Benchmarking AI language models against tumor boards for oncology treatment decisions
Official title: Benchmarking AI for Clinical Oncology decisioNmaking (BEACON): A Prospective, Multicentre, Blinded Evaluation of Frontier Large Language Models Against Multidisciplinary Tumour Board Recommendations in Oncology Treatment Planning
Run by Assistance Publique - Hôpitaux de Paris · NCT07739121 · Data updated 2026-11-12
This project tests whether frontier AI language models can match the treatment recommendations of multidisciplinary tumor boards for common cancers using standardized synthetic cases.
Quick facts
| Study type | Observational |
|---|---|
| Ages | 18 Years and up |
| Sex | All |
| Enrollment | 100 (estimated) |
| Sponsor | Assistance Publique - Hôpitaux de Paris Academic / other |
| Locations | 5 sites (Boulogne-Billancourt and 4 other locations) |
| Status | Recruiting |
| Started | May 1, 2026 |
| Main results expected | December 31, 2026 |
| Study ends | March 31, 2027 |
| Registry record updated | October 7, 2026 |
| Trial ID | NCT07739121 on ClinicalTrials.gov |
Is this trial a fit for you?
✓Likely a good fit
Ideal candidates are synthetic oncology cases within breast, lung, urological, digestive, or gynaecological cancers with complete structured data and resolvable clinical questions.
✕Probably not a fit
Patients whose care falls outside the predefined localizations or whose data are incomplete or ambiguous may not benefit.
A plain-English reading of the study's rules. Only the study team can confirm whether you qualify.
What this trial studies
BEACON is a prospective, multicentre benchmark that compares six advanced language models with human tumor boards in oncology planning. Standardized synthetic cases across five cancer localizations are input identically to two independent boards and to the AI models. Recommendations are broken into five domains and scored for concordance to evaluate alignment between models and human decision-making, with an independent blinded adjudication panel. The study design controls for input variability and uses automated scoring to produce objective comparisons.
Why it matters
Potential benefit: If successful, AI models could support or augment oncology decision-making and reduce variability in treatment planning.
How similar studies have gone: Previous benchmarking work suggests potential alignment between AI recommendations and human experts, though this exact design is novel due to its blinded, multicentre, synthetic-case approach.
Who can join: the full rules
Show full inclusion / exclusion criteria
Inclusion Criteria: * Synthetic oncology case within one of the five predefined localisations (breast, lung, urological, digestive, gynaecological). * Complete structured schema: UICC 8th-edition stage, biomarkers, ECOG performance status, comorbidities and a standardised clinical question. * A clinically answerable treatment-planning question that is mappable to the locked guideline matrix. Exclusion Criteria: * Case outside the five predefined localisations. * Incomplete, internally inconsistent or ambiguous schema. * Duplicate or near-duplicate of an existing case in the set. * Question not resolvable by current guidelines.
Where it's running
-
Hopital Européen Georges Pompidou
Paris, France Recruiting -
Hôpital Ambroise Paré AP-HP
Boulogne-Billancourt, France Not yet recruiting -
Hopital Tenon
Paris, France Not yet recruiting -
Hôpital Cochin AP-HP
Paris, France Not yet recruiting -
Hôpital Saint Antoine AP-HP
Paris, France Not yet recruiting
Study contacts
- Study coordinator: Jean-Emmanuel Bibault, MD PhD
- Email: jean-emmanuel.bibault@aphp.fr
- Phone: 01 56 09 34 06
Similar studies
Other studies for the same condition or treatment. Recruiting studies are listed first.
-
Methylscape test in body fluids for multi-cancer detection in Colombia
A simple blood, urine, or saliva test called Methylscape is being tested to see if it can detect multiple cancers and identify where the signal is coming from, using people with cancer and healthy volunteers in Colombia.
-
Evaluating HTL0039732 for Advanced Solid Tumours
This study is testing a new drug called HTL0039732, alone and with another treatment, to see if it can help people with advanced solid tumors by boosting their immune response against cancer.
-
Preventing Dato‑DXd–related mouth sores with dexamethasone mouthwash
This study will see if using an alcohol‑free dexamethasone mouthwash can prevent mouth sores in adults with advanced breast cancer or non‑squamous non‑small cell lung cancer who are receiving datopotumab deruxtecan.
-
CD70-targeted immunoPET Imaging of Kidney Cancer
This study is testing a new imaging technique that uses a special marker to help doctors see kidney cancer more clearly.
-
Using vorolanib for treating advanced kidney cancer
This study is testing if the drug vorolanib can help people with advanced kidney cancer live longer when used as a second treatment option.
-
Understanding lymphedema after urologic surgery
This study looks at how often and how badly lymphedema affects people after they have surgery for urologic cancer that involves removing lymph nodes.
How to take part
- Review the eligibility criteria above with your treating physician.
- Check the official listing on ClinicalTrials.gov for the latest contact details and recruitment status.
- Contact the study coordinator or principal investigator to ask for pre-screening. Pre-screening is free and never obligates you to enroll.