratory disease management
|
||||
- GRADE methodology
|
||||
|
||||
**ACR (American College of Rheumatology)**
|
||||
- Rheumatic disease guidelines
|
||||
- Conditionally recommended vs strongly recommended
|
||||
|
||||
**KDIGO (Kidney Disease: Improving Global Outcomes)**
|
||||
- Chronic kidney disease, dialysis, transplant
|
||||
- GRADE-based recommendations
|
||||
|
||||
## GRADE Methodology
|
||||
|
||||
### Assessing Quality of Evidence
|
||||
|
||||
**Initial Quality Assignment**
|
||||
|
||||
**Randomized Controlled Trials**: Start at HIGH quality (⊕⊕⊕⊕)
|
||||
|
||||
**Observational Studies**: Start at LOW quality (⊕⊕○○)
|
||||
|
||||
### Factors Decreasing Quality (Downgrade)
|
||||
|
||||
**Risk of Bias** (-1 or -2 levels)
|
||||
- Lack of allocation concealment
|
||||
- Lack of blinding
|
||||
- Incomplete outcome data
|
||||
- Selective outcome reporting
|
||||
- Other sources of bias
|
||||
|
||||
**Inconsistency** (-1 or -2 levels)
|
||||
- Unexplained heterogeneity in results across studies
|
||||
- Wide variation in effect estimates
|
||||
- Non-overlapping confidence intervals
|
||||
- High I² statistic in meta-analysis (>50-75%)
|
||||
|
||||
**Indirectness** (-1 or -2 levels)
|
||||
- Different population than target (younger patients in trials, applying to elderly)
|
||||
- Different intervention (higher dose in trial than used in practice)
|
||||
- Different comparator (placebo in trial, comparing to active treatment)
|
||||
- Surrogate outcomes (PFS) when interested in survival (OS)
|
||||
|
||||
**Imprecision** (-1 or -2 levels)
|
||||
- Wide confidence intervals crossing threshold of benefit/harm
|
||||
- Small sample size, few events
|
||||
- Optimal information size (OIS) not met
|
||||
- Rule of thumb: <300 events for continuous outcomes, <200 events for dichotomous
|
||||
|
||||
**Publication Bias** (-1 level)
|
||||
- Funnel plot asymmetry (if ≥10 studies)
|
||||
- Known unpublished studies with negative results
|
||||
- Selective outcome reporting
|
||||
- Industry-sponsored studies only
|
||||
|
||||
### Factors Increasing Quality (Upgrade - Observational Only)
|
||||
|
||||
**Large Magnitude of Effect** (+1 or +2 levels)
|
||||
- +1: RR >2 or <0.5 (moderate effect)
|
||||
- +2: RR >5 or <0.2 (large effect)
|
||||
- No plausible confounders would reduce effect
|
||||
|
||||
**Dose-Response Gradient** (+1 level)
|
||||
- Clear dose-response or duration-response relationship
|
||||
- Strengthens causal inference
|
||||
|
||||
**All Plausible Confounders Would Reduce Effect** (+1 level)
|
||||
- Observed effect despite confounders biasing toward null
|
||||
- Rare, requires careful justification
|
||||
|
||||
### Final Quality Rating
|
||||
|
||||
After adjustments, assign final quality:
|
||||
- **High (⊕⊕⊕⊕)**: Very confident in effect estimate
|
||||
- **Moderate (⊕⊕⊕○)**: Moderately confident; true effect likely close to estimate
|
||||
- **Low (⊕⊕○○)**: Limited confidence; true effect may be substantially different
|
||||
- **Very Low (⊕○○○)**: Very little confidence; true effect likely substantially different
|
||||
|
||||
## Systematic Reviews and Meta-Analyses
|
||||
|
||||
### PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses)
|
||||
|
||||
**Search Strategy**
|
||||
- **Databases**: PubMed/MEDLINE, Embase, Cochrane Library, Web of Science
|
||||
- **Search Terms**: PICO (Population, Intervention, Comparator, Outcome)
|
||||
- **Date Range**: Typically last 10-20 years or comprehensive
|
||||
- **Language**: English only or all languages with translation
|
||||
- **Grey Literature**: Conference abstracts, trial registries, unpublished data
|
||||
|
||||
**Study Selection**
|
||||
```
|
||||
PRISMA Flow Diagram:
|
||||
|
||||
Records identified through database searching (n=2,450)
|
||||
Additional records through other sources (n=15)
|
||||
↓
|
||||
Records after duplicates removed (n=1,823)
|
||||
↓
|
||||
Records screened (title/abstract) (n=1,823) → Excluded (n=1,652)
|
||||
↓ - Not relevant topic (n=1,120)
|
||||
Full-text articles assessed (n=171) - Animal studies (n=332)
|
||||
↓ - Reviews (n=200)
|
||||
Studies included in qualitative synthesis (n=38) → Excluded (n=133)
|
||||
↓ - Wrong population (n=42)
|
||||
Studies included in meta-analysis (n=24) - Wrong intervention (n=35)
|
||||
- No outcomes reported (n=28)
|
||||
- Duplicate data (n=18)
|
||||
- Poor quality (n=10)
|
||||
```
|
||||
|
||||
**Data Extraction**
|
||||
- Study characteristics: Design, sample size, population, intervention
|
||||
- Results: Outcomes, effect sizes, confidence intervals, p-values
|
||||
- Quality assessment: Risk of bias tool (Cochrane RoB 2.0 for RCTs)
|
||||
- Dual extraction: Two reviewers independently, resolve disagreements
|
||||
|
||||
### Meta-Analysis Methods
|
||||
|
||||
**Fixed-Effect Model**
|
||||
- **Assumption**: Single true effect size shared by all studies
|
||||
- **Weighting**: By inverse variance (larger studies have more weight)
|
||||
- **Application**: When heterogeneity is low (I² <25%)
|
||||
- **Interpretation**: Estimate of common effect across studies
|
||||
|
||||
**Random-Effects Model**
|
||||
- **Assumption**: True effect varies across studies (distribution of effects)
|
||||
- **Weighting**: By inverse variance + between-study variance
|
||||
- **Application**: When heterogeneity moderate to high (I² ≥25%)
|
||||
- **Interpretation**: Estimate of average effect (center of distribution)
|
||||
- **Wider CI**: Accounts for heterogeneity, more conservative
|
||||
|
||||
**Heterogeneity Assessment**
|
||||
|
||||
**I² Statistic**
|
||||
- Percentage of variability due to heterogeneity rather than chance
|
||||
- I² = 0-25%: Low heterogeneity
|
||||
- I² = 25-50%: Moderate heterogeneity
|
||||
- I² = 50-75%: Substantial heterogeneity
|
||||
- I² = 75-100%: Considerable heterogeneity
|
||||
|
||||
**Q Test (Cochran's Q)**
|
||||
- Test for heterogeneity
|
||||
- p<0.10 suggests significant heterogeneity (liberal threshold)
|
||||
- Low power when few studies, use I² as primary measure
|
||||
|
||||
**Tau² (τ²)**
|
||||
- Estimate of between-study variance
|
||||
- Used in random-effects weighting
|
||||
|
||||
**Subgroup Analysis**
|
||||
- Explore sources of heterogeneity
|
||||
- Pre-specified subgroups: Disease stage, biomarker status, treatment regimen
|
||||
- Test for interaction between subgroups
|
||||
|
||||
**Forest Plot Interpretation**
|
||||
```
|
||||
Study n HR (95% CI) Weight
|
||||
─────────────────────────────────────────────────────────────
|
||||
Trial A 2018 450 0.62 (0.45-0.85) ●───┤ 28%
|
||||
Trial B 2019 320 0.71 (0.49-1.02) ●────┤ 22%
|
||||
Trial C 2020 580 0.55 (0.41-0.74) ●──┤ 32%
|
||||
Trial D 2021 210 0.88 (0.56-1.38) ●──────┤ 18%
|
||||
|
||||
Overall (RE model) 1560 0.65 (0.53-0.80) ◆──┤
|
||||
Heterogeneity: I²=42%, p=0.16
|
||||
|
||||
0.25 0.5 1.0 2.0 4.0
|
||||
Favors Treatment Favors Control
|
||||
```
|
||||
|
||||
## Guideline Integration
|
||||
|
||||
### Concordance Checking
|
||||
|
||||
**Multi-Guideline Comparison**
|
||||
```
|
||||
Recommendation: First-line treatment for advanced NSCLC, PD-L1 ≥50%
|
||||
|
||||
Guideline Version Recommendation Strength
|
||||
─────────────────────────────────────────────────────────────────────────────
|
||||
NCCN v4.2024 Pembrolizumab monotherapy (preferred) Category 1
|
||||
ESMO 2023 Pembrolizumab monotherapy (preferred) I, A
|
||||
ASCO 2022 Endorses NCCN guidelines Strong
|
||||
NICE (UK) 2023 Pembrolizumab approved Recommended
|
||||
|
||||
Synthesis: Strong consensus across guidelines for pembrolizumab monotherapy.
|
||||
Alternative: Pembrolizumab + chemotherapy also Category 1/I-A recommended.
|
||||
```
|
||||
|
||||
**Discordance Resolution**
|
||||
- Identify differences and reasons (geography, cost, access, evidence interpretation)
|
||||
- Note date of each guideline (newer may incorporate recent trials)
|
||||
- Consider regional applicability
|
||||
- Favor guidelines with most rigorous methodology (GRADE-based)
|
||||
|
||||
### Regulatory Approval Landscape
|
||||
|
||||
**FDA Approvals**
|
||||
- Track indication-specific approvals
|
||||
- Accelerated approval vs full approval
|
||||
- Post-marketing requirements
|
||||
- Contraindications and warnings
|
||||
|
||||
**EMA (European Medicines Agency)**
|
||||
- May differ from FDA in approved indications
|
||||
- Conditional marketing authorization
|
||||
- Additional monitoring (black triangle)
|
||||
|
||||
**Regional Variations**
|
||||
- Health Technology Assessment (HTA) agencies
|
||||
- NICE (UK): Cost-effectiveness analysis, QALY thresholds
|
||||
- CADTH (Canada): Therapeutic review and recommendations
|
||||
- PBAC (Australia): Reimbursement decisions
|
||||
|
||||
## Real-World Evidence (RWE)
|
||||
|
||||
### Sources of RWE
|
||||
|
||||
**Electronic Health Records (EHR)**
|
||||
- Clinical data from routine practice
|
||||
- Large patient numbers
|
||||
- Heterogeneous populations (more generalizable than RCTs)
|
||||
- Limitations: Missing data, inconsistent documentation, selection bias
|
||||
|
||||
**Claims Databases**
|
||||
- Administrative claims for billing/reimbursement
|
||||
- Large scale (millions of patients)
|
||||
- Outcomes: Mortality, hospitalizations, procedures
|
||||
- Limitations: Lack clinical detail (labs, imaging, biomarkers)
|
||||
|
||||
**Cancer Registries**
|
||||
- **SEER (Surveillance, Epidemiology, and End Results)**: US cancer registry
|
||||
- **NCDB (National Cancer Database)**: Hospital registry data
|
||||
- Population-level survival, treatment patterns
|
||||
- Limited treatment detail, no toxicity data
|
||||
|
||||
**Prospective Cohorts**
|
||||
- Framingham Heart Study, Nurses' Health Study
|
||||
- Long-term follow-up, rich covariate data
|
||||
- Expensive, time-consuming
|
||||
|
||||
### RWE Applications
|
||||
|
||||
**Comparative Effectiveness**
|
||||
- Compare treatments in real-world settings (less strict eligibility than RCTs)
|
||||
- Complement RCT data with broader populations
|
||||
- Example: Effectiveness of immunotherapy in elderly, poor PS patients excluded from trials
|
||||
|
||||
**Safety Signal Detection**
|
||||
- Rare adverse events not detected in trials
|
||||
- Long-term toxicities
|
||||
- Drug-drug interactions in polypharmacy
|
||||
- Postmarketing surveillance
|
||||
|
||||
**Treatment Patterns and Access**
|
||||
- Guideline adherence in community practice
|
||||
- Time to treatment initiation
|
||||
- Disparities in care delivery
|
||||
- Off-label use prevalence
|
||||
|
||||
**Limitations of RWE**
|
||||
- **Confounding by indication**: Sicker patients receive more aggressive treatment
|
||||
- **Immortal time bias**: Time between events affecting survival estimates
|
||||
- **Missing data**: Incomplete or inconsistent data collection
|
||||
- **Causality**: Association does not prove causation without randomization
|
||||
|
||||
**Strengthening RWE**
|
||||
- **Propensity score matching**: Balance baseline characteristics between groups
|
||||
- **Multivariable adjustment**: Adjust for measured confounders in Cox model
|
||||
- **Sensitivity analyses**: Test robustness to unmeasured confounding
|
||||
- **Instrumental variables**: Use natural experiments to approximate randomization
|
||||
|
||||
## Meta-Analysis Techniques
|
||||
|
||||
### Binary Outcomes (Response Rate, Event Rate)
|
||||
|
||||
**Effect Measures**
|
||||
- **Risk Ratio (RR)**: Ratio of event probabilities
|
||||
- **Odds Ratio (OR)**: Ratio of odds (less intuitive)
|
||||
- **Risk Difference (RD)**: Absolute difference in event rates
|
||||
|
||||
**Example Calculation**
|
||||
```
|
||||
Study 1:
|
||||
- Treatment A: 30/100 responded (30%)
|
||||
- Treatment B: 15/100 responded (15%)
|
||||
- RR = 0.30/0.15 = 2.0 (95% CI 1.15-3.48)
|
||||
- RD = 0.30 - 0.15 = 0.15 or 15% (95% CI 4.2%-25.8%)
|
||||
- NNT = 1/RD = 1/0.15 = 6.7 (treat 7 patients to get 1 additional response)
|
||||
```
|
||||
|
||||
**Pooling Methods**
|
||||
- **Mantel-Haenszel**: Common fixed-effect method
|
||||
- **DerSimonian-Laird**: Random-effects method
|
||||
- **Peto**: For rare events (event rate <1%)
|
||||
|
||||
### Time-to-Event Outcomes (Survival, PFS)
|
||||
|
||||
**Hazard Ratio Pooling**
|
||||
- Extract HR and 95% CI (or log(HR) and SE) from each study
|
||||
- Weight by inverse variance
|
||||
- Pool using generic inverse variance method
|
||||
- Report pooled HR with 95% CI, heterogeneity statistics
|
||||
|
||||
**When HR Not Reported**
|
||||
- Extract from Kaplan-Meier curves (Parmar method, digitizing software)
|
||||
- Calculate from log-rank p-value and event counts
|
||||
- Request from study authors
|
||||
|
||||
### Continuous Outcomes (Quality of Life, Lab Values)
|
||||
|
||||
**Standardized Mean Difference (SMD)**
|
||||
- Application: Different scales used across studies
|
||||
- SMD = (Mean₁ - Mean₂) / Pooled SD
|
||||
- Interpretation: Cohen's d effect size (0.2 small, 0.5 medium, 0.8 large)
|
||||
|
||||
**Mean Difference (MD)**
|
||||
- Application: Same scale/unit used across studies
|
||||
- MD = Mean₁ - Mean₂
|
||||
- More directly interpretable than SMD
|
||||
|
||||
## Network Meta-Analysis
|
||||
|
||||
### Purpose
|
||||
|
||||
Compare multiple treatments simultaneously when no head-to-head trials exist
|
||||
|
||||
**Example Scenario**
|
||||
- Drug A vs placebo (Trial 1)
|
||||
- Drug B vs placebo (Trial 2)
|
||||
- Drug C vs Drug A (Trial 3)
|
||||
- **Question**: How does Drug B compare to Drug C? (no direct comparison)
|
||||
|
||||
### Methods
|
||||
|
||||
**Fixed-Effect Network Meta-Analysis**
|
||||
- Assumes consistency (transitivity): A vs B effect = (A vs C effect) - (B vs C effect)
|
||||
- Provides indirect comparison estimates
|
||||
- Ranks treatments by P-score or SUCRA
|
||||
|
||||
**Random-Effects Network Meta-Analysis**
|
||||
- Allows heterogeneity between studies
|
||||
- More conservative estimates
|
||||
|
||||
**Consistency Checking**
|
||||
- Compare direct vs indirect evidence for same comparison
|
||||
- Node-splitting analysis
|
||||
- Loop consistency (if closed loops in network)
|
||||
|
||||
### Interpretation Cautions
|
||||
|
||||
- **Transitivity assumption**: May not hold if studies differ in important ways
|
||||
- **Indirect evidence**: Less reliable than direct head-to-head trials
|
||||
- **Rankings**: Probabilistic, not definitive ordering
|
||||
- **Clinical judgment**: Consider beyond statistical rankings
|
||||
|
||||
## Evidence Tables
|
||||
|
||||
### Constructing Evidence Summary Tables
|
||||
|
||||
**PICO Framework**
|
||||
- **P (Population)**: Patient characteristics, disease stage, biomarker status
|
||||
- **I (Intervention)**: Treatment regimen, dose, schedule
|
||||
- **C (Comparator)**: Control arm (placebo, standard of care)
|
||||
- **O (Outcomes)**: Primary and secondary endpoints
|
||||
|
||||
**Evidence Table Template**
|
||||
```
|
||||
Study Design n Population Intervention vs Comparator Outcome Result Quality
|
||||
────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
|
||||
Smith 2020 RCT 450 Advanced NSCLC Drug A 10mg vs Median PFS 12 vs 6 months High
|
||||
EGFR+ standard chemo (95% CI) (10-14 vs 5-7) ⊕⊕⊕⊕
|
||||
HR (95% CI) 0.48 (0.36-0.64)
|
||||
p-value p<0.001
|
||||
|
||||
ORR 65% vs 35%
|
||||
Grade 3-4 AEs 42% vs 38%
|
||||
|
||||
Jones 2021 RCT 380 Advanced NSCLC Drug A 10mg vs Median PFS 10 vs 5.5 months High
|
||||
EGFR+ placebo HR (95% CI) 0.42 (0.30-0.58) ⊕⊕⊕⊕
|
||||
p-value p<0.001
|
||||
|
||||
Pooled Effect Pooled HR 0.45 (0.36-0.57) High
|
||||
(Meta-analysis) I² 12% (low heterogeneity) ⊕⊕⊕⊕
|
||||
```
|
||||
|
||||
### Evidence to Decision Framework
|
||||
|
||||
**Benefits and Harms**
|
||||
- Magnitude of desirable effects (ORR, PFS, OS improvement)
|
||||
- Magnitude of undesirable effects (toxicity, quality of life impact)
|
||||
- Balance of benefits and harms
|
||||
- Net benefit calculation
|
||||
|
||||
**Values and Preferences**
|
||||
- How do patients value outcomes? (survival vs quality of life)
|
||||
- Variability in patient values
|
||||
- Shared decision-making importance
|
||||
|
||||
**Resource Considerations**
|
||||
- Cost of intervention
|
||||
- Cost-effectiveness ($/QALY)
|
||||
- Budget impact
|
||||
- Equity and access
|
||||
|
||||
**Feasibility and Acceptability**
|
||||
- Is treatment available in practice settings?
|
||||
- Route of administration feasible? (oral vs IV vs subcutaneous)
|
||||
- Monitoring requirements realistic?
|
||||
- Patient and provider acceptability
|
||||
|
||||
## Guideline Concordance Documentation
|
||||
|
||||
### Synthesizing Multiple Guidelines
|
||||
|
||||
**Concordant Recommendations**
|
||||
```
|
||||
Clinical Question: Treatment for HER2+ metastatic breast cancer, first-line
|
||||
|
||||
Guideline Summary:
|
||||
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
|
||||
NCCN v3.2024 (Category 1):
|
||||
Preferred: Pertuzumab + trastuzumab + taxane
|
||||
Alternative: T-DM1, other HER2-targeted combinations
|
||||
|
||||
ESMO 2022 (Grade I, A):
|
||||
Preferred: Pertuzumab + trastuzumab + docetaxel
|
||||
Alternative: Trastuzumab + chemotherapy (if pertuzumab unavailable)
|
||||
|
||||
ASCO 2020 Endorsement:
|
||||
Endorses NCCN guidelines, recommends pertuzumab-based first-line
|
||||
|
||||
Synthesis:
|
||||
Strong consensus for pertuzumab + trastuzumab + taxane as first-line standard.
|
||||
Evidence: CLEOPATRA trial (Swain 2015): median OS 56.5 vs 40.8 months (HR 0.68, p<0.001)
|
||||
|
||||
Recommendation:
|
||||
Pertuzumab 840 mg IV loading then 420 mg + trastuzumab 8 mg/kg loading then 6 mg/kg
|
||||
+ docetaxel 75 mg/m² every 3 weeks until progression.
|
||||
|
||||
Strength: Strong (GRADE 1A)
|
||||
Evidence: High-quality, multiple RCTs, guideline concordance
|
||||
```
|
||||
|
||||
**Discordant Recommendations**
|
||||
```
|
||||
Clinical Question: Adjuvant osimertinib for resected EGFR+ NSCLC
|
||||
|
||||
NCCN v4.2024 (Category 1):
|
||||
Osimertinib 80 mg daily × 3 years after adjuvant chemotherapy
|
||||
Evidence: ADAURA trial (median DFS not reached vs 28 months, HR 0.17)
|
||||
|
||||
ESMO 2023 (II, B):
|
||||
Osimertinib may be considered
|
||||
Note: Cost-effectiveness concerns, OS data immature
|
||||
|
||||
NICE (UK) 2022:
|
||||
Not recommended for routine use
|
||||
Reason: QALY analysis unfavorable at current pricing
|
||||
|
||||
Synthesis:
|
||||
Efficacy demonstrated in phase 3 trial (ADAURA), FDA/EMA approved.
|
||||
Guideline discordance based on cost-effectiveness, not clinical efficacy.
|
||||
|
||||
US practice: NCCN Category 1, widely adopted
|
||||
European/UK: Variable adoption based on national HTA decisions
|
||||
|
||||
Recommendation Context-Dependent:
|
||||
US: Strong recommendation if accessible (GRADE 1B)
|
||||
Countries with cost constraints: Conditional recommendation (GRADE 2B)
|
||||
```
|
||||
|
||||
## Quality Assessment Tools
|
||||
|
||||
### RCT Quality Assessment (Cochrane Risk of Bias 2.0)
|
||||
|
||||
**Domains**
|
||||
1. **Bias from randomization process**: Sequence generation, allocation concealment
|
||||
2. **Bias from deviations from intended interventions**: Blinding, protocol adherence
|
||||
3. **Bias from missing outcome data**: Attrition, intention-to-treat analysis
|
||||
4. **Bias in outcome measurement**: Blinded assessment, objective outcomes
|
||||
5. **Bias in selection of reported result**: Selective reporting, outcome switching
|
||||
|
||||
**Judgment**: Low risk, some concerns, high risk (for each domain)
|
||||
|
||||
**Overall Risk of Bias**: Based on highest-risk domain
|
||||
|
||||
### Observational Study Quality (Newcastle-Ottawa Scale)
|
||||
|
||||
**Selection (max 4 stars)**
|
||||
- Representativeness of exposed cohort
|
||||
- Selection of non-exposed cohort
|
||||
- Ascertainment of exposure
|
||||
- Outcome not present at start
|
||||
|
||||
**Comparability (max 2 stars)**
|
||||
- Comparability of cohorts (design/analysis adjustment for confounders)
|
||||
|
||||
**Outcome (max 3 stars)**
|
||||
- Assessment of outcome
|
||||
- Follow-up duration adequate
|
||||
- Adequacy of follow-up (low attrition)
|
||||
|
||||
**Total Score**: 0-9 stars
|
||||
- **High quality**: 7-9 stars
|
||||
- **Moderate quality**: 4-6 stars
|
||||
- **Low quality**: 0-3 stars
|
||||
|
||||
## Translating Evidence to Recommendations
|
||||
|
||||
### Recommendation Development Process
|
||||
|
||||
**Step 1: PICO Question Formulation**
|
||||
```
|
||||
Example PICO:
|
||||
P - Population: Adults with type 2 diabetes and cardiovascular disease
|
||||
I - Intervention: SGLT2 inhibitor (empagliflozin)
|
||||
C - Comparator: Placebo (added to standard care)
|
||||
O - Outcomes: Major adverse cardiovascular events (3P-MACE), hospitalization for heart failure
|
||||
```
|
||||
|
||||
**Step 2: Systematic Evidence Review**
|
||||
- Identify all relevant studies
|
||||
- Assess quality using standardized tools
|
||||
- Extract outcome data
|
||||
- Synthesize findings (narrative or meta-analysis)
|
||||
|
||||
**Step 3: GRADE Evidence Rating**
|
||||
- Start at high (RCTs) or low (observational)
|
||||
- Downgrade for risk of bias, inconsistency, indirectness, imprecision, publication bias
|
||||
- Upgrade for large effect, dose-response, confounders reducing effect (observational only)
|
||||
- Assign final quality rating
|
||||
|
||||
**Step 4: Recommendation Strength Determination**
|
||||
|
||||
**Strong Recommendation (Grade 1)**
|
||||
- Desirable effects clearly outweigh undesirable effects
|
||||
- High or moderate quality evidence
|
||||
- Little variability in patient values
|
||||
- Intervention cost-effective
|
||||
|
||||
**Conditional Recommendation (Grade 2)**
|
||||
- Trade-offs: Desirable and undesirable effects closely balanced
|
||||
- Low or very low quality evidence
|
||||
- Substantial variability in patient values/preferences
|
||||
- Uncertain cost-effectiveness
|
||||
|
||||
**Step 5: Wording the Recommendation**
|
||||
```
|
||||
Strong: "We recommend..."
|
||||
Example: "We recommend SGLT2 inhibitor therapy for adults with type 2 diabetes and
|
||||
established cardiovascular disease to reduce risk of hospitalization for heart failure
|
||||
and cardiovascular death (Strong recommendation, high-quality evidence - GRADE 1A)."
|
||||
|
||||
Conditional: "We suggest..."
