Safety and Efficacy of KN046 in Combination with KN026 in Patients with Advanced HER2-Positive Breast Cancer: A Phase II Trial Open Access

Breast cancer is the most common cancer type among women worldwide, with HER2-positive breast cancer accounting for 15% to 20% of all breast cancer cases (1, 2). The introduction of HER2-targeted therapies, particularly the antibody–drug conjugates trastuzumab emtansine and trastuzumab deruxtecan, as well as the tyrosine kinase inhibitors pyrotinib and tucatinib, has significantly improved the survival outcomes for patients with HER2-positive metastatic breast cancer (3–5). Despite these advances, metastatic patients ultimately develop resistance to currently approved anti-HER2 therapies, highlighting the significant challenge and urgent need for developing novel therapeutic strategies to improve patient outcomes.

Several studies have demonstrated that anti-HER2 antibodies such as trastuzumab and pertuzumab can stimulate both innate and adaptive immunity responses (6, 7). Preclinical research has shown that dual blockade of HER2 and PD-1/PD-L1 significantly enhances the antitumor effect of anti-HER2 antibodies in mice bearing HER2-positive breast carcinomas (8, 9). However, early-phase trials of first-generation PD-1/PD-L1 mAb pembrolizumab or durvalumab combined with trastuzumab showed very limited clinical benefits in patients with trastuzumab-resistant, advanced HER2-positive breast cancer (10, 11). There were also several previous efforts such as IMpassion050 and Neo-PATH studies to combine HER2-targeting agents with first-generation immune checkpoint inhibitors in HER2-positive disease in the neoadjuvant setting, although the combinational treatment did not increase pathologic complete response (CR) rates (12, 13).

KN026 is a bispecific anti-HER2 antibody consisting of the heavy-chain variable domains of trastuzumab and pertuzumab on each arm, along with the fragment crystallizable region from trastuzumab. It may enhance tumor cell binding, suppress ligand-dependent and ligand-independent tumor growth, and promote internalization of HER2 receptors (14, 15). A prior phase I study of KN026 monotherapy at dosages of 20 mg/kg every 2 weeks or 30 mg/kg every 3 weeks achieved an objective response rate (ORR) of 28.1% with a tolerable safety profile in 57 patients with HER2-positive metastatic breast cancer who had progressed on prior anti-HER2 therapies (16). Additionally, KN046 is a novel bispecific antibody targeting PD-L1 and cytotoxic T lymphocyte–associated protein 4, which shows a higher affinity for PD-L1, and mediates the depletion of regulatory T cells in the tumor microenvironment (17). A recent phase I trial of KN046 monotherapy in treating patients with advanced solid tumors who had failed standard treatments observed an ORR of 12.5% (18). Here, we report the efficacy and safety profile of the breast cancer cohort in the KN026-203 study, a multiarmed phase II trial evaluating the combination of KN026 and KN046 in treating advanced HER2-positive solid tumors that have failed more than one line of prior systemic therapies.

This was an open-label, phase II, multicenter study (ClinicalTrials.gov number: NCT04521179) designed to evaluate the efficacy, safety, and tolerability of KN026 in combination with KN046 in subjects with HER2-positive solid tumors, including gastric/gastroesophageal junction cancer, breast cancer, and other HER2-positive solid tumors. This analysis focused on the breast cancer cohort. See Supplementary Table S1 for data on representativeness of our study population. The main inclusion criteria were as follows: (i) patients ages 18 to 75 years; (ii) histologically or cytologically confirmed, metastatic or locally advanced unresectable HER2-positive breast cancer. HER2-positive was defined as HER2 IHC 3+ or HER2 IHC 2+ with gene amplification: confirmed by in situ molecular hybridization or next-generation sequencing (RRID: SCR_005182) with an HER2/CEP17 ratio >2.0 or next-generation sequencing HER2 copy number >6; (iii) disease progression (PD) during or after prior treatment with ≥1 line of HER2-targeted therapy in the advanced setting. Patients who relapsed within 12 months after completing neoadjuvant/adjuvant chemotherapy were considered first-line treatment failures; (iv) Eastern Cooperative Oncology Group performance status of 0 or 1, with measurable lesions according to RECIST v1.1; and (v) adequate organ and bone marrow function. Exclusion criteria included untreated active brain metastases or leptomeningeal metastases (however, patients with treated brain metastases could be included if the metastases were stable, defined by at least 4 weeks of neuroimaging prior to first dosing showing stability of lesions, with no evidence of new neurologic symptoms, or if symptoms had returned to baseline and no new or enlarging brain metastases were present). Patients with a history of autoimmune disease, active infections, recent treatment with a systemic immunostimulatory agent (within 4 weeks before enrollment), or a history of severe allergic reactions to mAbs were also excluded. Complete inclusion and exclusion criteria are listed in the trial protocol (Supplementary File S1).

All enrolled patients received KN026 at 30 mg/kg every 3 weeks in combination with KN046 at 5 mg/kg every 3 weeks, until PD, occurrence of intolerable adverse events (AE), or withdrawal of consent. A loading dose of KN026 was administered on day 8 of cycle 1. This dose level was confirmed to be effective in a phase I study (19). Dose delays were allowed for the management of AEs, but dose modifications were not permitted.

