Mapping PSMA-617 sensitivity in prostate cancer PDX models to enable combination therapy and resistance studies

Prostate cancer represents the second most frequently diagnosed malignancy and ranks fifth in cancer-related mortality worldwide among men. Therapeutic strategies include the FDA-approved radioligand therapy (RLT) [¹⁷⁷Lu]Lu-vipivotide tetraxetan ([¹⁷⁷Lu]Lu-PSMA-617, Pluvicto), indicated for PSMA-positive metastatic castration-resistant prostate cancer (mCRPC) following progression on hormone therapy and/or chemotherapy. Given the favourable toxicity profile of [¹⁷⁷Lu]Lu-PSMA-617, clinical investigations are exploring its application earlier in the therapeutic sequence. Nevertheless, a substantial proportion of patients exhibit only transient responses or fail to respond due to intrinsic or acquired resistance mechanisms, underscoring the need for combination regimens incorporating chemotherapeutics, radiosensitizers, targeted agents, and/or immunotherapies. Robust preclinical models are essential for evaluating such combination strategies, as they enable controlled assessment of treatment efficacy within biologically relevant systems. Patient-derived xenografts (PDXs) retain the genomic, histopathologic, and pharmacologic characteristics of the original tumors, providing reliable predictive power for clinical outcomes. Consequently, well-validated PDX models are critical for elucidating synergistic interactions and guiding translational application of novel combinatorial approaches. We utilized a panel of prostate cancer PDX models selected based on PSMA expression, quantified by immunohistochemistry and AI-driven image analysis (Visiopharm). PSMA-expressing tumors validated by IHC before the treatment were treated with ~1 mCi (30-37 MBq) of [¹⁷⁷Lu]Lu-PSMA-617 and tumor uptake of was evaluated using SPECT/CT. Tumor growth was monitored longitudinally, and therapeutic efficacy was assessed using tumor growth inhibition (TGI) and time to progression (TTP). To model acquired resistance, a [¹⁷⁷Lu]Lu-PSMA-617-resistant PDX (ST1273/RTR; XenoSTART) was derived from an initially sensitive tumor (ST1273; XenoSTART). This model tolerated up to 30 MBq of [¹⁷⁷Lu]Lu-PSMA-617, while PSMA expression and radioligand uptake remained comparable between parental and resistant tumors. As proof-of-concept, PARP inhibition with olaparib was evaluated as a radiosensitizer in ST1273/RTR. Tumor relapse occurred 29 days post-[¹⁷⁷Lu]Lu-PSMA-617 monotherapy, whereas combination treatment extended relapse to 49 days, demonstrating enhanced antitumor efficacy. Collectively, these fully characterized prostate PDX models-validated for PSMA expression and [¹⁷⁷Lu]Lu-PSMA-617 responsiveness-integrated with comprehensive clinical data, establish a robust, translationally relevant platform for preclinical evaluation of innovative combinatorial strategies in RLT.