NCT07227233

Artificial Intelligence and Machine Learning to Guide CDK4/6 Inhibitor Rechallenge in Breast Cancer.

Artificial Intelligence and Machine Learning-Enhanced Biomarker-dRiven CDK4/6 Inhibitor Rechallenge in HR+ HER2- Advanced Breast Tumors.

• Status: Recruiting
• Phase: Phase 2
• Study type: Interventional
• Enrollment: 105 (estimated)
• Sponsor: University of California, San Diego · Academic / Other
• Sex: All
• Age: 18 Years
• Healthy volunteers: Not accepted

Summary

The goal of this clinical trial is to learn if an artificial intelligence model will be able to select patients with advanced breast cancer who may respond to a second cyclin-dependent kinase 4/6 (CDK4/6) inhibitor after they have progressed on the first CDK4/6 inhibitor. Patients for this study need to have hormone receptor positive (HR+) human epidermal growth factor receptor 2 negative (HER2-) advanced breast cancer. The main questions the study aims to answer are: If the artificial intelligence model says that a patient's tumor should respond to the second CDK4/6 inhibitor, and that patient receives the second CDK4/6 inhibitor together with fulvestrant (an endocrine therapy also called Faslodex), will it take longer for the tumor to get worse than if the patient receives another type of therapy? Will the tumor respond better? Will the therapy be safe? Researchers will compare the combination of a second CDK4/6 inhibitor plus fulvestrant to the therapy chosen by the physician. Participants will: Take the assigned therapy based on the way the therapy is usually prescribed. Visit the clinic once every month for checkups, tests, and questionnaires. Keep a diary of the pills they take at home.

Detailed description

This is a phase II, open-label, randomized controlled single-site study to assess the clinical utility of an artificial intelligence model, called palbo-VNN, to guide treatment selection in patients with hormone receptor positive (HR+) human epidermal growth factor receptor 2 negative (HER2-) advanced (unresectable or metastatic) breast cancer who have progressed after prior treatment with a cyclin-dependent kinase 4/6 (CDK4/6) inhibitor. The investigators hypothesize that: (1) the model will be able to predict who will respond to a second CDK4/6 inhibitors combined with fulvestrant (Faslodex); and (2) that the participants treated with a second CDK4/6 inhibitors plus fulvestrant will have better outcomes (longer progression free survival, higher overall response rate, higher clinical benefit rate, and better quality of life) than the participants treated based on the physician's choice. Next-generation sequencing of the tumor will be performed using biopsies obtained after the first CDK4/6 inhibitor treatment and within six months of study enrollment. The next-generation sequencing data will be analyzed with the palbo-VNN model. Participants with tumors predicted to respond to a second CDK4/6 inhibitor (CDK4/6i sensitive) will be randomized. Half of the randomized participants will receive a second CDK4/6 inhibitor plus fulvestrant; the other half will receive the treatment of physician's choice, excluding CDK4/6 inhibitors. After 6 months of therapy, the investigators will evaluate the progression free survival rate of the randomized participants, as well as other measures of efficacy, such as overall response rate, clinical benefit rate, and quality of life. They will also do studies to correlate treatment efficacy with molecular alterations of the tumor. These correlative studies may allow to improve the predictive ability of the palbo-VNN model in the future. The investigators will also evaluate treatment safety.

Conditions

• Breast Cancer

Interventions

Type	Name	Description
DRUG	A CDK4/6 inhibitor different from the first that the participant had received, plus fulvestrant.	A CDK4/6 inhibitor different from the first that the participant had received, plus fulvestrant.
DRUG	Treatment of physician's choice	Treatment of physician's choice, excluding CDK4/6 inhibitors.

Timeline

• Start date: 2026-03-31
• Primary completion: 2027-12-01
• Completion: 2028-12-01
• First posted: 2025-11-12
• Last updated: 2026-04-09

Locations

1 site across 1 country: United States

Regulatory

• FDA-regulated drug study