An anti-EpCAM x anti-CD3 bispecific antibody, M701, for the treatment of malignant pleural effusion due to NSCLC: A prospective randomized controlled phase II trial.

Abstract

12061 Background: Malignant Pleural Effusion (MPE) is a common complication in patients with advanced NSCLC, associated with poor prognosis, severe symptoms, and reduced quality of life. Indwelling pleural catheter (IPC) is the first-line recommended therapy to alleviate MPE-related symptoms; however, no effective pharmacotherapy has been established in current clinical guidelines for MPE management. Methods: Patients with advanced NSCLC and symptomatic MPE who had progressed following at least 1 line of systemic treatment were eligible. Patients were randomized 1:1 to experimental arm (Arm E: IPC + M701 IP at 25 μg on day 1, 400 μg on days 4, 7, 10) or control arm (Arm C: IPC + cisplatin IP 40 mg on days 1, 7 or IPC alone), stratified by prior intrapleural chemotherapy (yes/no). Catheters were removed after day 10. Primary endpoint was puncture-free survival (PuFS), defined as time from catheter removal to next puncture or death. Secondary endpoints included time to next puncture (TTNP), overall survival (OS), MPE objective response rate (ORR), investigator-assessed MPE symptoms (Likert scale), and patient-reported symptoms (LC-13 scale). Results: As of December 22, 2025, 92 patients were randomized to Arm E (n=45) or Arm C (cisplatin, n=41; thoracentesis, n=6). Baseline characteristics were generally balanced between the two arms. The proportions of patients positive for actionable genomic alterations (AGA) were 77.77% (n=35) in Arm E and 72.34% (n=34) in Arm C, respectively. M701 demonstrated a significant improvement trend in PuFS and TTNP compared with cisplatin (see Table 1). The MPE ORR was 46.7% for Arm E and 34.0% for Arm C (p=0.217). Patients with prior intrapleural (IP) chemotherapy or without AGA had a more favorable treatment response in Arm E relative to Arm C, whereas the remaining patients did not. Dyspnea symptoms also demonstrated a trend toward greater improvement in Arm E, as assessed by both investigators and patients. OS data were immature, and no significant between-group difference was observed (HR=0.94, p=0.855). Grade 3 or higher treatment-related adverse events (AEs) occurred in 2.2% of patients in Arm E and 6.4% in Arm C. Only one serious adverse event (SAE), grade 2 fever, was deemed related to M701. Conclusions: M701 IP infusion showed a promising efficacy on re-accumulation of MPE compared with Cisplatin infusion, and was well tolerated, supporting further clinical development, especially in patients with no AGA or patients who received IP chemotherapy priorly. Clinical trial information: NCT05543330 . Efficacy of the FAS set and subgroup. Population mPuFS (days) HR P value mTTNP (days) HR P value Full analysis(45 vs 47) 170 vs 72 0.72 0.212 NR vs 86 0.05 0.035 Prior IP chemotherapy or no AGA(28 vs 32) 253 vs 55 0.58 0.099 NR vs 65 0.26 0.004 With AGA and no prior IP chemotherapy(17 vs 15) 124 vs 114 1.09 0.832 170 vs NR 1.34 0.582