Immune checkpoint inhibitors have altered the treatment landscape significantly in several cancers, yet not enough for many cancer patients. T cell costimulatory receptors have been pursued as targets for the next generation of cancer immunotherapies, however, sufficient clinical efficacy has not yet been achieved. CD137 (TNFRSF9, 4-1BB) provides co-stimulatory signals and activates cytotoxic effects of CD8⁺ T cells and helps to form memory T cells. In addition, CD137 signalling can activate NK cells and dendritic cells which further supports cytotoxic T cell activation. An agonistic monoclonal antibody to CD137, urelumab, provided promising clinical efficacy signals but the responses were achieved above the maximum tolerated dose. Utomilumab is another CD137 monoclonal antibody to CD137 but is not as potent as urelumab. Recent advances in antibody engineering technologies have enabled mitigation of the hepato-toxicity that hampered clinical application of urelumab and have enabled to maintain similar potency to urelumab. Next generation CD137 targeting molecules currently in clinical trials support T cell and NK cell expansion in patient samples. CD137 targeting molecules in combination with checkpoint inhibitors or ADCC-enhancing monoclonal antibodies have been sought to improve both clinical safety and efficacy. Further investigation on patient samples will be required to provide insights to understand compensating pathways for future combination strategies involving CD137 targeting agents to optimize and maintain the T cell activation status in tumors.

Purpose: CB307 is a trispecific variable heavy-chain antibody fragment against prostate-specific membrane antigen (PSMA), CD137, and human serum albumin. It is designed to mitigate hepatotoxicity by activating T cells only in PSMA-positive tumors and to increase drug half-life through albumin binding. This phase I study investigated the safety and tolerability of CB307 as monotherapy or in combination with pembrolizumab. Patients and Methods: Patients who were heavily pretreated with PSMA-positive solid tumors were enrolled in the dose-escalation phase of CB307 monotherapy. The additional safety and efficacy of CB307 were assessed in the CB307 monotherapy expansion cohort and in combination with pembrolizumab. Results: CB307 was administered to 75 patients. CB307 was given once every 7 days as monotherapy (N = 50) or in combination with pembrolizumab (N = 25). Two dose-limiting toxicities (grade 3 transient transaminitis) were observed. A total of three grade 3 transaminitis events (one in the monotherapy cohort and two in the combination cohort) were observed, and none involved bilirubin elevation. Durable RECIST responses were observed in two patients with metastatic castration-resistant prostate cancer enrolled in the 800 mg CB307 monotherapy and in one patient in the combination cohort (overall response rates: 11.1% and 7.1%, respectively). A disease control rate (DCR) of 50% was observed in patients enrolled in the 800 mg CB307 monotherapy cohort and 42.9% in the combination cohort. In a post hoc analysis, the response was numerically better in patients who had not received chemotherapy in the 6 months prior to starting CB307 (ORR = 20%, DCR = 60% vs. ORR = 0%, DCR = 37.5%). CB307 induces cytotoxic cell expansion in tumors and PD-L1 expression. Conclusions: CB307 was well tolerated as a monotherapy and in combination with pembrolizumab, and tumor responses were observed in patients with metastatic castration-resistant prostate cancer.