FDG PET/CT Beyond Mismatch Exclusion in [177Lu]Lu-PSMA Therapy: Baseline Prognosis and Response Assessment

Purpose To evaluate baseline FDG-PET/CT for prognostic stratification and assess the utility of FDG-PET/CT after two cycles, in mismatch-negative mCRPC treated with [ 177 Lu]Lu-PSMA. Methods Patients without PSMA-/FDG+ mismatch receiving ≥2 cycles of [ 177 Lu]Lu-PSMA, with baseline and post-cycle-2 PSMA and FDG-PET/CT, were included. Total tumor volumes (TTV) were segmented. Metabolic burden at baseline was assessed as FDG/PSMA-TTV ratio. Baseline FDG-PET parameters were measured before therapy, whereas the appearance of FDG-avid new lesions (FDG-NLs) and changes in FDG-PET parameters were assessed after 2 cycles. Cox proportional hazard models were used for survival analyses. Interim response evaluation agreement between FDG-PET/CT, PSA, and PSMA-PET/CT was evaluated using Cohen’s kappa. C-index compared prognostic discrimination. Results One-hundred patients were included (median follow-up 37.7 months; median OS 17.5 months). At baseline, FDG-TTV independently predicted worse OS (HR2.61, p=0.001), as did FDG/PSMA-TTV ratio (HR1.72, p=0.026); FDG-SUV metrics were not prognostic. At interim, FDG-NLs were independently associated with worse OS (HR2.68, p<0.0001). Reductions in FDG-TTV, FDG-SUVmax, and FDG-SUVmean were independently associated with improved OS (p≤0.021). Substantial agreement was found between FDG-NLs and PSA progression (82%, κ=0.61), and moderate with RECIP progression (79%, κ=0.54), and no differences between them (C-indices, p>0.05). In an exploratory subgroup of patients with stable disease (RECIP and PSA), FDG-NLs were associated with shorter OS (HR12.99, p=0.0024). Conclusion In [ 177 Lu]Lu-PSMA-treated patients with mismatch-negative mCRPC, baseline FDG-TTV provides independent prognostic information for OS. Interim FDG-PET/CT adds limited additional value for response assessment when compared to PSA and RECIP. However, in patients with PSA and RECIP stable disease, appearance of FDG-NLs identifies a patient subgroup with poor outcomes. Prospective validation is warranted.