Phase II study evaluating the safety and efficacy of neratinib and trastuzumab biosimilar in patients with <i>HER2</i> mutated advanced solid tumors: KCSG AL20‐17/KM23 trial

Human epidermal growth factor receptor 2 (HER2) overexpression and amplification activate key pathways driving tumor progression, leading to HER2-targeted therapies. However, HER2 signaling can also be aberrantly activated by somatic mutations, independent of overexpression or amplification, contributing to tumorigenesis [1]. These mutations, found in domains such as the extracellular (ECD), transmembrane (TMD)/juxtamembrane (JMD), and tyrosine kinase (KD) regions, occur across cancers, from melanoma (1%) to bladder cancer (≤12%) [1, 2], suggesting a significant population could benefit from HER2-targeted treatments. Neratinib, an irreversible pan-HER tyrosine kinase inhibitor, has shown efficacy in HER2-mutated cancers and is recommended in National Comprehensive Cancer Network guidelines [3]. However, resistance mechanisms, such as secondary HER2 mutations or amplifications, may limit its efficacy [4, 5], highlighting the need for combination therapies. Preclinical and clinical studies demonstrate that combining neratinib with trastuzumab enhances HER2 signaling inhibition and improves outcomes in HER2-mutated non-small cell lung cancer (NSCLC) and hormone receptor-positive/HER2-negative breast cancer [6]. Here, we evaluated dual HER2 inhibition with neratinib and trastuzumab biosimilar (Herzuma®) in heavily pretreated patients with HER2-mutated solid tumors (excluding HER2 amplifications). We also explored circulating tumor DNA (ctDNA) testing as a biomarker for treatment suitability and identified genetic alterations linked to treatment response. Detailed methods are provided in the Supplementary Materials and Methods. A total of 40 patients received the study treatment and were followed up (Supplementary Figure S1). The patients were heavily pretreated, with 40% receiving more than three lines of systemic therapy prior to this study, including adjuvant or concurrent chemoradiotherapy, with the latest regimen being in the salvage setting. The baseline patient characteristics are summarized in Supplementary Table S1. Among the enrolled patients, 23 unique HER2 mutations were identified, with 3 patients harboring 2 different mutations each, resulting in 43 mutations overall (Supplementary Table S2). HER2 mutations were detected via tissue in 38 patients and ctDNA in the remaining 2 patients. The KD was the most frequently mutated site (26/43 mutations, 60.5%), followed by TMD/JMD (10/43 mutations, 23.3%), and ECD (7/43 mutations, 16.3%). Most patients with NSCLC (10/17, 58.8%) harbored exon 20 insertion mutations. Of the 40 treated patients, 39 were evaluable for efficacy. The objective response rate (ORR) was 25.6% (95% confidence interval [CI], 12.9-41.2), with 10 partial responses (PR) and a median duration of response (DOR) of 11.2 months (95% CI, 3.9-18.4). Stratified by tumor type, the ORR was 23.5% in NSCLC, 40% in biliary cancer, and 66.7% in breast cancer. One patient with vulvar cancer and one external auditory canal cancer patient achieved a PR, as well. All responders (best percentage change ≤ -30%) had HER2 KD mutations, including four with exon 20 insertions, except one patient with an ECD mutation (S310Y) (Figure 1A). Three NSCLC patients with exon 20 insertions achieved PR. Seven patients had stable disease (SD) for more than 6 months, resulting in a clinical benefit rate (CBR) of 43.6%. Median progression-free survival (PFS) and overall survival (OS) were 3.4 (95% CI, 1.6-5.3) and 9.5 months (95% CI, 2.9-16.0), respectively. Notably, the patient with vulvar cancer had a PFS of 17.5 months during the study period, and as of the treatment cut-off date, the patient was still receiving treatment. Patients with KD mutations had significantly better outcomes than those with ECD or TMD/JMD mutations, with longer PFS (5.33 vs. 2.27 months; hazard ratio [HR] = 0.29 P = 0.006) and OS (not reached vs. 5.98 months; HR = 0.34, P = 0.03) (Figure 1B-C, Supplementary Figure S2). Efficacy and survival outcomes of neratinib and trastuzumab in heavily pretreated HER2-mutated advanced cancers. (A) Best percentage change in target lesion size, assessed by RECIST v1.1. (B-C) Kaplan-Meier curves for PFS (B) and OS based on HER2 mutation location (C). (D) Genomic landscape and treatment outcomes for 32 patients with baseline ctDNA evaluation. (E-F) Kaplan-Meier curves for PFS (E) and OS stratified by ctDNA fraction (F). Abbreviations: RECIST, response evaluation criteria in solid tumor; KD, kinase domain; ECD, extracellular domain; TMD/JMD, transmembrane domain/juxtamembrane domain; PFS, progression-free survival; HR, hazard ratio; CI, confidence interval; OS, overall survival; No., number; PR, partial response; SD, stable disease; PD, progressive disease; ctDNA, circulating tumor DNA; SNV, single nucleotide variant; CNV, copy number variant. Exploratory biomarker analysis for baseline ctDNA was performed on samples from 32 patients after excluding 8 samples that did not meet quality control criteria. HER2 mutations were identified in all but 3 cases, consistent with prior tumor/ctDNA findings. In these 3 patients, where baseline ctDNA did not detect HER2 mutations, 2 achieved SD and 1 experienced PD, suggesting