|
||||
Example: "We suggest considering GLP-1 receptor agonist therapy for adults with type 2
|
||||
diabetes and CKD to reduce risk of kidney disease progression (Conditional recommendation,
|
||||
moderate-quality evidence - GRADE 2B)."
|
||||
```
|
||||
|
||||
## Incorporating Emerging Evidence
|
||||
|
||||
### Early-Phase Trial Data
|
||||
|
||||
**Phase 1 Trials**
|
||||
- Purpose: Dose-finding, safety
|
||||
- Outcomes: Maximum tolerated dose (MTD), dose-limiting toxicities (DLTs), pharmacokinetics
|
||||
- Evidence level: Very low (expert opinion, case series)
|
||||
- Clinical application: Investigational only, clinical trial enrollment
|
||||
|
||||
**Phase 2 Trials**
|
||||
- Purpose: Preliminary efficacy signal
|
||||
- Design: Single-arm (ORR primary endpoint) or randomized (PFS comparison)
|
||||
- Evidence level: Low to moderate
|
||||
- Clinical application: May support off-label use in refractory settings, clinical trial enrollment preferred
|
||||
|
||||
**Phase 3 Trials**
|
||||
- Purpose: Confirmatory efficacy and safety
|
||||
- Design: Randomized controlled trial, OS or PFS primary endpoint
|
||||
- Evidence level: High (if well-designed and executed)
|
||||
- Clinical application: Regulatory approval basis, guideline recommendations
|
||||
|
||||
**Phase 4 Trials**
|
||||
- Purpose: Post-marketing surveillance, additional indications
|
||||
- Evidence level: Variable (depends on design)
|
||||
- Clinical application: Safety monitoring, expanded usage
|
||||
|
||||
### Breakthrough Therapy Designation
|
||||
|
||||
**FDA Fast-Track Programs**
|
||||
- **Breakthrough Therapy**: Preliminary evidence of substantial improvement over existing therapy
|
||||
- **Accelerated Approval**: Approval based on surrogate endpoint (PFS, ORR)
|
||||
- Post-marketing requirement: Confirmatory OS trial
|
||||
- **Priority Review**: Shortened FDA review time (6 vs 10 months)
|
||||
|
||||
**Implications for Guidelines**
|
||||
- May receive NCCN Category 2A before phase 3 data mature
|
||||
- Upgrade to Category 1 when confirmatory data published
|
||||
- Monitor for post-market confirmatory trial results
|
||||
|
||||
### Updating Recommendations
|
||||
|
||||
**Triggers for Update**
|
||||
- New phase 3 trial results (major journal publication)
|
||||
- FDA/EMA approval for new indication or agent
|
||||
- Guideline update from NCCN, ASCO, ESMO
|
||||
- Safety alert or drug withdrawal
|
||||
- Meta-analysis changing effect estimates
|
||||
|
||||
**Rapid Update Process**
|
||||
- Critical appraisal of new evidence
|
||||
- Assess impact on current recommendations
|
||||
- Revise evidence grade and recommendation strength if needed
|
||||
- Disseminate update to users
|
||||
- Version control and change log
|
||||
|
||||
## Conflicts of Interest and Bias
|
||||
|
||||
### Identifying Potential Bias
|
||||
|
||||
**Study Sponsorship**
|
||||
- **Industry-sponsored**: May favor sponsor's product (publication bias, outcome selection)
|
||||
- **Academic**: May favor investigator's hypothesis
|
||||
- **Independent**: Government funding (NIH, PCORI)
|
||||
|
||||
**Author Conflicts of Interest**
|
||||
- Consulting fees, research funding, stock ownership
|
||||
- Disclosure statements required by journals
|
||||
- ICMJE Form for Disclosure of Potential COI
|
||||
|
||||
**Mitigating Bias**
|
||||
- Register trials prospectively (ClinicalTrials.gov)
|
||||
- Pre-specify primary endpoint and analysis plan
|
||||
- Independent data monitoring committee (IDMC)
|
||||
- Blinding of outcome assessors
|
||||
- Intention-to-treat analysis
|
||||
|
||||
### Transparency in Evidence Synthesis
|
||||
|
||||
**Pre-Registration**
|
||||
- PROSPERO for systematic reviews
|
||||
- Pre-specify PICO, search strategy, outcomes, analysis plan
|
||||
- Prevents post-hoc changes to avoid negative findings
|
||||
|
||||
**Reporting Checklists**
|
||||
- PRISMA for systematic reviews/meta-analyses
|
||||
- CONSORT for RCTs
|
||||
- STROBE for observational studies
|
||||
|
||||
**Data Availability**
|
||||
- Individual patient data (IPD) sharing increases transparency
|
||||
- Repositories: ClinicalTrials.gov results database, journal supplements
|
||||
|
||||
## Practical Application
|
||||
|
||||
### Evidence Summary for Clinical Document
|
||||
|
||||
```
|
||||
EVIDENCE SYNTHESIS: Osimertinib for EGFR-Mutated NSCLC
|
||||
|
||||
Clinical Question:
|
||||
Should adults with treatment-naïve advanced NSCLC harboring EGFR exon 19 deletion
|
||||
or L858R mutation receive osimertinib versus first-generation EGFR TKIs?
|
||||
|
||||
Evidence Review:
|
||||
┌──────────────────────────────────────────────────────────────────────┐
|
||||
│ FLAURA Trial (Soria et al., NEJM 2018) │
|
||||
├──────────────────────────────────────────────────────────────────────┤
|
||||
│ Design: Phase 3 RCT, double-blind, 1:1 randomization │
|
||||
│ Population: EGFR exon 19 del or L858R, stage IIIB/IV, ECOG 0-1 │
|
||||
│ Sample Size: n=556 (279 osimertinib, 277 comparator) │
|
||||
│ Intervention: Osimertinib 80 mg PO daily │
|
||||
│ Comparator: Gefitinib 250 mg or erlotinib 150 mg PO daily │
|
||||
│ Primary Endpoint: PFS by investigator assessment │
|
||||
│ Secondary: OS, ORR, DOR, CNS progression, safety │
|
||||
│ │
|
||||
│ Results: │
|
||||
│ - Median PFS: 18.9 vs 10.2 months (HR 0.46, 95% CI 0.37-0.57, p<0.001)│
|
||||
│ - Median OS: 38.6 vs 31.8 months (HR 0.80, 95% CI 0.64-1.00, p=0.046)│
|
||||
│ - ORR: 80% vs 76% (p=0.24) │
|
||||
│ - Grade ≥3 AEs: 34% vs 45% │
|
||||
│ - Quality: High (well-designed RCT, low risk of bias) │
|
||||
└──────────────────────────────────────────────────────────────────────┘
|
||||
|
||||
Guideline Recommendations:
|
||||
NCCN v4.2024: Category 1 preferred
|
||||
ESMO 2022: Grade I, A
|
||||
ASCO 2022: Endorsed
|
||||
|
||||
GRADE Assessment:
|
||||
Quality of Evidence: ⊕⊕⊕⊕ HIGH
|
||||
- Randomized controlled trial
|
||||
- Low risk of bias (allocation concealment, blinding, ITT analysis)
|
||||
- Consistent results (single large trial, consistent with phase 2 data)
|
||||
- Direct evidence (target population and outcomes)
|
||||
- Precise estimate (narrow CI, sufficient events)
|
||||
- No publication bias concerns
|
||||
|
||||
Balance of Benefits and Harms:
|
||||
- Large PFS benefit (8.7 month improvement, HR 0.46)
|
||||
- OS benefit (6.8 month improvement, HR 0.80)
|
||||
- Similar ORR, improved tolerability (lower grade 3-4 AEs)
|
||||
- Desirable effects clearly outweigh undesirable effects
|
||||
|
||||
Patient Values: Little variability (most patients value survival extension)
|
||||
|
||||
Cost: Higher cost than first-gen TKIs, but widely accessible in developed countries
|
||||
|
||||
FINAL RECOMMENDATION:
|
||||
Osimertinib 80 mg PO daily is recommended as first-line therapy for adults with
|
||||
advanced NSCLC harboring EGFR exon 19 deletion or L858R mutation.
|
||||
|
||||
Strength: STRONG (Grade 1)
|
||||
Quality of Evidence: HIGH (⊕⊕⊕⊕)
|
||||
GRADE: 1A
|
||||
```
|
||||
|
||||
## Keeping Current
|
||||
|
||||
### Literature Surveillance
|
||||
|
||||
**Automated Alerts**
|
||||
- PubMed My NCBI (save searches, email alerts)
|
||||
- Google Scholar alerts for specific topics
|
||||
- Journal table of contents alerts (NEJM, Lancet, JCO)
|
||||
- Guideline update notifications (NCCN, ASCO, ESMO email lists)
|
||||
|
||||
**Conference Monitoring**
|
||||
- ASCO Annual Meeting (June)
|
||||
- ESMO Congress (September)
|
||||
- ASH Annual Meeting (December, hematology)
|
||||
- AHA Scientific Sessions (November, cardiology)
|
||||
- Plenary and press releases for practice-changing trials
|
||||
|
||||
**Trial Results Databases**
|
||||
- ClinicalTrials.gov results database
|
||||
- FDA approval letters and reviews
|
||||
- EMA European public assessment reports (EPARs)
|
||||
|
||||
### Critical Appraisal Workflow
|
||||
|
||||
**Weekly Review**
|
||||
1. Screen new publications (title/abstract)
|
||||
2. Full-text review of relevant studies
|
||||
3. Quality assessment using checklists
|
||||
4. Extract key findings
|
||||
5. Assess impact on current recommendations
|
||||
|
||||
**Monthly Synthesis**
|
||||
1. Review accumulated evidence
|
||||
2. Identify practice-changing findings
|
||||
3. Update evidence tables
|
||||
4. Revise recommendations if warranted
|
||||
5. Disseminate updates to clinical teams
|
||||
|
||||
**Annual Comprehensive Review**
|
||||
1. Systematic review of guideline updates
|
||||
2. Re-assess all recommendations
|
||||
3. Incorporate year's evidence
|
||||
4. Major version release
|
||||
5. Continuing education activities
|
||||
|
||||
640
skills/clinical-decision-support/references/outcome_analysis.md
Normal file
640
skills/clinical-decision-support/references/outcome_analysis.md
Normal file
@@ -0,0 +1,640 @@
|
||||
# Outcome Analysis and Statistical Methods Guide
|
||||
|
||||
## Overview
|
||||
|
||||
Rigorous outcome analysis is essential for clinical decision support documents. This guide covers survival analysis, response assessment, statistical testing, and data visualization for patient cohort analyses and treatment evaluation.
|
||||
|
||||
## Survival Analysis
|
||||
|
||||
### Kaplan-Meier Method
|
||||
|
||||
**Overview**
|
||||
- Non-parametric estimator of survival function from time-to-event data
|
||||
- Handles censored observations (patients alive at last follow-up)
|
||||
- Provides survival probability at each time point
|
||||
- Generates characteristic step-function survival curves
|
||||
|
||||
**Key Concepts**
|
||||
|
||||
**Censoring**
|
||||
- **Right censoring**: Most common - patient alive at last follow-up or study end
|
||||
- **Left censoring**: Rare in clinical studies
|
||||
- **Interval censoring**: Event occurred between two assessment times
|
||||
- **Informative vs non-informative**: Censoring should be independent of outcome
|
||||
|
||||
**Survival Function S(t)**
|
||||
- S(t) = Probability of surviving beyond time t
|
||||
- S(0) = 1.0 (100% alive at time zero)
|
||||
- S(t) decreases as time increases
|
||||
- Step decreases at each event time
|
||||
|
||||
**Median Survival**
|
||||
- Time point where S(t) = 0.50
|
||||
- 50% of patients alive, 50% have had event
|
||||
- Reported with 95% confidence interval
|
||||
- "Not reached (NR)" if fewer than 50% events
|
||||
|
||||
**Survival Rates at Fixed Time Points**
|
||||
- 1-year survival rate, 2-year survival rate, 5-year survival rate
|
||||
- Read from K-M curve at specific time point
|
||||
- Report with 95% CI: S(t) ± 1.96 × SE
|
||||
|
||||
**Calculation Example**
|
||||
```
|
||||
Time Events At Risk Survival Probability
|
||||
0 0 100 1.000
|
||||
3 2 100 0.980 (98/100)
|
||||
5 1 95 0.970 (97/100 × 95/98)
|
||||
8 3 87 0.936 (94/100 × 92/95 × 84/87)
|
||||
...
|
||||
```
|
||||
|
||||
### Log-Rank Test
|
||||
|
||||
**Purpose**: Compare survival curves between two or more groups
|
||||
|
||||
**Null Hypothesis**: No difference in survival distributions between groups
|
||||
|
||||
**Test Statistic**
|
||||
- Compares observed vs expected events in each group at each time point
|
||||
- Weights all time points equally
|
||||
- Follows chi-square distribution with df = k-1 (k groups)
|
||||
|
||||
**Reporting**
|
||||
- Chi-square statistic, degrees of freedom, p-value
|
||||
- Example: χ² = 6.82, df = 1, p = 0.009
|
||||
- Interpretation: Significant difference in survival curves
|
||||
|
||||
**Assumptions**
|
||||
- Censoring is non-informative and independent
|
||||
- Proportional hazards (constant HR over time)
|
||||
- If non-proportional, consider time-varying effects
|
||||
|
||||
**Alternatives for Non-Proportional Hazards**
|
||||
- **Gehan-Breslow test**: Weights early events more heavily
|
||||
- **Peto-Peto test**: Modifies Gehan-Breslow weighting
|
||||
- **Restricted mean survival time (RMST)**: Difference in area under K-M curve
|
||||
|
||||
### Cox Proportional Hazards Regression
|
||||
|
||||
**Purpose**: Multivariable survival analysis, estimate hazard ratios adjusting for covariates
|
||||
|
||||
**Model**: h(t|X) = h₀(t) × exp(β₁X₁ + β₂X₂ + ... + βₚXₚ)
|
||||
- h(t|X): Hazard rate for individual with covariates X
|
||||
- h₀(t): Baseline hazard function (unspecified)
|
||||
- exp(β): Hazard ratio for one-unit change in covariate
|
||||
|
||||
**Hazard Ratio Interpretation**
|
||||
- HR = 1.0: No effect
|
||||
- HR > 1.0: Increased risk (harmful)
|
||||
- HR < 1.0: Decreased risk (beneficial)
|
||||
- HR = 0.50: 50% reduction in hazard (risk of event)
|
||||
|
||||
**Example Output**
|
||||
```
|
||||
Variable HR 95% CI p-value
|
||||
Treatment (B vs A) 0.62 0.43-0.89 0.010
|
||||
Age (per 10 years) 1.15 1.02-1.30 0.021
|
||||
ECOG PS (2 vs 0-1) 1.85 1.21-2.83 0.004
|
||||
Biomarker+ (vs -) 0.71 0.48-1.05 0.089
|
||||
```
|
||||
|
||||
**Proportional Hazards Assumption**
|
||||
- Hazard ratio constant over time
|
||||
- Test: Schoenfeld residuals, log-minus-log plots
|
||||
- Violation: Time-varying effects, consider stratification or time-dependent covariates
|
||||
|
||||
**Multivariable vs Univariable**
|
||||
- **Univariable**: One covariate at a time, unadjusted HRs
|
||||
- **Multivariable**: Multiple covariates simultaneously, adjusted HRs
|
||||
- Report both: Univariable for all variables, multivariable for final model
|
||||
|
||||
**Model Selection**
|
||||
- **Forward selection**: Start with empty model, add significant variables
|
||||
- **Backward elimination**: Start with all variables, remove non-significant
|
||||
- **Clinical judgment**: Include known prognostic factors regardless of p-value
|
||||
- **Parsimony**: Avoid overfitting, rule of thumb 1 variable per 10-15 events
|
||||
|
||||
## Response Assessment
|
||||
|
||||
### RECIST v1.1 (Response Evaluation Criteria in Solid Tumors)
|
||||
|
||||
**Target Lesions**
|
||||
- Select up to 5 lesions total (maximum 2 per organ)
|
||||
- Measurable: ≥10 mm longest diameter (≥15 mm for lymph nodes short axis)
|
||||
- Sum of longest diameters (SLD) at baseline
|
||||
|
||||
**Response Categories**
|
||||
|
||||
**Complete Response (CR)**
|
||||
- Disappearance of all target and non-target lesions
|
||||
- Lymph nodes must regress to <10 mm short axis
|
||||
- Confirmation required at ≥4 weeks
|
||||
|
||||
**Partial Response (PR)**
|
||||
- ≥30% decrease in SLD from baseline
|
||||
- No new lesions or unequivocal progression of non-target lesions
|
||||
- Confirmation required at ≥4 weeks
|
||||
|
||||
**Stable Disease (SD)**
|
||||
- Neither PR nor PD criteria met
|
||||
- Minimum duration typically 6-8 weeks from baseline
|
||||
|
||||
**Progressive Disease (PD)**
|
||||
- ≥20% increase in SLD AND ≥5 mm absolute increase from smallest SLD (nadir)
|
||||
- OR appearance of new lesions
|
||||
- OR unequivocal progression of non-target lesions
|
||||
|
||||
**Example Calculation**
|
||||
```
|
||||
Baseline SLD: 80 mm (4 target lesions)
|
||||
Week 6 SLD: 52 mm
|
||||
|
||||
Percent change: (52 - 80)/80 × 100% = -35%
|
||||
Classification: Partial Response (≥30% decrease)
|
||||
|
||||
Week 12 SLD: 48 mm (nadir)
|
||||
Week 18 SLD: 62 mm
|
||||
|
||||
Percent change from nadir: (62 - 48)/48 × 100% = +29%
|
||||
Absolute change: 62 - 48 = 14 mm
|
||||
Classification: Progressive Disease (>20% AND ≥5 mm increase)
|
||||
```
|
||||
|
||||
### iRECIST (Immune RECIST)
|
||||
|
||||
**Purpose**: Account for atypical response patterns with immunotherapy
|
||||
|
||||
**Modifications from RECIST v1.1**
|
||||
|
||||
**iUPD (Immune Unconfirmed Progressive Disease)**
|
||||
- Initial increase in tumor burden or new lesions
|
||||
- Requires confirmation at next assessment (≥4 weeks later)
|
||||
- Continue treatment if clinically stable
|
||||
|
||||
**iCPD (Immune Confirmed Progressive Disease)**
|
||||
- Confirmed progression at repeat imaging
|
||||
- Discontinue immunotherapy
|
||||
|
||||
**Pseudoprogression**
|
||||
- Initial apparent progression followed by response
|
||||
- Mechanism: Immune cell infiltration increases tumor size
|
||||
- Incidence: 5-10% of patients on immunotherapy
|
||||
- Management: Continue treatment if patient clinically stable
|
||||
|
||||
**New Lesions**
|
||||
- Record size and location but continue treatment
|
||||
- Do not automatically classify as PD
|
||||
- Confirm progression if new lesions grow or additional new lesions appear
|
||||
|
||||
### Other Response Criteria
|
||||
|
||||
**Lugano Classification (Lymphoma)**
|
||||
- **PET-based**: Deauville 5-point scale
|
||||
- Score 1-3: Negative (metabolic CR)
|
||||
- Score 4-5: Positive (residual disease)
|
||||
- **CT-based**: If PET not available
|
||||
- **Bone marrow**: Required for staging in some lymphomas
|
||||
|
||||
**RANO (Response Assessment in Neuro-Oncology)**
|
||||
- **Glioblastoma-specific**: Accounts for pseudoprogression with radiation/temozolomide
|
||||
- **Enhancing disease**: Bidimensional measurements (product of perpendicular diameters)
|
||||
- **Non-enhancing disease**: FLAIR changes assessed separately
|
||||
- **Corticosteroid dose**: Must document, increase may indicate progression
|
||||
|
||||
**mRECIST (Modified RECIST for HCC)**
|
||||
- **Viable tumor**: Enhancing portion only (arterial phase enhancement)
|
||||
- **Necrosis**: Non-enhancing areas excluded from measurements
|
||||
- **Application**: Hepatocellular carcinoma with arterial enhancement
|
||||
|
||||
## Outcome Metrics
|
||||
|
||||
### Efficacy Endpoints
|
||||
|
||||
**Overall Survival (OS)**
|
||||
- **Definition**: Time from randomization/treatment start to death from any cause
|
||||
- **Advantages**: Objective, not subject to assessment bias, regulatory gold standard
|
||||
- **Disadvantages**: Requires long follow-up, affected by subsequent therapies
|
||||
- **Censoring**: Last known alive date
|
||||
- **Analysis**: Kaplan-Meier, log-rank test, Cox regression
|
||||
|
||||
**Progression-Free Survival (PFS)**
|
||||
- **Definition**: Time from randomization to progression (RECIST) or death
|
||||
- **Advantages**: Earlier readout than OS, direct treatment effect
|
||||
- **Disadvantages**: Requires regular imaging, subject to assessment timing
|
||||
- **Censoring**: Last tumor assessment without progression
|
||||
- **Sensitivity Analysis**: Assess impact of censoring assumptions
|
||||
|
||||
**Objective Response Rate (ORR)**
|
||||
- **Definition**: Proportion of patients achieving CR or PR (best response)
|
||||
- **Denominator**: Evaluable patients (baseline measurable disease)
|
||||
- **Reporting**: Percentage with 95% CI (exact binomial method)
|
||||
- **Duration**: Time from first response to progression (DOR)
|
||||
- **Advantage**: Binary endpoint, no censoring complications
|
||||
|
||||
**Disease Control Rate (DCR)**
|
||||
- **Definition**: CR + PR + SD (stable disease ≥6-8 weeks)
|
||||
- **Less Stringent**: Captures clinical benefit beyond objective response
|
||||
- **Reporting**: Percentage with 95% CI
|
||||
|
||||
**Duration of Response (DOR)**
|
||||
- **Definition**: Time from first CR or PR to progression (among responders only)
|
||||
- **Population**: Subset analysis of responders
|
||||
- **Analysis**: Kaplan-Meier among responders
|
||||
- **Reporting**: Median DOR with 95% CI
|
||||
|
||||
**Time to Treatment Failure (TTF)**
|
||||
- **Definition**: Time from start to discontinuation for any reason (progression, toxicity, death, patient choice)
|
||||
- **Advantage**: Reflects real-world treatment duration
|
||||
- **Components**: PFS + toxicity-related discontinuations
|
||||
|
||||
### Safety Endpoints
|
||||
|
||||
**Adverse Events (CTCAE v5.0)**
|
||||
|
||||
**Grading**
|
||||
- **Grade 1**: Mild, asymptomatic or mild symptoms, clinical intervention not indicated
|
||||
- **Grade 2**: Moderate, minimal/local intervention indicated, age-appropriate ADL limitation
|
||||
- **Grade 3**: Severe or medically significant, not immediately life-threatening, hospitalization/prolongation indicated, disabling, self-care ADL limitation
|
||||
- **Grade 4**: Life-threatening consequences, urgent intervention indicated
|
||||
- **Grade 5**: Death related to adverse event
|
||||
|
||||
**Reporting Standards**
|
||||
```
|
||||
Adverse Event Summary Table:
|
||||
|
||||
AE Term (MedDRA) Any Grade, n (%) Grade 3-4, n (%) Grade 5, n (%)
|
||||
Trt A Trt B Trt A Trt B Trt A Trt B
|
||||
─────────────────────────────────────────────────────────────────────────
|
||||
Hematologic
|
||||
Anemia 45 (90%) 42 (84%) 8 (16%) 6 (12%) 0 0
|
||||
Neutropenia 35 (70%) 38 (76%) 15 (30%) 18 (36%) 0 0
|
||||
Thrombocytopenia 28 (56%) 25 (50%) 6 (12%) 4 (8%) 0 0
|
||||
Febrile neutropenia 4 (8%) 6 (12%) 4 (8%) 6 (12%) 0 0
|
||||
|
||||
Gastrointestinal
|
||||
Nausea 42 (84%) 40 (80%) 2 (4%) 1 (2%) 0 0
|
||||
Diarrhea 31 (62%) 28 (56%) 5 (10%) 3 (6%) 0 0
|
||||
Mucositis 18 (36%) 15 (30%) 3 (6%) 2 (4%) 0 0
|
||||
|
||||
Any AE 50 (100%) 50 (100%) 38 (76%) 35 (70%) 1 (2%) 0
|
||||
```
|
||||
|
||||
**Serious Adverse Events (SAEs)**
|
||||
- SAE incidence and type
|
||||
- Relationship to treatment (related vs unrelated)
|
||||
- Outcome (resolved, ongoing, fatal)
|
||||
- Causality assessment (definite, probable, possible, unlikely, unrelated)
|
||||
|
||||
**Treatment Modifications**
|
||||
- Dose reductions: n (%), reason
|
||||
- Dose delays: n (%), duration
|
||||
- Discontinuations: n (%), reason (toxicity vs progression vs other)
|
||||
- Relative dose intensity: (actual dose delivered / planned dose) × 100%
|
||||
|
||||
## Statistical Analysis Methods
|
||||
|
||||
### Comparing Continuous Outcomes
|
||||
|
||||
**Independent Samples t-test**
|
||||
- **Application**: Compare means between two independent groups (normally distributed)
|
||||
- **Assumptions**: Normal distribution, equal variances (or use Welch's t-test)
|
||||
- **Reporting**: Mean ± SD for each group, mean difference (95% CI), t-statistic, df, p-value
|
||||
- **Example**: Mean age 62.3 ± 8.4 vs 58.7 ± 9.1 years, difference 3.6 years (95% CI 0.2-7.0, p=0.038)
|
||||
|
