靶向CD39和TGF-beta的双特异性抗体ES014作为单药治疗晚期实体瘤患者的首次人体研究

Background: ES014 is a first-in-class bifunctional antibody-ligand trap designed to target the adenosine and TGF-beta pathways. By simultaneously targeting CD39 and TGF-beta, ES014 can inhibit the production of immunosuppressive adenosine, promote the formation of immunostimulatory ATP, and neutralize immunosuppressive cytokine TGF-beta. We conducted a multicenter phase I study to evaluate the safety, tolerability, pharmacokinetics and preliminary antitumor activity of ES014 in patients with advanced solid tumors. Methods: This is a phase I, open-label, multicenter study with two parts: a dose escalation part and a cohort expansion part. The dose escalation part included 20, 70, 200, 700 and 1400 mg by an mTPI (modified Toxicity Probability Interval) design. Eligible patients were defined as advanced solid tumor patients with progressive disease after standard therapies. ES014 was administrated intravenously every 2 weeks. Baseline tumor CD39 expression was assessed by immunohistochemistry (IHC) staining and scored using the combined positive score (CPS) for correlation analysis with clinical responses. Additional IHC studies were conducted in samples obtained outside of the phase 1 study. The expression of biomarkers associated with the following pathways were assessed: immune checkpoint inhibition, the adenosine pathway, the TGF-beta pathway and the gamma-secretase pathway. Results: As of December 22, 2025, 75 patients received ES014 treatment. Among the 75 patients, the median age was 62.2 years. 54 (72%) were males. 22 patients (29%) experienced treatment-emergent adverse events (TEAEs) ≥Grade 3 in severity. The overall incidence of treatment-related adverse events (TRAEs) that resulted in drug discontinuation was low (4%). No patient had a TRAE that resulted in death. The most common TRAEs were anaemia (39%), rash (19%), and pruritus (17%). Among the 5 patients with desmoid tumor (DT), all received ES014 treatment of 1400 mg. 2 achieved confirmed PRs and 3 achieved stable disease, resulting in an ORR of 40% and a DCR of 100%. Time to response in the 2 responders were 8.1 and 6.8 months, respectively. Duration of treatment of the 2 responders were 13.7 and 10.5 months, respectively, and treatment was ongoing in both responders. The baseline CD39 CPS for the two responders were 90 and 10, respectively. The baseline CD39 CPS for the other patients were 5, 100 and <1, respectively. Pharmacokinetic profiles of DT patients were comparable to those of other patients at the same level in the initial cycles. In the additional biomarker studies, CD39 was highly expressed (CPS>10) in 82% (27/33) of DT samples. CD73 expression was detected (CPS>10) in 61.5% (8/13) of DT samples. p-Smad2, beta-catenin and HES1 were strongly stained (CPS >70) in all (13/13) DT samples. PD-1 and PD-L1 staining were negative in all (13/13) DT samples. Conclusions: ES014 monotherapy showed preliminary antitumor activities in DT and a good safety, tolerability, and PK profile. There appears to be a correlation between CD39 expression and clinical response.