Anbenitamab in combination with chemotherapy in patients with HER2‐positive gastric or gastroesophageal junction carcinoma who failed previous therapy containing trastuzumab: a multicenter, phase II study (KC‐WISE 01)

In 2022, gastric cancer (GC) ranked as the fifth most common cancer and the third leading cause of cancer death in China, with 358,672 new cases and 260,372 deaths, accounting for 37.0% and 39.4% of global cases, respectively [1]. Previous studies have shown that 25.9% and 36.5% of GC patients in China were diagnosed at stages III and IV, respectively, with 5-year overall survival (OS) rates of 33.0% for stage III and 5.5% for stage IV [2, 3]. A previous study reported that human epidermal growth factor receptor 2 (HER2) positivity was found in 8.8% of Chinese patients with gastric adenocarcinoma [4]. Building on the ToGA trial [5], which established trastuzumab plus chemotherapy as the standard first-line treatment for HER2-positive GC, the KEYNOTE-811 trial [6] demonstrated superior progression-free survival (PFS) combined with immune checkpoint inhibitors, establishing this regimen as the current standard of care. Recent interim-analysis of the DESTINY-Gastric04 trial revealed that trastuzumab deruxtecan significantly improved overall survival compared to chemotherapy-based regimen in patients with HER2-positive gastric/gastroesophageal junction (GC/GEJ) carcinoma, leading to its establishment as second-line therapy [7]. Furthermore, zanidatamab, a HER2-targeted bispecific antibody against extracellular domains (ECDs) 2 and 4, has shown promising efficacy as both a monotherapy and in combination with chemotherapy for HER2-overexpressing GC in the second-line setting [8]. Anbenitamab, also known as SYS6092 (KN026), is a bispecific antibody that simultaneously binds to two distinct HER2 epitopes, the same extracellular domains targeted by trastuzumab (domain IV) and pertuzumab (domain II) [9]. Here, we present the results of a phase II study (NCT05427383) evaluating the safety and efficacy of anbenitamab plus chemotherapy in patients with HER2-positive advanced GC who failed previous therapy containing trastuzumab. Patients with locally advanced or metastatic HER2-positive GC/GEJ carcinoma who failed previous therapy containing trastuzumab were assigned to receive anbenitamab (30 mg/kg, Day 1, every 3 weeks [Q3W]) plus paclitaxel (175 mg/m2, Day 1, Q3W) or irinotecan (125 mg/m2, Day 1 and Day 8, Q3W) at investigators' discretion based on previous treatment. Full methodology is detailed in the Supplementary Materials. Between April 7, 2022, and January 12, 2023, 39 patients were enrolled across 19 hospitals in China (Supplementary Table S1) and were assigned to receive either anbenitamab plus paclitaxel (n = 20) or anbenitamab plus irinotecan (n = 19; Supplementary Figure S1). Baseline patient characteristics are detailed in Supplementary Table S2. Eastern Cooperative Oncology Group performance status (ECOG PS) scores of 0 and 1 were reported in 7 (17.9%) and 32 (82.1%) patients, respectively. HER2 immunohistochemistry 3+ was observed in 34 (87.2%) patients. Metastatic involvement of ≥ 3 organs was present in 17 (43.6%) patients. Meanwhile, 10 (25.6%) patients had received ≥ 2 prior lines of therapy. At the data cutoff date (March 26, 2024), median follow-up duration was 17.0 months (95% confidence interval [CI], 14.3-19.1). In the anbenitamab plus paclitaxel arm, median duration of exposure was 112 days (range, 21-614), with mean relative dose intensities of 90.0% (standard deviation [SD], 13.1%) for anbenitamab and 86.3% (SD, 13.7%) for paclitaxel, respectively. In the anbenitamab plus irinotecan arm, median duration of exposure was 281 days (range, 21-663), with mean relative dose intensities of 86.7% (SD, 13.3%) for anbenitamab and 58.9% (SD, 18.0%) for irinotecan, respectively. Twenty patients underwent dose reduction and 33 had dose interruption. As of the cutoff date, 32 patients discontinued treatment due to disease progression (n = 20), withdrawal of consent (n = 5), death (n = 4), or investigators' discretion (n = 3) (Supplementary Figure S1). During the safety observation period (n = 6 per treatment arm), one dose-limiting toxicity of grade 3 febrile neutropenia was observed in a patient treated with anbenitamab plus paclitaxel. Therefore, an additional 27 patients were enrolled and received treatment with anbenitamab plus paclitaxel or irinotecan. All patients (n = 39) experienced at least one treatment-related adverse event (TRAE), of which grade ≥ 3 TRAEs occurred in 25 (64.1%) patients (Supplementary Table S3). The most common grade ≥ 3 TRAEs were neutrophil count decreased (33.3%), white blood cell count decreased (28.2%), anemia (17.9%), and fatigue (10.3%) (Supplementary Table S4). Serious TRAEs were reported in 8 (20.5%) patients, with neutrophil count decreased (5.1%) being the most common (Supplementary Table S5). TRAEs leading to treatment discontinuation occurred in 3 (7.7%) patients. One patient was reported as an adverse event of special interest (grade 2 reduced ejection fraction), which was determined not to be related to the treatment. At the cutoff date, 3 (7.7%) patients died during the treatment period, defined as death occurring within 30 days after the last dose, due to heart attack, gastrointestinal hemorrhage and unknown reason. Of these, none were considered to be related to the treatment by the investigator. A total of 37 patients were eligible for response evaluation by an independent review committee (IRC). Of those, 14 (37.8%) patients achieved confirmed partial response, 9 (24.3%) patients had stable disease, 6 (16.2%) patients achieved Non-complete response (CR)/Non-progressive disease (PD), and 8 (21.6%) patients had progressive disease. The confirmed objective response rate (ORR) was 37.8% (95% CI, 22.5%-55.2%), while the disease control rate (DCR) was 78.4% (95% CI, 61.8%-90.2%; Supplementary Table S6, Figure 1A and 1B). Comparable outcomes were observed upon investigator assessment. Among 37 patients evaluable for response, confirmed partial response was achieved in 17 (45.9%) patients, stable disease in 12 (32.4%) patients, and progressive disease in 7 (18.9%) patients. One patient was deemed not evaluable due to < 42 days between two stable disease assessments. The confirmed ORR was 45.9% (95% CI, 29.5%-63.1%), while the DCR was 78.4% (95% CI, 61.8%-90.2%). The median time to response and duration of response (DoR), as assessed by IRC, were 1.4 months (interquartile range, 1.4-1.5) and 11.7 months (95% CI, 6.0-15.6), respectively. Median PFS was 8.6 months (95% CI, 3.8-13.1) per IRC assessment (Figure 1C), and 8.8 months (95% CI, 4.1-11.3) per investigator assessment (Supplementary Table S6). As of March 26, 2024, 22 deaths had occurred, with a median OS of 13.2 months (95% CI, 10.6-20.5) (Figure 1D). No anti-drug antibodies or neutralizing antibodies were detected in patients receiving anbenitamab plus paclitaxel or irinotecan. In the present study, anbenitamab plus paclitaxel or irinotecan demonstrated a tolerable safety profile and promising antitumor efficacy. The confirmed IRC-assessed ORR was 37.8%, with a median PFS of 8.6 months and a median OS of 13.2 months. These efficacy outcomes are comparable to those observed in the recent DESTINY-Gastric04 trial [7], in which trastuzumab deruxtecan reported a median PFS of 6.7 months and a median OS of 14.7 months. Both regimens represent promising treatment options in the second-line setting for HER2-positive GC. These outcomes may stem from: (1) Anbenitamab's bispecific mechanism, enabling stronger HER2 binding and superior preclinical antitumor activity versus trastuzumab plus pertuzumab [9], with induced receptor clustering driving HER2 degradation [10]; and (2) its improved safety profile, particularly reduced cardiotoxicity and gastrointestinal toxicities, facilitating sustained full-dose combination regimens. Nonetheless, there are several limitations to this study. Firstly, although this study provided initial insights, its non-comparative, single-arm nature and unblinded design inherently restricted causal inferences between intervention