||||
**Mann-Whitney U Test (Wilcoxon Rank-Sum)**
|
||||
- **Application**: Compare medians between two groups (non-normal distribution)
|
||||
- **Non-parametric**: No distributional assumptions
|
||||
- **Reporting**: Median [IQR] for each group, median difference, U-statistic, p-value
|
||||
- **Example**: Median time to response 6.2 [4.1-8.3] vs 8.5 [5.9-11.2] weeks, p=0.042
|
||||
|
||||
**ANOVA (Analysis of Variance)**
|
||||
- **Application**: Compare means across three or more groups
|
||||
- **Output**: F-statistic, p-value (overall test)
|
||||
- **Post-hoc**: If significant, pairwise comparisons with Tukey or Bonferroni correction
|
||||
- **Example**: Treatment effect varied by biomarker subgroup (F=4.32, df=2, p=0.016)
|
||||
|
||||
### Comparing Categorical Outcomes
|
||||
|
||||
**Chi-Square Test for Independence**
|
||||
- **Application**: Compare proportions between two or more groups
|
||||
- **Assumptions**: Expected count ≥5 in at least 80% of cells
|
||||
- **Reporting**: n (%) for each cell, χ², df, p-value
|
||||
- **Example**: ORR 45% vs 30%, χ²=6.21, df=1, p=0.013
|
||||
|
||||
**Fisher's Exact Test**
|
||||
- **Application**: 2×2 tables when expected count <5
|
||||
- **Exact p-value**: No large-sample approximation
|
||||
- **Two-sided vs one-sided**: Typically report two-sided
|
||||
- **Example**: SAE rate 3/20 (15%) vs 8/22 (36%), Fisher's exact p=0.083
|
||||
|
||||
**McNemar's Test**
|
||||
- **Application**: Paired categorical data (before/after, matched pairs)
|
||||
- **Example**: Response before vs after treatment switch in same patients
|
||||
|
||||
### Sample Size and Power
|
||||
|
||||
**Power Analysis Components**
|
||||
- **Alpha (α)**: Type I error rate, typically 0.05 (two-sided)
|
||||
- **Beta (β)**: Type II error rate, typically 0.10 or 0.20
|
||||
- **Power**: 1 - β, typically 0.80 or 0.90 (80-90% power)
|
||||
- **Effect size**: Expected difference (HR, mean difference, proportion difference)
|
||||
- **Sample size**: Number of patients or events needed
|
||||
|
||||
**Survival Study Sample Size**
|
||||
- Events-driven: Need sufficient events (deaths, progressions)
|
||||
- Rule of thumb: 80% power requires approximately 165 events for HR=0.70 (α=0.05, two-sided)
|
||||
- Accrual time + follow-up time determines calendar time
|
||||
|
||||
**Response Rate Study**
|
||||
```
|
||||
Example: Detect ORR difference 45% vs 30% (15 percentage points)
|
||||
- α = 0.05 (two-sided)
|
||||
- Power = 0.80
|
||||
- Sample size: n = 94 per group (188 total)
|
||||
- With 10% dropout: n = 105 per group (210 total)
|
||||
```
|
||||
|
||||
## Data Visualization
|
||||
|
||||
### Survival Curves
|
||||
|
||||
**Kaplan-Meier Plot Best Practices**
|
||||
|
||||
```python
|
||||
# Key elements for publication-quality survival curve
|
||||
1. X-axis: Time (months or years), starts at 0
|
||||
2. Y-axis: Survival probability (0 to 1.0 or 0% to 100%)
|
||||
3. Step function: Survival curve with steps at event times
|
||||
4. 95% CI bands: Shaded region around survival curve (optional but recommended)
|
||||
5. Number at risk table: Below x-axis showing n at risk at time intervals
|
||||
6. Censoring marks: Vertical tick marks (|) at censored observations
|
||||
7. Legend: Clearly identify each curve
|
||||
8. Log-rank p-value: Prominently displayed
|
||||
9. Median survival: Horizontal line at 0.50, labeled
|
||||
10. Follow-up: Median follow-up time reported
|
||||
```
|
||||
|
||||
**Number at Risk Table Format**
|
||||
```
|
||||
Number at risk
|
||||
Group A 50 42 35 28 18 10 5
|
||||
Group B 48 38 29 19 12 6 2
|
||||
Time 0 6 12 18 24 30 36 (months)
|
||||
```
|
||||
|
||||
**Hazard Ratio Annotation**
|
||||
```
|
||||
On plot: HR 0.62 (95% CI 0.43-0.89), p=0.010
|
||||
Or in caption: Log-rank test p=0.010; Cox model HR=0.62 (95% CI 0.43-0.89)
|
||||
```
|
||||
|
||||
### Waterfall Plots
|
||||
|
||||
**Purpose**: Visualize individual patient responses to treatment
|
||||
|
||||
**Construction**
|
||||
- **X-axis**: Individual patients (anonymized patient IDs)
|
||||
- **Y-axis**: Best % change from baseline tumor burden
|
||||
- **Bars**: Vertical bars, one per patient
|
||||
- Positive values: Tumor growth
|
||||
- Negative values: Tumor shrinkage
|
||||
- **Ordering**: Sorted from best response (left) to worst (right)
|
||||
- **Color coding**:
|
||||
- Green/blue: CR or PR (≥30% decrease)
|
||||
- Yellow: SD (-30% to +20%)
|
||||
- Red: PD (≥20% increase)
|
||||
- **Reference lines**: Horizontal lines at +20% (PD), -30% (PR)
|
||||
- **Annotations**: Biomarker status, response duration (symbols)
|
||||
|
||||
**Example Annotations**
|
||||
```
|
||||
■ = Biomarker-positive
|
||||
○ = Biomarker-negative
|
||||
* = Ongoing response
|
||||
† = Progressed
|
||||
```
|
||||
|
||||
### Forest Plots
|
||||
|
||||
**Purpose**: Display subgroup analyses with hazard ratios and confidence intervals
|
||||
|
||||
**Construction**
|
||||
- **Y-axis**: Subgroup categories
|
||||
- **X-axis**: Hazard ratio (log scale), vertical line at HR=1.0
|
||||
- **Points**: HR estimate for each subgroup
|
||||
- **Horizontal lines**: 95% confidence interval
|
||||
- **Square size**: Proportional to sample size or precision
|
||||
- **Overall effect**: Diamond at bottom, width represents 95% CI
|
||||
|
||||
**Subgroups to Display**
|
||||
```
|
||||
Subgroup n HR (95% CI) Favors A Favors B
|
||||
──────────────────────────────────────────────────────────────────────────
|
||||
Overall 300 0.65 (0.48-0.88) ●────┤
|
||||
Age
|
||||
<65 years 180 0.58 (0.39-0.86) ●────┤
|
||||
≥65 years 120 0.78 (0.49-1.24) ●──────┤
|
||||
Sex
|
||||
Male 175 0.62 (0.43-0.90) ●────┤
|
||||
Female 125 0.70 (0.44-1.12) ●─────┤
|
||||
Biomarker Status
|
||||
Positive 140 0.45 (0.28-0.72) ●───┤
|
||||
Negative 160 0.89 (0.59-1.34) ●──────┤
|
||||
p-interaction=0.041
|
||||
|
||||
0.25 0.5 1.0 2.0
|
||||
Hazard Ratio
|
||||
```
|
||||
|
||||
**Interaction Testing**
|
||||
- Test whether treatment effect differs across subgroups
|
||||
- p-interaction <0.05 suggests heterogeneity
|
||||
- Pre-specify subgroups to avoid data mining
|
||||
|
||||
### Spider Plots
|
||||
|
||||
**Purpose**: Display longitudinal tumor burden changes over time for individual patients
|
||||
|
||||
**Construction**
|
||||
- **X-axis**: Time from treatment start (weeks or months)
|
||||
- **Y-axis**: % change from baseline tumor burden
|
||||
- **Lines**: One line per patient connecting assessments
|
||||
- **Color coding**: By response category or biomarker status
|
||||
- **Reference lines**: 0% (no change), +20% (PD threshold), -30% (PR threshold)
|
||||
|
||||
**Clinical Insights**
|
||||
- Identify delayed responders (initial SD then PR)
|
||||
- Detect early progression (rapid upward trajectory)
|
||||
- Assess depth of response (maximum tumor shrinkage)
|
||||
- Duration visualization (when lines cross PD threshold)
|
||||
|
||||
### Swimmer Plots
|
||||
|
||||
**Purpose**: Display treatment duration and response for individual patients
|
||||
|
||||
**Construction**
|
||||
- **X-axis**: Time from treatment start (weeks or months)
|
||||
- **Y-axis**: Individual patients (one row per patient)
|
||||
- **Bars**: Horizontal bars representing treatment duration
|
||||
- **Symbols**:
|
||||
- ● Start of treatment
|
||||
- ▼ Ongoing treatment (arrow)
|
||||
- ■ Progressive disease (end of bar)
|
||||
- ◆ Death
|
||||
- | Dose modification
|
||||
- **Color**: Response status (CR=green, PR=blue, SD=yellow, PD=red)
|
||||
|
||||
**Example**
|
||||
```
|
||||
Patient ID |0 3 6 9 12 15 18 21 24 months
|
||||
──────────────|──────────────────────────────────────────
|
||||
Pt-001 ●═══PR═══════════|════════PR══════════▼
|
||||
Pt-002 ●═══PR═══════════════PD■
|
||||
Pt-003 ●══════SD══════════PD■
|
||||
Pt-004 ●PR══════════════════════════════════PR▼
|
||||
...
|
||||
```
|
||||
|
||||
## Confidence Intervals
|
||||
|
||||
### Interpretation
|
||||
|
||||
**95% Confidence Interval**
|
||||
- Range of plausible values for true population parameter
|
||||
- If study repeated 100 times, 95 of the 95% CIs would contain true value
|
||||
- **Not**: 95% probability true value within this interval (frequentist, not Bayesian)
|
||||
|
||||
**Relationship to p-value**
|
||||
- If 95% CI excludes null value (HR=1.0, difference=0), p<0.05
|
||||
- If 95% CI includes null value, p≥0.05
|
||||
- CI provides more information: magnitude and precision of effect
|
||||
|
||||
**Precision**
|
||||
- **Narrow CI**: High precision, large sample size
|
||||
- **Wide CI**: Low precision, small sample size or high variability
|
||||
- **Example**: HR 0.65 (95% CI 0.62-0.68) very precise; HR 0.65 (0.30-1.40) imprecise
|
||||
|
||||
### Calculation Methods
|
||||
|
||||
**Hazard Ratio CI**
|
||||
- From Cox regression output
|
||||
- Standard error of log(HR) → exp(log(HR) ± 1.96×SE)
|
||||
- Example: HR=0.62, SE(logHR)=0.185 → 95% CI (0.43, 0.89)
|
||||
|
||||
**Survival Rate CI (Greenwood Formula)**
|
||||
- SE(S(t)) = S(t) × sqrt(Σ[d_i / (n_i × (n_i - d_i))])
|
||||
- 95% CI: S(t) ± 1.96 × SE(S(t))
|
||||
- Can use complementary log-log transformation for better properties
|
||||
|
||||
**Proportion CI (Exact Binomial)**
|
||||
- For ORR, DCR: Use exact method (Clopper-Pearson) for small samples
|
||||
- Wilson score interval: Better properties than normal approximation
|
||||
- Example: 12/30 responses → ORR 40% (95% CI 22.7-59.4%)
|
||||
|
||||
## Censoring and Missing Data
|
||||
|
||||
### Types of Censoring
|
||||
|
||||
**Right Censoring**
|
||||
- **End of study**: Patient alive at study termination (administrative censoring)
|
||||
- **Loss to follow-up**: Patient stops attending visits
|
||||
- **Withdrawal**: Patient withdraws consent
|
||||
- **Competing risk**: Death from unrelated cause (in disease-specific survival)
|
||||
|
||||
**Handling Censoring**
|
||||
- **Assumption**: Non-informative - censoring independent of event probability
|
||||
- **Sensitivity Analysis**: Assess impact if assumption violated
|
||||
- Best case: All censored patients never progress
|
||||
- Worst case: All censored patients progress immediately after censoring
|
||||
- Actual result should fall between best/worst case
|
||||
|
||||
### Missing Data
|
||||
|
||||
**Mechanisms**
|
||||
- **MCAR (Missing Completely at Random)**: Missingness unrelated to any variable
|
||||
- **MAR (Missing at Random)**: Missingness related to observed but not unobserved variables
|
||||
- **NMAR (Not Missing at Random)**: Missingness related to the missing value itself
|
||||
|
||||
**Handling Strategies**
|
||||
- **Complete case analysis**: Exclude patients with missing data (biased if not MCAR)
|
||||
- **Multiple imputation**: Generate multiple plausible datasets, analyze each, pool results
|
||||
- **Maximum likelihood**: Estimate parameters using all available data
|
||||
- **Sensitivity analysis**: Assess robustness to missing data assumptions
|
||||
|
||||
**Response Assessment Missing Data**
|
||||
- **Unevaluable for response**: Baseline measurable disease but post-baseline assessment missing
|
||||
- Exclude from ORR denominator or count as non-responder (sensitivity analysis)
|
||||
- **PFS censoring**: Last adequate tumor assessment date if later assessments missing
|
||||
|
||||
## Reporting Standards
|
||||
|
||||
### CONSORT Statement (RCTs)
|
||||
|
||||
**Flow Diagram**
|
||||
- Assessed for eligibility (n=)
|
||||
- Randomized (n=)
|
||||
- Allocated to intervention (n=)
|
||||
- Lost to follow-up (n=, reasons)
|
||||
- Discontinued intervention (n=, reasons)
|
||||
- Analyzed (n=)
|
||||
|
||||
**Baseline Table**
|
||||
- Demographics and clinical characteristics
|
||||
- Baseline prognostic factors
|
||||
- Show balance between arms
|
||||
|
||||
**Outcomes Table**
|
||||
- Primary endpoint results with CI and p-value
|
||||
- Secondary endpoints
|
||||
- Safety summary
|
||||
|
||||
### STROBE Statement (Observational Studies)
|
||||
|
||||
**Study Design**: Cohort, case-control, or cross-sectional
|
||||
|
||||
**Participants**: Eligibility, sources, selection methods, sample size
|
||||
|
||||
**Variables**: Clearly define outcomes, exposures, predictors, confounders
|
||||
|
||||
**Statistical Methods**: Describe all methods, handling of missing data, sensitivity analyses
|
||||
|
||||
**Results**: Participant flow, descriptive data, outcome data, main results, other analyses
|
||||
|
||||
### Reproducible Research Practices
|
||||
|
||||
**Statistical Analysis Plan (SAP)**
|
||||
- Pre-specify all analyses before data lock
|
||||
- Primary and secondary endpoints
|
||||
- Analysis populations (ITT, per-protocol, safety)
|
||||
- Statistical tests and models
|
||||
- Subgroup analyses (pre-specified)
|
||||
- Interim analyses (if planned)
|
||||
- Multiple testing procedures
|
||||
|
||||
**Transparency**
|
||||
- Report all pre-specified analyses
|
||||
- Distinguish pre-specified from post-hoc exploratory
|
||||
- Report both positive and negative results
|
||||
- Provide access to anonymized individual patient data (when possible)
|
||||
|
||||
## Software and Tools
|
||||
|
||||
### R Packages for Survival Analysis
|
||||
- **survival**: Core package (Surv, survfit, coxph, survdiff)
|
||||
- **survminer**: Publication-ready Kaplan-Meier plots (ggsurvplot)
|
||||
- **rms**: Regression modeling strategies
|
||||
- **flexsurv**: Flexible parametric survival models
|
||||
|
||||
### Python Libraries
|
||||
- **lifelines**: Kaplan-Meier, Cox regression, survival curves
|
||||
- **scikit-survival**: Machine learning for survival analysis
|
||||
- **matplotlib**: Custom survival curve plotting
|
||||
|
||||
### Statistical Software
|
||||
- **R**: Most comprehensive for survival analysis
|
||||
- **Stata**: Medical statistics, good for epidemiology
|
||||
- **SAS**: Industry standard for clinical trials
|
||||
- **GraphPad Prism**: User-friendly for basic analyses
|
||||
- **SPSS**: Point-and-click interface, limited survival features
|
||||
|
||||
@@ -0,0 +1,427 @@
|
||||
# Patient Cohort Analysis Guide
|
||||
|
||||
## Overview
|
||||
|
||||
Patient cohort analysis involves systematically studying groups of patients to identify patterns, compare outcomes, and derive clinical insights. In pharmaceutical and clinical research settings, cohort analysis is essential for understanding treatment effectiveness, biomarker correlations, and patient stratification.
|
||||
|
||||
## Patient Stratification Methods
|
||||
|
||||
### Biomarker-Based Stratification
|
||||
|
||||
**Genomic Biomarkers**
|
||||
- **Mutations**: Driver mutations (EGFR, KRAS, BRAF), resistance mutations (T790M)
|
||||
- **Copy Number Variations**: Amplifications (HER2, MET), deletions (PTEN, RB1)
|
||||
- **Gene Fusions**: ALK, ROS1, NTRK, RET rearrangements
|
||||
- **Tumor Mutational Burden (TMB)**: High (≥10 mut/Mb) vs low TMB
|
||||
- **Microsatellite Instability**: MSI-high vs MSS/MSI-low
|
||||
|
||||
**Expression Biomarkers**
|
||||
- **IHC Scores**: PD-L1 TPS (<1%, 1-49%, ≥50%), HER2 (0, 1+, 2+, 3+)
|
||||
- **RNA Expression**: Gene signatures, pathway activity scores
|
||||
- **Protein Levels**: Ki-67 proliferation index, hormone receptors (ER/PR)
|
||||
|
||||
**Molecular Subtypes**
|
||||
- **Breast Cancer**: Luminal A, Luminal B, HER2-enriched, Triple-negative
|
||||
- **Glioblastoma**: Proneural, neural, classical, mesenchymal
|
||||
- **Lung Adenocarcinoma**: Terminal respiratory unit, proximal inflammatory, proximal proliferative
|
||||
- **Colorectal Cancer**: CMS1-4 (consensus molecular subtypes)
|
||||
|
||||
### Demographic Stratification
|
||||
|
||||
- **Age Groups**: Pediatric (<18), young adult (18-39), middle-age (40-64), elderly (65-79), very elderly (≥80)
|
||||
- **Sex/Gender**: Male, female, sex-specific biomarkers
|
||||
- **Race/Ethnicity**: FDA-recognized categories, ancestry-informative markers
|
||||
- **Geographic Location**: Regional variation in disease prevalence
|
||||
|
||||
### Clinical Stratification
|
||||
|
||||
**Disease Characteristics**
|
||||
- **Stage**: TNM staging (I, II, III, IV), Ann Arbor (lymphoma)
|
||||
- **Grade**: Well-differentiated (G1), moderately differentiated (G2), poorly differentiated (G3), undifferentiated (G4)
|
||||
- **Histology**: Adenocarcinoma vs squamous vs other subtypes
|
||||
- **Disease Burden**: Tumor volume, number of lesions, organ involvement
|
||||
|
||||
**Patient Status**
|
||||
- **Performance Status**: ECOG (0-4), Karnofsky (0-100)
|
||||
- **Comorbidities**: Charlson Comorbidity Index, organ dysfunction
|
||||
- **Prior Treatment**: Treatment-naïve, previously treated, lines of therapy
|
||||
- **Response to Prior Therapy**: Responders vs non-responders, progressive disease
|
||||
|
||||
### Risk Stratification
|
||||
|
||||
**Prognostic Scores**
|
||||
- **Cancer**: AJCC staging, Gleason score, Nottingham grade
|
||||
- **Cardiovascular**: Framingham risk, TIMI, GRACE, CHADS2-VASc
|
||||
- **Liver Disease**: Child-Pugh class, MELD score
|
||||
- **Renal Disease**: eGFR categories, albuminuria stages
|
||||
|
||||
**Composite Risk Models**
|
||||
- Low risk: Good prognosis, less aggressive treatment
|
||||
- Intermediate risk: Moderate prognosis, standard treatment
|
||||
- High risk: Poor prognosis, intensive treatment or clinical trials
|
||||
|
||||
## Cluster Analysis and Subgroup Identification
|
||||
|
||||
### Unsupervised Clustering
|
||||
|
||||
**Methods**
|
||||
- **K-means**: Partition-based clustering with pre-defined number of clusters
|
||||
- **Hierarchical Clustering**: Agglomerative or divisive, creates dendrogram
|
||||
- **DBSCAN**: Density-based clustering, identifies outliers
|
||||
- **Consensus Clustering**: Robust cluster identification across multiple runs
|
||||
|
||||
**Applications**
|
||||
- Molecular subtype discovery (e.g., GBM mesenchymal-immune-active cluster)
|
||||
- Patient phenotype identification
|
||||
- Treatment response patterns
|
||||
- Multi-omic data integration
|
||||
|
||||
### Supervised Classification
|
||||
|
||||
**Approaches**
|
||||
- **Pre-defined Criteria**: Clinical guidelines, established biomarker cut-points
|
||||
- **Machine Learning**: Random forests, support vector machines for prediction
|
||||
- **Neural Networks**: Deep learning for complex pattern recognition
|
||||
- **Validated Signatures**: Published gene expression panels (Oncotype DX, MammaPrint)
|
||||
|
||||
### Validation Requirements
|
||||
|
||||
- **Internal Validation**: Cross-validation, bootstrap resampling
|
||||
- **External Validation**: Independent cohort confirmation
|
||||
- **Clinical Validation**: Prospective trial confirmation of utility
|
||||
- **Analytical Validation**: Assay reproducibility, inter-lab concordance
|
||||
|
||||
## Outcome Metrics
|
||||
|
||||
### Survival Endpoints
|
||||
|
||||
**Overall Survival (OS)**
|
||||
- Definition: Time from treatment start (or randomization) to death from any cause
|
||||
- Censoring: Last known alive date for patients lost to follow-up
|
||||
- Reporting: Median OS, 1-year/2-year/5-year OS rates, hazard ratio
|
||||
- Gold Standard: Primary endpoint for regulatory approval
|
||||
|
||||
**Progression-Free Survival (PFS)**
|
||||
- Definition: Time from treatment start to disease progression or death
|
||||
- Assessment: RECIST v1.1, iRECIST (for immunotherapy)
|
||||
- Advantages: Earlier readout than OS, direct measure of treatment benefit
|
||||
- Limitations: Requires imaging, subject to assessment timing
|
||||
|
||||
**Disease-Free Survival (DFS)**
|
||||
- Definition: Time from complete response to recurrence or death (adjuvant setting)
|
||||
- Application: Post-surgery, post-curative treatment
|
||||
- Synonyms: Recurrence-free survival (RFS), event-free survival (EFS)
|
||||
|
||||
### Response Endpoints
|
||||
|
||||
**Objective Response Rate (ORR)**
|
||||
- Definition: Proportion achieving complete response (CR) or partial response (PR)
|
||||
- Measurement: RECIST v1.1 criteria (≥30% tumor shrinkage for PR)
|
||||
- Reporting: ORR with 95% confidence interval
|
||||
- Advantage: Earlier endpoint than survival
|
||||
|
||||
**Duration of Response (DOR)**
|
||||
- Definition: Time from first response (CR/PR) to progression
|
||||
- Population: Responders only
|
||||
- Clinical Relevance: Durability of treatment benefit
|
||||
- Reporting: Median DOR among responders
|
||||
|
||||
**Disease Control Rate (DCR)**
|
||||
- Definition: CR + PR + stable disease (SD)
|
||||
- Threshold: SD must persist ≥6-8 weeks typically
|
||||
- Application: Less stringent than ORR, captures clinical benefit
|
||||
|
||||
### Quality of Life and Functional Status
|
||||
|
||||
**Performance Status**
|
||||
- **ECOG Scale**: 0 (fully active) to 4 (bedridden)
|
||||
- **Karnofsky Scale**: 100% (normal) to 0% (dead)
|
||||
- **Assessment Frequency**: Baseline and each cycle
|
||||
|
||||
**Patient-Reported Outcomes (PROs)**
|
||||
- **Symptom Scales**: EORTC QLQ-C30, FACT-G
|
||||
- **Disease-Specific**: FACT-L (lung), FACT-B (breast)
|
||||
- **Toxicity**: PRO-CTCAE for adverse events
|
||||
- **Reporting**: Change from baseline, clinically meaningful differences
|
||||
|
||||
### Safety and Tolerability
|
||||
|
||||
**Adverse Events (AEs)**
|
||||
- **Grading**: CTCAE v5.0 (Grade 1-5)
|
||||
- **Attribution**: Related vs unrelated to treatment
|
||||
- **Serious AEs (SAEs)**: Death, life-threatening, hospitalization, disability
|
||||
- **Reporting**: Incidence, severity, time to onset, resolution
|
||||
|
||||
**Treatment Modifications**
|
||||
- **Dose Reductions**: Proportion requiring dose decrease
|
||||
- **Dose Delays**: Treatment interruptions, cycle delays
|
||||
- **Discontinuations**: Treatment termination due to toxicity
|
||||
- **Relative Dose Intensity**: Actual dose / planned dose ratio
|
||||
|
||||
## Statistical Methods for Group Comparisons
|
||||
|
||||
### Continuous Variables
|
||||
|
||||
**Parametric Tests (Normal Distribution)**
|
||||
- **Two Groups**: Independent t-test, paired t-test
