Background: Prostate cancer (CaP) is the most common cancer in US men. Disseminated CaP invariably progresses to metastatic castrate-resistant prostate cancer (mCRPC). Current treatment options for mCRPC usually lead to therapeutic resistance, and novel therapies are urgently needed. PSMA is a prostate-specific antigen over-expressed on most mCRPC. Antibodies against PSMA have been used to create T-cell engaging bispecific Abs (TCEs) and chimeric antigen receptor T cells, but all such approaches to date induce frequent/severe cytokine release syndrome (CRS). We combined a high-affinity αPSMA moiety with a low-activating αCD3 binder to create TNB-585; in preclinical studies, TNB-585 showed equivalent anti-tumor efficacy but much reduced cytokine secretion compared to PSMA-targeted TCEs with a strongly activating αCD3 domain. TNB-585 also has a full length silenced Fc domain, conferring a 3-week half-life. A phase 1 study investigating the safety, pharmacokinetics (PK), anti-drug antibodies (ADA) and preliminary activity of TNB-585 in patients with mCRPC is ongoing and described. Methods: TNB585.001 (NCT04740034) is an open-label, multi-center study of TNB-585 in patients with mCRPC. The study is divided into escalation (Arm A, N=24) and expansion (Arm B, N=30) arms. Subjects who have received 2 or more prior lines of therapy are eligible. Prior exposure to PSMA-targeted therapy is permitted, as are well-controlled HBV, HCV, and HIV infection; subjects with secondary malignancies that do not interfere with the study may also be enrolled. Other key inclusion/exclusion criteria include EGFR of > 30ml/min and ECOG ≤ 2. TNB-585 is administered as an intravenous infusion every 3 weeks. Subjects must be admitted for 48 hours after their 1st dose; TNB-585 is given on an outpatient basis thereafter. Dose escalation is proceeding in Arm A via single patient cohorts until the onset of toxicity or activity; thereafter subjects enroll using a BOIN design. Arm B will start once the maximum tolerated dose (MTD) / recommended phase 2 dose (RP2D) has been selected. Subjects will be treated until progression or unacceptable toxicity. In Arm A, occurrence of dose limiting toxicities (DLTs) will drive identification of the MTD (or RP2D) based on the BOIN escalation and de-escalation boundaries (λe of 0.236 and a λd of 0.358). In Arm B accrual will be suspended if more than 33% of subjects experience a DLT event. Adverse events (AEs), laboratory profiles, and vital signs will be assessed throughout the study. AEs are graded according to the NCI CTCAE, version 5.0. The activity endpoints (per PCWG3/RECIST1.1) include overall response rate, PSA50, PSA30, CTC counts, progression free survival and overall survival. The relationship between PSMA expression (via PSMA-PET) and activity will be assessed. Clinical trial information: NCT04740034