Imaging expression of prostate-specific membrane antigen and response to PSMA-targeted β-emitting radionuclide therapies in metastatic castration-resistant prostate cancer.

Background Prostate-specific membrane antigen (PSMA)-targeted radionuclide therapy (TRT) has demonstrated efficacy and tolerability with a dose-response effect in phase I/II trials in men with metastatic castration-resistant prostate cancer (mCRPC). The need for positive PSMA imaging before PSMA-TRT to select patients is largely practiced, but its utility is not proven. Given target heterogeneity, developing a biomarker to identify the optimal patient population remains an unmet need. The aim of this study was to assess PSMA uptake by imaging and response to PSMA-TRT. Methods We performed an analysis of men with mCRPC enrolled in sequential prospective phase I/II trials of PSMA-TRT. Each patient had baseline PSMA imaging by planar 111 In and/or 177 Lu SPECT (N = 171) or 68 Ga-PSMA-11 PET/CT (N = 44), but the results were not used to include/exclude treatment. Semiquantitative imaging scores (IS) on a 0-4 scale were assigned based on PSMA uptake in tumors compared to liver uptake. We compared the ≥50% PSA decline response proportions between low (0-1) and high (2-4) PSMA IS using the χ2 -test. We used multivariable logistic regression analysis to understand the relationship between independent and dependent variables, including IS, radionuclide activity (dose) administered, CALGB (Halabi) prognostic risk score, prior taxane use. Results 215 men with progressive mCRPC received PSMA-TRT as follows: 177 Lu-J591 (n = 137), 177 Lu-PSMA-617 (n = 44), 90 Y-J591 (n = 28), 177 Lu-J591 + 177 Lu-PSMA-617 (n = 6). High PSMA expression (IS 2-4) was found in 160 (74.4%) patients and was significantly associated with more frequent ≥ 50% PSA reduction (26.2 vs. 7.3%, p = .006). On multivariate logistic regression analysis, higher IS was associated with a ≥50% decrease in PSA, even after accounting for CALGB (Halabi) prognostic score, the dose administered, and previous taxane use (OR, 4.72; 95% CI, 1.71-16.85; p = .006). Patients with low PSMA expression (N = 55, 24.7%) were less likely to respond. Thirteen of 26 (50%) with no PSMA uptake (IS = 0) had post-PSMA-TRT PSA decline with 2 (7.7%) having ≥ 50% PSA declines. Conclusion Collectively, the data provide evidence in favor of the hypothesis that patients with high PSMA uptake and high administered radionuclide dose correlate with a higher chance of response.