|
||||
- **Multiple Groups**: ANOVA (analysis of variance), repeated measures ANOVA
|
||||
- **Reporting**: Mean ± SD, mean difference with 95% CI, p-value
|
||||
|
||||
**Non-Parametric Tests (Non-Normal Distribution)**
|
||||
- **Two Groups**: Mann-Whitney U test (Wilcoxon rank-sum)
|
||||
- **Paired Data**: Wilcoxon signed-rank test
|
||||
- **Multiple Groups**: Kruskal-Wallis test
|
||||
- **Reporting**: Median [IQR], median difference, p-value
|
||||
|
||||
### Categorical Variables
|
||||
|
||||
**Chi-Square Test**
|
||||
- **Application**: Compare proportions between ≥2 groups
|
||||
- **Assumptions**: Expected count ≥5 in each cell
|
||||
- **Reporting**: Proportions, chi-square statistic, df, p-value
|
||||
|
||||
**Fisher's Exact Test**
|
||||
- **Application**: 2x2 tables with small sample sizes (expected count <5)
|
||||
- **Advantage**: Exact p-value, no large-sample approximation
|
||||
- **Limitation**: Computationally intensive for large tables
|
||||
|
||||
### Survival Analysis
|
||||
|
||||
**Kaplan-Meier Method**
|
||||
- **Application**: Estimate survival curves with censored data
|
||||
- **Output**: Survival probability at each time point, median survival
|
||||
- **Visualization**: Step function curves with 95% CI bands
|
||||
|
||||
**Log-Rank Test**
|
||||
- **Application**: Compare survival curves between groups
|
||||
- **Null Hypothesis**: No difference in survival distributions
|
||||
- **Reporting**: Chi-square statistic, df, p-value
|
||||
- **Limitation**: Assumes proportional hazards
|
||||
|
||||
**Cox Proportional Hazards Model**
|
||||
- **Application**: Multivariable survival analysis
|
||||
- **Output**: Hazard ratio (HR) with 95% CI for each covariate
|
||||
- **Interpretation**: HR > 1 (increased risk), HR < 1 (decreased risk)
|
||||
- **Assumptions**: Proportional hazards (test with Schoenfeld residuals)
|
||||
|
||||
### Effect Sizes
|
||||
|
||||
**Hazard Ratio (HR)**
|
||||
- Definition: Ratio of hazard rates between groups
|
||||
- Interpretation: HR = 0.5 means 50% reduction in risk
|
||||
- Reporting: HR (95% CI), p-value
|
||||
- Example: HR = 0.65 (0.52-0.81), p<0.001
|
||||
|
||||
**Odds Ratio (OR)**
|
||||
- Application: Case-control studies, logistic regression
|
||||
- Interpretation: OR > 1 (increased odds), OR < 1 (decreased odds)
|
||||
- Reporting: OR (95% CI), p-value
|
||||
|
||||
**Risk Ratio (RR) / Relative Risk**
|
||||
- Application: Cohort studies, clinical trials
|
||||
- Interpretation: RR = 2.0 means 2-fold increased risk
|
||||
- More intuitive than OR for interpreting probabilities
|
||||
|
||||
### Multiple Testing Corrections
|
||||
|
||||
**Bonferroni Correction**
|
||||
- Method: Divide α by number of tests (α/n)
|
||||
- Example: 5 tests → significance threshold = 0.05/5 = 0.01
|
||||
- Conservative: Reduces Type I error but increases Type II error
|
||||
|
||||
**False Discovery Rate (FDR)**
|
||||
- Method: Benjamini-Hochberg procedure
|
||||
- Interpretation: Expected proportion of false positives among significant results
|
||||
- Less Conservative: More power than Bonferroni
|
||||
|
||||
**Family-Wise Error Rate (FWER)**
|
||||
- Method: Control probability of any false positive
|
||||
- Application: When even one false positive is problematic
|
||||
- Examples: Bonferroni, Holm-Bonferroni
|
||||
|
||||
## Biomarker Correlation with Outcomes
|
||||
|
||||
### Predictive Biomarkers
|
||||
|
||||
**Definition**: Biomarkers that identify patients likely to respond to a specific treatment
|
||||
|
||||
**Examples**
|
||||
- **PD-L1 ≥50%**: Predicts response to pembrolizumab monotherapy (NSCLC)
|
||||
- **HER2 3+**: Predicts response to trastuzumab (breast cancer)
|
||||
- **EGFR mutations**: Predicts response to EGFR TKIs (lung cancer)
|
||||
- **BRAF V600E**: Predicts response to vemurafenib (melanoma)
|
||||
- **MSI-H/dMMR**: Predicts response to immune checkpoint inhibitors
|
||||
|
||||
**Analysis**
|
||||
- Stratified analysis: Compare treatment effect within biomarker-positive vs negative
|
||||
- Interaction test: Test if treatment effect differs by biomarker status
|
||||
- Reporting: HR in biomarker+ vs biomarker-, interaction p-value
|
||||
|
||||
### Prognostic Biomarkers
|
||||
|
||||
**Definition**: Biomarkers that predict outcome regardless of treatment
|
||||
|
||||
**Examples**
|
||||
- **High Ki-67**: Poor prognosis independent of treatment (breast cancer)
|
||||
- **TP53 mutation**: Poor prognosis in many cancers
|
||||
- **Low albumin**: Poor prognosis marker (many diseases)
|
||||
- **Elevated LDH**: Poor prognosis (melanoma, lymphoma)
|
||||
|
||||
**Analysis**
|
||||
- Compare outcomes across biomarker levels in untreated or uniformly treated cohort
|
||||
- Multivariable Cox model adjusting for other prognostic factors
|
||||
- Validate in independent cohorts
|
||||
|
||||
### Continuous Biomarker Analysis
|
||||
|
||||
**Cut-Point Selection**
|
||||
- **Data-Driven**: Maximally selected rank statistics, ROC curve analysis
|
||||
- **Literature-Based**: Established clinical cut-points
|
||||
- **Median/Tertiles**: Simple divisions for exploration
|
||||
- **Validation**: Cut-points must be validated in independent cohort
|
||||
|
||||
**Continuous Analysis**
|
||||
- Treat biomarker as continuous variable in Cox model
|
||||
- Report HR per unit increase or per standard deviation
|
||||
- Spline curves to assess non-linear relationships
|
||||
- Advantage: No information loss from dichotomization
|
||||
|
||||
## Data Presentation
|
||||
|
||||
### Baseline Characteristics Table (Table 1)
|
||||
|
||||
**Standard Format**
|
||||
```
|
||||
Characteristic Group A (n=50) Group B (n=45) p-value
|
||||
Age, years (median [IQR]) 62 [54-68] 59 [52-66] 0.34
|
||||
Sex, n (%)
|
||||
Male 30 (60%) 28 (62%) 0.82
|
||||
Female 20 (40%) 17 (38%)
|
||||
ECOG PS, n (%)
|
||||
0-1 42 (84%) 39 (87%) 0.71
|
||||
2 8 (16%) 6 (13%)
|
||||
Biomarker+, n (%) 23 (46%) 21 (47%) 0.94
|
||||
```
|
||||
|
||||
**Key Principles**
|
||||
- Report all clinically relevant baseline variables
|
||||
- Use appropriate summary statistics (mean±SD for normal, median[IQR] for skewed)
|
||||
- Include sample size for each group
|
||||
- Report p-values for group comparisons (but baseline imbalances expected by chance)
|
||||
- Do NOT adjust baseline p-values for multiple testing
|
||||
|
||||
### Efficacy Outcomes Table
|
||||
|
||||
**Response Outcomes**
|
||||
```
|
||||
Outcome Group A (n=50) Group B (n=45) p-value
|
||||
ORR, n (%) [95% CI] 25 (50%) [36-64] 15 (33%) [20-48] 0.08
|
||||
Complete Response 3 (6%) 1 (2%)
|
||||
Partial Response 22 (44%) 14 (31%)
|
||||
DCR, n (%) [95% CI] 40 (80%) [66-90] 35 (78%) [63-89] 0.79
|
||||
Median DOR, months (95% CI) 8.2 (6.1-11.3) 6.8 (4.9-9.7) 0.12
|
||||
```
|
||||
|
||||
**Survival Outcomes**
|
||||
```
|
||||
Endpoint Group A Group B HR (95% CI) p-value
|
||||
Median PFS, months (95% CI) 10.2 (8.3-12.1) 6.5 (5.1-7.9) 0.62 (0.41-0.94) 0.02
|
||||
12-month PFS rate 42% 28%
|
||||
Median OS, months (95% CI) 21.3 (17.8-NR) 15.7 (12.4-19.1) 0.71 (0.45-1.12) 0.14
|
||||
12-month OS rate 68% 58%
|
||||
```
|
||||
|
||||
### Safety and Tolerability Table
|
||||
|
||||
**Adverse Events**
|
||||
```
|
||||
Adverse Event Any Grade, n (%) Grade 3-4, n (%)
|
||||
Group A Group B Group A Group B
|
||||
Fatigue 35 (70%) 32 (71%) 3 (6%) 2 (4%)
|
||||
Nausea 28 (56%) 25 (56%) 1 (2%) 1 (2%)
|
||||
Neutropenia 15 (30%) 18 (40%) 8 (16%) 10 (22%)
|
||||
Thrombocytopenia 12 (24%) 14 (31%) 4 (8%) 6 (13%)
|
||||
Hepatotoxicity 8 (16%) 6 (13%) 2 (4%) 1 (2%)
|
||||
Treatment discontinuation 6 (12%) 8 (18%) - -
|
||||
```
|
||||
|
||||
### Visualization Formats
|
||||
|
||||
**Survival Curves**
|
||||
- Kaplan-Meier plots with 95% CI bands
|
||||
- Number at risk table below x-axis
|
||||
- Log-rank p-value and HR prominently displayed
|
||||
- Clear legend identifying groups
|
||||
|
||||
**Forest Plots**
|
||||
- Subgroup analysis showing HR with 95% CI for each subgroup
|
||||
- Test for interaction assessing heterogeneity
|
||||
- Overall effect at bottom
|
||||
|
||||
**Waterfall Plots**
|
||||
- Individual patient best response (% change from baseline)
|
||||
- Ordered from best to worst response
|
||||
- Color-coded by response category (CR, PR, SD, PD)
|
||||
- Biomarker status annotation
|
||||
|
||||
**Swimmer Plots**
|
||||
- Time on treatment for each patient
|
||||
- Response duration for responders
|
||||
- Treatment modifications marked
|
||||
- Ongoing treatments indicated with arrow
|
||||
|
||||
## Quality Control and Validation
|
||||
|
||||
### Data Quality Checks
|
||||
|
||||
- **Completeness**: Missing data patterns, loss to follow-up
|
||||
- **Consistency**: Cross-field validation, logical checks
|
||||
- **Outliers**: Identify and investigate extreme values
|
||||
- **Duplicates**: Patient ID verification, enrollment checks
|
||||
|
||||
### Statistical Assumptions
|
||||
|
||||
- **Normality**: Shapiro-Wilk test, Q-Q plots for continuous variables
|
||||
- **Proportional Hazards**: Schoenfeld residuals for Cox models
|
||||
- **Independence**: Check for clustering, matched data
|
||||
- **Missing Data**: Assess mechanism (MCAR, MAR, NMAR), handle appropriately
|
||||
|
||||
### Reporting Standards
|
||||
|
||||
- **CONSORT**: Randomized controlled trials
|
||||
- **STROBE**: Observational studies
|
||||
- **REMARK**: Tumor marker prognostic studies
|
||||
- **STARD**: Diagnostic accuracy studies
|
||||
- **TRIPOD**: Prediction model development/validation
|
||||
|
||||
## Clinical Interpretation
|
||||
|
||||
### Translating Statistics to Clinical Meaning
|
||||
|
||||
**Statistical Significance vs Clinical Significance**
|
||||
- p<0.05 does not guarantee clinical importance
|
||||
- Small effects can be statistically significant with large samples
|
||||
- Large effects can be non-significant with small samples
|
||||
- Consider effect size magnitude and confidence interval width
|
||||
|
||||
**Number Needed to Treat (NNT)**
|
||||
- NNT = 1 / absolute risk reduction
|
||||
- Example: 10% vs 5% event rate → ARR = 5% → NNT = 20
|
||||
- Interpretation: Treat 20 patients to prevent 1 event
|
||||
- Useful for communicating treatment benefit
|
||||
|
||||
**Minimal Clinically Important Difference (MCID)**
|
||||
- Pre-defined threshold for meaningful clinical benefit
|
||||
- OS: Often 2-3 months in oncology
|
||||
- PFS: Context-dependent, often 1.5-3 months
|
||||
- QoL: 10-point change on 100-point scale
|
||||
- Response rate: Often 10-15 percentage point difference
|
||||
|
||||
### Contextualization
|
||||
|
||||
- Compare to historical controls or standard of care
|
||||
- Consider patient population characteristics
|
||||
- Account for prior treatment exposure
|
||||
- Evaluate toxicity trade-offs
|
||||
- Assess quality of life impact
|
||||
|
||||
@@ -0,0 +1,521 @@
|
||||
# Treatment Recommendations Guide
|
||||
|
||||
## Overview
|
||||
|
||||
Evidence-based treatment recommendations provide clinicians with systematic guidance for therapeutic decision-making. This guide covers the development, grading, and presentation of clinical recommendations in pharmaceutical and healthcare settings.
|
||||
|
||||
## Evidence Grading Systems
|
||||
|
||||
### GRADE (Grading of Recommendations Assessment, Development and Evaluation)
|
||||
|
||||
**Quality of Evidence Levels**
|
||||
|
||||
**High Quality (⊕⊕⊕⊕)**
|
||||
- Further research very unlikely to change confidence in estimate
|
||||
- Criteria: Well-designed RCTs with consistent results, no serious limitations
|
||||
- Example: Multiple large RCTs showing similar treatment effects
|
||||
|
||||
**Moderate Quality (⊕⊕⊕○)**
|
||||
- Further research likely to have important impact on confidence
|
||||
- Criteria: RCTs with limitations OR very strong evidence from observational studies
|
||||
- Example: Single RCT or multiple RCTs with some inconsistency
|
||||
|
||||
**Low Quality (⊕⊕○○)**
|
||||
- Further research very likely to have important impact on confidence
|
||||
- Criteria: Observational studies OR RCTs with serious limitations
|
||||
- Example: Case-control studies, cohort studies with confounding
|
||||
|
||||
**Very Low Quality (⊕○○○)**
|
||||
- Estimate of effect very uncertain
|
||||
- Criteria: Case series, expert opinion, or very serious limitations
|
||||
- Example: Mechanistic reasoning, unsystematic clinical observations
|
||||
|
||||
**Strength of Recommendation**
|
||||
|
||||
**Strong Recommendation (Grade 1)**
|
||||
- Benefits clearly outweigh risks and burdens (or vice versa)
|
||||
- Wording: "We recommend..."
|
||||
- Implications: Most patients should receive recommended course
|
||||
- Symbol: ↑↑ (strong for) or ↓↓ (strong against)
|
||||
|
||||
**Conditional/Weak Recommendation (Grade 2)**
|
||||
- Trade-offs exist; benefits and risks closely balanced
|
||||
- Wording: "We suggest..."
|
||||
- Implications: Different choices for different patients; shared decision-making
|
||||
- Symbol: ↑ (weak for) or ↓ (weak against)
|
||||
|
||||
**GRADE Notation Examples**
|
||||
- **1A**: Strong recommendation, high-quality evidence
|
||||
- **1B**: Strong recommendation, moderate-quality evidence
|
||||
- **2A**: Weak recommendation, high-quality evidence
|
||||
- **2B**: Weak recommendation, moderate-quality evidence
|
||||
- **2C**: Weak recommendation, low- or very low-quality evidence
|
||||
|
||||
### Oxford Centre for Evidence-Based Medicine (CEBM) Levels
|
||||
|
||||
**Level 1: Systematic Review/Meta-Analysis**
|
||||
- 1a: SR of RCTs
|
||||
- 1b: Individual RCT with narrow confidence interval
|
||||
- 1c: All-or-none studies (all patients died before treatment, some survive after)
|
||||
|
||||
**Level 2: Cohort Studies**
|
||||
- 2a: SR of cohort studies
|
||||
- 2b: Individual cohort study (including low-quality RCT)
|
||||
- 2c: Outcomes research, ecological studies
|
||||
|
||||
**Level 3: Case-Control Studies**
|
||||
- 3a: SR of case-control studies
|
||||
- 3b: Individual case-control study
|
||||
|
||||
**Level 4: Case Series**
|
||||
- Case series, poor-quality cohort, or case-control studies
|
||||
|
||||
**Level 5: Expert Opinion**
|
||||
- Mechanism-based reasoning, expert opinion without critical appraisal
|
||||
|
||||
**Grades of Recommendation**
|
||||
- **Grade A**: Consistent level 1 studies
|
||||
- **Grade B**: Consistent level 2 or 3 studies, or extrapolations from level 1
|
||||
- **Grade C**: Level 4 studies or extrapolations from level 2 or 3
|
||||
- **Grade D**: Level 5 evidence or inconsistent/inconclusive studies
|
||||
|
||||
## Treatment Sequencing and Line-of-Therapy
|
||||
|
||||
### First-Line Therapy
|
||||
|
||||
**Selection Criteria**
|
||||
- **Standard of Care**: Guideline-recommended based on phase 3 trials
|
||||
- **Patient Factors**: Performance status, comorbidities, organ function
|
||||
- **Disease Factors**: Stage, molecular profile, aggressiveness
|
||||
- **Goals**: Cure (adjuvant/neoadjuvant), prolonged remission, symptom control
|
||||
|
||||
**First-Line Options Documentation**
|
||||
```
|
||||
First-Line Treatment Options:
|
||||
|
||||
Option 1: Regimen A (NCCN Category 1, ESMO I-A)
|
||||
- Evidence: Phase 3 RCT (n=1000), median PFS 12 months vs 8 months (HR 0.6, p<0.001)
|
||||
- Population: PD-L1 ≥50%, EGFR/ALK negative
|
||||
- Toxicity Profile: Immune-related AEs (15% grade 3-4)
|
||||
- Recommendation Strength: 1A (strong, high-quality evidence)
|
||||
|
||||
Option 2: Regimen B (NCCN Category 1, ESMO I-A)
|
||||
- Evidence: Phase 3 RCT (n=800), median PFS 10 months vs 8 months (HR 0.7, p=0.003)
|
||||
- Population: All patients, no biomarker selection
|
||||
- Toxicity Profile: Hematologic toxicity (25% grade 3-4)
|
||||
- Recommendation Strength: 1A (strong, high-quality evidence)
|
||||
```
|
||||
|
||||
### Second-Line and Beyond
|
||||
|
||||
**Second-Line Selection**
|
||||
- **Prior Response**: Duration of response to first-line
|
||||
- **Progression Pattern**: Oligoprogression vs widespread progression
|
||||
- **Residual Toxicity**: Recovery from first-line toxicities
|
||||
- **Biomarker Evolution**: Acquired resistance mechanisms
|
||||
- **Clinical Trial Availability**: Novel agents in development
|
||||
|
||||
**Treatment History Documentation**
|
||||
```
|
||||
Prior Therapies:
|
||||
1. First-Line: Pembrolizumab (12 cycles)
|
||||
- Best Response: Partial response (-45% tumor burden)
|
||||
- PFS: 14 months
|
||||
- Discontinuation Reason: Progressive disease
|
||||
- Residual Toxicity: Grade 1 hypothyroidism (on levothyroxine)
|
||||
|
||||
2. Second-Line: Docetaxel + ramucirumab (6 cycles)
|
||||
- Best Response: Stable disease
|
||||
- PFS: 5 months
|
||||
- Discontinuation Reason: Progressive disease
|
||||
- Residual Toxicity: Grade 2 peripheral neuropathy
|
||||
|
||||
Current Consideration: Third-Line Options
|
||||
- Clinical trial vs platinum-based chemotherapy
|
||||
```
|
||||
|
||||
### Maintenance Therapy
|
||||
|
||||
**Indications**
|
||||
- Consolidation after response to induction therapy
|
||||
- Prevention of progression without continuous cytotoxic treatment
|
||||
- Bridging to definitive therapy (e.g., transplant)
|
||||
|
||||
**Evidence Requirements**
|
||||
- PFS benefit demonstrated in randomized trials
|
||||
- Tolerable long-term toxicity profile
|
||||
- Quality of life preserved or improved
|
||||
|
||||
## Biomarker-Guided Therapy Selection
|
||||
|
||||
### Companion Diagnostics
|
||||
|
||||
**FDA-Approved Biomarker-Drug Pairs**
|
||||
|
||||
**Required Testing (Treatment-Specific)**
|
||||
- **ALK rearrangement → Alectinib, Brigatinib, Lorlatinib** (NSCLC)
|
||||
- **EGFR exon 19 del/L858R → Osimertinib** (NSCLC)
|
||||
- **BRAF V600E → Dabrafenib + Trametinib** (Melanoma, NSCLC, CRC)
|
||||
- **HER2 amplification/3+ → Trastuzumab, Pertuzumab** (Breast, Gastric)
|
||||
- **PD-L1 ≥50% → Pembrolizumab monotherapy** (NSCLC first-line)
|
||||
|
||||
**Complementary Diagnostics (Informative but not Required)**
|
||||
- **PD-L1 1-49%**: Combination immunotherapy preferred
|
||||
- **TMB-high**: May predict immunotherapy benefit (investigational)
|
||||
- **MSI-H/dMMR**: Pembrolizumab approved across tumor types
|
||||
|
||||
### Biomarker Testing Algorithms
|
||||
|
||||
**NSCLC Biomarker Panel**
|
||||
```
|
||||
Reflex Testing at Diagnosis:
|
||||
✓ EGFR mutations (exons 18, 19, 20, 21)
|
||||
✓ ALK rearrangement (IHC or FISH)
|
||||
✓ ROS1 rearrangement (FISH or NGS)
|
||||
✓ BRAF V600E mutation
|
||||
✓ PD-L1 IHC (22C3 or SP263)
|
||||
✓ Consider: Comprehensive NGS panel
|
||||
|
||||
If EGFR+ on Osimertinib progression:
|
||||
✓ Liquid biopsy for T790M (if first/second-gen TKI)
|
||||
✓ Tissue biopsy for resistance mechanisms
|
||||
✓ MET amplification, HER2 amplification, SCLC transformation
|
||||
```
|
||||
|
||||
**Breast Cancer Biomarker Algorithm**
|
||||
```
|
||||
Initial Diagnosis:
|
||||
✓ ER/PR IHC
|
||||
✓ HER2 IHC and FISH (if 2+)
|
||||
✓ Ki-67 proliferation index
|
||||
|
||||
If Metastatic ER+/HER2-:
|
||||
✓ ESR1 mutations (liquid biopsy after progression on AI)
|
||||
✓ PIK3CA mutations (for alpelisib eligibility)
|
||||
✓ BRCA1/2 germline testing (for PARP inhibitor eligibility)
|
||||
✓ PD-L1 testing (if considering immunotherapy combinations)
|
||||
```
|
||||
|
||||
### Actionable Alterations
|
||||
|
||||
**Tier I: FDA-Approved Targeted Therapy**
|
||||
- Strong evidence from prospective trials
|
||||
- Guideline-recommended
|
||||
- Examples: EGFR exon 19 deletion, HER2 amplification, ALK fusion
|
||||
|
||||
**Tier II: Clinical Trial or Off-Label Use**
|
||||
- Emerging evidence, clinical trial preferred
|
||||
- Examples: NTRK fusion (larotrectinib), RET fusion (selpercatinib)
|
||||
|
||||
**Tier III: Biological Plausibility**
|
||||
- Preclinical evidence only
|
||||
- Clinical trial enrollment strongly recommended
|
||||
- Examples: Novel kinase fusions, rare resistance mutations
|
||||
|
||||
## Combination Therapy Protocols
|
||||
|
||||
### Rationale for Combinations
|
||||
|
||||
**Mechanisms**
|
||||
- **Non-Overlapping Toxicity**: Maximize dose intensity of each agent
|
||||
- **Synergistic Activity**: Enhanced efficacy beyond additive effects
|
||||
- **Complementary Mechanisms**: Target multiple pathways simultaneously
|
||||
- **Prevent Resistance**: Decrease selection pressure for resistant clones
|
||||
|
||||
**Combination Design Principles**
|
||||
- **Sequential**: Induction then consolidation (different regimens)
|
||||
- **Concurrent**: Administered together for synergy
|
||||
- **Alternating**: Rotate regimens to minimize resistance
|
||||
- **Intermittent**: Pulse dosing vs continuous exposure
|
||||
|
||||
### Drug Interaction Assessment
|
||||
|
||||
**Pharmacokinetic Interactions**
|
||||
- **CYP450 Induction/Inhibition**: Check for drug-drug interactions
|
||||
- **Transporter Interactions**: P-gp, BCRP, OATP substrates/inhibitors
|
||||
- **Protein Binding**: Highly protein-bound drugs (warfarin caution)
|
||||
- **Renal/Hepatic Clearance**: Avoid multiple renally cleared agents
|
||||
|
||||
**Pharmacodynamic Interactions**
|
||||
- **Additive Toxicity**: Avoid overlapping adverse events (e.g., QTc prolongation)
|
||||
- **Antagonism**: Ensure mechanisms are complementary, not opposing
|
||||
- **Dose Modifications**: Pre-defined dose reduction schedules for combinations
|
||||
|
||||
### Combination Documentation
|
||||
|
||||
```
|
||||
Combination Regimen: Drug A + Drug B
|
||||
|
||||
Rationale:
|
||||
- Phase 3 RCT demonstrated PFS benefit (16 vs 11 months, HR 0.62, p<0.001)
|
||||
- Complementary mechanisms: Drug A (VEGF inhibitor) + Drug B (immune checkpoint inhibitor)
|
||||
- Non-overlapping toxicity profiles
|
||||
|
||||
Dosing:
|
||||
- Drug A: 10 mg/kg IV every 3 weeks
|
||||
- Drug B: 1200 mg IV every 3 weeks
|
||||
- Continue until progression or unacceptable toxicity
|
||||
|
||||
Key Toxicities:
|
||||
- Hypertension (Drug A): 30% grade 3-4, manage with antihypertensives
