LLM-Assisted Patient-to-Trial Matching Navigator
Cuts manual screening effort by prioritizing likely-eligible trials with criterion-level explanations Evidence basis: TrialGPT reported criterion-level matching near expert review with strong recall; pilot results showed faster screening with similar decision quality; broader fairness validation is still needed
The Problem
“LLM-Assisted Patient-to-Trial Matching Navigator”
Organizations face these key challenges:
Cuts manual screening effort by prioritizing likely-eligible trials with criterion-level explanations
Impact When Solved
The Shift
Human Does
- •Review patient records against trial eligibility criteria manually
- •Compare candidate trials and prioritize likely matches in spreadsheets
- •Discuss unclear eligibility cases and make final screening decisions
- •Document screening outcomes and perform retrospective quality checks
Automation
- •No AI-supported matching or prioritization
- •No criterion-level explanation generation
- •No automated triage of likely-eligible trials
Human Does
- •Review AI-prioritized trial matches and confirm final eligibility decisions
- •Assess criterion-level explanations for unclear or borderline cases
- •Handle exceptions, missing information, and escalation decisions
AI Handles
- •Analyze patient information against trial criteria
- •Prioritize likely-eligible trials for human review
- •Generate criterion-level match explanations for each recommendation
- •Flag uncertain or low-confidence matches for closer review
Operating Intelligence
How LLM-Assisted Patient-to-Trial Matching Navigator runs once it is live
AI runs the first three steps autonomously.
Humans own every decision.
The system gets smarter each cycle.
Who is in control at each step
Each column marks the operating owner for that step. AI-led actions sit above the divider, human decisions and feedback loops sit below it.
Step 1
Assemble Context
Step 2
Analyze
Step 3
Recommend
Step 4
Human Decision
Step 5
Execute
Step 6
Feedback
AI lead
Autonomous execution
Human lead
Approval, override, feedback
AI handles assembly, analysis, and execution. The human gate sits at the decision point. Every cycle refines future recommendations.
The Loop
6 steps
Assemble Context
Combine the relevant records, signals, and constraints.
Analyze
Evaluate options, risk, and likely outcomes.
Recommend
Present a ranked recommendation with supporting rationale.
Human Decision
A human accepts, edits, or rejects the recommendation.
Authority gates · 1
The system must not make the final patient eligibility decision without review and approval by a clinical trial coordinator or study team reviewer [S1].
Why this step is human
The decision carries real-world consequences that require professional judgment and accountability.
Execute
Carry out the approved action in the operating workflow.
Feedback
Outcome data improves future recommendations.
1 operating angles mapped
Operational Depth
Technologies
Technologies commonly used in LLM-Assisted Patient-to-Trial Matching Navigator implementations: