A Preregistered Multi-Cohort Evaluation of the FATHOM AI System for Molecular Testing Prioritization to Support Clinical Trial Enrollment

NCT07814872 | Not Yet Recruiting

• 1 other identifier: FATHOM
• Study Type: observational
• Target: 30,000
• Locations: 1 country
• Sites: 1

Brief Summary

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.

Conditions

Neoplasms (Cancer / Tumors); Colorectal Neoplasms; Lung Neoplasms; Endometrial Neoplasms; Breast Neoplasms; Pancreatic Neoplasms; Head and Neck Neoplasms; Kidney Neoplasms; Glioma; Ovarian Neoplasms; Stomach Neoplasms

Keywords

Deep Learning; Agentic AI; Digital Pathology; Artificial Intelligence; Multi-omics; Biomarkers; Cancer

Outcome Measures

Primary Outcomes (1)

• Proportion of prespecified evaluation scenarios in which an AI-generated deployment policy demonstrates effective screening enrichment or rule-out performance

  An evaluation scenario consists of one molecular marker evaluated in one study cohort. For each prespecified scenario, the study assesses whether the AI system generates a policy that either prioritizes patients more likely to carry the marker for confirmatory testing or identifies patients who may safely be spared testing, compared with testing everyone. The outcome is the proportion of scenarios in which the policy meets these performance criteria.

  Periprocedural (at the time of pathology slide evaluation)

Secondary Outcomes (4)

• Per-scenario performance of each AI-generated deployment policy

  Periprocedural (at the time of pathology slide evaluation)

• Temporal generalizability for trials first posted on or after January 1, 2026

  Periprocedural (at the time of pathology slide evaluation)

• Temporal performance for trials first posted on or before December 31, 2025

  Periprocedural (at the time of pathology slide evaluation)

• Proportion of evaluation scenarios in which a non-default AI-generated deployment policy demonstrates effective screening enrichment or rule-out performance

  Periprocedural (at the time of pathology slide evaluation)

Study Arms (1)

Archived evaluation cohorts

Patient records and data from archived multi-institutional cohorts with routine H\&E whole-slide images and molecular profiles. No intervention is assigned, and no patient is contacted.

Eligibility Criteria

• Sex: all
• Healthy Volunteers: No
• Age Groups: Child (0-17), Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

Patient records from archived, multi-institutional cohorts of patients with histologically confirmed cancers, relevant molecular profiling results, and at least one diagnostic hematoxylin and eosin (H\&E) whole-slide image. For the evaluation of a given policy, patients whose slides were used to train that policy's classifier are excluded. No patients are enrolled or contacted; enrollment counts refer to patient records analyzed.

You may qualify if:

• Patients with a histologically confirmed cancer
• Availability of relevant molecular profiling results
• At least one diagnostic hematoxylin and eosin (H\&E) whole-slide image

You may not qualify if:

• 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

Sponsors & Collaborators

• Harvard Medical School (HMS and HSDM): lead

Study Sites (1)

Harvard Medical School

Boston, Massachusetts, 02115, United States

Location

MeSH Terms

Conditions

Neoplasms; Colorectal Neoplasms; Lung Neoplasms; Endometrial Neoplasms; Breast Neoplasms; Pancreatic Neoplasms; Head and Neck Neoplasms; Kidney Neoplasms; Glioma; Ovarian Neoplasms; Stomach Neoplasms

Condition Hierarchy (Ancestors)

Intestinal Neoplasms; Gastrointestinal Neoplasms; Digestive System Neoplasms; Neoplasms by Site; Digestive System Diseases; Gastrointestinal Diseases; Colonic Diseases; Intestinal Diseases; Rectal Diseases; Respiratory Tract Neoplasms; Thoracic Neoplasms; Lung Diseases; Respiratory Tract Diseases; Uterine Neoplasms; Genital Neoplasms, Female; Urogenital Neoplasms; Uterine Diseases; Genital Diseases, Female; Female Urogenital Diseases; Female Urogenital Diseases and Pregnancy Complications; Urogenital Diseases; Genital Diseases; Breast Diseases; Skin Diseases; Skin and Connective Tissue Diseases; Endocrine Gland Neoplasms; Pancreatic Diseases; Endocrine System Diseases; Urologic Neoplasms; Kidney Diseases; Urologic Diseases; Male Urogenital Diseases; Neoplasms, Neuroepithelial; Neuroectodermal Tumors; Neoplasms, Germ Cell and Embryonal; Neoplasms by Histologic Type; Neoplasms, Glandular and Epithelial; Neoplasms, Nerve Tissue; Ovarian Diseases; Adnexal Diseases; Gonadal Disorders; Stomach Diseases

Central Study Contacts

Chi-Kang Pai

CONTACT

6172334924; chikang_pai@fas.harvard.edu

Study Design

• Study Type: observational
• Observational Model: COHORT
• Time Perspective: RETROSPECTIVE
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Associate Professor

Study Record Dates

• First Submitted: September 6, 2026
• First Posted: September 11, 2026
• Study Start: September 1, 2026
• Primary Completion (Estimated): November 1, 2026
• Study Completion (Estimated): December 1, 2026
• Last Updated: September 11, 2026

Record last verified: 2026-09

Data Sharing

• IPD Sharing: Will not share

Locations

US (1)