|
||||
- Immune-related AEs (Drug B): 15% grade 3-4, corticosteroid management
|
||||
- No significant pharmacokinetic interactions observed
|
||||
|
||||
Monitoring:
|
||||
- Blood pressure: Daily for first month, then weekly
|
||||
- Thyroid function: Every 6 weeks
|
||||
- Liver enzymes: Before each cycle
|
||||
- Imaging: Every 6 weeks (RECIST v1.1)
|
||||
```
|
||||
|
||||
## Monitoring and Follow-up Schedules
|
||||
|
||||
### On-Treatment Monitoring
|
||||
|
||||
**Laboratory Monitoring**
|
||||
```
|
||||
Test Baseline Cycle 1 Cycle 2+ Rationale
|
||||
CBC with differential ✓ Weekly Day 1 Myelosuppression risk
|
||||
Comprehensive panel ✓ Day 1 Day 1 Electrolytes, renal, hepatic
|
||||
Thyroid function ✓ - Q6 weeks Immunotherapy
|
||||
Lipase/amylase ✓ - As needed Pancreatitis risk
|
||||
Troponin/BNP ✓* - As needed Cardiotoxicity risk
|
||||
(*if cardiotoxic agent)
|
||||
```
|
||||
|
||||
**Imaging Assessment**
|
||||
```
|
||||
Modality Baseline Follow-up Criteria
|
||||
CT chest/abd/pelvis ✓ Every 6-9 weeks RECIST v1.1
|
||||
Brain MRI ✓* Every 12 weeks If CNS metastases
|
||||
Bone scan ✓** Every 12-24 weeks If bone metastases
|
||||
PET/CT ✓*** Response assessment Lymphoma (Lugano criteria)
|
||||
(*if CNS mets, **if bone mets, ***if PET-avid tumor)
|
||||
```
|
||||
|
||||
**Clinical Assessment**
|
||||
```
|
||||
Assessment Frequency Notes
|
||||
ECOG performance status Every visit Decline may warrant dose modification
|
||||
Vital signs Every visit Blood pressure for anti-VEGF agents
|
||||
Weight Every visit Cachexia, fluid retention
|
||||
Symptom assessment Every visit PRO-CTCAE questionnaire
|
||||
Physical exam Every visit Target lesions, new symptoms
|
||||
```
|
||||
|
||||
### Dose Modification Guidelines
|
||||
|
||||
**Hematologic Toxicity**
|
||||
```
|
||||
ANC and Platelet Counts Action
|
||||
ANC ≥1.5 AND platelets ≥100k Treat at full dose
|
||||
ANC 1.0-1.5 OR platelets 75-100k Delay 1 week, recheck
|
||||
ANC 0.5-1.0 OR platelets 50-75k Delay treatment, G-CSF support, reduce dose 20%
|
||||
ANC <0.5 OR platelets <50k Hold treatment, G-CSF, transfusion PRN, reduce 40%
|
||||
|
||||
Febrile Neutropenia Hold treatment, hospitalize, antibiotics, G-CSF
|
||||
Reduce dose 20-40% on recovery, consider prophylactic G-CSF
|
||||
```
|
||||
|
||||
**Non-Hematologic Toxicity**
|
||||
```
|
||||
Adverse Event Grade 1 Grade 2 Grade 3 Grade 4
|
||||
Diarrhea Continue Continue with Hold until ≤G1, Hold, hospitalize
|
||||
loperamide reduce 20% Consider discontinuation
|
||||
Rash Continue Continue with Hold until ≤G1, Discontinue
|
||||
topical Rx reduce 20%
|
||||
Hepatotoxicity Continue Repeat in 1 wk, Hold until ≤G1, Discontinue permanently
|
||||
hold if worsening reduce 20-40%
|
||||
Pneumonitis Continue Hold, consider Hold, corticosteroids, Discontinue, high-dose
|
||||
corticosteroids discontinue if no improvement steroids
|
||||
```
|
||||
|
||||
### Post-Treatment Surveillance
|
||||
|
||||
**Disease Monitoring**
|
||||
```
|
||||
Time After Treatment Imaging Frequency Labs Clinical
|
||||
Year 1 Every 3 months Every 3 months Every 3 months
|
||||
Year 2 Every 3-4 months Every 3-4 months Every 3-4 months
|
||||
Years 3-5 Every 6 months Every 6 months Every 6 months
|
||||
Year 5+ Annually Annually Annually
|
||||
|
||||
Earlier imaging if symptoms suggest recurrence
|
||||
```
|
||||
|
||||
**Survivorship Care**
|
||||
```
|
||||
Surveillance Frequency Duration
|
||||
Disease monitoring Per schedule above Lifelong or until recurrence
|
||||
Late toxicity screening Annually Lifelong
|
||||
- Cardiac function Every 1-2 years If anthracycline/trastuzumab
|
||||
- Pulmonary function As clinically indicated If bleomycin/radiation
|
||||
- Neuropathy Symptom-based Peripheral neuropathy history
|
||||
- Secondary malignancy Age-appropriate screening Lifelong (increased risk)
|
||||
Genetic counseling One time If hereditary cancer syndrome
|
||||
Psychosocial support As needed Depression, anxiety, PTSD screening
|
||||
```
|
||||
|
||||
## Special Populations
|
||||
|
||||
### Elderly Patients (≥65-70 years)
|
||||
|
||||
**Considerations**
|
||||
- **Reduced organ function**: Adjust for renal/hepatic impairment
|
||||
- **Polypharmacy**: Drug-drug interaction risk
|
||||
- **Frailty**: Geriatric assessment (G8, VES-13, CARG score)
|
||||
- **Goals of care**: Quality of life vs survival, functional independence
|
||||
|
||||
**Modifications**
|
||||
- Dose reductions: 20-25% reduction for frail patients
|
||||
- Longer intervals: Every 4 weeks instead of every 3 weeks
|
||||
- Less aggressive regimens: Single-agent vs combination therapy
|
||||
- Supportive care: Increased monitoring, G-CSF prophylaxis
|
||||
|
||||
### Renal Impairment
|
||||
|
||||
**Dose Adjustments by eGFR**
|
||||
```
|
||||
eGFR (mL/min/1.73m²) Category Action
|
||||
≥90 Normal Standard dosing
|
||||
60-89 Mild Standard dosing (most agents)
|
||||
30-59 Moderate Dose reduce renally cleared drugs 25-50%
|
||||
15-29 Severe Dose reduce 50-75%, avoid nephrotoxic agents
|
||||
<15 (dialysis) ESRD Avoid most agents, case-by-case decisions
|
||||
```
|
||||
|
||||
**Renally Cleared Agents Requiring Adjustment**
|
||||
- Carboplatin (Calvert formula: AUC × [GFR + 25])
|
||||
- Methotrexate (reduce dose 50-75% if CrCl <60)
|
||||
- Capecitabine (reduce dose 25-50% if CrCl 30-50)
|
||||
|
||||
### Hepatic Impairment
|
||||
|
||||
**Dose Adjustments by Bili and AST/ALT**
|
||||
```
|
||||
Category Bilirubin AST/ALT Action
|
||||
Normal ≤ULN ≤ULN Standard dosing
|
||||
Mild (Child A) 1-1.5× ULN Any Reduce dose 25% for hepatically metabolized
|
||||
Moderate (Child B) 1.5-3× ULN Any Reduce dose 50%, consider alternative
|
||||
Severe (Child C) >3× ULN Any Avoid most agents, case-by-case
|
||||
```
|
||||
|
||||
**Hepatically Metabolized Agents Requiring Adjustment**
|
||||
- Docetaxel (reduce 25-50% if bilirubin elevated)
|
||||
- Irinotecan (reduce 50% if bilirubin 1.5-3× ULN)
|
||||
- Tyrosine kinase inhibitors (most metabolized by CYP3A4, reduce by 50%)
|
||||
|
||||
### Pregnancy and Fertility
|
||||
|
||||
**Contraception Requirements**
|
||||
- Effective contraception required during treatment and 6-12 months after
|
||||
- Two methods recommended for highly teratogenic agents
|
||||
- Male patients: Contraception if partner of childbearing potential
|
||||
|
||||
**Fertility Preservation**
|
||||
- Oocyte/embryo cryopreservation (females, before gonadotoxic therapy)
|
||||
- Sperm banking (males, before alkylating agents, platinum)
|
||||
- GnRH agonists (ovarian suppression, controversial efficacy)
|
||||
- Referral to reproductive endocrinology before treatment
|
||||
|
||||
**Pregnancy Management**
|
||||
- Avoid chemotherapy in first trimester (organogenesis)
|
||||
- Selective agents safe in second/third trimester (case-by-case)
|
||||
- Multidisciplinary team: oncology, maternal-fetal medicine, neonatology
|
||||
|
||||
## Clinical Trial Considerations
|
||||
|
||||
### When to Recommend Clinical Trials
|
||||
|
||||
**Ideal Scenarios**
|
||||
- No standard therapy available (rare diseases, refractory settings)
|
||||
- Multiple equivalent standard options (patient preference for novel agent)
|
||||
- Standard therapy failed (second-line and beyond)
|
||||
- High-risk disease (adjuvant trials for improved outcomes)
|
||||
|
||||
**Trial Selection Criteria**
|
||||
- **Phase**: Phase 1 (dose-finding, safety), Phase 2 (efficacy signal), Phase 3 (comparative effectiveness)
|
||||
- **Eligibility**: Match patient to inclusion/exclusion criteria
|
||||
- **Mechanism**: Novel vs established mechanism, biological rationale
|
||||
- **Sponsor**: Academic vs industry, trial design quality
|
||||
- **Logistics**: Distance to trial site, visit frequency, out-of-pocket costs
|
||||
|
||||
### Shared Decision-Making
|
||||
|
||||
**Informing Patients**
|
||||
- Natural history without treatment
|
||||
- Standard treatment options with evidence, benefits, risks
|
||||
- Clinical trial options (if available)
|
||||
- Goals of care alignment
|
||||
- Patient values and preferences
|
||||
|
||||
**Decision Aids**
|
||||
- Visual representations of benefit (icon arrays)
|
||||
- Number needed to treat calculations
|
||||
- Quality of life trade-offs
|
||||
- Decisional conflict scales
|
||||
|
||||
## Documentation Standards
|
||||
|
||||
### Treatment Plan Documentation
|
||||
|
||||
```
|
||||
TREATMENT PLAN
|
||||
|
||||
Diagnosis: [Disease, stage, molecular profile]
|
||||
|
||||
Goals of Therapy:
|
||||
☐ Curative intent
|
||||
☐ Prolonged disease control
|
||||
☑ Palliation and quality of life
|
||||
|
||||
Recommended Regimen: [Name] (NCCN Category 1, GRADE 1A)
|
||||
|
||||
Evidence Basis:
|
||||
- Primary study: [Citation], Phase 3 RCT, n=XXX
|
||||
- Primary endpoint: PFS 12 months vs 8 months (HR 0.6, 95% CI 0.45-0.80, p<0.001)
|
||||
- Secondary endpoints: OS 24 vs 20 months (HR 0.75, p=0.02), ORR 60% vs 40%
|
||||
- Safety: Grade 3-4 AEs 35%, discontinuation rate 12%
|
||||
|
||||
Dosing Schedule:
|
||||
- Drug A: XX mg IV day 1
|
||||
- Drug B: XX mg PO days 1-21
|
||||
- Cycle length: 21 days
|
||||
- Planned cycles: Until progression or unacceptable toxicity
|
||||
|
||||
Premedications:
|
||||
- Dexamethasone 8 mg IV (anti-emetic)
|
||||
- Ondansetron 16 mg IV (anti-emetic)
|
||||
- Diphenhydramine 25 mg IV (hypersensitivity prophylaxis)
|
||||
|
||||
Monitoring Plan: [See schedule above]
|
||||
|
||||
Dose Modification Plan: [See guidelines above]
|
||||
|
||||
Alternative Options Discussed:
|
||||
- Option 2: [Alternative regimen], GRADE 1B
|
||||
- Clinical trial: [Trial name/number], Phase 2, novel agent
|
||||
- Best supportive care
|
||||
|
||||
Patient Decision: Proceed with recommended regimen
|
||||
|
||||
Informed Consent: Obtained for chemotherapy, risks/benefits discussed
|
||||
|
||||
Date: [Date]
|
||||
Provider: [Name, credentials]
|
||||
```
|
||||
|
||||
## Quality Metrics
|
||||
|
||||
### Treatment Recommendation Quality Indicators
|
||||
|
||||
- Evidence grading provided for all recommendations
|
||||
- Multiple options presented when equivalent evidence exists
|
||||
- Toxicity profiles clearly described
|
||||
- Monitoring plans specified
|
||||
- Dose modification guidelines included
|
||||
- Special populations addressed (elderly, renal/hepatic impairment)
|
||||
- Clinical trial options mentioned when appropriate
|
||||
- Shared decision-making documented
|
||||
- Goals of care aligned with treatment intensity
|
||||
|
||||
384
skills/clinical-decision-support/scripts/biomarker_classifier.py
Executable file
384
skills/clinical-decision-support/scripts/biomarker_classifier.py
Executable file
@@ -0,0 +1,384 @@
|
||||
#!/usr/bin/env python3
|
||||
"""
|
||||
Biomarker-Based Patient Stratification and Classification
|
||||
|
||||
Performs patient stratification based on biomarker profiles with:
|
||||
- Binary classification (biomarker+/-)
|
||||
- Multi-class molecular subtypes
|
||||
- Continuous biomarker scoring
|
||||
- Correlation with clinical outcomes
|
||||
|
||||
Dependencies: pandas, numpy, scipy, scikit-learn (optional for clustering)
|
||||
"""
|
||||
|
||||
import pandas as pd
|
||||
import numpy as np
|
||||
from scipy import stats
|
||||
import argparse
|
||||
from pathlib import Path
|
||||
|
||||
|
||||
def classify_binary_biomarker(data, biomarker_col, threshold,
|
||||
above_label='Biomarker+', below_label='Biomarker-'):
|
||||
"""
|
||||
Binary classification based on biomarker threshold.
|
||||
|
||||
Parameters:
|
||||
data: DataFrame
|
||||
biomarker_col: Column name for biomarker values
|
||||
threshold: Cut-point value
|
||||
above_label: Label for values >= threshold
|
||||
below_label: Label for values < threshold
|
||||
|
||||
Returns:
|
||||
DataFrame with added 'biomarker_class' column
|
||||
"""
|
||||
|
||||
data = data.copy()
|
||||
data['biomarker_class'] = data[biomarker_col].apply(
|
||||
lambda x: above_label if x >= threshold else below_label
|
||||
)
|
||||
|
||||
return data
|
||||
|
||||
|
||||
def classify_pd_l1_tps(data, pd_l1_col='pd_l1_tps'):
|
||||
"""
|
||||
Classify PD-L1 Tumor Proportion Score into clinical categories.
|
||||
|
||||
Categories:
|
||||
- Negative: <1%
|
||||
- Low: 1-49%
|
||||
- High: >=50%
|
||||
|
||||
Returns:
|
||||
DataFrame with 'pd_l1_category' column
|
||||
"""
|
||||
|
||||
data = data.copy()
|
||||
|
||||
def categorize(tps):
|
||||
if tps < 1:
|
||||
return 'PD-L1 Negative (<1%)'
|
||||
elif tps < 50:
|
||||
return 'PD-L1 Low (1-49%)'
|
||||
else:
|
||||
return 'PD-L1 High (≥50%)'
|
||||
|
||||
data['pd_l1_category'] = data[pd_l1_col].apply(categorize)
|
||||
|
||||
# Distribution
|
||||
print("\nPD-L1 TPS Distribution:")
|
||||
print(data['pd_l1_category'].value_counts())
|
||||
|
||||
return data
|
||||
|
||||
|
||||
def classify_her2_status(data, ihc_col='her2_ihc', fish_col='her2_fish'):
|
||||
"""
|
||||
Classify HER2 status based on IHC and FISH results (ASCO/CAP guidelines).
|
||||
|
||||
IHC Scores: 0, 1+, 2+, 3+
|
||||
FISH: Positive, Negative (if IHC 2+)
|
||||
|
||||
Classification:
|
||||
- HER2-positive: IHC 3+ OR IHC 2+/FISH+
|
||||
- HER2-negative: IHC 0/1+ OR IHC 2+/FISH-
|
||||
- HER2-low: IHC 1+ or IHC 2+/FISH- (subset of HER2-negative)
|
||||
|
||||
Returns:
|
||||
DataFrame with 'her2_status' and 'her2_low' columns
|
||||
"""
|
||||
|
||||
data = data.copy()
|
||||
|
||||
def classify_her2(row):
|
||||
ihc = row[ihc_col]
|
||||
fish = row.get(fish_col, None)
|
||||
|
||||
if ihc == '3+':
|
||||
status = 'HER2-positive'
|
||||
her2_low = False
|
||||
elif ihc == '2+':
|
||||
if fish == 'Positive':
|
||||
status = 'HER2-positive'
|
||||
her2_low = False
|
||||
elif fish == 'Negative':
|
||||
status = 'HER2-negative'
|
||||
her2_low = True # HER2-low
|
||||
else:
|
||||
status = 'HER2-equivocal (FISH needed)'
|
||||
her2_low = False
|
||||
elif ihc == '1+':
|
||||
status = 'HER2-negative'
|
||||
her2_low = True # HER2-low
|
||||
else: # IHC 0
|
||||
status = 'HER2-negative'
|
||||
her2_low = False
|
||||
|
||||
return pd.Series({'her2_status': status, 'her2_low': her2_low})
|
||||
|
||||
data[['her2_status', 'her2_low']] = data.apply(classify_her2, axis=1)
|
||||
|
||||
print("\nHER2 Status Distribution:")
|
||||
print(data['her2_status'].value_counts())
|
||||
print(f"\nHER2-low (IHC 1+ or 2+/FISH-): {data['her2_low'].sum()} patients")
|
||||
|
||||
return data
|
||||
|
||||
|
||||
def classify_breast_cancer_subtype(data, er_col='er_positive', pr_col='pr_positive',
|
||||
her2_col='her2_positive'):
|
||||
"""
|
||||
Classify breast cancer into molecular subtypes.
|
||||
|
||||
Subtypes:
|
||||
- HR+/HER2-: Luminal (ER+ and/or PR+, HER2-)
|
||||
- HER2+: Any HER2-positive (regardless of HR status)
|
||||
- Triple-negative: ER-, PR-, HER2-
|
||||
|
||||
Returns:
|
||||
DataFrame with 'bc_subtype' column
|
||||
"""
|
||||
|
||||
data = data.copy()
|
||||
|
||||
def get_subtype(row):
|
||||
er = row[er_col]
|
||||
pr = row[pr_col]
|
||||
her2 = row[her2_col]
|
||||
|
||||
if her2:
|
||||
if er or pr:
|
||||
return 'HR+/HER2+ (Luminal B HER2+)'
|
||||
else:
|
||||
return 'HR-/HER2+ (HER2-enriched)'
|
||||
elif er or pr:
|
||||
return 'HR+/HER2- (Luminal)'
|
||||
else:
|
||||
return 'Triple-Negative'
|
||||
|
||||
data['bc_subtype'] = data.apply(get_subtype, axis=1)
|
||||
|
||||
print("\nBreast Cancer Subtype Distribution:")
|
||||
print(data['bc_subtype'].value_counts())
|
||||
|
||||
return data
|
||||
|
||||
|
||||
def correlate_biomarker_outcome(data, biomarker_col, outcome_col, biomarker_type='binary'):
|
||||
"""
|
||||
Assess correlation between biomarker and clinical outcome.
|
||||
|
||||
Parameters:
|
||||
biomarker_col: Biomarker variable
|
||||
outcome_col: Outcome variable
|
||||
biomarker_type: 'binary', 'categorical', 'continuous'
|
||||
|
||||
Returns:
|
||||
Statistical test results
|
||||
"""
|
||||
|
||||
print(f"\nCorrelation Analysis: {biomarker_col} vs {outcome_col}")
|
||||
print("="*60)
|
||||
|
||||
# Remove missing data
|
||||
analysis_data = data[[biomarker_col, outcome_col]].dropna()
|
||||
|
||||
if biomarker_type == 'binary' or biomarker_type == 'categorical':
|
||||
# Cross-tabulation
|
||||
contingency = pd.crosstab(analysis_data[biomarker_col], analysis_data[outcome_col])
|
||||
print("\nContingency Table:")
|
||||
print(contingency)
|
||||
|
||||
# Chi-square test
|
||||
chi2, p_value, dof, expected = stats.chi2_contingency(contingency)
|
||||
|
||||
print(f"\nChi-square test:")
|
||||
print(f" χ² = {chi2:.2f}, df = {dof}, p = {p_value:.4f}")
|
||||
|
||||
# Odds ratio if 2x2 table
|
||||
if contingency.shape == (2, 2):
|
||||
a, b = contingency.iloc[0, :]
|
||||
c, d = contingency.iloc[1, :]
|
||||
or_value = (a * d) / (b * c) if b * c > 0 else np.inf
|
||||
|
||||
# Confidence interval for OR (log method)
|
||||
log_or = np.log(or_value)
|
||||
se_log_or = np.sqrt(1/a + 1/b + 1/c + 1/d)
|
||||
ci_lower = np.exp(log_or - 1.96 * se_log_or)
|
||||
ci_upper = np.exp(log_or + 1.96 * se_log_or)
|
||||
|
||||
print(f"\nOdds Ratio: {or_value:.2f} (95% CI {ci_lower:.2f}-{ci_upper:.2f})")
|
||||
|
||||
elif biomarker_type == 'continuous':
|
||||
# Correlation coefficient
|
||||
r, p_value = stats.pearsonr(analysis_data[biomarker_col], analysis_data[outcome_col])
|
||||
|
||||
print(f"\nPearson correlation:")
|
||||
print(f" r = {r:.3f}, p = {p_value:.4f}")
|
||||
|
||||
# Also report Spearman for robustness
|
||||
rho, p_spearman = stats.spearmanr(analysis_data[biomarker_col], analysis_data[outcome_col])
|
||||
print(f"Spearman correlation:")
|
||||
print(f" ρ = {rho:.3f}, p = {p_spearman:.4f}")
|
||||
|
||||
return p_value
|
||||
|
||||
|
||||
def stratify_cohort_report(data, stratification_var, output_dir='stratification_report'):
|
||||
"""
|
||||
Generate comprehensive stratification report.
|
||||
|
||||
Parameters:
|
||||
data: DataFrame with patient data
|
||||
stratification_var: Column name for stratification
|
||||
output_dir: Output directory for reports
|
||||
"""
|
||||
|
||||
output_dir = Path(output_dir)
|
||||
output_dir.mkdir(parents=True, exist_ok=True)
|
||||
|
||||
print(f"\nCOHORT STRATIFICATION REPORT")
|
||||
print("="*60)
|
||||
print(f"Stratification Variable: {stratification_var}")
|
||||
print(f"Total Patients: {len(data)}")
|
||||
|
||||
# Group distribution
|
||||
distribution = data[stratification_var].value_counts()
|
||||
print(f"\nGroup Distribution:")
|
||||
for group, count in distribution.items():
|
||||
pct = count / len(data) * 100
|
||||
print(f" {group}: {count} ({pct:.1f}%)")
|
||||
|
||||
# Save distribution
|
||||
distribution.to_csv(output_dir / 'group_distribution.csv')
|
||||
|
||||
# Compare baseline characteristics across groups
|
||||
print(f"\nBaseline Characteristics by {stratification_var}:")
|
||||
|
||||
results = []
|
||||
|
||||
# Continuous variables
|
||||
continuous_vars = data.select_dtypes(include=[np.number]).columns.tolist()
|
||||
continuous_vars = [v for v in continuous_vars if v != stratification_var]
|
||||
|
||||
for var in continuous_vars[:5]: # Limit to first 5 for demo
|
||||
print(f"\n{var}:")
|
||||
for group in distribution.index:
|
||||
group_data = data[data[stratification_var] == group][var].dropna()
|
||||
print(f" {group}: median {group_data.median():.1f} [IQR {group_data.quantile(0.25):.1f}-{group_data.quantile(0.75):.1f}]")
|
||||
|
||||
# Statistical test
|
||||
if len(distribution) == 2:
|
||||
groups_list = distribution.index.tolist()
|
||||
g1 = data[data[stratification_var] == groups_list[0]][var].dropna()
|
||||
g2 = data[data[stratification_var] == groups_list[1]][var].dropna()
|
||||
_, p_value = stats.mannwhitneyu(g1, g2, alternative='two-sided')
|
||||
print(f" p-value: {p_value:.4f}")
|
||||
|
||||
results.append({
|
||||
'Variable': var,
|
||||
'Test': 'Mann-Whitney U',
|
||||
'p_value': p_value,
|
||||
'Significant': 'Yes' if p_value < 0.05 else 'No'
|
||||
})
|
||||
|
||||
# Save results
|
||||
if results:
|
||||
df_results = pd.DataFrame(results)
|
||||
df_results.to_csv(output_dir / 'statistical_comparisons.csv', index=False)
|
||||
print(f"\nStatistical comparison results saved to: {output_dir}/statistical_comparisons.csv")
|
||||
|
||||
print(f"\nStratification report complete! Files saved to {output_dir}/")
|
||||
|
||||
|
||||
def main():
|
||||
parser = argparse.ArgumentParser(description='Biomarker-based patient classification')
|
||||
parser.add_argument('input_file', type=str, nargs='?', default=None,
|
||||
help='CSV file with patient and biomarker data')
|
||||
parser.add_argument('-b', '--biomarker', type=str, default=None,
|
||||
help='Biomarker column name for stratification')
|
||||
parser.add_argument('-t', '--threshold', type=float, default=None,
|
||||
help='Threshold for binary classification')
|
||||
parser.add_argument('-o', '--output-dir', type=str, default='stratification',
|
||||
help='Output directory')
|
||||
parser.add_argument('--example', action='store_true',
|
||||
help='Run with example data')
|
||||
|
||||
args = parser.parse_args()
|
||||
|
||||
# Example data if requested
|
||||
if args.example or args.input_file is None:
|
||||
print("Generating example dataset...")
|
||||
np.random.seed(42)
|
||||
n = 80
|
||||
|
||||
data = pd.DataFrame({
|
||||
'patient_id': [f'PT{i:03d}' for i in range(1, n+1)],
|
||||
'age': np.random.normal(62, 10, n),
|
||||
'sex': np.random.choice(['Male', 'Female'], n),
|
||||
'pd_l1_tps': np.random.exponential(20, n), # Exponential distribution for PD-L1
|
||||
'tmb': np.random.exponential(8, n), # Mutations per Mb
|
||||
'her2_ihc': np.random.choice(['0', '1+', '2+', '3+'], n, p=[0.6, 0.2, 0.15, 0.05]),
|
||||
'response': np.random.choice(['Yes', 'No'], n, p=[0.4, 0.6]),
|
||||
})
|
||||
|
||||
# Simulate correlation: higher PD-L1 -> better response
|
||||
data.loc[data['pd_l1_tps'] >= 50, 'response'] = np.random.choice(['Yes', 'No'],
|
||||
(data['pd_l1_tps'] >= 50).sum(),
|
||||
p=[0.65, 0.35])
|
||||
else:
|
||||
print(f"Loading data from {args.input_file}...")
