Tooluniverse Clinical Trial Matching

What it does

This skill matches patients to relevant clinical trials using ToolUniverse's database, accelerating recruitment and improving patient access to innovative treatments. It transforms a patient’s molecular profile and clinical characteristics into prioritized clinical trial recommendations by searching ClinicalTrials.gov and cross-referencing with databases like CIViC, OpenTargets, ChEMBL, and FDA. Each recommendation is evidence-graded and scored quantitatively, prioritizing matches based on specific biomarkers first, then disease stage and prior treatments.

When to use it

• Finding clinical trials for a patient with NSCLC and an EGFR L858R mutation.
• Identifying trials for a patient with BRAF V600E melanoma who has failed ipilimumab treatment.
• Locating basket trials targeting NTRK fusions.
• Discovering post-CDK4/6 inhibitor clinical trials for breast cancer patients with HER2 amplification.
• Finding clinical trials for KRAS G12C colorectal cancer.

Key capabilities

• Molecular Matching Priority: Prioritizes trials based on specific mutations, then disease stage and prior treatments.
• Evidence-Graded Recommendations: Each trial match has an evidence tier (T1-T4).
• Quantitative Scoring: Provides a Trial Match Score (0-100) for each trial.
• Eligibility Evaluation: Parses and evaluates clinical trial inclusion/exclusion criteria.
• Actionable Output: Includes clear next steps, contact information, and enrollment status.
• Source Referencing: Every statement cites the tool or database source used.
• Completeness Checklist: Provides a checklist showing analysis coverage.

Example prompts

• "What clinical trials are available for my NSCLC with EGFR L858R?"
• "Patient has BRAF V600E melanoma, failed ipilimumab - what trials?"
• "Find basket trials for NTRK fusion."

Tips & gotchas

• English-first queries: Always use English terms when requesting clinical trial matches. The skill will respond in the user’s language.
• Prioritize databases: When uncertain about scientific facts, search relevant databases first rather than relying on reasoning.
• Computation over description: The skill uses Python code (pandas, scipy, statsmodels, matplotlib) to perform calculations and data analysis; it will execute code and report results instead of describing the process.