Analysis of Molecular Imaging and Laboratory Baseline Biomarkers in PSMA-RLT: Whole-Body Total Lesion PSMA (TLP) Predicts Overall Survival

The aim of this retrospective study was to identify pre-therapeutic predictive laboratory and molecular imaging biomarkers for response and overall survival (OS) in patients with metastatic castration-resistant prostate cancer (mCRPC) treated with prostate-specific membrane antigen (PSMA)-targeted radioligand therapy (RLT). Pre-therapeutic laboratory and [68Ga]Ga-PSMA-11 PET/CT data of n = 102 mCRPC patients receiving [177Lu]Lu-PSMA-617 RLT within a prospective registry (REALITY Study, NCT04833517) were analyzed including laboratory parameters such as alkaline phosphatase (ALP), prostate-specific antigen (PSA), gamma glutamyl transferase (GGT), glutamate oxaloacetate transaminase (GOT), glutamate pyruvate transaminase (GPT), neuron specific enolase (NSE), hemoglobin (Hb), and imaging parameters such as maximum standardized uptake value of the tumor lesions (SUVmax), the mean standardized uptake value of all tumor lesions (SUVmean), the whole-body molecular tumor volume (MTV), and the whole-body total lesion PSMA (TLP). Mann–Whitney U test, univariate and multivariable Cox-regression were performed to test for association of the parameters with response and OS. The SUVmean of all lesions was significantly different between responders and non-responders (SUVmean responders 8.95 ± 2.83 vs. non-responders 7.88 ± 4.46, p = 0.003), whereas all other tested biochemical and imaging parameters did not reveal significant differences. Hb and the molecular imaging parameters MTV and TLP showed a significant association with OS (p = 0.013, p = 0.005; p = 0.009) in univariant Cox regression; however, only TLP remained significant in multivariable analysis (Hazard ratio 1.033, p = 0.009). This study demonstrates a statistically significant association between the quantitative PET/CT imaging parameter SUVmean and PSA response, as well as between the baseline TLP and OS of mCRPC patients undergoing RLT.