A Preregistered Multi-Cohort Evaluation of the FATHOM AI System for Molecular Testing Prioritization to Support Clinical Trial Enrollment
Who can join
As published by the study on ClinicalTrials.gov. "Not stated" means the study hasn't said — ask the study team.
- Age
- No age limits stated
- Sex
- All sexes
- Healthy volunteers
- Not accepted
- Compensation
- Not stated
Other requirements are listed under Eligibility below. Only the study team can confirm whether you can join.
What to expect
No one can promise a trial will help you personally — that is what the research is trying to find out.
What participation involves
Not stated in the study description — ask the study team about visits and procedures.
- Compensation
- Not stated
- Study-related exams and procedures are typically provided at no cost. This is how many trials work in general — confirm the details for this study with the study team.
- Participants are monitored closely by the research team throughout the study.
- Some participants may receive access to an investigational treatment, which is still being studied.
- Your participation helps researchers learn whether this approach could help future patients.
About this study
In plain language
This study tests whether an AI system called FATHOM can use cancer tissue images to help choose who should get tests for markers needed to join cancer trials. It uses saved images and test results from people with cancer. It does not enroll or contact patients, change their care, or affect whether they join a trial.
Many clinical trials evaluating cancer treatments require patients to undergo testing for specific molecular markers as part of eligibility screening, typically using immunohistochemistry or sequencing. Because relatively few patients may carry a required marker, trial investigators often test large numbers of patients to identify the few who may ultimately qualify for enrollment. Pathology laboratories routinely produce hematoxylin-and-eosin (H\&E) slides during cancer diagnosis. Pathology foundation models-large neural networks pretrained on millions of histology images-have shown promise in predicting molecular characteristics from these slides. Researchers can use these models to build classifiers that predict specific molecular markers and prioritize patients for confirmatory testing. This study evaluates FATHOM (Facilitating Accrual through Tumor Histology and Omics Matching), an autonomous research system powered by large multimodal models. Its agents read registered clinical trial records, identify molecular markers used as enrollment criteria, build prediction models using pathology foundation models, select the individual models or model combinations that best meet prespecified criteria, set their decision thresholds, and determine whether to deploy them. Together, a prediction model, its decision threshold, and the decision to deploy it constitute an AI prediction policy. Before FATHOM runs, the investigators preregister the clinical trial records that its agents may read, the cutoff date that defines which trial information they may use, the rules governing the agents, and the analysis plan. The system timestamps and locks each policy immediately after an agent produces it. The investigators then apply the policies to archived patient slides and compare their predictions with existing molecular marker results. The primary outcome is the proportion of prespecified evaluation scenarios in which an agent-generated policy, compared with universal molecular testing, either enriches the population selected for confirmatory testing with marker-positive patients or safely spares patients from confirmatory testing while meeting prespecified performance criteria. This study analyzes existing pathology images and clinical trial records only. It does not enroll or contact patients, influence patient care, or affect participation in any clinical trial.
Compensation is read from the study's own description and may be missing — always confirm with the study team.
Eligibility
Who can join
- Cancer confirmed by looking at tissue under a microscope.
- Available results from tests for relevant molecular markers, such as genes or proteins.
- At least one complete digital image of a tissue slide stained with H&E, dyes used to help diagnose cancer.
Who can't join
- Slides that are poor quality or cannot be read, judged without using the AI results.
- For each AI model being tested, patients whose slides were used to train that model.
Simplified for readability. The official criteria below are what the study team uses.
Official eligibility criteria
Inclusion Criteria: * Patients with a histologically confirmed cancer * Availability of relevant molecular profiling results * At least one diagnostic hematoxylin and eosin (H\&E) whole-slide image Exclusion Criteria: * Poor-quality or unreadable slides, assessed independently of model output * Patients whose slides were used to train a policy's classifier, for that policy's evaluation
Study sites (1)
MA
- Harvard Medical School · Boston, MA 02115Chi-Kang Pai · Study contact · 6172334924 · chikang_pai@fas.harvard.eduKun-Hsing Yu · Principal investigatorMore trials in Boston