AI Test Changes Prostate Cancer Decisions

A prospective study examined whether biomarker results could refine treatment selection

October 12, 2026
4 min

The Pathologist

Overview

Interim findings from the prospective, multicenter ASTuTE implementation study indicate that a multimodal AI prostate test changed short-term androgen deprivation therapy (ST-ADT) decisions for men receiving radiotherapy for intermediate-risk prostate cancer. Among 786 patients, planned ST-ADT use fell by a net 20.7% after testing; the study assessed decision changes, not long-term cancer control or survival.

Background

ST-ADT can improve the effectiveness of radiotherapy, but its benefits vary among patients and treatment can cause sexual dysfunction, hot flashes, fatigue, osteoporosis, metabolic changes, cardiovascular complications, and mood disorders. Treatment decisions have traditionally used clinical and pathologic factors, including PSA, tumor stage, Gleason score, and the proportion of biopsy cores containing cancer. The test combines these factors with information from a digitized biopsy slide and reports both 10-year distant-metastasis risk and estimated benefit from adding four to six months of ST-ADT. Current guidance supports four to six months of ADT with radiotherapy for unfavorable intermediate-risk disease, while ADT is generally not routinely used with radiotherapy for favorable intermediate-risk disease.

Data Highlights

Measure	Interim ASTuTE finding
Study population	786 patients
Initially recommended ST-ADT	230 patients
Initially recommended ST-ADT who decided against it after testing	176 patients (approximately three-quarters)
Patients not initially planned for ST-ADT for whom testing supported adding it	21 patients
Net change in planned ST-ADT use	Decrease of 20.7%
Primary endpoint	Proportion of shared treatment recommendations that changed after testing
Planned longer-term follow-up	Registry linkage for quality of life and disease control after a median follow-up of 5 years

Key Findings

ASTuTE was a prospective, multicenter implementation study involving 786 patients receiving radiotherapy for intermediate-risk prostate cancer.
Clinicians and patients recorded their planned use of ST-ADT before receiving the test results and reconsidered the decision afterward.
Of 230 patients initially recommended ST-ADT, 176 decided against treatment after testing.
Testing supported adding ST-ADT for 21 patients who had not initially planned to receive it; overall planned use declined by a net 20.7%.
The test combines clinical factors with analysis of a digitized hematoxylin-and-eosin-stained biopsy slide reviewed by a certified pathologist for suitability. It provides a 10-year distant-metastasis risk estimate and a prediction of benefit from four to six months of ST-ADT.
Investigators reported that clinicians became less likely to recommend ST-ADT before receiving test results as the study progressed, but the reason was not established. Long-term outcomes remain to be evaluated, so the interim findings do not establish effects on cancer control, recurrence, or mortality.

Clinical Implications

The interim results show that the test was associated with changes in shared treatment recommendations in both directions, but they do not demonstrate improved patient outcomes or establish the safety of withholding or adding ST-ADT based on the test. The source notes that further assessment is needed of analytical reproducibility, slide-quality requirements, performance across patient groups, and integration into routine digital pathology workflows.

Conclusion

In ASTuTE, a multimodal AI prostate test influenced ST-ADT decisions for men receiving radiotherapy for intermediate-risk disease, with a net reduction in planned treatment use. Whether these decision changes affect quality of life or disease control awaits longer-term follow-up.