Abstract
593
Introduction: 177Lutetium-PSMA (LuPSMA) is a radioligand with high specificity for prostate-specific membrane antigen (PSMA) enabling targeted treatment of prostate cancer. Favorable activity reported in Phase II clinical trials in patients with metastatic castration-resistant prostate cancer (mCRPC) led to an ongoing Phase III clinical trial (VISION: NCT03511664). However, treatment non-response occurs in up to 50% of patients. Herein, we investigated prognostic biomarkers of non-response and disease progression in advanced prostate cancer patients receiving LuPSMA. Materials & Methods: Patients with mCRPC across six centers worldwide who progressed after standard therapies and started LuPSMA until January, 2019 were considered for this analysis. Endpoints included non-response to LuPSMA defined as ≥25% PSA increase at 12 weeks after treatment initiation and PSA progression-free survival (PCWG3). Sixteen prognostic parameters including demographics, clinical history, and patient’s characteristics at treatment initiation were comprised in the uni- and multivariable Cox regression models. In addition, PSMA-targeted PET-derived whole-body tumor burden parameters were also included, i.e. PSMA-avid tumor volume (PSMA-TV), PSMA-SUVmean, and PSMA-SUVmax. Tumor segmentation was performed using qPSMA software. Patients were included if all 16 prognostic parameters as well as endpoint data were available.
Results: Of 352 screened patients, we identified 267 eligible patients who received 758 cycles of LuPSMA; 103 patients were treated under prospective clinical trials, while 151 patients had access to compassionate use programs. Median baseline PSA level was 126 (IQR 37-368) ng/ml. Median baseline PSMA-TV was 560 (IQR 198-1283) ml. PSA decrease ≥ 50% was achieved in 118 of 258 patients (46%, 95%CI 40-52%), including 50 patients (19%, 95%CI 15-24%) with ≥ 90% decrease. At a median follow-up of 20.0 months, median overall survival was 13.2 (95%CI 11.8-14.6) months. Results from the multivariable Cox models are detailed in Table 1. A longer time since diagnosis of PC, lower National Comprehensive Cancer Network (NCCN) risk score, higher PSMA-SUVmean, higher PSMA-TV, higher baseline haemoglobin, and the non-visceral involvement were associated with a longer PSA progression-free survival (p<0.05), while a shorter time since diagnosis of prostate cancer, lower PSMA-SUVmeanand lower PSMA-TV were associated with an increased risk of inherent resistance to LuPSMA (p<0.05).
Conclusions: In heavily pre-treated patients with mCRPC, prognostic markers may aid in improved patient stratification towards radioligand therapy with LuPSMA. PSMA PET-derived tumor burden parameters are associated with treatment outcome. Lower baseline PSMA-uptake and lower PSMA-avid tumor burden appear to be associated with less favorable treatment response. Our Findings need to be corroborated in prospective studies.