limited clinical benefit from dual anti-HER2 therapy (Figure 1D). The median ctDNA fraction was 5.7% (range 0.1-72.0), with a cut-off of 11.5% identified as most informative. Higher ctDNA fractions were linked to shorter PFS (4.87 vs. 2.01 months; HR = 3.81, P = 0.02) and OS (13.71 vs. 4.64 months; HR = 4.96, P = 0.036) (Figure 1E-F). Common pathogenic mutations included TP53 (n = 20), EGFR amplification (n = 6), PIK3CA amplification (n = 3), HER2 amplification (n = 3), and KRAS mutations (n = 3). Alterations in EGFR, BRAF, KRAS, or PTEN appeared associated with poorer treatment response and shorter PFS, though statistical analysis was limited by low sample size and mutation frequency (Figure 1D, Supplementary Figure S3). During the study, 35 (87.5%) patients experienced treatment-related adverse events (AEs), with 12 (30%) having grade ≥3 AEs. Diarrhea was the most common AE (77.5%), leading to neratinib discontinuation in three patients due to grade 3 diarrhea. The CONTROL trial later demonstrated that a dose escalation strategy effectively reduces diarrhea severity [7]. While this was not implemented in our study due to timing, future adoption could improve safety and efficacy. No cardiac events were observed. Most AEs were manageable, though one patient with underlying lung cancer experienced grade 5 pneumonitis, with unclear causality. AE details are in Supplementary Table S3. Despite the noted limitations of our study, including its small sample size, single-arm design, and heterogeneous patient population, our findings provide meaningful insights into dual HER2 inhibition with neratinib and trastuzumab biosimilar in HER2-mutated tumors. While recent studies on HER2 antibody-drug conjugates (ADCs) have demonstrated promising efficacy in HER2-mutated cancers [8, 9], their accessibility may be limited in certain settings. Dual anti-HER2 therapy with neratinib and trastuzumab can serve as an alternative treatment strategy, particularly for patients with disease progression or resistance following HER2-targeted ADC, expanding therapeutic options and addressing unmet needs in HER2 mutated cancers. Furthermore, our exploratory analysis of ctDNA as a prognostic biomarker offers a practical tool to optimize patient selection and monitor treatment response. In conclusion, dual anti-HER2 therapy with neratinib and trastuzumab demonstrated promising efficacy and a manageable safety profile in heavily pretreated patients with HER2-mutated cancers. The identification of ctDNA as a prognostic biomarker further enhances its clinical utility, supporting dual HER2 blockade as a valuable treatment option deserving further validation. Kyoungmin Lee and Kyung-Hun Lee: methodology, formal analysis, validation, and writing-original draft. Jeesun Yoon: methodology, investigations, and writing-review & editing. Dong-Wan Kim, Yoon Ji Choi, Soohyeon Lee, Ju Won Kim, Kyong Hwa Park, Wonyoung Choi, Youngjoo Lee, Hyewon Ryu, Dong-Hoe Koo, Yun‑Gyoo Lee, Hei-Cheul Jeung, Min-Young Lee, Namsu Lee, Myoung Joo Kang, Jieun Lee, Sook Hee Hong, and Eun Joo Kang: data collection and resources. In Hae Park: conceptualization, methodology, writing-review & editing, and project administration. All authors have read and approved the manuscript, and IHP is responsible for submitting the manuscript for publication. We thank the patients and their families as well as the investigators who participated in this study. We also thank the Korean Cancer Study Group and KOSMOS trial group. Neratinib (Nerlynx®) was supported by Bixink Therapeutics and Herzuma® by Celltrion, Republic of Korea. Dong-Wan Kim reports research funding to his institution from Alpha Biopharma, Amgen, AstraZeneca, BMS, Boehringer-Ingelheim, Bridge BioTherapeutics, Chong Keun Dang, Daiichi-Sankyo, GSK, Hanmi, IMBDx, InnoN, IQVIA, Janssen, Merck, Merus, Mirati Therapeutics, MSD, Novartis, ONO Pharmaceutical, Pfizer, Roche/Genentech, Takeda, TP Therapeutics, Xcovery, and Yuhan. He has also received medical writing assistance from Amgen, AstraZeneca, BMS, Boehringer Ingelheim, Bridge BioTherapeutics, Chong Keun Dang, Daiichi-Sankyo, GSK, IMBDx, Janssen, Merus, Mirati Therapeutics, MSD, Meck, Novartis, Pfizer, Roche, Takeda, and Yuhan. No other potential conflicts of interest were reported. This research was supported by grants from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), the Ministry of Health & Welfare, Republic of Korea (grant numbers: HI17C2206 and HA22C0012). Additional support was provided by a grant from the National R&D Program for Cancer Control, National Cancer Center (NCC), Ministry of Health & Welfare, Republic of Korea (grant number: HA22C0052), through the KOSMOS Molecular Tumor Board. The study was conducted in compliance with the principles of good clinical practice, adhering to the International Conference on Harmonization guidelines, and the ethical principles outlined in the Declaration of Helsinki. Approval for this study was obtained from the independent ethics committees or institutional review boards at each participating site (IRB number of the principal investigator's institution: 2021GR0117), and all patients provided written informed consent prior to participation. ClinicalTrials.gov, number NCT06083662. The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials. 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.