Background: Ephrin receptors and their membrane-localized ligands induce bidirectional signaling and facilitate tumor-stroma interactions. Expression of EphB4 is increased in prostate cancer tissue and cell lines and retained in castration resistant states, and can promote cell migration, invasion, and metastases. Blocking the EphB4-EphrinB2 pathway, which can be accomplished by soluble EphB4 conjugated to human serum albumin (sEphB4-HSA), has efficacy in preclinical models of aggressive prostate cancer. A phase I clinical trial of sEphB4-HSA led to response or stable disease in 56% of patients, with no grade 4 or 5 related adverse events, and combination pembrolizumab sEphB4-HSA led to a 52% response rate in EphrinB2 expressing urothelial cancer. We hypothesized that targeting the EphB4-EphrinB2 pathway may serve as a therapeutic target in the treatment of metastatic castration resistant prostate cancer (mCRPC). Methods: We conducted a single arm, phase II trial in patients with progressive mCRPC and treatment with at least one second generation androgen receptor (AR)-targeted therapy but no more than three prior therapies for mCRPC. On Day 1 of each cycle patients received sEphB4-HSA 1000 mg IV, with cycle length 14 days cycles 1-6 and cycle length 21 days for cycle 7 and beyond. The primary endpoint was confirmed prostate specific antigen (PSA) response rate (confirmed decrease in PSA by > 50%). We employed a Simon two stage Minimax design, requiring two or more responses among the first 15 patients to enroll an additional ten patients. Results: Fourteen eligible patients enrolled in the study. Median age was 73.5 years (range 52-83), patients had a median baseline PSA value of 65.11 ng/mL (range 7.77-2850 ng/mL) and received a median of three prior therapies (range 1-3) for mCRPC. Ten patients received prior taxane for mCRPC or hormone sensitive prostate cancer. The median length of treatment with sEphB4-HSA was 6.5 weeks (range 2-35 weeks). The potentially treatment-related adverse events (AEs) that occurred in more than 25% of patients were hypertension (10 patients) and fatigue (7 patients). Three patients experienced a serious adverse event potentially related to therapy, including one patient with a grade 5 event (cerebral vascular accident) possibly related to study drug. No patient had a confirmed PSA response, and the study was stopped for futility. Thirteen patients had PSA progression ( > 25% increase in PSA), and one patient withdrew due to toxicity prior to having an evaluable PSA response. The median time to PSA progression was 28 days (95% CI 28-64 days), and median time to radiologic progression was 55 days (95% CI 55 days-NR). Of three patients with measurable disease, two had stable disease and one had progressive disease. Conclusions: In patients with mCRPC who progressed on prior second generation AR-targeted therapy, sEphB4-HSA monotherapy had no discernable anti-tumor activity. Clinical trial information: NCT04033432.