|
||||
data = pd.read_csv(args.input_file)
|
||||
|
||||
print(f"Dataset: {len(data)} patients")
|
||||
print(f"Columns: {list(data.columns)}")
|
||||
|
||||
# PD-L1 classification example
|
||||
if 'pd_l1_tps' in data.columns or args.biomarker == 'pd_l1_tps':
|
||||
data = classify_pd_l1_tps(data, 'pd_l1_tps')
|
||||
|
||||
# Correlate with response if available
|
||||
if 'response' in data.columns:
|
||||
correlate_biomarker_outcome(data, 'pd_l1_category', 'response', biomarker_type='categorical')
|
||||
|
||||
# HER2 classification if columns present
|
||||
if 'her2_ihc' in data.columns:
|
||||
if 'her2_fish' not in data.columns:
|
||||
# Add placeholder FISH for IHC 2+
|
||||
data['her2_fish'] = np.nan
|
||||
data = classify_her2_status(data, 'her2_ihc', 'her2_fish')
|
||||
|
||||
# Generic binary classification if threshold provided
|
||||
if args.biomarker and args.threshold is not None:
|
||||
print(f"\nBinary classification: {args.biomarker} with threshold {args.threshold}")
|
||||
data = classify_binary_biomarker(data, args.biomarker, args.threshold)
|
||||
print(data['biomarker_class'].value_counts())
|
||||
|
||||
# Generate stratification report
|
||||
if args.biomarker:
|
||||
stratify_cohort_report(data, args.biomarker, output_dir=args.output_dir)
|
||||
elif 'pd_l1_category' in data.columns:
|
||||
stratify_cohort_report(data, 'pd_l1_category', output_dir=args.output_dir)
|
||||
|
||||
# Save classified data
|
||||
output_path = Path(args.output_dir) / 'classified_data.csv'
|
||||
data.to_csv(output_path, index=False)
|
||||
print(f"\nClassified data saved to: {output_path}")
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
main()
|
||||
|
||||
|
||||
# Example usage:
|
||||
# python biomarker_classifier.py data.csv -b pd_l1_tps -t 50 -o classification/
|
||||
# python biomarker_classifier.py --example
|
||||
#
|
||||
# Input CSV format:
|
||||
# patient_id,pd_l1_tps,tmb,her2_ihc,response,pfs_months,event
|
||||
# PT001,55.5,12.3,1+,Yes,14.2,1
|
||||
# PT002,8.2,5.1,0,No,6.5,1
|
||||
# ...
|
||||
|
||||
447
skills/clinical-decision-support/scripts/build_decision_tree.py
Executable file
447
skills/clinical-decision-support/scripts/build_decision_tree.py
Executable file
@@ -0,0 +1,447 @@
|
||||
#!/usr/bin/env python3
|
||||
"""
|
||||
Build Clinical Decision Tree Flowcharts in TikZ Format
|
||||
|
||||
Generates LaTeX/TikZ code for clinical decision algorithms from
|
||||
simple text or YAML descriptions.
|
||||
|
||||
Dependencies: pyyaml (optional, for YAML input)
|
||||
"""
|
||||
|
||||
import argparse
|
||||
from pathlib import Path
|
||||
import json
|
||||
|
||||
|
||||
class DecisionNode:
|
||||
"""Represents a decision point in the clinical algorithm."""
|
||||
|
||||
def __init__(self, question, yes_path=None, no_path=None, node_id=None):
|
||||
self.question = question
|
||||
self.yes_path = yes_path
|
||||
self.no_path = no_path
|
||||
self.node_id = node_id or self._generate_id(question)
|
||||
|
||||
def _generate_id(self, text):
|
||||
"""Generate clean node ID from text."""
|
||||
return ''.join(c for c in text if c.isalnum())[:15].lower()
|
||||
|
||||
|
||||
class ActionNode:
|
||||
"""Represents an action/outcome in the clinical algorithm."""
|
||||
|
||||
def __init__(self, action, urgency='routine', node_id=None):
|
||||
self.action = action
|
||||
self.urgency = urgency # 'urgent', 'semiurgent', 'routine'
|
||||
self.node_id = node_id or self._generate_id(action)
|
||||
|
||||
def _generate_id(self, text):
|
||||
return ''.join(c for c in text if c.isalnum())[:15].lower()
|
||||
|
||||
|
||||
def generate_tikz_header():
|
||||
"""Generate TikZ preamble with style definitions."""
|
||||
|
||||
tikz = """\\documentclass[10pt]{article}
|
||||
\\usepackage[margin=0.5in, landscape]{geometry}
|
||||
\\usepackage{tikz}
|
||||
\\usetikzlibrary{shapes,arrows,positioning}
|
||||
\\usepackage{xcolor}
|
||||
|
||||
% Color definitions
|
||||
\\definecolor{urgentred}{RGB}{220,20,60}
|
||||
\\definecolor{actiongreen}{RGB}{0,153,76}
|
||||
\\definecolor{decisionyellow}{RGB}{255,193,7}
|
||||
\\definecolor{routineblue}{RGB}{100,181,246}
|
||||
\\definecolor{headerblue}{RGB}{0,102,204}
|
||||
|
||||
% TikZ styles
|
||||
\\tikzstyle{startstop} = [rectangle, rounded corners=8pt, minimum width=3cm, minimum height=1cm,
|
||||
text centered, draw=black, fill=headerblue!20, font=\\small\\bfseries]
|
||||
\\tikzstyle{decision} = [diamond, minimum width=3cm, minimum height=1.2cm, text centered,
|
||||
draw=black, fill=decisionyellow!40, font=\\small, aspect=2, inner sep=0pt,
|
||||
text width=3.5cm]
|
||||
\\tikzstyle{process} = [rectangle, rounded corners=4pt, minimum width=3.5cm, minimum height=0.9cm,
|
||||
text centered, draw=black, fill=actiongreen!20, font=\\small]
|
||||
\\tikzstyle{urgent} = [rectangle, rounded corners=4pt, minimum width=3.5cm, minimum height=0.9cm,
|
||||
text centered, draw=urgentred, line width=1.5pt, fill=urgentred!15,
|
||||
font=\\small\\bfseries]
|
||||
\\tikzstyle{routine} = [rectangle, rounded corners=4pt, minimum width=3.5cm, minimum height=0.9cm,
|
||||
text centered, draw=black, fill=routineblue!20, font=\\small]
|
||||
\\tikzstyle{arrow} = [thick,->,>=stealth]
|
||||
\\tikzstyle{urgentarrow} = [ultra thick,->,>=stealth,color=urgentred]
|
||||
|
||||
\\begin{document}
|
||||
|
||||
\\begin{center}
|
||||
{\\Large\\bfseries Clinical Decision Algorithm}\\\\[10pt]
|
||||
{\\large [TITLE TO BE SPECIFIED]}
|
||||
\\end{center}
|
||||
|
||||
\\vspace{10pt}
|
||||
|
||||
\\begin{tikzpicture}[node distance=2.2cm and 3.5cm, auto]
|
||||
|
||||
"""
|
||||
|
||||
return tikz
|
||||
|
||||
|
||||
def generate_tikz_footer():
|
||||
"""Generate TikZ closing code."""
|
||||
|
||||
tikz = """
|
||||
\\end{tikzpicture}
|
||||
|
||||
\\end{document}
|
||||
"""
|
||||
|
||||
return tikz
|
||||
|
||||
|
||||
def simple_algorithm_to_tikz(algorithm_text, output_file='algorithm.tex'):
|
||||
"""
|
||||
Convert simple text-based algorithm to TikZ flowchart.
|
||||
|
||||
Input format (simple question-action pairs):
|
||||
START: Chief complaint
|
||||
Q1: High-risk criteria present? -> YES: Immediate action (URGENT) | NO: Continue
|
||||
Q2: Risk score >= 3? -> YES: Admit ICU | NO: Outpatient management (ROUTINE)
|
||||
END: Final outcome
|
||||
|
||||
Parameters:
|
||||
algorithm_text: Multi-line string with algorithm
|
||||
output_file: Path to save .tex file
|
||||
"""
|
||||
|
||||
tikz_code = generate_tikz_header()
|
||||
|
||||
# Parse algorithm text
|
||||
lines = [line.strip() for line in algorithm_text.strip().split('\n') if line.strip()]
|
||||
|
||||
node_defs = []
|
||||
arrow_defs = []
|
||||
|
||||
previous_node = None
|
||||
node_counter = 0
|
||||
|
||||
for line in lines:
|
||||
if line.startswith('START:'):
|
||||
# Start node
|
||||
text = line.replace('START:', '').strip()
|
||||
node_id = 'start'
|
||||
node_defs.append(f"\\node [startstop] ({node_id}) {{{text}}};")
|
||||
previous_node = node_id
|
||||
node_counter += 1
|
||||
|
||||
elif line.startswith('END:'):
|
||||
# End node
|
||||
text = line.replace('END:', '').strip()
|
||||
node_id = 'end'
|
||||
|
||||
# Position relative to previous
|
||||
if previous_node:
|
||||
node_defs.append(f"\\node [startstop, below=of {previous_node}] ({node_id}) {{{text}}};")
|
||||
arrow_defs.append(f"\\draw [arrow] ({previous_node}) -- ({node_id});")
|
||||
|
||||
elif line.startswith('Q'):
|
||||
# Decision node
|
||||
parts = line.split(':', 1)
|
||||
if len(parts) < 2:
|
||||
continue
|
||||
|
||||
question_part = parts[1].split('->')[0].strip()
|
||||
node_id = f'q{node_counter}'
|
||||
|
||||
# Add decision node
|
||||
if previous_node:
|
||||
node_defs.append(f"\\node [decision, below=of {previous_node}] ({node_id}) {{{question_part}}};")
|
||||
arrow_defs.append(f"\\draw [arrow] ({previous_node}) -- ({node_id});")
|
||||
else:
|
||||
node_defs.append(f"\\node [decision] ({node_id}) {{{question_part}}};")
|
||||
|
||||
# Parse YES and NO branches
|
||||
if '->' in line:
|
||||
branches = line.split('->')[1].split('|')
|
||||
|
||||
for branch in branches:
|
||||
branch = branch.strip()
|
||||
|
||||
if branch.startswith('YES:'):
|
||||
yes_action = branch.replace('YES:', '').strip()
|
||||
yes_id = f'yes{node_counter}'
|
||||
|
||||
# Check urgency
|
||||
if '(URGENT)' in yes_action:
|
||||
style = 'urgent'
|
||||
yes_action = yes_action.replace('(URGENT)', '').strip()
|
||||
arrow_style = 'urgentarrow'
|
||||
elif '(ROUTINE)' in yes_action:
|
||||
style = 'routine'
|
||||
yes_action = yes_action.replace('(ROUTINE)', '').strip()
|
||||
arrow_style = 'arrow'
|
||||
else:
|
||||
style = 'process'
|
||||
arrow_style = 'arrow'
|
||||
|
||||
node_defs.append(f"\\node [{style}, left=of {node_id}] ({yes_id}) {{{yes_action}}};")
|
||||
arrow_defs.append(f"\\draw [{arrow_style}] ({node_id}) -- node[above] {{Yes}} ({yes_id});")
|
||||
|
||||
elif branch.startswith('NO:'):
|
||||
no_action = branch.replace('NO:', '').strip()
|
||||
no_id = f'no{node_counter}'
|
||||
|
||||
# Check urgency
|
||||
if '(URGENT)' in no_action:
|
||||
style = 'urgent'
|
||||
no_action = no_action.replace('(URGENT)', '').strip()
|
||||
arrow_style = 'urgentarrow'
|
||||
elif '(ROUTINE)' in no_action:
|
||||
style = 'routine'
|
||||
no_action = no_action.replace('(ROUTINE)', '').strip()
|
||||
arrow_style = 'arrow'
|
||||
else:
|
||||
style = 'process'
|
||||
arrow_style = 'arrow'
|
||||
|
||||
node_defs.append(f"\\node [{style}, right=of {node_id}] ({no_id}) {{{no_action}}};")
|
||||
arrow_defs.append(f"\\draw [{arrow_style}] ({node_id}) -- node[above] {{No}} ({no_id});")
|
||||
|
||||
previous_node = node_id
|
||||
node_counter += 1
|
||||
|
||||
# Add all nodes and arrows to TikZ
|
||||
tikz_code += '\n'.join(node_defs) + '\n\n'
|
||||
tikz_code += '% Arrows\n'
|
||||
tikz_code += '\n'.join(arrow_defs) + '\n'
|
||||
|
||||
tikz_code += generate_tikz_footer()
|
||||
|
||||
# Save to file
|
||||
with open(output_file, 'w') as f:
|
||||
f.write(tikz_code)
|
||||
|
||||
print(f"TikZ flowchart saved to: {output_file}")
|
||||
print(f"Compile with: pdflatex {output_file}")
|
||||
|
||||
return tikz_code
|
||||
|
||||
|
||||
def json_to_tikz(json_file, output_file='algorithm.tex'):
|
||||
"""
|
||||
Convert JSON decision tree specification to TikZ flowchart.
|
||||
|
||||
JSON format:
|
||||
{
|
||||
"title": "Algorithm Title",
|
||||
"nodes": {
|
||||
"start": {"type": "start", "text": "Patient presentation"},
|
||||
"q1": {"type": "decision", "text": "Criteria met?", "yes": "action1", "no": "q2"},
|
||||
"action1": {"type": "action", "text": "Immediate intervention", "urgency": "urgent"},
|
||||
"q2": {"type": "decision", "text": "Score >= 3?", "yes": "action2", "no": "action3"},
|
||||
"action2": {"type": "action", "text": "Admit ICU"},
|
||||
"action3": {"type": "action", "text": "Outpatient", "urgency": "routine"}
|
||||
},
|
||||
"start_node": "start"
|
||||
}
|
||||
"""
|
||||
|
||||
with open(json_file, 'r') as f:
|
||||
spec = json.load(f)
|
||||
|
||||
tikz_code = generate_tikz_header()
|
||||
|
||||
# Replace title
|
||||
title = spec.get('title', 'Clinical Decision Algorithm')
|
||||
tikz_code = tikz_code.replace('[TITLE TO BE SPECIFIED]', title)
|
||||
|
||||
nodes = spec['nodes']
|
||||
start_node = spec.get('start_node', 'start')
|
||||
|
||||
# Generate nodes (simplified layout - vertical)
|
||||
node_defs = []
|
||||
arrow_defs = []
|
||||
|
||||
# Track positioning
|
||||
previous_node = None
|
||||
level = 0
|
||||
|
||||
def add_node(node_id, position_rel=None):
|
||||
"""Recursively add nodes."""
|
||||
|
||||
if node_id not in nodes:
|
||||
return
|
||||
|
||||
node = nodes[node_id]
|
||||
node_type = node['type']
|
||||
text = node['text']
|
||||
|
||||
# Determine TikZ style
|
||||
if node_type == 'start' or node_type == 'end':
|
||||
style = 'startstop'
|
||||
elif node_type == 'decision':
|
||||
style = 'decision'
|
||||
elif node_type == 'action':
|
||||
urgency = node.get('urgency', 'normal')
|
||||
if urgency == 'urgent':
|
||||
style = 'urgent'
|
||||
elif urgency == 'routine':
|
||||
style = 'routine'
|
||||
else:
|
||||
style = 'process'
|
||||
else:
|
||||
style = 'process'
|
||||
|
||||
# Position node
|
||||
if position_rel:
|
||||
node_def = f"\\node [{style}, {position_rel}] ({node_id}) {{{text}}};"
|
||||
else:
|
||||
node_def = f"\\node [{style}] ({node_id}) {{{text}}};"
|
||||
|
||||
node_defs.append(node_def)
|
||||
|
||||
# Add arrows for decision nodes
|
||||
if node_type == 'decision':
|
||||
yes_target = node.get('yes')
|
||||
no_target = node.get('no')
|
||||
|
||||
if yes_target:
|
||||
# Determine arrow style based on target urgency
|
||||
target_node = nodes.get(yes_target, {})
|
||||
arrow_style = 'urgentarrow' if target_node.get('urgency') == 'urgent' else 'arrow'
|
||||
arrow_defs.append(f"\\draw [{arrow_style}] ({node_id}) -| node[near start, above] {{Yes}} ({yes_target});")
|
||||
|
||||
if no_target:
|
||||
target_node = nodes.get(no_target, {})
|
||||
arrow_style = 'urgentarrow' if target_node.get('urgency') == 'urgent' else 'arrow'
|
||||
arrow_defs.append(f"\\draw [{arrow_style}] ({node_id}) -| node[near start, above] {{No}} ({no_target});")
|
||||
|
||||
# Simple layout - just list nodes (manual positioning in JSON works better for complex trees)
|
||||
for node_id in nodes.keys():
|
||||
add_node(node_id)
|
||||
|
||||
tikz_code += '\n'.join(node_defs) + '\n\n'
|
||||
tikz_code += '% Arrows\n'
|
||||
tikz_code += '\n'.join(arrow_defs) + '\n'
|
||||
|
||||
tikz_code += generate_tikz_footer()
|
||||
|
||||
# Save
|
||||
with open(output_file, 'w') as f:
|
||||
f.write(tikz_code)
|
||||
|
||||
print(f"TikZ flowchart saved to: {output_file}")
|
||||
return tikz_code
|
||||
|
||||
|
||||
def create_example_json():
|
||||
"""Create example JSON specification for testing."""
|
||||
|
||||
example = {
|
||||
"title": "Acute Chest Pain Management Algorithm",
|
||||
"nodes": {
|
||||
"start": {
|
||||
"type": "start",
|
||||
"text": "Patient with\\nchest pain"
|
||||
},
|
||||
"q1": {
|
||||
"type": "decision",
|
||||
"text": "STEMI\\ncriteria?",
|
||||
"yes": "stemi_action",
|
||||
"no": "q2"
|
||||
},
|
||||
"stemi_action": {
|
||||
"type": "action",
|
||||
"text": "Activate cath lab\\nAspirin, heparin\\nPrimary PCI",
|
||||
"urgency": "urgent"
|
||||
},
|
||||
"q2": {
|
||||
"type": "decision",
|
||||
"text": "High-risk\\nfeatures?",
|
||||
"yes": "admit",
|
||||
"no": "q3"
|
||||
},
|
||||
"admit": {
|
||||
"type": "action",
|
||||
"text": "Admit CCU\\nSerial troponins\\nEarly angiography"
|
||||
},
|
||||
"q3": {
|
||||
"type": "decision",
|
||||
"text": "TIMI\\nscore 0-1?",
|
||||
"yes": "lowrisk",
|
||||
"no": "moderate"
|
||||
},
|
||||
"lowrisk": {
|
||||
"type": "action",
|
||||
"text": "Observe 6-12h\\nStress test\\nOutpatient f/u",
|
||||
"urgency": "routine"
|
||||
},
|
||||
"moderate": {
|
||||
"type": "action",
|
||||
"text": "Admit telemetry\\nMedical management\\nRisk stratification"
|
||||
}
|
||||
},
|
||||
"start_node": "start"
|
||||
}
|
||||
|
||||
return example
|
||||
|
||||
|
||||
def main():
|
||||
parser = argparse.ArgumentParser(description='Build clinical decision tree flowcharts')
|
||||
parser.add_argument('-i', '--input', type=str, default=None,
|
||||
help='Input file (JSON format)')
|
||||
parser.add_argument('-o', '--output', type=str, default='clinical_algorithm.tex',
|
||||
help='Output .tex file')
|
||||
parser.add_argument('--example', action='store_true',
|
||||
help='Generate example algorithm')
|
||||
parser.add_argument('--text', type=str, default=None,
|
||||
help='Simple text algorithm (see format in docstring)')
|
||||
|
||||
args = parser.parse_args()
|
||||
|
||||
if args.example:
|
||||
print("Generating example algorithm...")
|
||||
example_spec = create_example_json()
|
||||
|
||||
# Save example JSON
|
||||
with open('example_algorithm.json', 'w') as f:
|
||||
json.dump(example_spec, f, indent=2)
|
||||
print("Example JSON saved to: example_algorithm.json")
|
||||
|
||||
# Generate TikZ from example
|
||||
json_to_tikz('example_algorithm.json', args.output)
|
||||
|
||||
elif args.text:
|
||||
print("Generating algorithm from text...")
|
||||
simple_algorithm_to_tikz(args.text, args.output)
|
||||
|
||||
elif args.input:
|
||||
print(f"Generating algorithm from {args.input}...")
|
||||
if args.input.endswith('.json'):
|
||||
json_to_tikz(args.input, args.output)
|
||||
else:
|
||||
with open(args.input, 'r') as f:
|
||||
text = f.read()
|
||||
simple_algorithm_to_tikz(text, args.output)
|
||||
|
||||
else:
|
||||
print("No input provided. Use --example to generate example, --text for simple text, or -i for JSON input.")
|
||||
print("\nSimple text format:")
|
||||
print("START: Patient presentation")
|
||||
print("Q1: Criteria met? -> YES: Action (URGENT) | NO: Continue")
|
||||
print("Q2: Score >= 3? -> YES: Admit | NO: Outpatient (ROUTINE)")
|
||||
print("END: Follow-up")
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
main()
|
||||
|
||||
|
||||
# Example usage:
|
||||
# python build_decision_tree.py --example
|
||||
# python build_decision_tree.py -i algorithm_spec.json -o my_algorithm.tex
|
||||
#
|
||||
# Then compile:
|
||||
# pdflatex clinical_algorithm.tex
|
||||
|
||||
524
skills/clinical-decision-support/scripts/create_cohort_tables.py
Executable file
524
skills/clinical-decision-support/scripts/create_cohort_tables.py
Executable file
@@ -0,0 +1,524 @@
|
||||
#!/usr/bin/env python3
|
||||
"""
|
||||
Generate Clinical Cohort Tables for Baseline Characteristics and Outcomes
|
||||
|
||||
Creates publication-ready tables with:
|
||||
- Baseline demographics (Table 1 style)
|
||||
- Efficacy outcomes
|
||||
- Safety/adverse events
|
||||
- Statistical comparisons between groups
|
||||
|
||||
Dependencies: pandas, numpy, scipy
|
||||
"""
|
||||
|
||||
import pandas as pd
|
||||
import numpy as np
|
||||
from scipy import stats
|
||||
from pathlib import Path
|
||||
import argparse
|
||||
|
||||
|
||||
def calculate_p_value(data, variable, group_col='group', var_type='categorical'):
|
||||
"""
|
||||
Calculate appropriate p-value for group comparison.
|
||||
|
||||
Parameters:
|
||||
data: DataFrame
|
||||
variable: Column name to compare
|
||||
group_col: Grouping variable
|
||||
var_type: 'categorical', 'continuous_normal', 'continuous_nonnormal'
|
||||
|
||||
Returns:
|
||||
p-value (float)
|
||||
"""
|
||||
|
||||
groups = data[group_col].unique()
|
||||
|
||||
if len(groups) != 2:
|
||||
return np.nan # Only handle 2-group comparisons
|
||||
|
||||
group1_data = data[data[group_col] == groups[0]][variable].dropna()
|
||||
group2_data = data[data[group_col] == groups[1]][variable].dropna()
|
||||
|
||||
if var_type == 'categorical':
|
||||
# Chi-square or Fisher's exact test
|
||||
contingency = pd.crosstab(data[variable], data[group_col])
|
||||
|
||||
# Check if Fisher's exact is needed (expected count < 5)
|
||||
if contingency.min().min() < 5:
|
||||
# Fisher's exact (2x2 only)
|
||||
if contingency.shape == (2, 2):
|
||||
_, p_value = stats.fisher_exact(contingency)
|
||||
else:
|
||||
# Use chi-square but note limitation
|
||||
_, p_value, _, _ = stats.chi2_contingency(contingency)
|
||||
else:
|
||||
_, p_value, _, _ = stats.chi2_contingency(contingency)
|
||||
|
||||
elif var_type == 'continuous_normal':
|
||||
# Independent t-test
|
||||
_, p_value = stats.ttest_ind(group1_data, group2_data, equal_var=False)
|
||||
|
||||
elif var_type == 'continuous_nonnormal':
|
||||
# Mann-Whitney U test
|
||||
_, p_value = stats.mannwhitneyu(group1_data, group2_data, alternative='two-sided')
|
||||
|
||||
else:
|
||||
raise ValueError("var_type must be 'categorical', 'continuous_normal', or 'continuous_nonnormal'")
|
||||
|
||||
return p_value
|
||||
|
||||
|
||||
def format_continuous_variable(data, variable, group_col, distribution='normal'):
|
||||
"""
|
||||
Format continuous variable for table display.
|
||||
|
||||
Returns:
|
||||
Dictionary with formatted strings for each group and p-value
|
||||
"""
|
||||
|
||||
groups = data[group_col].unique()
|
||||
results = {}
|
||||
|
||||
for group in groups:
|
||||
group_data = data[data[group_col] == group][variable].dropna()
|
||||
|
||||
if distribution == 'normal':
|
||||
# Mean ± SD
|
||||
mean = group_data.mean()
|
||||
std = group_data.std()
|
||||
results[group] = f"{mean:.1f} ± {std:.1f}"
|
||||
else:
|
||||
# Median [IQR]
|
||||
median = group_data.median()
|
||||
q1 = group_data.quantile(0.25)
|
||||
q3 = group_data.quantile(0.75)
|
||||
results[group] = f"{median:.1f} [{q1:.1f}-{q3:.1f}]"
|
||||
|
||||
# Calculate p-value
|
||||
var_type = 'continuous_normal' if distribution == 'normal' else 'continuous_nonnormal'
|
||||
p_value = calculate_p_value(data, variable, group_col, var_type)
|
||||
results['p_value'] = f"{p_value:.3f}" if p_value < 0.001 else f"{p_value:.2f}" if p_value < 1.0 else "—"
|
||||
|
||||
return results
|
||||
|
||||
|
||||
def format_categorical_variable(data, variable, group_col):
|
||||
"""
|
||||
Format categorical variable for table display.
