UC38: Clinical Trial Matching Agent

The Problem

Only 5% of cancer patients enroll in clinical trials, despite trials offering access to cutting-edge therapies. A major barrier is manual matching: oncologists lack time to screen thousands of trials for eligibility, and patient-trial matching requires matching genomic data, comorbidities, prior treatments, and complex inclusion/exclusion criteria. Genomically informed trials are under-enrolled, prolonging development timelines and limiting treatment options.

What the Agent Does

Oncologist triggers agent on patient electronic chart (or scheduled batch for eligible patients)
Agent retrieves patient data: EHR history (diagnoses, prior treatments, performance status), lab results (CBC, metabolic panel), genomic/molecular data (tumor sequencing, mutation profile)
Queries ClinicalTrials.gov API and aggregates inclusion/exclusion criteria from trial records
Performs eligibility matching: compares patient genomic markers (EGFR, ALK, BRCA, etc.) against trial genetic requirements; cross-references comorbidities and prior therapies
Returns ranked list of matching trials with confidence scores, trial phase, enrollment status, location, and contact information
Provides structured summary for physician review and patient discussion

Data Requirements

Primary Data Sources: EHR longitudinal record (diagnoses, treatment history, comorbidities, performance status, lab results), genomic files (tumor sequencing variants, pathology), ClinicalTrials.gov API (trial criteria, phase, status, location)
Classification: PHI/PII (genomic data, diagnosis, treatment history) + research data (trial criteria are public)
Quality Metrics: Trial eligibility matching accuracy >95%, genomic variant interpretation concordance with established databases (ClinVar, OncoKB) >99%
Complexity: High (genomic interpretation, multi-criteria logic, natural language processing of trial inclusion/exclusion criteria)
Systems: Epic (EHR), Tempus (AI oncology platform), Foundation Medicine (genomic analysis)

Score Breakdown

Criterion	Weight	Score (1-5)	Weighted
Time Recaptured	15%	5	0.75
Error Reduction	10%	3	0.30
Cost Avoidance	10%	3	0.30
Strategic Leverage	5%	3	0.15
Data Availability	15%	4	0.60
Process Clarity	15%	4	0.60
Ease of Implementation	10%	3	0.30
Fallback Available	10%	3	0.30
Audience (Int/Ext)	10%	4	0.40
Composite	100%		3.65

Why It Scores Well

Clinical Impact: Increases trial enrollment by 30 to 50% in early pilots; expands treatment options for patients
Research Acceleration: Faster enrollment timelines shorten drug development cycles
Evidence Base: Genomic matching algorithms validated in peer-reviewed literature
Regulatory Fit: ClinicalTrials.gov is public; FDA oversight of trial eligibility determination is light
Scalability: One-time development; agents process unlimited patients

Regulatory Alignment

FDA: Clinical trial eligibility determination is not typically regulated as a medical device (information-only to physician)
IRB Requirements: Must be reviewed by institutional IRB if integrated into trial recruitment workflow; patient consent for genomic data use
HIPAA: Patient genomic data is PHI; must be de-identified when transmitted to external trial databases
GINA (Genetic Information Nondiscrimination Act): Protects against genetic data discrimination; agent must not expose genetic data to insurers
State Genomic Privacy Laws: CA, VA have specific genetic data privacy statutes

Sprint Factory Fit

3 to 4 sprints (EHR API integration, ClinicalTrials.gov API integration, genomic data pipeline, eligibility algorithm, clinical governance review)

Sprints Required: 3 to 4 sprints (EHR API integration, ClinicalTrials.gov API integration, genomic data pipeline, eligibility algorithm, clinical governance review)
Timeline: 8 to 10 weeks from requirements to pilot
Governance: Require oncology leadership and IRB sign-off; audit trial matching accuracy in first 50 cases
Success Metrics: Trial identification rate >80% of eligible patients; physician acceptance >70%; enrollment increase >20%

Comparable Implementations

Deploy This Use Case with the Sprint Factory

From zero to a governed, production agent in 6 weeks.

Sprint Factory Schedule a Briefing

Related Use Cases

On-Demand Clinical

AI Clinical Scribe & Note Generator

Score: 4.3/5.0

On-Demand Revenue Cycle Management

Denial Management & Appeal Generation

Score: 4.2/5.0

On-Demand PBM

Pharmacy Help Desk (Tier 1)

Score: 3.9/5.0

Governance Risks to Consider

Before deploying this use case, review these agentic AI risks from the Corvair Risk Catalogue. Each is scored on the DAMAGE framework and mapped to regulatory expectations.