A team led by Dr. Ekta Khurana at Weill Cornell Medicine has been awarded a $1 million, two-year Challenge Award from the Prostate Cancer Foundation (PCF) to develop an artificial intelligence system for the early detection of treatment-resistant prostate cancer subtypes. The research will involve collaboration with scientists at Memorial Sloan Kettering Cancer Center.
The PCF Challenge Award supports research that may not otherwise receive funding, focusing on cross-disciplinary teams tackling complex problems in prostate cancer. Dr. Khurana, an associate professor of systems and computational biomedicine, will work alongside Dr. Iman Hajirasouliha, also of Weill Cornell Medicine, and physician-scientist Dr. Yu Chen and pathologist Dr. Anuradha Gopalan from Memorial Sloan Kettering Cancer Center. The team’s combined expertise spans pathology, genomics, computational science, and artificial intelligence.
“In this project, we will develop an AI system that can identify patients whose tumors may be developing these subtypes early on, so they can be enrolled in clinical trials for drugs that might help them,” Dr. Khurana said. She is also a member of the Englander Institute for Precision Medicine and the Sandra and Edward Meyer Cancer Center at Weill Cornell Medicine.
Approximately 300,000 new cases of prostate cancer are diagnosed annually in the United States, with a lifetime risk of about 12% for men. Although initial treatment often involves reducing testosterone levels or directly inhibiting androgen receptor signaling, many tumors eventually become resistant to these therapies. This resistance often stems from the development of subtypes that can grow independently of androgen signaling. Currently, clinicians lack reliable methods to identify these treatment-resistant subtypes early in the disease process.
Dr. Khurana and her colleagues previously identified four distinct subtypes of castration-resistant prostate cancer (CRPC) in a 2022 study published in Science. Three of these subtypes can proliferate without the need for androgen receptors, rendering standard hormone-based therapies ineffective. A Department of Defense grant, awarded in 2023, is also supporting Dr. Khurana’s research into the evolution of prostate cancer cell resistance to hormone-blocking therapy.
For the current PCF-funded project, researchers will train AI models using a large dataset of pathology slides from prostate tumors, alongside gene activity patterns and treatment outcomes. The AI will be designed to predict tumor subtype, or a mix of subtypes, and likely treatment responses based on analysis of the pathology slides.
If the AI model demonstrates sufficient accuracy – minimizing both false-negative and false-positive results – clinicians could use it to identify patients suitable for clinical trials of experimental treatments. Conversely, the AI could help avoid administering standard treatments to patients for whom they are unlikely to be effective. The researchers plan to validate the model through clinical trials following its development.
