OncoEducate: A pilot study of generative AI to support patient education in GU cancer care.

This prospective two-phase pilot study developed and evaluated OncoEducate, a clinician-supervised generative AI application designed to deliver standardized, plain-language treatment education to patients with advanced genitourinary cancers at the point of care. In Phase I, six oncology clinicians rated AI-generated handouts for nine commonly used palliative-intent regimens and found them to be accurate and clinically appropriate across multiple domains. In Phase II, 20 patients with advanced kidney, prostate, or urothelial cancers received clinician-reviewed handouts and completed follow-up surveys, demonstrating feasibility and strong patient engagement with the materials.

These findings represent an early but meaningful proof of concept for using generative AI to enhance patient-clinician communication in oncology. Ensuring that patients understand their diagnosis, treatment regimen, and treatment intent is foundational to informed consent and shared decision-making, yet delivering consistent, personalized education at scale remains a persistent challenge in busy clinical settings. OncoEducate illustrates how clinician-supervised AI tools can help bridge that gap without replacing the human judgment that remains essential to high-quality cancer care.