Real-world outcomes of [177Lu]Lu-PSMA-I&T in [18F]FDG-positive metastatic castration-resistant prostate cancer: factors related to response and survival

Background Little is known about predictors of response to radionuclide therapy with PSMA-ligands in patients with [ 18 F]FDG-positive metastatic castration-resistant prostate cancer (mCRPC). We assessed the correlation between baseline characteristics, including dual tracer PET parameters, and response to [ 177 Lu]Lu-PSMA-I&T in a cohort of patients with [ 18 F]FDG-positive mCRPC. Prognostic factors related to progression-free survival (PFS) and overall survival (OS) were also investigated. Methods mCRPC patients who underwent [ 68 Ga]Ga-PSMA-11 and [ 18 F]FDG PET/CT prior to [ 177 Lu]Lu-PSMA-I&T were retrospectively evaluated. Only [ 18 F]FDG-positive patients were included in the analysis. A semi-automatic segmentation tool was applied to measure the whole-body maximum standardized uptake value (SUV max ), mean standardized uptake value (SUV mean ), metabolic tumor volume (MTV), and total lesion uptake (TLU) on both PET/CTs. PSA response was defined as ≥ 30% and ≥ 50% decline. Imaging response, assessed two months after the last cycle, was defined according to RECIST 1.1 and/or PSMA PET progression (PPP) criteria. Clinical, biochemical, and imaging-based factors were correlated to response to treatment, PFS, and OS. Results Twenty-five [ 18 F]FDG-positive patients who received [ 177 Lu]Lu-PSMA-I&T were included. PSA30 and PSA50 responses were achieved in 11/25 (44%) and in 9/25 (36%) patients, respectively. In the univariate analysis [ 68 Ga]Ga-PSMA-11 SUV mean ( p = 0.022) and [ 68 Ga]Ga-PSMA/[ 18 F]FDG SUV mean ratio ( p = 0.021) were significantly associated with PSA30 response. [ 68 Ga]Ga-PSMA-11 SUV mean was also significant in the multivariate model. ROC analysis indicated an optimal [ 68 Ga]Ga-PSMA-11 SUV mean cutoff value of 10.23 (AUC = 0.741), and a best discriminating [ 68 Ga]Ga-PSMA/[ 18 F]FDG SUV mean cutoff ratio of 2.08 (AUC = 0.727). In the univariate analysis for radiological PFS (rPFS), high ALP ( p = 0.023), [ 68 Ga]Ga-PSMA-11 MTV ( p = 0.011), [ 68 Ga]Ga-PSMA-11 TLU ( p = 0.015), and [ 18 F]FDG MTV ( p = 0.013) were significantly associated with shorter rPFS. [ 68 Ga]Ga-PSMA-11 MTV showed a borderline association with rPFS in the multivariate model ( p = 0.061). Hemoglobin levels ( p = 0.008), ALP ( p = 0.018) and PSA ( p = 0.023) values before treatment were associated with OS, as well as [ 68 Ga]Ga-PSMA-11 MTV ( p = 0.003), [ 68 Ga]Ga-PSMA-11 TLU ( p = 0.025), and [ 18 F]FDG MTV ( p = 0.024). However, no independent predictors of OS were identified in the multivariate analysis. Conclusions Our preliminary results suggest that a whole-body [ 68 Ga]Ga-PSMA-11 SUV mean higher than 10 and a [ 68 Ga]Ga-PSMA-11/[ 18 F]FDG SUV mean ratio > 2 may serve as useful predictors for identifying patients likely to achieve at least a 30% reduction in PSA levels, even among [ 18 F]FDG-positive mCRPC patients. Furthermore, our findings support the prognostic significance of tumor burden either calculated by imaging-based volumetric parameters or by known biochemical markers of disease.