Objective:

Assess whether a multi-modal AI prostate test changes decisions about short-term androgen deprivation therapy (ST-ADT) for patients receiving radiotherapy for intermediate-risk prostate cancer.

Approach:

• Study design: Prospective, multicenter implementation study of 786 patients in the Australian ASTuTE trial.
• Decision process: Clinicians and patients recorded plans for ST-ADT before testing and reconsidered them after receiving the test results.
• Test: The test combines clinical factors with information from a digitized biopsy slide. It estimates 10-year distant-metastasis risk and predicts likely benefit from adding four to six months of ST-ADT to radiotherapy.

Key Findings:

• Of 230 patients initially recommended ST-ADT, 176 decided against it after testing.
• Testing supported adding ST-ADT for 21 patients who had not initially planned to receive it.
• Across the cohort, planned ST-ADT use fell by a net 20.7%.
• Clinicians became less likely to recommend ST-ADT before receiving test results as the study progressed; the reason was not established.

Interpretation:

Interim findings show that the test influenced treatment decisions in both directions. The study demonstrated an effect on decision-making, not improved patient outcomes.

Limitations:

• Longer-term outcomes, including quality of life and disease control, remain to be assessed through registry linkage after a median follow-up of five years.
• The findings do not establish whether avoiding ST-ADT maintains cancer control or whether adding it reduces recurrence or mortality.
• Further assessment is needed of analytical reproducibility, slide-quality requirements, performance across patient groups, and integration into routine digital pathology workflows.

Conclusion:

The ASTuTE interim findings indicate that a pathology-derived AI test can change ST-ADT recommendations in real-world care; effects on patient outcomes have not yet been established.

Sources:

• Article context provided