|
|
|
|
|
|
|
PEER-TO-PEER CLINICAL CONVERSATION |
|
|
|
|
|
|
Multimodal AI Test Predicts Prostate Cancer Metastasis and Treatment Response
|
|
Timothy Showalter, MD, MPH
|
| Leslie Ballas speaks with Timothy Showalter about a multimodal artificial intelligence-based prostate cancer test. Dr. Showalter explains how ArteraAI analyzes H&E-stained biopsy slides using computer vision combined with clinical factors to generate personalized risk stratification for 10-year metastasis likelihood and hormone therapy benefit prediction.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Highlights from the 2025 American Society of Clinical Oncology Annual Meeting |
|
|
|
|
| Updated Prostate Cancer Risk Groups by PSMA-PET PROMISE (PPP2): Results from an International Multi-Centre Registry Study |
| Wolfgang Fendler, MD |
| Wolfgang Fendler presented updated PSMA-PET PROMISE nomograms (PPP2), validated in over 6,000 prostate cancer patients across 20 global centers. These nomograms accurately stratify patients into low, intermediate, and high-risk groups for overall survival, outperforming NCCN and EAU risk models. Performance was consistent across imaging agents and PROMISE versions, reinforcing PPP2's clinical utility. |
|
|
|
|
|
| REASSURE – Long-Term Safety of Radium-223 in mCRPC: 7-Year Follow-up from the Largest Global Prospective Study |
| A. Oliver Sartor, MD |
| Oliver Sartor presented final 7-year data from the REASSURE study, the largest global prospective analysis of long-term safety for radium-223 in mCRPC. The study confirmed a low incidence of secondary malignancies (2%), manageable hematologic safety—even post-taxanes—and reduced fracture rates with bone-protective agents. Median overall survival was 15.6 months, aligning with real-world benchmarks. |
|
|
|
|
|
| Uptake of Targeted Therapy in a Large Cohort of Patients with Advanced Prostate Cancer and Germline Pathogenic Variants |
| Hiba M. Khan, MD, MPH |
| Hiba Khan presented data on targeted therapy uptake in men with advanced prostate cancer and germline HRR/MMR mutations, revealing that fewer than 25% of eligible patients received appropriate therapies. Black men were more likely to receive PARP inhibitors than White men but had lower odds of getting platinum chemo or immunotherapy, highlighting disparities in treatment patterns. |
|
|
|
|
|
|
|
|
|
|
| 'One Button Push’ Fully Automated PSMA PET Quantification: Correlation with Progression Free and Overall Survival in Patients Undergoing 177Lu PSMA Therapy for mCRPC
|
| Louise Emmett, MBChB, FRACP, FAANMS, MD
|
| Louise Emmett presented results showing that fully automated PSMA PET quantification ('one button push') accurately predicts PSA progression-free survival and overall survival in men receiving 177Lu-PSMA therapy for mCRPC. In a cohort of 139 patients, higher SUVmean and lower tumor volume were associated with significantly better outcomes, with results closely matching those from semi-automated analysis.
|
|
|
|
|
|
| How Low Do You Need to Go? Association Between Various Prostate-Specific Antigen Response Measures and Clinical Outcomes in Metastatic Castration‑sensitive Prostate Cancer in the Veteran Health Administration Data |
| Stephen Freedland, MD |
| Stephen Freedland presented a VHA study showing that reaching a PSA level <0.2 ng/mL within 9 months of starting ADT in mCSPC is strongly associated with significantly better survival and lower risk of disease progression. In contrast, while a ≥90% PSA decline was linked to improved overall survival, it did not correlate with progression risk. Patients receiving ADT + ARPIs were more likely to achieve the optimal PSA response, underscoring the importance of treatment intensification in this population. |
|
|
|
|
|
| 177Lu-PSMA-617 Consolidation Therapy Post Docetaxel in Patients with De-Novo High-Volume Metastatic Hormone-Sensitive Prostate Cancer: A Randomized, Phase 2 Trial |
| Ashwani Sood, MBBS, DRM, DNB |
| Ashwani Sood presented a phase II trial showing that two cycles of 177Lu-PSMA-617 consolidation therapy after docetaxel significantly improved PSA responses and progression-free survival in patients with de novo, high-volume mHSPC and residual disease. At 6 months, 60% of patients receiving 177Lu-PSMA-617 achieved a PSA ≤0.2 ng/mL versus 13% in the control arm, with improved radiographic (18 vs 9 months) and PSA progression-free survival (15 vs 9 months). |
|
|
|
|
|
| Real-World Patient Characteristics, Treatment Patterns, and Overall Survival in Metastatic Hormone-Sensitive Prostate Cancer: Insights by PTEN Status |
| Dana Rathkopf, MD |
| Dana Rathkopf presented a real-world analysis showing that patients with metastatic hormone-sensitive prostate cancer harboring PTEN alterations had significantly worse overall survival compared to those without such alterations. PTEN-altered tumors were also more likely to harbor TP53 and RB1 mutations and responded less favorably to first-line therapies like ADT, ARPI, or docetaxel. |
|
|
|
|
|
| External Validation of a Pathology-Based Multimodal Artificial Intelligence Biomarker for Predicting Prostate Cancer Outcomes after Prostatectomy |
| Chien-Kuang Cornelia Ding, MD, PhD |
| Chien-Kuang Cornelia Ding presented the first external validation of a multimodal artificial intelligence (MMAI) model that predicts prostate cancer outcomes after radical prostatectomy using pathology slides and clinical data. In a UCSF cohort of 640 patients, the MMAI model independently predicted adverse outcomes—including metastasis, bone metastasis, disease progression, and prostate cancer–specific mortality—more accurately than the widely used CAPRA-S score, particularly in patients eligible for salvage therapy. |
|
|
|
|
|
| A Large Language Model (LLM)-Based Multi-Agent Framework for Risk Stratification and Treatment Recommendations in Localized Prostate Cancer |
| Umair Ayub, PhD, MS |
| Umair Ayub presented a large language model (LLM)-based multi-agent framework for risk stratification and treatment recommendations in localized prostate cancer, using GPT-4 and a rule-based algorithm. In a Mayo Clinic cohort of 858 patients, the model outperformed clinicians in risk classification and delivered highly accurate NCCN-based treatment plans, especially when using retrieval-augmented generation (RAG) to reduce hallucinations. |
|
|
|
|
|
|
|
|
|
|
