Large language model support for gynecologic oncology treatment decisions
Medical Students and Their Perception of Large Language Models (LLMs) in Gynecologic Oncology
This project will test whether giving medical students access to a local large language model helps them make treatment recommendations for gynecologic oncology cases compared with using guideline PDFs.
Quick facts
| Phase | Not applicable |
|---|---|
| Study type | Interventional |
| Enrollment | 68 (estimated) |
| Ages | 18 Years and up |
| Sex | All |
| Sponsor | Philipps University Marburg Academic / other |
| Locations | 1 site (Marburg) |
| Trial ID | NCT06865534 on ClinicalTrials.gov |
What this trial studies
In this interventional comparison, medical students who have begun clinical training will use either a locally deployed large language model or standard guideline PDF documents to formulate treatment recommendations for oncology case vignettes. Their recommendations will be compared to decisions made by a conventional multidisciplinary tumor board to measure concordance. The study uses on‑premise models (e.g., Llama) and retrieval-augmented techniques to keep data secure while providing guideline-relevant suggestions. Outcomes include agreement with tumor board recommendations and analysis of when the model helps or introduces errors.
Who should consider this trial
Good fit: Ideal participants are medical students who have started clinical coursework and can attend the study activities at the Marburg site.
Not a fit: Patients with very rare, highly individualized, or non-guideline-driven conditions may not benefit because the tool focuses on synthesizing guideline-based recommendations.
Why it matters
Potential benefit: If successful, this approach could help trainees produce treatment plans that better match expert tumor board recommendations, potentially improving care consistency and reducing decision errors.
How similar studies have performed: Earlier small studies show large language models can help answer medical questions and draft documentation, but their use specifically for oncology treatment decision support is only preliminarily explored with mixed results.
Eligibility criteria
Show full inclusion / exclusion criteria
Inclusion Criteria: \- Medical students having started with clinical subjects Exclusion Criteria: \- Not being a medical student
Where this trial is running
Marburg
- Institute for Digital Medicine, University Hospital of Giessen and Marburg, Philipps University Marburg — Marburg, Germany (Recruiting)
Study contacts
- Principal investigator: Sebastian Griewing, MD PhD — Philipps University Marburg
- Study coordinator: Sebastian Griewing, MD PhD
- Email: s.griewing@uni-marburg.de
- Phone: 0049 06421 586 2589
How to participate
- Review the eligibility criteria above with your treating physician.
- Visit the official trial page on ClinicalTrials.gov for the most current contact information and recruitment status.
- Contact the listed study coordinator or principal investigator to request pre-screening. Pre-screening is free and never obligates you to enroll.