An AI pathology classifier picked out who gained from enzalutamide in ENZAMET

Design

Prespecified biomarker analysis of a randomised phase 3 trial

Population

492 men with metastatic hormone-sensitive prostate cancer and digitised tissue from ENZAMET

Primary outcome

Overall survival by classifier status and treatment

Effect

Classifier-negative HR 0.50 (95% CI 0.35 to 0.73); positive HR 1.04 (0.67 to 1.62); interaction P = 0.014

ENZAMET showed that adding enzalutamide to androgen deprivation improves survival in metastatic hormone-sensitive prostate cancer. This biomarker study, published in Clinical Cancer Research on 15 October 2026, applied an artificial-intelligence classifier of routine H&E slides, developed and locked in an earlier trial, to 492 ENZAMET participants with digitised tissue (median follow-up 68.8 months).

The classifier labelled 64% as negative and 36% as positive. In classifier-negative disease, enzalutamide improved survival against a non-steroidal antiandrogen (HR 0.50, 95% CI 0.35 to 0.73; five-year survival 76.5% vs 58.3%). In classifier-positive disease there was little evidence of benefit (HR 1.04, 0.67 to 1.62). The interaction was significant (P = 0.014), and the pattern held in low-volume disease.

Only 44% of trial participants had usable tissue, and the result needs confirmation in other trials before it can guide treatment. It is not a reason to withhold enzalutamide now. But it shows that standard slides may carry predictive information that current clinical factors miss.

• Continue to offer treatment intensification in metastatic hormone-sensitive prostate cancer according to current guidelines
• Do not withhold enzalutamide on the basis of this classifier outside a trial
• Make sure diagnostic biopsy slides are retained and available for future biomarker testing
• Watch for validation of AI-based pathology classifiers in other trials

Why it matters

It points towards selecting intensification by tumour biology read from routine slides.

Don't overread it

A biomarker analysis in under half the trial population; it does not justify withholding enzalutamide.

The statistics, in plain English

A hazard ratio of 0.50 means the death rate was halved with enzalutamide in classifier-negative patients. An HR of 1.04 with an interval from 0.67 to 1.62 in positive patients means benefit there was not shown, though it was not ruled out.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.