|
|
|
|
|
|
|
PEER-TO-PEER CLINICAL CONVERSATIONS |
|
|
|
|
|
|
AI Applications in Prostate Cancer Pathology and Prognostics
|
Tamara Lotan, MD
Tamara Lotan discusses artificial intelligence applications in prostate cancer pathology. Dr. Lotan describes three main categories of AI tools: diagnostic algorithms for tumor annotation, prognostic models for outcome prediction, and predictive tools for therapy response.
|
|
|
|
|
|
|
|
|
|
|
|
Genomic Testing in Prostate Cancer: Management, Applications, and Treatment Selection
|
Joaquin Mateo, MD, PhD
Neeraj Agarwal speaks with Joaquin Mateo about genomics in prostate cancer treatment. Dr. Mateo reflects on progress over the past decade, highlighting that beyond bringing drugs to clinic, genomics enables individualized treatment decisions and helps patients on an individual basis.
|
|
|
|
|
|
|
|
|
|
|
|
Integrating Genomic Prognostic and Hallmark Signatures to Predict Outcomes in Active Surveillance
|
Kevin Shee, MD, PhD
Kevin Shee analyzes Decipher® GRID whole-transcriptome data in 500 active surveillance patients from UCSF. Cox proportional regression identified Long and Yu signatures predicting upgrade risk beyond CAPRA score adjustment.
|
|
|
|
|
|
|
|
|
|
|
| Prostate Cancer Diagnostics in the AI Age: Fast, Crowded, and (Hopefully) Open
|
| Matthew Cooperberg, MD, MPH
|
| AI is moving into prostate cancer diagnostics very quickly, especially in pathology and MRI interpretation, but most tools are still designed to assist clinicians rather than replace them. The strongest open-source examples—PANDA for biopsy grading and PI-CAI for MRI—show that AI can improve accuracy and consistency, yet the article argues that broader validation, transparency, and head-to-head benchmarking are still needed before these systems become routine standard-of-care tools.
|
|
|
|
|
|
| Artificial Intelligence in Diagnostic, Prognostic, and Predictive Genomic Biomarkers for Prostate Cancer: Ready for Prime Time? - Beyond the Abstract
|
| Andrey Bazarkin, Mark Taratkin, Stanislav Vovdenko et al.
|
| Bazarkin et al. concluded that AI can help discover diagnostic, prognostic, and treatment-response genomic biomarkers in prostate cancer, with promising performance for detecting tumor-related genes and predicting recurrence, metastasis, survival, and therapy toxicity. The big limitation is that most models are still research-stage rather than clinic-ready, because they lack prospective validation, standardized implementation, and clear proof of real-world benefit.
|
|
|
|
|
|
|
|
|
|
|
| Association Between Genomic Classifier Scores and Initial Management of Localized Prostate Cancer in a Population-Based Cohort in the United States
|
| Michael Leapman, MD, MHS
|
| In this population-based U.S. cohort, higher genomic classifier scores were associated with lower odds of choosing active surveillance and a greater likelihood of immediate treatment in men with low- or favorable intermediate-risk localized prostate cancer. The effect was seen in both low- and intermediate-risk groups, and it was strongest in the intermediate-risk patients, suggesting the genomic result is influencing real-world initial management decisions.
|
|
|
|
|
|
| Stratifying Risk in Active Surveillance for Prostate Cancer Using Decipher Genomic Classifiers and Gleason Grade Group
|
| Anthony Zhang, MD
|
| Anthony Zhang discusses a study which found that adding Decipher genomic risk to Gleason Grade Group improved risk stratification for men on active surveillance, especially by separating GG1 patients with low versus high genomic risk. In practical terms, GG1 with high Decipher risk and GG2 with low Decipher risk behaved as higher-risk groups and may warrant closer surveillance or earlier treatment, while GG1 low and intermediate genomic risk remained relatively favorable.
|
|
|
|
|
|
| Predictive Value of Decipher and MRI for Upgrading of Prostate Cancer Patients on Active Surveillance
|
| Marina Schnauss, PhD
|
| Marina Schnauss presents a study which found that Decipher added predictive value beyond MRI for men on active surveillance, because higher Decipher scores were associated with both future biopsy upgrading and adverse pathology at prostatectomy. In contrast, PI-RADS 4-5 was not a significant predictor once Grade Group, PSA density, and Decipher were included, suggesting genomic risk may be more informative than MRI alone for some patients.
|
|
|
|
|
|
|
|
|
|
|
