skills.rest
mims-harvard
mims-harvardOFFICIAL
Explore all skills by this author →
Website𝕏@marinkazitnikGitHub

tooluniverse-immunotherapy-response-prediction

Predicts immune checkpoint inhibitor response by integrating TMB, MSI, PD-L1, and mutation biomarkers.

1.7k|254|Updated Mar 3, 2025
Search Tags:#biomarker#oncology#tmb#immunotherapy#checkpoint inhibitor#pd-l1#response prediction
One-click install
npx skills add https://github.com/mims-harvard/ToolUniverse --skill tooluniverse-immunotherapy-response-prediction
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: tooluniverse-immunotherapy-response-prediction
Source: https://github.com/mims-harvard/ToolUniverse/tree/main/plugins/tooluniverse/skills/tooluniverse-immunotherapy-response-prediction
Command: npx skills add https://github.com/mims-harvard/ToolUniverse --skill tooluniverse-immunotherapy-response-prediction

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Oncology teams need to decide whether a patient will benefit from immune checkpoint inhibitors, but response depends on many biomarkers (TMB, MSI, PD-L1, resistance mutations) scattered across databases. This Skill integrates them into a single quantitative ICI Response Score with drug-specific recommendations.

Core Features & Use Cases

  • Multi-Biomarker Scoring: Combines TMB, MSI status, PD-L1 expression, neoantigen burden, and sensitivity/resistance mutations into a 0-100 ICI Response Score with transparent component breakdown.
  • Drug-Specific Recommendations: Selects among pembrolizumab, nivolumab, atezolizumab, and other ICIs using cancer-specific algorithms backed by FDA approvals and clinical trial evidence.
  • Resistance Risk Assessment: Flags resistance mutations (STK11, JAK1/2, B2M, KEAP1) and pathway-level resistance mechanisms with mitigation strategies.
  • Use Case: Given "NSCLC, TMB 25 mut/Mb, PD-L1 80%, no STK11 mutation", the Skill computes a HIGH response score and recommends pembrolizumab monotherapy with KEYNOTE-024 evidence, a monitoring plan, and a full evidence-graded report.

Quick Start

Ask the agent to predict immunotherapy response for a patient with melanoma, BRAF V600E, TMB 15 mut/Mb, and PD-L1 50%.

Frequently Asked Questions about tooluniverse-immunotherapy-response-prediction

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I predict immunotherapy response from TMB and PD-L1?▼

Provide the cancer type plus TMB value and PD-L1 expression, and the Skill classifies each biomarker, applies cancer-specific thresholds, and sums component scores into a 0-100 ICI Response Score. Scores of 70-100 indicate HIGH response likelihood, 40-69 MODERATE, and below 40 LOW.

Which checkpoint inhibitor should I use for NSCLC with high PD-L1?▼

For NSCLC with PD-L1 TPS >=50% and no STK11 or EGFR mutations, the Skill recommends pembrolizumab monotherapy based on KEYNOTE-024 evidence. For PD-L1 1-49% it recommends pembrolizumab plus chemotherapy, verified against FDA indications.

What input format does the immunotherapy prediction skill accept?▼

It requires a cancer type plus at least a mutation list or TMB value, such as "Melanoma, BRAF V600E, TMB 15, PD-L1 50%, MSS". Optional inputs include MSI status, immune infiltration data, HLA type, and prior treatments.

Does the skill account for immunotherapy resistance mutations?▼

Yes, it applies score penalties for resistance mutations including STK11 (-10), JAK1/2 (-10 each), B2M (-15), KEAP1 (-5), PTEN (-5), and EGFR (-5), and assesses pathway-level resistance in IFN-gamma signaling and antigen presentation.

What are the limitations of biomarker-based ICI response prediction?▼

Predictions depend on available biomarkers; missing TMB, MSI, or PD-L1 data lowers confidence and forces neutral scoring. Some cancers like RCC respond to ICIs despite low TMB, so the Skill applies cancer-specific context rather than universal thresholds.

Related Skills

Explore More Education & Research Skills→
biomaps-infrabiomaps-infra

db-cbioportal

Query cBioPortal REST API for cancer genomics data using Python requests and pandas.

Official
Advanced
kamenskamens
26

hstack-summarize-research

Synthesize medical research into structured sections for clinician discussions.

Community
Advanced
ammawlaammawla
26

cross-reference

Link ENCODE genomic experiments to PubMed, ClinVar, GWAS Catalog, GTEx, and Ensembl.

Community
Advanced
OwnLabAIOwnLabAI
21

research-grants

Generate agency-specific grant proposal frameworks with templates and budgeting guidance.

Official
Advanced