|
||||
|
||||
Returns:
|
||||
List of dictionaries for each category with counts and percentages
|
||||
"""
|
||||
|
||||
groups = data[group_col].unique()
|
||||
categories = data[variable].dropna().unique()
|
||||
|
||||
results = []
|
||||
|
||||
for category in categories:
|
||||
row = {'category': category}
|
||||
|
||||
for group in groups:
|
||||
group_data = data[data[group_col] == group]
|
||||
count = (group_data[variable] == category).sum()
|
||||
total = group_data[variable].notna().sum()
|
||||
percentage = (count / total * 100) if total > 0 else 0
|
||||
row[group] = f"{count} ({percentage:.0f}%)"
|
||||
|
||||
results.append(row)
|
||||
|
||||
# Calculate p-value for overall categorical variable
|
||||
p_value = calculate_p_value(data, variable, group_col, 'categorical')
|
||||
results[0]['p_value'] = f"{p_value:.3f}" if p_value < 0.001 else f"{p_value:.2f}" if p_value < 1.0 else "—"
|
||||
|
||||
return results
|
||||
|
||||
|
||||
def generate_baseline_table(data, group_col='group', output_file='table1_baseline.csv'):
|
||||
"""
|
||||
Generate Table 1: Baseline characteristics.
|
||||
|
||||
Customize the variables list for your specific cohort.
|
||||
"""
|
||||
|
||||
groups = data[group_col].unique()
|
||||
|
||||
# Initialize results list
|
||||
table_rows = []
|
||||
|
||||
# Header row
|
||||
header = {
|
||||
'Characteristic': 'Characteristic',
|
||||
**{group: f"{group} (n={len(data[data[group_col]==group])})" for group in groups},
|
||||
'p_value': 'p-value'
|
||||
}
|
||||
table_rows.append(header)
|
||||
|
||||
# Age (continuous)
|
||||
if 'age' in data.columns:
|
||||
age_results = format_continuous_variable(data, 'age', group_col, distribution='nonnormal')
|
||||
row = {'Characteristic': 'Age, years (median [IQR])'}
|
||||
for group in groups:
|
||||
row[group] = age_results[group]
|
||||
row['p_value'] = age_results['p_value']
|
||||
table_rows.append(row)
|
||||
|
||||
# Sex (categorical)
|
||||
if 'sex' in data.columns:
|
||||
table_rows.append({'Characteristic': 'Sex, n (%)', **{g: '' for g in groups}, 'p_value': ''})
|
||||
sex_results = format_categorical_variable(data, 'sex', group_col)
|
||||
for sex_row in sex_results:
|
||||
row = {'Characteristic': f" {sex_row['category']}"}
|
||||
for group in groups:
|
||||
row[group] = sex_row[group]
|
||||
row['p_value'] = sex_row.get('p_value', '')
|
||||
table_rows.append(row)
|
||||
|
||||
# ECOG Performance Status (categorical)
|
||||
if 'ecog_ps' in data.columns:
|
||||
table_rows.append({'Characteristic': 'ECOG PS, n (%)', **{g: '' for g in groups}, 'p_value': ''})
|
||||
ecog_results = format_categorical_variable(data, 'ecog_ps', group_col)
|
||||
for ecog_row in ecog_results:
|
||||
row = {'Characteristic': f" {ecog_row['category']}"}
|
||||
for group in groups:
|
||||
row[group] = ecog_row[group]
|
||||
row['p_value'] = ecog_row.get('p_value', '')
|
||||
table_rows.append(row)
|
||||
|
||||
# Convert to DataFrame and save
|
||||
df_table = pd.DataFrame(table_rows)
|
||||
df_table.to_csv(output_file, index=False)
|
||||
print(f"Baseline characteristics table saved to: {output_file}")
|
||||
|
||||
return df_table
|
||||
|
||||
|
||||
def generate_efficacy_table(data, group_col='group', output_file='table2_efficacy.csv'):
|
||||
"""
|
||||
Generate efficacy outcomes table.
|
||||
|
||||
Expected columns:
|
||||
- best_response: CR, PR, SD, PD
|
||||
- Additional binary outcomes (response, disease_control, etc.)
|
||||
"""
|
||||
|
||||
groups = data[group_col].unique()
|
||||
table_rows = []
|
||||
|
||||
# Header
|
||||
header = {
|
||||
'Outcome': 'Outcome',
|
||||
**{group: f"{group} (n={len(data[data[group_col]==group])})" for group in groups},
|
||||
'p_value': 'p-value'
|
||||
}
|
||||
table_rows.append(header)
|
||||
|
||||
# Objective Response Rate (ORR = CR + PR)
|
||||
if 'best_response' in data.columns:
|
||||
for group in groups:
|
||||
group_data = data[data[group_col] == group]
|
||||
cr_pr = ((group_data['best_response'] == 'CR') | (group_data['best_response'] == 'PR')).sum()
|
||||
total = len(group_data)
|
||||
orr = cr_pr / total * 100
|
||||
|
||||
# Calculate exact binomial CI (Clopper-Pearson)
|
||||
ci_lower, ci_upper = _binomial_ci(cr_pr, total)
|
||||
|
||||
if group == groups[0]:
|
||||
orr_row = {'Outcome': 'ORR, n (%) [95% CI]'}
|
||||
|
||||
orr_row[group] = f"{cr_pr} ({orr:.0f}%) [{ci_lower:.0f}-{ci_upper:.0f}]"
|
||||
|
||||
# P-value for ORR difference
|
||||
contingency = pd.crosstab(
|
||||
data['best_response'].isin(['CR', 'PR']),
|
||||
data[group_col]
|
||||
)
|
||||
_, p_value, _, _ = stats.chi2_contingency(contingency)
|
||||
orr_row['p_value'] = f"{p_value:.3f}" if p_value >= 0.001 else "<0.001"
|
||||
table_rows.append(orr_row)
|
||||
|
||||
# Individual response categories
|
||||
for response in ['CR', 'PR', 'SD', 'PD']:
|
||||
row = {'Outcome': f" {response}"}
|
||||
for group in groups:
|
||||
group_data = data[data[group_col] == group]
|
||||
count = (group_data['best_response'] == response).sum()
|
||||
total = len(group_data)
|
||||
pct = count / total * 100
|
||||
row[group] = f"{count} ({pct:.0f}%)"
|
||||
row['p_value'] = ''
|
||||
table_rows.append(row)
|
||||
|
||||
# Disease Control Rate (DCR = CR + PR + SD)
|
||||
if 'best_response' in data.columns:
|
||||
dcr_row = {'Outcome': 'DCR, n (%) [95% CI]'}
|
||||
for group in groups:
|
||||
group_data = data[data[group_col] == group]
|
||||
dcr_count = group_data['best_response'].isin(['CR', 'PR', 'SD']).sum()
|
||||
total = len(group_data)
|
||||
dcr = dcr_count / total * 100
|
||||
ci_lower, ci_upper = _binomial_ci(dcr_count, total)
|
||||
dcr_row[group] = f"{dcr_count} ({dcr:.0f}%) [{ci_lower:.0f}-{ci_upper:.0f}]"
|
||||
|
||||
# P-value
|
||||
contingency = pd.crosstab(
|
||||
data['best_response'].isin(['CR', 'PR', 'SD']),
|
||||
data[group_col]
|
||||
)
|
||||
_, p_value, _, _ = stats.chi2_contingency(contingency)
|
||||
dcr_row['p_value'] = f"{p_value:.3f}" if p_value >= 0.001 else "<0.001"
|
||||
table_rows.append(dcr_row)
|
||||
|
||||
# Save table
|
||||
df_table = pd.DataFrame(table_rows)
|
||||
df_table.to_csv(output_file, index=False)
|
||||
print(f"Efficacy table saved to: {output_file}")
|
||||
|
||||
return df_table
|
||||
|
||||
|
||||
def generate_safety_table(data, ae_columns, group_col='group', output_file='table3_safety.csv'):
|
||||
"""
|
||||
Generate adverse events table.
|
||||
|
||||
Parameters:
|
||||
data: DataFrame with AE data
|
||||
ae_columns: List of AE column names (each should have values 0-5 for CTCAE grades)
|
||||
group_col: Grouping variable
|
||||
output_file: Output CSV path
|
||||
"""
|
||||
|
||||
groups = data[group_col].unique()
|
||||
table_rows = []
|
||||
|
||||
# Header
|
||||
header = {
|
||||
'Adverse Event': 'Adverse Event',
|
||||
**{f'{group}_any': f'Any Grade' for group in groups},
|
||||
**{f'{group}_g34': f'Grade 3-4' for group in groups}
|
||||
}
|
||||
|
||||
for ae in ae_columns:
|
||||
if ae not in data.columns:
|
||||
continue
|
||||
|
||||
row = {'Adverse Event': ae.replace('_', ' ').title()}
|
||||
|
||||
for group in groups:
|
||||
group_data = data[data[group_col] == group][ae].dropna()
|
||||
total = len(group_data)
|
||||
|
||||
# Any grade (Grade 1-5)
|
||||
any_grade = (group_data > 0).sum()
|
||||
any_pct = any_grade / total * 100 if total > 0 else 0
|
||||
row[f'{group}_any'] = f"{any_grade} ({any_pct:.0f}%)"
|
||||
|
||||
# Grade 3-4
|
||||
grade_34 = (group_data >= 3).sum()
|
||||
g34_pct = grade_34 / total * 100 if total > 0 else 0
|
||||
row[f'{group}_g34'] = f"{grade_34} ({g34_pct:.0f}%)"
|
||||
|
||||
table_rows.append(row)
|
||||
|
||||
# Save table
|
||||
df_table = pd.DataFrame(table_rows)
|
||||
df_table.to_csv(output_file, index=False)
|
||||
print(f"Safety table saved to: {output_file}")
|
||||
|
||||
return df_table
|
||||
|
||||
|
||||
def generate_latex_table(df, caption, label='table'):
|
||||
"""
|
||||
Convert DataFrame to LaTeX table code.
|
||||
|
||||
Returns:
|
||||
String with LaTeX table code
|
||||
"""
|
||||
|
||||
latex_code = "\\begin{table}[H]\n"
|
||||
latex_code += "\\centering\n"
|
||||
latex_code += "\\small\n"
|
||||
latex_code += "\\begin{tabular}{" + "l" * len(df.columns) + "}\n"
|
||||
latex_code += "\\toprule\n"
|
||||
|
||||
# Header
|
||||
header_row = " & ".join([f"\\textbf{{{col}}}" for col in df.columns])
|
||||
latex_code += header_row + " \\\\\n"
|
||||
latex_code += "\\midrule\n"
|
||||
|
||||
# Data rows
|
||||
for _, row in df.iterrows():
|
||||
# Handle indentation for subcategories (lines starting with spaces)
|
||||
first_col = str(row.iloc[0])
|
||||
if first_col.startswith(' '):
|
||||
first_col = '\\quad ' + first_col.strip()
|
||||
|
||||
data_row = [first_col] + [str(val) if pd.notna(val) else '—' for val in row.iloc[1:]]
|
||||
latex_code += " & ".join(data_row) + " \\\\\n"
|
||||
|
||||
latex_code += "\\bottomrule\n"
|
||||
latex_code += "\\end{tabular}\n"
|
||||
latex_code += f"\\caption{{{caption}}}\n"
|
||||
latex_code += f"\\label{{tab:{label}}}\n"
|
||||
latex_code += "\\end{table}\n"
|
||||
|
||||
return latex_code
|
||||
|
||||
|
||||
def _binomial_ci(successes, trials, confidence=0.95):
|
||||
"""
|
||||
Calculate exact binomial confidence interval (Clopper-Pearson method).
|
||||
|
||||
Returns:
|
||||
Lower and upper bounds as percentages
|
||||
"""
|
||||
|
||||
if trials == 0:
|
||||
return 0.0, 0.0
|
||||
|
||||
alpha = 1 - confidence
|
||||
|
||||
# Use beta distribution
|
||||
from scipy.stats import beta
|
||||
|
||||
if successes == 0:
|
||||
lower = 0.0
|
||||
else:
|
||||
lower = beta.ppf(alpha/2, successes, trials - successes + 1)
|
||||
|
||||
if successes == trials:
|
||||
upper = 1.0
|
||||
else:
|
||||
upper = beta.ppf(1 - alpha/2, successes + 1, trials - successes)
|
||||
|
||||
return lower * 100, upper * 100
|
||||
|
||||
|
||||
def create_example_data():
|
||||
"""Create example dataset for testing."""
|
||||
|
||||
np.random.seed(42)
|
||||
n = 100
|
||||
|
||||
data = pd.DataFrame({
|
||||
'patient_id': [f'PT{i:03d}' for i in range(1, n+1)],
|
||||
'group': np.random.choice(['Biomarker+', 'Biomarker-'], n),
|
||||
'age': np.random.normal(62, 10, n),
|
||||
'sex': np.random.choice(['Male', 'Female'], n),
|
||||
'ecog_ps': np.random.choice(['0-1', '2'], n, p=[0.8, 0.2]),
|
||||
'stage': np.random.choice(['III', 'IV'], n, p=[0.3, 0.7]),
|
||||
'best_response': np.random.choice(['CR', 'PR', 'SD', 'PD'], n, p=[0.05, 0.35, 0.40, 0.20]),
|
||||
'fatigue_grade': np.random.choice([0, 1, 2, 3], n, p=[0.3, 0.4, 0.2, 0.1]),
|
||||
'nausea_grade': np.random.choice([0, 1, 2, 3], n, p=[0.4, 0.35, 0.20, 0.05]),
|
||||
'neutropenia_grade': np.random.choice([0, 1, 2, 3, 4], n, p=[0.5, 0.2, 0.15, 0.10, 0.05]),
|
||||
})
|
||||
|
||||
return data
|
||||
|
||||
|
||||
def main():
|
||||
parser = argparse.ArgumentParser(description='Generate clinical cohort tables')
|
||||
parser.add_argument('input_file', type=str, nargs='?', default=None,
|
||||
help='CSV file with cohort data (if not provided, uses example data)')
|
||||
parser.add_argument('-o', '--output-dir', type=str, default='tables',
|
||||
help='Output directory (default: tables)')
|
||||
parser.add_argument('--group-col', type=str, default='group',
|
||||
help='Column name for grouping variable')
|
||||
parser.add_argument('--example', action='store_true',
|
||||
help='Generate tables using example data')
|
||||
|
||||
args = parser.parse_args()
|
||||
|
||||
# Create output directory
|
||||
output_dir = Path(args.output_dir)
|
||||
output_dir.mkdir(parents=True, exist_ok=True)
|
||||
|
||||
# Load or create data
|
||||
if args.example or args.input_file is None:
|
||||
print("Generating example dataset...")
|
||||
data = create_example_data()
|
||||
else:
|
||||
print(f"Loading data from {args.input_file}...")
|
||||
data = pd.read_csv(args.input_file)
|
||||
|
||||
print(f"Dataset: {len(data)} patients, {len(data[args.group_col].unique())} groups")
|
||||
print(f"Groups: {data[args.group_col].value_counts().to_dict()}")
|
||||
|
||||
# Generate Table 1: Baseline characteristics
|
||||
print("\nGenerating baseline characteristics table...")
|
||||
baseline_table = generate_baseline_table(
|
||||
data,
|
||||
group_col=args.group_col,
|
||||
output_file=output_dir / 'table1_baseline.csv'
|
||||
)
|
||||
|
||||
# Generate LaTeX code for baseline table
|
||||
latex_code = generate_latex_table(
|
||||
baseline_table,
|
||||
caption="Baseline patient demographics and clinical characteristics",
|
||||
label="baseline"
|
||||
)
|
||||
with open(output_dir / 'table1_baseline.tex', 'w') as f:
|
||||
f.write(latex_code)
|
||||
print(f"LaTeX code saved to: {output_dir}/table1_baseline.tex")
|
||||
|
||||
# Generate Table 2: Efficacy outcomes
|
||||
if 'best_response' in data.columns:
|
||||
print("\nGenerating efficacy outcomes table...")
|
||||
efficacy_table = generate_efficacy_table(
|
||||
data,
|
||||
group_col=args.group_col,
|
||||
output_file=output_dir / 'table2_efficacy.csv'
|
||||
)
|
||||
|
||||
latex_code = generate_latex_table(
|
||||
efficacy_table,
|
||||
caption="Treatment efficacy outcomes by group",
|
||||
label="efficacy"
|
||||
)
|
||||
with open(output_dir / 'table2_efficacy.tex', 'w') as f:
|
||||
f.write(latex_code)
|
||||
|
||||
# Generate Table 3: Safety (identify AE columns)
|
||||
ae_columns = [col for col in data.columns if col.endswith('_grade')]
|
||||
if ae_columns:
|
||||
print("\nGenerating safety table...")
|
||||
safety_table = generate_safety_table(
|
||||
data,
|
||||
ae_columns=ae_columns,
|
||||
group_col=args.group_col,
|
||||
output_file=output_dir / 'table3_safety.csv'
|
||||
)
|
||||
|
||||
latex_code = generate_latex_table(
|
||||
safety_table,
|
||||
caption="Treatment-emergent adverse events by group (CTCAE v5.0)",
|
||||
label="safety"
|
||||
)
|
||||
with open(output_dir / 'table3_safety.tex', 'w') as f:
|
||||
f.write(latex_code)
|
||||
|
||||
print(f"\nAll tables generated successfully in {output_dir}/")
|
||||
print("Files created:")
|
||||
print(" - table1_baseline.csv / .tex")
|
||||
print(" - table2_efficacy.csv / .tex (if response data available)")
|
||||
print(" - table3_safety.csv / .tex (if AE data available)")
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
main()
|
||||
|
||||
|
||||
# Example usage:
|
||||
# python create_cohort_tables.py cohort_data.csv -o tables/
|
||||
# python create_cohort_tables.py --example # Generate example tables
|
||||
#
|
||||
# Input CSV format:
|
||||
# patient_id,group,age,sex,ecog_ps,stage,best_response,fatigue_grade,nausea_grade,...
|
||||
# PT001,Biomarker+,65,Male,0-1,IV,PR,1,0,...
|
||||
# PT002,Biomarker-,58,Female,0-1,III,SD,2,1,...
|
||||
# ...
|
||||
|
||||
422
skills/clinical-decision-support/scripts/generate_survival_analysis.py
Executable file
422
skills/clinical-decision-support/scripts/generate_survival_analysis.py
Executable file
@@ -0,0 +1,422 @@
|
||||
#!/usr/bin/env python3
|
||||
"""
|
||||
Generate Kaplan-Meier Survival Curves for Clinical Decision Support Documents
|
||||
|
||||
This script creates publication-quality survival curves with:
|
||||
- Kaplan-Meier survival estimates
|
||||
- 95% confidence intervals
|
||||
- Log-rank test statistics
|
||||
- Hazard ratios with confidence intervals
|
||||
- Number at risk tables
|
||||
- Median survival annotations
|
||||
|
||||
Dependencies: lifelines, matplotlib, pandas, numpy
|
||||
"""
|
||||
|
||||
import pandas as pd
|
||||
import numpy as np
|
||||
import matplotlib.pyplot as plt
|
||||
from lifelines import KaplanMeierFitter
|
||||
from lifelines.statistics import logrank_test, multivariate_logrank_test
|
||||
from lifelines import CoxPHFitter
|
||||
import argparse
|
||||
from pathlib import Path
|
||||
|
||||
|
||||
def load_survival_data(filepath):
|
||||
"""
|
||||
Load survival data from CSV file.
|
||||
|
||||
Expected columns:
|
||||
- patient_id: Unique patient identifier
|
||||
- time: Survival time (months or days)
|
||||
- event: Event indicator (1=event occurred, 0=censored)
|
||||
- group: Stratification variable (e.g., 'Biomarker+', 'Biomarker-')
|
||||
- Optional: Additional covariates for Cox regression
|
||||
|
||||
Returns:
|
||||
pandas.DataFrame
|
||||
"""
|
||||
df = pd.read_csv(filepath)
|
||||
|
||||
# Validate required columns
|
||||
required_cols = ['patient_id', 'time', 'event', 'group']
|
||||
missing = [col for col in required_cols if col not in df.columns]
|
||||
if missing:
|
||||
raise ValueError(f"Missing required columns: {missing}")
|
||||
|
||||
# Convert event to boolean if needed
|
||||
df['event'] = df['event'].astype(bool)
|
||||
|
||||
return df
|
||||
|
||||
|
||||
def calculate_median_survival(kmf):
|
||||
"""Calculate median survival with 95% CI."""
|
||||
median = kmf.median_survival_time_
|
||||
ci = kmf.confidence_interval_survival_function_
|
||||
|
||||
# Find time when survival crosses 0.5
|
||||
if median == np.inf:
|
||||
return None, None, None
|
||||
|
||||
# Get CI at median
|
||||
idx = np.argmin(np.abs(kmf.survival_function_.index - median))
|
||||
lower_ci = ci.iloc[idx]['KM_estimate_lower_0.95']
|
||||
upper_ci = ci.iloc[idx]['KM_estimate_upper_0.95']
|
||||
|
||||
return median, lower_ci, upper_ci
|
||||
|
||||
|
||||
def generate_kaplan_meier_plot(data, time_col='time', event_col='event',
|
||||
group_col='group', output_path='survival_curve.pdf',
|
||||
title='Kaplan-Meier Survival Curve',
|
||||
xlabel='Time (months)', ylabel='Survival Probability'):
|
||||
"""
|
||||
Generate Kaplan-Meier survival curve comparing groups.
|
||||
|
||||
Parameters:
|
||||
data: DataFrame with survival data
|
||||
time_col: Column name for survival time
|
||||
event_col: Column name for event indicator
|
||||
group_col: Column name for stratification
|
||||
output_path: Path to save figure
|
||||
title: Plot title
|
||||
xlabel: X-axis label (specify units)
|
||||
ylabel: Y-axis label
|
||||
"""
|
||||
|
||||
# Create figure and axis
|
||||
fig, ax = plt.subplots(figsize=(10, 6))
|
||||
|
||||
# Get unique groups
|
||||
groups = data[group_col].unique()
|
||||
|
||||
# Colors for groups (colorblind-friendly)
|
||||
colors = ['#0173B2', '#DE8F05', '#029E73', '#CC78BC', '#CA9161']
|
||||
|
||||
kmf_models = {}
|
||||
median_survivals = {}
|
||||
|
||||
# Plot each group
|
||||
for i, group in enumerate(groups):
|
||||
group_data = data[data[group_col] == group]
|
||||
|
||||
# Fit Kaplan-Meier
|
||||
kmf = KaplanMeierFitter()
|
||||
kmf.fit(group_data[time_col], group_data[event_col], label=str(group))
|
||||
|
||||
# Plot survival curve
|
||||
kmf.plot_survival_function(ax=ax, ci_show=True, color=colors[i % len(colors)],
|
||||
linewidth=2, alpha=0.8)
|
||||
|
||||
# Store model
|
||||
kmf_models[group] = kmf
|
||||
|
||||
# Calculate median survival
|
||||
median, lower, upper = calculate_median_survival(kmf)
|
||||
median_survivals[group] = (median, lower, upper)
|
||||
|
||||
# Log-rank test
|
||||
if len(groups) == 2:
|
||||
group1_data = data[data[group_col] == groups[0]]
|
||||
group2_data = data[data[group_col] == groups[1]]
|
||||
|
||||
results = logrank_test(
|
||||
group1_data[time_col], group2_data[time_col],
|
||||
group1_data[event_col], group2_data[event_col]
|
||||
)
|
||||
|
||||
p_value = results.p_value
|
||||
test_statistic = results.test_statistic
|
||||
|
||||
# Add log-rank test result to plot
|
||||
ax.text(0.02, 0.15, f'Log-rank test:\np = {p_value:.4f}',
|
||||
transform=ax.transAxes, fontsize=10,
|
||||
verticalalignment='top',
|
||||
bbox=dict(boxstyle='round', facecolor='wheat', alpha=0.5))
|
||||
else:
|
||||
# Multivariate log-rank for >2 groups
|
||||
results = multivariate_logrank_test(data[time_col], data[group_col], data[event_col])
|
||||
p_value = results.p_value
|
||||
test_statistic = results.test_statistic
|
||||
|
||||
ax.text(0.02, 0.15, f'Log-rank test:\np = {p_value:.4f}\n({len(groups)} groups)',
|
||||
transform=ax.transAxes, fontsize=10,
|
||||
verticalalignment='top',
|
||||
bbox=dict(boxstyle='round', facecolor='wheat', alpha=0.5))
|
||||
|
||||
# Add median survival annotations
|
||||
y_pos = 0.95
|
||||
for group, (median, lower, upper) in median_survivals.items():
|
||||
if median is not None:
|
||||
ax.text(0.98, y_pos, f'{group}: {median:.1f} months (95% CI {lower:.1f}-{upper:.1f})',
|
||||
transform=ax.transAxes, fontsize=9, ha='right',
|
||||
verticalalignment='top')
|
||||
else:
|
||||
ax.text(0.98, y_pos, f'{group}: Not reached',
|
||||
transform=ax.transAxes, fontsize=9, ha='right',
|
||||
verticalalignment='top')
|
||||
y_pos -= 0.05
|
||||
|
||||
# Formatting
|
||||
ax.set_xlabel(xlabel, fontsize=12, fontweight='bold')
|
||||
ax.set_ylabel(ylabel, fontsize=12, fontweight='bold')
|
||||
ax.set_title(title, fontsize=14, fontweight='bold', pad=15)
|
||||
ax.legend(loc='lower left', frameon=True, fontsize=10)
|
||||
ax.grid(True, alpha=0.3, linestyle='--')
|
||||
ax.set_ylim([0, 1.05])
|
||||
|
||||
plt.tight_layout()
|
||||
|
||||
# Save figure
|
||||
plt.savefig(output_path, dpi=300, bbox_inches='tight')
|
||||
print(f"Survival curve saved to: {output_path}")
|
||||
|
||||
# Also save as PNG for easy viewing
|
||||
png_path = Path(output_path).with_suffix('.png')
|
||||
plt.savefig(png_path, dpi=300, bbox_inches='tight')
|
||||
print(f"PNG version saved to: {png_path}")
|
||||
|
||||
plt.close()
|
||||
|
||||
return kmf_models, p_value
|
||||
|
||||
|
||||
def generate_number_at_risk_table(data, time_col='time', event_col='event',
|
||||
group_col='group', time_points=None):
|
||||
"""
|
||||
Generate number at risk table for survival analysis.
