Abstract

Preliminary results from the 2025 American Society of Clinical Oncology (ASCO) Annual Meeting, published in the Journal of Clinical Oncology, highlight the role of SelectPSMA, Nucs AI’s AI-powered PSMA-PET/CT analysis tool, in predicting outcomes for metastatic castration-resistant prostate cancer (mCRPC) patients treated with [177Lu]Lu-PSMA radiopharmaceuticals. This prospective, multicenter study addresses the unmet need for better patient selection, as only 46% of patients achieved a PSA response in the phase 3 VISION trial.

Methods

VISION-eligible patients with mCRPC who had progressed on taxane-based chemotherapy and ARSIs underwent baseline [68Ga]Ga-PSMA-11 PET/CT before treatment with [177Lu]Lu-PSMA-617 or [177Lu]Lu-PSMA-I&T. SelectPSMA was used to classify patients as responders (PSMA-R) or non-responders (PSMA-NR). This interim analysis reports exclusively on the University of Grenoble-Alpes cohort (IRB: CEMEN 202406). Primary outcomes included PSA50 response (≥50% PSA decline), PSA progression-free survival (PSA-PFS), and overall survival (OS). Associations were evaluated using Fisher’s exact test and Kaplan-Meier analysis.

Results

Of 72 patients screened, 60 (83%) were enrolled between August 2023 and September 2024. Most (95%) had prior taxane therapy, and all received ARSIs. Median follow-up was 9.5 months. At last follow-up, 63% of patients achieved PSA50, and the median PSA-PFS was 5.1 months (95% CI: 3.3–7.1). SelectPSMA classified 9/60 (15%) as non-responders, who had significantly lower PSA50 rates (22% vs. 71%; p=0.009) and shorter PSA-PFS (1.2 vs. 6.6 months; p<0.001) compared to responders. OS data were immature at this stage.

Conclusion

Early findings demonstrate that SelectPSMA can identify patients with lower likelihood of PSA response and shorter progression-free survival following [177Lu]Lu-PSMA therapy. These results provide preliminary evidence for the value of AI-based imaging biomarkers in optimizing patient selection for PSMA-targeted therapies. Full multicenter analysis with mature OS data is ongoing.

Reference: Djaileb L, Mercier A, Rovera G, et al. Artificial intelligence to predict outcome after [177Lu]Lu-PSMA for metastatic castration-resistant prostate cancer: Preliminary results from a multicentric prospective study. J Clin Oncol. 2025;43(16_suppl):17077. doi:10.1200/JCO.2025.43.16_suppl.e17077

Source (click to view link): ASCO GU Annual Meeting 2025 | DOI: 10.1200/JCO.2025.43.16_suppl.e17077

‍