and outcomes. Secondly, the sample size of this study was relatively small, limiting the generalizability of the findings. Thirdly, although tumor samples following progression after trastuzumab-containing regimen treatment were preferred per study protocol, the majority of samples ultimately collected for the HER2 level test were archived samples obtained before disease progression during previous trastuzumab treatment. This approach might potentially result in the undetected loss of HER2 overexpression, which could possibly dilute the treatment efficacy. In conclusion, anbenitamab in combination with chemotherapy demonstrated a manageable safety profile and promising anti-tumor activity. A phase III study is in progress evaluating anbenitamab in combination with chemotherapy versus chemotherapy alone in patients with HER2-positive GC/GEJ carcinoma. All authors had access to the data used to prepare the manuscript and participated in the writing or critical review and editing of the manuscript. Jianming Xu and Yanping Liu conceptualized, designed and planned the study. Chuanhua Zhao, Jun Zhao, Yigui Chen, Bo Liu, Yangfeng Du, Ying Cheng, Chenglin Li, Jingdong Zhang, Mudan Yang, Ying Liu, Yuxian Bai, Suyi Li, Ruixing Zhang and Fangling Ning were involved in data acquisition. Qi Zhang provided administrative, logistical or technical support. Yijiao Xie, Yuping An and Kai Zou analyzed the data. Jianming Xu, Yanping Liu, Yijiao Xie and Yuping An interpreted the results. Jianming Xu drafted the manuscript. Chuanhua Zhao, Jun Zhao, Yigui Chen, Bo Liu, Yangfeng Du, Ying Cheng, Chenglin Li, Jingdong Zhang, Mudan Yang, Ying Liu, Yuxian Bai, Suyi Li, Ruixing Zhang, Fangling Ning, Yanping Liu, Yuping An, Kai Zou and Yijiao Xie reviewed and revised the manuscript for important intellectual content. All authors approved the final draft of the manuscript. We thank all the patients who participated in this study. We would like to thank Lei Wang, an employee of the Shanghai JMT-Bio Technology Co., Ltd, for the medical writing assistance. Yanping Liu, Kai Zou, Qi Zhang, Yijiao Xie and Yiping An are employees of Shanghai JMT-Bio Technology Co., Ltd. Other authors have nothing to declare. This study was supported by Shanghai JMT-Bio Technology Co., Ltd. The study protocol was approved by the Independent Ethics Committee of Chinese PLA General Hospital (C2021-102-01), Changzhi People's Hospital (LSPJ2022006), Fujian Provincial Cancer Hospital (2022-016-01), Cancer Hospital of Shandong First Medical University (SDZLEC2022-042-01), The First People's Hospital of Changde City (2022-016-01), Jilin Cancer Hospital (202201-011-01), Linyi People's Hospital (2022-lunlishencha-10), Liaoning Cancer Hospital and Institute (20220205), Shanxi Province Cancer Hospital (YW2022009), Affiliated Cancer Hospital of Zhengzhou University, Henan Cancer Hospital (L2022-Y010-002), Harbin Medical University Cancer Hospital (2022-91), Anhui Provincial Cancer Hospital, The First Affiliated Hospital of University of Science and Technology of China (2022-lunlishencha-08), The Fourth Hospital of Hebei Medical University (2022044), Binzhou Medical University Hospital (2022-001-01), Central Hospital Affiliated to Shandong Medical University (jizhongxinlunlilinshen2022-011-01), Qingdao Central Hospital, University of Health and Rehabilitation Science ([Y]SY202200401), Liaocheng People's Hospital (2022006), Suining Central Hospital (2022-lunliyijian-sy002-01), and Yanan University Xianyang Hospital (YDXYEC-YWPJ-2022-5) and conducted in accordance with the Declaration of Helsinki and Good Clinical Practices. Written informed consent was obtained from all patients before enrollment. The de-identified individual data generated during the current study are available from the corresponding author upon request by contacting [email protected]. All Requests will be reviewed by the corresponding author and the sponsor, Shanghai JMT-Bio Technology Co., Ltd. A signed data access agreement with the sponsor is required before data sharing. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.