|The following article features coverage from the ASCO Genitourinary Cancers Symposium 2022. Click here to read more of Cancer Therapy Advisor’s conference coverage.|
A novel digital pathology-based biomarker developed using artificial intelligence (AI) can successfully guide the use of androgen deprivation therapy (ADT) in men with intermediate-risk localized prostate cancer. These findings were reported at the ASCO Genitourinary Cancers Symposium 2022.
The current standard of care for men with intermediate- and high-risk localized prostate cancer treated with radiation is the addition of ADT, but no validated predictive markers for ADT use in this patient population currently exist.
Use of an AI-derived biomarker could determine which patients could be spared the cost and side effects of ADT.
Daniel E. Spratt, MD, of the Case Comprehensive Cancer Center at Case Western Reserve University in Cleveland, Ohio, and colleagues developed a multimodal deep-learning architecture to learn from both standard clinicopathologic variables and digital imaging histopathology data and identify differential outcomes by treatment type.
“The hypothesis is that there is a wealth of unused biologic information that currently we don’t recognize in our histopathology imagery,” Dr Spratt said.
He and his collaborators validated this approach in 1719 men who participated in a phase 3 randomized trial (NRG/RTOG 9408) comparing RT with or without 4 months of ADT for localized prostate cancer. Of these, 51% had National Comprehensive Cancer Network intermediate-risk disease. The cohort had a median follow-up of 17.6 years.
Among patients with AI-biomarker–positive disease, ADT plus RT was significantly associated with a 67% lower risk for distant metastasis compared with RT alone. In the AI-biomarker–negative group, adding ADT to RT did not improve outcomes compared with RT alone, Dr Spratt reported.
The 15-year distant metastasis and prostate cancer-specific mortality rate difference between RT alone and RT plus ADT was significantly greater in the AI-biomarker–positive group compared with the AI-biomarker–negative group (9.8% vs 0.4% and 8.2% vs 1.5%, respectively).
Approximately two-thirds of men with intermediate-risk prostate cancer could safely avoid ADT, regardless of prognosis, because of lack of relative and absolute benefit, Dr Spratt concluded.
Disclosures: Some study authors declared affiliations with biotech, pharmaceutical, and/or device companies. Please see the original reference for a full list of disclosures.
Read more of Cancer Therapy Advisor’s coverage of ASCO GU 2022 by visiting the conference page.
Spratt DE, Sun Y, van der Wal D, et al. An AI-derived digital pathology-based biomarker to predict the benefit of androgen deprivation therapy in localized prostate cancer with validation in NRG/RTOG 9408. Presented at ASCO GU 2022; February 17-19, 2022. Abstract 223.