5 Key Takeaways

• 1

  A multimodal AI-derived digital pathology biomarker was developed to predict distant metastasis in patients with biochemical recurrence after prostatectomy.

• 2

  The model integrates digitized pathology slides with clinical variables, enhancing risk stratification for patients post-radical prostatectomy.

• 3

  Validation of the model involved 533 patients with a median follow-up of nine years, showing an area under the curve of 0.74 for metastasis prediction.

• 4

  Patients were categorized into high- and low-risk groups, revealing a 10-year cumulative incidence of distant metastasis of 25% in high-risk patients.

• 5

  The study highlights the potential of archived specimens for prognostic information, with ongoing validation for prospective evaluation as a laboratory-developed test.