PSMA-targeted alpha therapies inhibit tumor growth in xenograft models of visceral and bone metastatic castration-resistant prostate cancer

Introduction: Prostate specific membrane antigen (PSMA)-targeted beta-emitting radionuclide therapy, lutetium-177-PSMA-617, was recently approved for the treatment of prostate cancer (Sartor et al., NEJM 2021). PSMA-targeted alpha-emitting thorium-227-pelgifatamab corixetan (²²⁷Th-pelgi) is currently in phase 1 development (NCT03724747). An actinium-225-pelgifatamab conjugate (²²⁵Ac-pelgi), which exhibits a higher dose rate and different chelator technology compared to ²²⁷Th-pelgi, is under development. ²²⁷Th-pelgi and ²²⁵Ac-pelgi are both PSMA‑targeted alpha therapies (PSMA-TAT) consisting of a human PSMA-targeting antibody linked to a chelator moiety, radiolabeled with thorium‑227 or actinium‑225, respectively. In this study, our aim was to evaluate the activity and potential differences of PSMA-TATs in preclinical models of visceral and bone metastases mimicking micrometastatic and metastatic prostate cancer.

Methods: We studied the efficacy of ²²⁷Th-pelgi and ²²⁵Ac-pelgi in the subcutaneous LNCaP model and in the intracardiac C4-2 and intratibial LNCaP models mimicking visceral and bone metastatic castration-resistant prostate cancer (CRPC). C4-2 cells were inoculated into the left cardiac ventricle of male NRG mice and LNCaP cells into the right tibiae or flank of male NOD.scid mice. Serum prostate-specific antigen (PSA) or tumor size measurements were used for monitoring tumor growth, randomization to treatment groups (n=8-12/group), and calculating T/C (treatment to control) values. The mice were treated with vehicle, ²²⁷Th-pelgi (150-300 kBq/kg) or ²²⁵Ac-pelgi (150-300 kBq/kg) with one or two (Q5W) i.v. injections. After the first dose, tumor growth was monitored for 3, 10 and 5 weeks in the intracardiac, intratibial and subcutaneous models, respectively. PSMA-staining of histologic sections and radiography were used to measure tumor burden and tumor-induced abnormal bone growth in the intracardiac and intratibial models, respectively.

Results: In both metastatic models, T/C ratios of PSA were 0.028 and 0.008 for ²²⁷Th-pelgi (300 kBq/kg), and 0.008 and 0.009 for ²²⁵Ac-pelgi (150 kBq/kg) in the intracardiac and intratibial models, respectively, compared with 0.228 and 0.118 for ²²⁷Th-pelgi and ²²⁵Ac-pelgi at 300 kBq/kg in the subcutaneous model. PSMA-TATs inhibited tumor growth when compared with vehicle as indicated by the mean PSA change relative to pre-treatment levels. In the intracardiac C4-2 model, the values for the ²²⁵Ac-pelgi (150 kBq/kg), ²²⁷Th-pelgi (300 kBq/kg) and vehicle groups were 65%, 361% and 8974%, respectively. The corresponding values in the intratibial LNCaP model were 19.7%, 33.8%, and 2797%, respectively. In the C4-2 model, ²²⁵Ac-pelgi (150 kBq/kg) decreased tumor burden in the liver, lungs, and brain while ²²⁷Th-pelgi (300 kBq/kg) reduced the tumor burden in the liver compared with vehicle. Also, ²²⁵Ac-pelgi (150 kBq/kg) decreased the number of liver micrometastases compared with vehicle. Furthermore, mice treated with PSMA-TATs exhibited stable or increased body weight indicating prevention of tumor-induced weight loss observed in the vehicle-treated mice in the C4-2 model. Compared with vehicle, PSMA-TATs were efficacious in inhibiting tumor-induced abnormal bone growth and tumor growth in the intratibial and subcutaneous LNCaP models, respectively.

Conclusions: The efficacy of both PSMA-TATs in the visceral and bone metastatic models was comparable or even better than in the subcutaneous setting. Despite the lower radioactive dose, ²²⁵Ac-pelgi showed more robust antitumor efficacy than ²²⁷Th-pelgi in the visceral metastasis model. These data suggest that ²²⁵Ac-pelgi is active in CRPC independent of metastatic growth site, including micrometastases. In conclusion, our results provide a proof of concept for investigating ²²⁵Ac-pelgi in the clinical setting in CRPC patients with visceral and/or bone metastases.