PSMA-targeted actinium-225 therapy in metastatic castration-resistant prostate cancer (mCRPC): Baseline and follow-up PSMA PET parameters associated with outcomes.

5070 Background: Prostate-specific membrane antigen (PSMA) is a validated therapeutic target in mCRPC. 225Ac-J591, an alpha-emitting radionuclide linked to an anti-PSMA antibody, has been studied in several early phase clinical trials. We evaluated baseline and follow-up (12 weeks post 225Ac-J591) PSMA PET metrics (SUVmax, SUVmean, and total tumor volume [TTV]) and their association with biochemical response (PSA50), overall survival (OS), and adverse events. Methods: Patients enrolled in phase I dose-escalation trials of 225Ac-J591, either as a single agent (NCT03276572, NCT04506567; n = 87) or in combination (with 177Lu-PSMA-I&T, NCT04886986, n = 18; or pembrolizumab and androgen receptor (AR) inhibitors, NCT04946370 n = 12), were included. 68Ga-PSMA-11 PET metrics were quantified using MIM Encore software. Outcomes assessed included PSA50, OS, and adverse events. Statistical analyses utilized univariate and multivariate logistic regression for PSA50, Cox regression for OS and Kruskal-Wallis and Wilcoxon rank sum tests for adverse events. Results: 117 patients were included with a median age of 71 yr and median baseline PSA of 58 ng/mL (range 0.67–9614). Prior therapies included chemotherapy (62%), >1 AR pathway inhibitor (53%), immunotherapy (42%), 177Lu-PSMA (28%), and radium-223 (17%). Metastatic sites included bone (88%), lymph nodes (63%), and visceral organs (29%). High CALGB (Halabi) risk in 56% of patients. Median baseline SUVmax, SUVmean, and TTV on PSMA-PET were 50 (31-85), 8.3 (5.7-11.8), and 308 (105-1066), respectively. Baseline SUVmean (OR 1.13, 95% CI 1.04–1.24, p = 0.006) and SUVmax (OR 1.01, 95% CI 1.00–1.02, p = 0.018) were associated with higher odds of PSA50, though only SUVmean remained significant on multivariable analysis (OR 1.11, p = 0.023). Baseline SUVmean and TTV were associated with OS (HR 0.95, p = 0.025; HR 1.000, p < 0.001), TTV remaining significant on multivariable analysis (HR 1.00, p = 0.007), along with prior chemotherapy (HR 1.45, p < 0.001) and CALGB high risk group (HR 1.84, p = 0.014). In multivariable analysis controlling for injected radioactivity, prior chemotherapy, and CALGB risk, TTV reduction was associated with PSA50 (OR 1.31, 95% CI 1.1–1.7, p = 0.015), but changes in SUVmean or SUVmax were not. Higher baseline TTV was more likely to have higher grade myelosuppression (anemia Gr 3/4 673 vs Gr 1/2 203, Gr0 284, p = 0.041), but less nausea and xerostomia. Conclusions: Higher baseline PSMA PET metrics, including SUVmean, SUVmax, and TTV are associated with PSA50 response and survival in patients receiving antibody-delivered alpha emitter 225Ac. Higher TTV, yielding more radionuclide delivery to tumor (most commonly bone) is associated with higher grade myelosuppression. However, the tumor antigen sink effect may decrease exposure to other PSMA+ organs leading to less xerostomia and nausea. Clinical trial information: NCT03276572 , NCT04506567 , NCT04886986 , NCT04946370 .