|
||||
|
||||
Parameters:
|
||||
data: DataFrame with survival data
|
||||
time_points: List of time points for risk table (if None, auto-generate)
|
||||
|
||||
Returns:
|
||||
DataFrame with number at risk at each time point
|
||||
"""
|
||||
|
||||
if time_points is None:
|
||||
# Auto-generate time points (every 6 months up to max time)
|
||||
max_time = data[time_col].max()
|
||||
time_points = np.arange(0, max_time + 6, 6)
|
||||
|
||||
groups = data[group_col].unique()
|
||||
risk_table = pd.DataFrame(index=time_points, columns=groups)
|
||||
|
||||
for group in groups:
|
||||
group_data = data[data[group_col] == group]
|
||||
|
||||
for t in time_points:
|
||||
# Number at risk = patients who haven't had event and haven't been censored before time t
|
||||
at_risk = len(group_data[group_data[time_col] >= t])
|
||||
risk_table.loc[t, group] = at_risk
|
||||
|
||||
return risk_table
|
||||
|
||||
|
||||
def calculate_hazard_ratio(data, time_col='time', event_col='event', group_col='group',
|
||||
reference_group=None):
|
||||
"""
|
||||
Calculate hazard ratio using Cox proportional hazards regression.
|
||||
|
||||
Parameters:
|
||||
data: DataFrame
|
||||
reference_group: Reference group for comparison (if None, uses first group)
|
||||
|
||||
Returns:
|
||||
Hazard ratio, 95% CI, p-value
|
||||
"""
|
||||
|
||||
# Encode group as binary for Cox regression
|
||||
groups = data[group_col].unique()
|
||||
if len(groups) != 2:
|
||||
print("Warning: Cox HR calculation assumes 2 groups. Using first 2 groups.")
|
||||
groups = groups[:2]
|
||||
|
||||
if reference_group is None:
|
||||
reference_group = groups[0]
|
||||
|
||||
# Create binary indicator (1 for comparison group, 0 for reference)
|
||||
data_cox = data.copy()
|
||||
data_cox['group_binary'] = (data_cox[group_col] != reference_group).astype(int)
|
||||
|
||||
# Fit Cox model
|
||||
cph = CoxPHFitter()
|
||||
cph.fit(data_cox[[time_col, event_col, 'group_binary']],
|
||||
duration_col=time_col, event_col=event_col)
|
||||
|
||||
# Extract results
|
||||
hr = np.exp(cph.params_['group_binary'])
|
||||
ci = np.exp(cph.confidence_intervals_.loc['group_binary'].values)
|
||||
p_value = cph.summary.loc['group_binary', 'p']
|
||||
|
||||
return hr, ci[0], ci[1], p_value
|
||||
|
||||
|
||||
def generate_report(data, output_dir, prefix='survival'):
|
||||
"""
|
||||
Generate comprehensive survival analysis report.
|
||||
|
||||
Creates:
|
||||
- Kaplan-Meier curves (PDF and PNG)
|
||||
- Number at risk table (CSV)
|
||||
- Statistical summary (TXT)
|
||||
- LaTeX table code (TEX)
|
||||
"""
|
||||
|
||||
output_dir = Path(output_dir)
|
||||
output_dir.mkdir(parents=True, exist_ok=True)
|
||||
|
||||
# Generate survival curve
|
||||
kmf_models, logrank_p = generate_kaplan_meier_plot(
|
||||
data,
|
||||
output_path=output_dir / f'{prefix}_kaplan_meier.pdf',
|
||||
title='Survival Analysis by Group'
|
||||
)
|
||||
|
||||
# Number at risk table
|
||||
risk_table = generate_number_at_risk_table(data)
|
||||
risk_table.to_csv(output_dir / f'{prefix}_number_at_risk.csv')
|
||||
|
||||
# Calculate hazard ratio
|
||||
hr, ci_lower, ci_upper, hr_p = calculate_hazard_ratio(data)
|
||||
|
||||
# Generate statistical summary
|
||||
with open(output_dir / f'{prefix}_statistics.txt', 'w') as f:
|
||||
f.write("SURVIVAL ANALYSIS STATISTICAL SUMMARY\n")
|
||||
f.write("=" * 60 + "\n\n")
|
||||
|
||||
groups = data['group'].unique()
|
||||
for group in groups:
|
||||
kmf = kmf_models[group]
|
||||
median = kmf.median_survival_time_
|
||||
|
||||
# Calculate survival rates at common time points
|
||||
try:
|
||||
surv_12m = kmf.survival_function_at_times(12).values[0]
|
||||
surv_24m = kmf.survival_function_at_times(24).values[0] if data['time'].max() >= 24 else None
|
||||
except:
|
||||
surv_12m = None
|
||||
surv_24m = None
|
||||
|
||||
f.write(f"Group: {group}\n")
|
||||
f.write(f" N = {len(data[data['group'] == group])}\n")
|
||||
f.write(f" Events = {data[data['group'] == group]['event'].sum()}\n")
|
||||
f.write(f" Median survival: {median:.1f} months\n" if median != np.inf else " Median survival: Not reached\n")
|
||||
if surv_12m is not None:
|
||||
f.write(f" 12-month survival rate: {surv_12m*100:.1f}%\n")
|
||||
if surv_24m is not None:
|
||||
f.write(f" 24-month survival rate: {surv_24m*100:.1f}%\n")
|
||||
f.write("\n")
|
||||
|
||||
f.write(f"Log-Rank Test:\n")
|
||||
f.write(f" p-value = {logrank_p:.4f}\n")
|
||||
f.write(f" Interpretation: {'Significant' if logrank_p < 0.05 else 'Not significant'} difference in survival\n\n")
|
||||
|
||||
if len(groups) == 2:
|
||||
f.write(f"Hazard Ratio ({groups[1]} vs {groups[0]}):\n")
|
||||
f.write(f" HR = {hr:.2f} (95% CI {ci_lower:.2f}-{ci_upper:.2f})\n")
|
||||
f.write(f" p-value = {hr_p:.4f}\n")
|
||||
f.write(f" Interpretation: {groups[1]} has {((1-hr)*100):.0f}% {'reduction' if hr < 1 else 'increase'} in risk\n")
|
||||
|
||||
# Generate LaTeX table code
|
||||
with open(output_dir / f'{prefix}_latex_table.tex', 'w') as f:
|
||||
f.write("% LaTeX table code for survival outcomes\n")
|
||||
f.write("\\begin{table}[H]\n")
|
||||
f.write("\\centering\n")
|
||||
f.write("\\small\n")
|
||||
f.write("\\begin{tabular}{lcccc}\n")
|
||||
f.write("\\toprule\n")
|
||||
f.write("\\textbf{Endpoint} & \\textbf{Group A} & \\textbf{Group B} & \\textbf{HR (95\\% CI)} & \\textbf{p-value} \\\\\n")
|
||||
f.write("\\midrule\n")
|
||||
|
||||
# Add median survival row
|
||||
for i, group in enumerate(groups):
|
||||
kmf = kmf_models[group]
|
||||
median = kmf.median_survival_time_
|
||||
if i == 0:
|
||||
f.write(f"Median survival, months (95\\% CI) & ")
|
||||
if median != np.inf:
|
||||
f.write(f"{median:.1f} & ")
|
||||
else:
|
||||
f.write("NR & ")
|
||||
else:
|
||||
if median != np.inf:
|
||||
f.write(f"{median:.1f} & ")
|
||||
else:
|
||||
f.write("NR & ")
|
||||
|
||||
f.write(f"{hr:.2f} ({ci_lower:.2f}-{ci_upper:.2f}) & {hr_p:.3f} \\\\\n")
|
||||
|
||||
# Add 12-month survival rate
|
||||
f.write("12-month survival rate (\\%) & ")
|
||||
for group in groups:
|
||||
kmf = kmf_models[group]
|
||||
try:
|
||||
surv_12m = kmf.survival_function_at_times(12).values[0]
|
||||
f.write(f"{surv_12m*100:.0f}\\% & ")
|
||||
except:
|
||||
f.write("-- & ")
|
||||
f.write("-- & -- \\\\\n")
|
||||
|
||||
f.write("\\bottomrule\n")
|
||||
f.write("\\end{tabular}\n")
|
||||
f.write(f"\\caption{{Survival outcomes by group (log-rank p={logrank_p:.3f})}}\n")
|
||||
f.write("\\end{table}\n")
|
||||
|
||||
print(f"\nAnalysis complete! Files saved to {output_dir}/")
|
||||
print(f" - Survival curves: {prefix}_kaplan_meier.pdf/png")
|
||||
print(f" - Statistics: {prefix}_statistics.txt")
|
||||
print(f" - LaTeX table: {prefix}_latex_table.tex")
|
||||
print(f" - Risk table: {prefix}_number_at_risk.csv")
|
||||
|
||||
|
||||
def main():
|
||||
parser = argparse.ArgumentParser(description='Generate Kaplan-Meier survival curves')
|
||||
parser.add_argument('input_file', type=str, help='CSV file with survival data')
|
||||
parser.add_argument('-o', '--output', type=str, default='survival_output',
|
||||
help='Output directory (default: survival_output)')
|
||||
parser.add_argument('-t', '--title', type=str, default='Kaplan-Meier Survival Curve',
|
||||
help='Plot title')
|
||||
parser.add_argument('-x', '--xlabel', type=str, default='Time (months)',
|
||||
help='X-axis label')
|
||||
parser.add_argument('-y', '--ylabel', type=str, default='Survival Probability',
|
||||
help='Y-axis label')
|
||||
parser.add_argument('--time-col', type=str, default='time',
|
||||
help='Column name for time variable')
|
||||
parser.add_argument('--event-col', type=str, default='event',
|
||||
help='Column name for event indicator')
|
||||
parser.add_argument('--group-col', type=str, default='group',
|
||||
help='Column name for grouping variable')
|
||||
|
||||
args = parser.parse_args()
|
||||
|
||||
# Load data
|
||||
print(f"Loading data from {args.input_file}...")
|
||||
data = load_survival_data(args.input_file)
|
||||
print(f"Loaded {len(data)} patients")
|
||||
print(f"Groups: {data[args.group_col].value_counts().to_dict()}")
|
||||
|
||||
# Generate analysis
|
||||
generate_report(
|
||||
data,
|
||||
output_dir=args.output,
|
||||
prefix='survival'
|
||||
)
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
main()
|
||||
|
||||
|
||||
# Example usage:
|
||||
# python generate_survival_analysis.py survival_data.csv -o figures/ -t "PFS by PD-L1 Status"
|
||||
#
|
||||
# Input CSV format:
|
||||
# patient_id,time,event,group
|
||||
# PT001,12.3,1,PD-L1+
|
||||
# PT002,8.5,1,PD-L1-
|
||||
# PT003,18.2,0,PD-L1+
|
||||
# ...
|
||||
|
||||
335
skills/clinical-decision-support/scripts/validate_cds_document.py
Executable file
335
skills/clinical-decision-support/scripts/validate_cds_document.py
Executable file
@@ -0,0 +1,335 @@
|
||||
#!/usr/bin/env python3
|
||||
"""
|
||||
Validate Clinical Decision Support Documents for Quality and Completeness
|
||||
|
||||
Checks for:
|
||||
- Evidence citations for all recommendations
|
||||
- Statistical reporting completeness
|
||||
- Biomarker nomenclature consistency
|
||||
- Required sections present
|
||||
- HIPAA de-identification
|
||||
- GRADE recommendation format
|
||||
|
||||
Dependencies: None (pure Python)
|
||||
"""
|
||||
|
||||
import re
|
||||
import argparse
|
||||
from pathlib import Path
|
||||
from collections import defaultdict
|
||||
|
||||
|
||||
class CDSValidator:
|
||||
"""Validator for clinical decision support documents."""
|
||||
|
||||
def __init__(self, filepath):
|
||||
self.filepath = filepath
|
||||
with open(filepath, 'r', encoding='utf-8', errors='ignore') as f:
|
||||
self.content = f.read()
|
||||
|
||||
self.errors = []
|
||||
self.warnings = []
|
||||
self.info = []
|
||||
|
||||
def validate_all(self):
|
||||
"""Run all validation checks."""
|
||||
|
||||
print(f"Validating: {self.filepath}")
|
||||
print("="*70)
|
||||
|
||||
self.check_required_sections()
|
||||
self.check_evidence_citations()
|
||||
self.check_recommendation_grading()
|
||||
self.check_statistical_reporting()
|
||||
self.check_hipaa_identifiers()
|
||||
self.check_biomarker_nomenclature()
|
||||
|
||||
return self.generate_report()
|
||||
|
||||
def check_required_sections(self):
|
||||
"""Check if required sections are present."""
|
||||
|
||||
# Cohort analysis required sections
|
||||
cohort_sections = [
|
||||
'cohort characteristics',
|
||||
'biomarker',
|
||||
'outcomes',
|
||||
'statistical analysis',
|
||||
'clinical implications',
|
||||
'references'
|
||||
]
|
||||
|
||||
# Treatment recommendation required sections
|
||||
rec_sections = [
|
||||
'evidence',
|
||||
'recommendation',
|
||||
'monitoring',
|
||||
'references'
|
||||
]
|
||||
|
||||
content_lower = self.content.lower()
|
||||
|
||||
# Check which document type
|
||||
is_cohort = 'cohort' in content_lower
|
||||
is_recommendation = 'recommendation' in content_lower
|
||||
|
||||
if is_cohort:
|
||||
missing = [sec for sec in cohort_sections if sec not in content_lower]
|
||||
if missing:
|
||||
self.warnings.append(f"Cohort analysis may be missing sections: {', '.join(missing)}")
|
||||
else:
|
||||
self.info.append("All cohort analysis sections present")
|
||||
|
||||
if is_recommendation:
|
||||
missing = [sec for sec in rec_sections if sec not in content_lower]
|
||||
if missing:
|
||||
self.errors.append(f"Recommendation document missing required sections: {', '.join(missing)}")
|
||||
else:
|
||||
self.info.append("All recommendation sections present")
|
||||
|
||||
def check_evidence_citations(self):
|
||||
"""Check that recommendations have citations."""
|
||||
|
||||
# Find recommendation statements
|
||||
rec_pattern = r'(recommend|should|prefer|suggest|consider)(.*?)(?:\n\n|\Z)'
|
||||
recommendations = re.findall(rec_pattern, self.content, re.IGNORECASE | re.DOTALL)
|
||||
|
||||
# Find citations
|
||||
citation_patterns = [
|
||||
r'\[\d+\]', # Numbered citations [1]
|
||||
r'\(.*?\d{4}\)', # Author year (Smith 2020)
|
||||
r'et al\.', # Et al citations
|
||||
r'NCCN|ASCO|ESMO', # Guideline references
|
||||
]
|
||||
|
||||
uncited_recommendations = []
|
||||
|
||||
for i, (_, rec_text) in enumerate(recommendations):
|
||||
has_citation = any(re.search(pattern, rec_text) for pattern in citation_patterns)
|
||||
|
||||
if not has_citation:
|
||||
snippet = rec_text[:60].strip() + '...'
|
||||
uncited_recommendations.append(snippet)
|
||||
|
||||
if uncited_recommendations:
|
||||
self.warnings.append(f"Found {len(uncited_recommendations)} recommendations without citations")
|
||||
for rec in uncited_recommendations[:3]: # Show first 3
|
||||
self.warnings.append(f" - {rec}")
|
||||
else:
|
||||
self.info.append(f"All {len(recommendations)} recommendations have citations")
|
||||
|
||||
def check_recommendation_grading(self):
|
||||
"""Check for GRADE-style recommendation strength."""
|
||||
|
||||
# Look for GRADE notation (1A, 1B, 2A, 2B, 2C)
|
||||
grade_pattern = r'GRADE\s*[12][A-C]|Grade\s*[12][A-C]|\(?\s*[12][A-C]\s*\)?'
|
||||
grades = re.findall(grade_pattern, self.content, re.IGNORECASE)
|
||||
|
||||
# Look for strong/conditional language
|
||||
strong_pattern = r'(strong|we recommend|should)'
|
||||
conditional_pattern = r'(conditional|weak|we suggest|may consider|could consider)'
|
||||
|
||||
strong_count = len(re.findall(strong_pattern, self.content, re.IGNORECASE))
|
||||
conditional_count = len(re.findall(conditional_pattern, self.content, re.IGNORECASE))
|
||||
|
||||
if grades:
|
||||
self.info.append(f"Found {len(grades)} GRADE-style recommendations")
|
||||
else:
|
||||
self.warnings.append("No GRADE-style recommendation grading found (1A, 1B, 2A, etc.)")
|
||||
|
||||
if strong_count > 0 or conditional_count > 0:
|
||||
self.info.append(f"Recommendation language: {strong_count} strong, {conditional_count} conditional")
|
||||
else:
|
||||
self.warnings.append("No clear recommendation strength language (strong/conditional) found")
|
||||
|
||||
def check_statistical_reporting(self):
|
||||
"""Check for proper statistical reporting."""
|
||||
|
||||
# Check for p-values
|
||||
p_values = re.findall(r'p\s*[=<>]\s*[\d.]+', self.content, re.IGNORECASE)
|
||||
|
||||
# Check for confidence intervals
|
||||
ci_pattern = r'95%\s*CI|confidence interval'
|
||||
cis = re.findall(ci_pattern, self.content, re.IGNORECASE)
|
||||
|
||||
# Check for hazard ratios
|
||||
hr_pattern = r'HR\s*[=:]\s*[\d.]+'
|
||||
hrs = re.findall(hr_pattern, self.content)
|
||||
|
||||
# Check for sample sizes
|
||||
n_pattern = r'n\s*=\s*\d+'
|
||||
sample_sizes = re.findall(n_pattern, self.content, re.IGNORECASE)
|
||||
|
||||
if not p_values:
|
||||
self.warnings.append("No p-values found - statistical significance not reported")
|
||||
else:
|
||||
self.info.append(f"Found {len(p_values)} p-values")
|
||||
|
||||
if hrs and not cis:
|
||||
self.warnings.append("Hazard ratios reported without confidence intervals")
|
||||
|
||||
if not sample_sizes:
|
||||
self.warnings.append("Sample sizes (n=X) not clearly reported")
|
||||
|
||||
# Check for common statistical errors
|
||||
if 'p=0.00' in self.content or 'p = 0.00' in self.content:
|
||||
self.warnings.append("Found p=0.00 (should report as p<0.001 instead)")
|
||||
|
||||
def check_hipaa_identifiers(self):
|
||||
"""Check for potential HIPAA identifiers."""
|
||||
|
||||
# 18 HIPAA identifiers (simplified check for common ones)
|
||||
identifiers = {
|
||||
'Names': r'Dr\.\s+[A-Z][a-z]+|Patient:\s*[A-Z][a-z]+',
|
||||
'Specific dates': r'\d{1,2}/\d{1,2}/\d{4}', # MM/DD/YYYY
|
||||
'Phone numbers': r'\d{3}[-.]?\d{3}[-.]?\d{4}',
|
||||
'Email addresses': r'[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}',
|
||||
'SSN': r'\d{3}-\d{2}-\d{4}',
|
||||
'MRN': r'MRN\s*:?\s*\d+',
|
||||
}
|
||||
|
||||
found_identifiers = []
|
||||
|
||||
for identifier_type, pattern in identifiers.items():
|
||||
matches = re.findall(pattern, self.content)
|
||||
if matches:
|
||||
found_identifiers.append(f"{identifier_type}: {len(matches)} instance(s)")
|
||||
|
||||
if found_identifiers:
|
||||
self.errors.append("Potential HIPAA identifiers detected:")
|
||||
for identifier in found_identifiers:
|
||||
self.errors.append(f" - {identifier}")
|
||||
self.errors.append(" ** Ensure proper de-identification before distribution **")
|
||||
else:
|
||||
self.info.append("No obvious HIPAA identifiers detected (basic check only)")
|
||||
|
||||
def check_biomarker_nomenclature(self):
|
||||
"""Check for consistent biomarker nomenclature."""
|
||||
|
||||
# Common biomarker naming issues
|
||||
issues = []
|
||||
|
||||
# Check for gene names (should be italicized in LaTeX)
|
||||
gene_names = ['EGFR', 'ALK', 'ROS1', 'BRAF', 'KRAS', 'HER2', 'TP53', 'BRCA1', 'BRCA2']
|
||||
for gene in gene_names:
|
||||
# Check if gene appears but not in italics (\textit{} or \emph{})
|
||||
if gene in self.content:
|
||||
if f'\\textit{{{gene}}}' not in self.content and f'\\emph{{{gene}}}' not in self.content:
|
||||
if '.tex' in self.filepath.suffix:
|
||||
issues.append(f"{gene} should be italicized in LaTeX (\\textit{{{gene}}})")
|
||||
|
||||
# Check for protein vs gene naming
|
||||
# HER2 (protein) vs ERBB2 (gene) - both valid
|
||||
# Check for mutation nomenclature (HGVS format)
|
||||
hgvs_pattern = r'p\.[A-Z]\d+[A-Z]' # e.g., p.L858R
|
||||
hgvs_mutations = re.findall(hgvs_pattern, self.content)
|
||||
|
||||
if hgvs_mutations:
|
||||
self.info.append(f"Found {len(hgvs_mutations)} HGVS protein nomenclature (e.g., p.L858R)")
|
||||
|
||||
# Warn about non-standard mutation format
|
||||
if 'EGFR mutation' in self.content and 'exon' not in self.content.lower():
|
||||
self.warnings.append("EGFR mutation mentioned - specify exon/variant (e.g., exon 19 deletion)")
|
||||
|
||||
if issues:
|
||||
self.warnings.extend(issues)
|
||||
|
||||
def generate_report(self):
|
||||
"""Generate validation report."""
|
||||
|
||||
print("\n" + "="*70)
|
||||
print("VALIDATION REPORT")
|
||||
print("="*70)
|
||||
|
||||
if self.errors:
|
||||
print(f"\n❌ ERRORS ({len(self.errors)}):")
|
||||
for error in self.errors:
|
||||
print(f" {error}")
|
||||
|
||||
if self.warnings:
|
||||
print(f"\n⚠️ WARNINGS ({len(self.warnings)}):")
|
||||
for warning in self.warnings:
|
||||
print(f" {warning}")
|
||||
|
||||
if self.info:
|
||||
print(f"\n✓ PASSED CHECKS ({len(self.info)}):")
|
||||
for info in self.info:
|
||||
print(f" {info}")
|
||||
|
||||
# Overall status
|
||||
print("\n" + "="*70)
|
||||
if self.errors:
|
||||
print("STATUS: ❌ VALIDATION FAILED - Address errors before distribution")
|
||||
return False
|
||||
elif self.warnings:
|
||||
print("STATUS: ⚠️ VALIDATION PASSED WITH WARNINGS - Review recommended")
|
||||
return True
|
||||
else:
|
||||
print("STATUS: ✓ VALIDATION PASSED - Document meets quality standards")
|
||||
return True
|
||||
|
||||
def save_report(self, output_file):
|
||||
"""Save validation report to file."""
|
||||
|
||||
with open(output_file, 'w') as f:
|
||||
f.write("CLINICAL DECISION SUPPORT DOCUMENT VALIDATION REPORT\n")
|
||||
f.write("="*70 + "\n")
|
||||
f.write(f"Document: {self.filepath}\n")
|
||||
f.write(f"Validated: {Path.cwd()}\n\n")
|
||||
|
||||
if self.errors:
|
||||
f.write(f"ERRORS ({len(self.errors)}):\n")
|
||||
for error in self.errors:
|
||||
f.write(f" - {error}\n")
|
||||
f.write("\n")
|
||||
|
||||
if self.warnings:
|
||||
f.write(f"WARNINGS ({len(self.warnings)}):\n")
|
||||
for warning in self.warnings:
|
||||
f.write(f" - {warning}\n")
|
||||
f.write("\n")
|
||||
|
||||
if self.info:
|
||||
f.write(f"PASSED CHECKS ({len(self.info)}):\n")
|
||||
for info in self.info:
|
||||
f.write(f" - {info}\n")
|
||||
|
||||
print(f"\nValidation report saved to: {output_file}")
|
||||
|
||||
|
||||
def main():
|
||||
parser = argparse.ArgumentParser(description='Validate clinical decision support documents')
|
||||
parser.add_argument('input_file', type=str, help='Document to validate (.tex, .md, .txt)')
|
||||
parser.add_argument('-o', '--output', type=str, default=None,
|
||||
help='Save validation report to file')
|
||||
parser.add_argument('--strict', action='store_true',
|
||||
help='Treat warnings as errors')
|
||||
|
||||
args = parser.parse_args()
|
||||
|
||||
# Validate
|
||||
validator = CDSValidator(args.input_file)
|
||||
passed = validator.validate_all()
|
||||
|
||||
# Save report if requested
|
||||
if args.output:
|
||||
validator.save_report(args.output)
|
||||
|
||||
# Exit code
|
||||
if args.strict and (validator.errors or validator.warnings):
|
||||
exit(1)
|
||||
elif validator.errors:
|
||||
exit(1)
|
||||
else:
|
||||
exit(0)
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
main()
|
||||
|
||||
|
||||
# Example usage:
|
||||
# python validate_cds_document.py cohort_analysis.tex
|
||||
# python validate_cds_document.py treatment_recommendations.tex -o validation_report.txt
|
||||
# python validate_cds_document.py document.tex --strict # Warnings cause failure
|
||||
|
||||
Reference in New Issue
Block a user