AI-Based Pathology Classifier Predicts Sensitivity to Enzalutamide in Metastatic Hormone-Sensitive Prostate Cancer: A Biomarker Analysis of the ENZAMET Trial.
Sebastian Medina, Naoto Tokuyama, Vaishnavi Putcha et al.
✦ AI-curated · Sources linked
In 30 seconds
This prespecified biomarker analysis of the randomised ENZAMET trial assessed whether a locked artificial intelligence pathology image classifier (APIC) identified differential overall-survival benefit from enzalutamide in metastatic hormone-sensitive prostate cancer. Among 492 participants with digitized tissue, APIC status significantly modified treatment benefit (interaction P = 0.014). In APIC-negative disease, enzalutamide versus a nonsteroidal antiandrogen was associated with improved survival (HR, 0.50; 95% CI, 0.35-0.73; 60-month OS, 76.5% vs. 58.3%); APIC-positive disease showed limited evidence of benefit (HR, 1.04; 95% CI, 0.67-1.62).
Key findings
- APIC classified 316 (64%) participants as negative and 176 (36%) as positive.
- The treatment-by-APIC interaction for overall survival was significant (interaction P = 0.014).
- In APIC-negative disease, 60-month OS was 76.5% with enzalutamide and 58.3% with NSAA (HR, 0.50; 95% CI, 0.35-0.73).
- In APIC-positive disease, 60-month OS was 55.6% with enzalutamide and 58.8% with NSAA (HR, 1.04; 95% CI, 0.67-1.62).
Why it matters
Treatment response to enzalutamide varied by APIC classification in this trial, suggesting that pathology-image features may help distinguish groups with different observed survival benefit. This is relevant to efforts to refine treatment selection in metastatic hormone-sensitive prostate cancer, while the findings remain specific to the biomarker analysis and require appropriate validation before guiding decisions.
What to keep in mind
Digitized tissue was available for 492 of 1,125 trial participants, so the biomarker analysis included a subset of the randomised population.