The study was sponsored by Jiangsu Alphamab Biopharmaceuticals Co., Ltd., and approved by the institutional review board at each participating site. Written informed consent was obtained from each subject before enrollment. This clinical trial was conducted according to the Declaration of Helsinki and approved by the Institutional Review Board of Sun Yat-sen Memorial Hospital, and written informed consent was obtained from all patients. All authors vouch for the accuracy and completeness of the data and the fidelity of the trial to the protocol.

The primary endpoint was the ORR, assessed by investigators according to RECIST v1.1. Patients with a first assessment of CR or partial response (PR) were required to confirm the response within 4 to 8 weeks. The secondary endpoints included duration of response (DoR, defined as the duration of persistent response for patients with confirmed PR or CR), progression-free survival (PFS), 6- and 12-month PFS rates, clinical benefit rate [CBR, defined as CR, PR, or stable disease (SD) ≥24 months], time to response, overall survival (OS), and the incidence and severity (as per NCI Common Terminology Criteria for Adverse Events, version 5.0) of AEs.

PD-L1 was measured using the 22C3 assay on the Dako Link 48 platform. PD-L1 status was recorded by the combined positive score (CPS). CPS ≥ 1 was defined as PD-L1 positive, and CPS < 1 was considered PD-L1 negative. Stromal tumor-infiltrating lymphocytes (TIL) were evaluated in hematoxylin and eosin sections following criteria proposed by the International Immuno-Oncology Biomarker Working Group (20).

This study aimed to enroll a total of 80 to 122 patients, with 30 to 36 patients in the HER2-positive breast cancer cohort, 30 to 60 patients in the HER2-positive gastric/gastroesophageal junction cancer cohort, and 20 to 26 patients in the other HER2-positive solid tumor cohort. As this study is an exploratory phase II trial, the sample size calculation is based on the estimate of the 95% confidence interval (CI) for ORR using the Clopper–Pearson method (details in Supplementary Table S2), instead of archiving a specific statistical power. We expected that the ORR of the treatment would be above 40%. Therefore, with 30 subjects enrolled in the breast cancer cohort, when more than or equal to 12 subjects achieved ORR, the lower limit of 95% CI of the ORR would be above the prespecified threshold of 20%. Considering a dropout rate of 20%, a total of 36 subjects would be enrolled in this study, and more than or equal to 15 subjects should achieve ORR.

The Clopper–Pearson method was used to calculate the two-sided 95% CI for ORR. For time-related events (DoR, PFS, OS, and time to response), the Kaplan–Meier method was used to calculate median times and 95% CIs. Efficacy and safety analyses were conducted for all patients who received at least one dose of the study drug. Post hoc biomarker analyses were performed to examine potential associations between efficacy and PD-L1 or stromal TIL status. All statistical analyses were conducted using SAS version 9.4 or GraphPad Prism software.

Due to patients’ privacy protection or ethical restrictions, the data generated in this study are not publicly available. The data are available upon request to the corresponding author.

Between December 7, 2020, and June 11, 2021, a total of 36 patients with HER2-positive breast cancer were enrolled at eight sites in China and received at least one dose of KN026 plus KN046. As of the data cutoff on November 10, 2024, five (13.9%) patients remained on treatment.

Among the 36 patients with HER2-positive breast cancer, the median age was 53.0 years. Overall, 41.7% of patients had liver metastasis. Furthermore, 41.7% of patients received ≥4 lines of prior anticancer therapy, and 36.1% of patients had received ≥3 lines of anti-HER2 therapy. All patient demographic and clinical characteristics are summarized in Table 1.

Among the 36 patients who received the study drug (KN026 and/or KN046), 17 (47.2%) patients had confirmed objective response, including two CRs (5.6%) and 15 PRs (41.7%); 12 (33.3%) patients were assessed as SD, including one patient with an SD duration exceeding 24 weeks. The confirmed ORR was 47.2% (95% CI, 30.4–64.5), with a median DoR of 15.2 months (95% CI, 4.1–not evaluable), and the median time to response was 1.5 months (95% CI, 1.4–2.8). The CBR was 50.0% (95% CI, 32.9–67.1; Table 2; Fig. 1A and B). Subgroup analysis indicated that the confirmed ORR was 66.7% in the 21 patients without liver metastasis and 61.5% in the 13 patients who received 1 to 2 lines prior treatment in the advanced setting. Furthermore, we enrolled three patients with brain metastases in this study; one had received a best response of PR that lasted for 35.9 weeks, one patient showed a best response of SD that lasted for 18.4 weeks, and the other one was observed as disease progression after two cycles of treatment. Therefore, the ORR in patients with brain metastases in this study was 66.7%.

The median PFS was 5.6 months (95% CI, 4.1–13.8). The 6- and 12-month PFS rates were 46.6% (95% CI, 29.2–62.3) and 36.6% (95% CI, 20.6–52.8), respectively (Fig. 2A). With a median follow-up time of 22.1 months, the median OS was 25.7 months (95% CI, 11.7–not evaluable), and the 12-month OS rate was 63.9% (95% CI, 46.1–77.2; Fig. 2B).

In addition, we performed subgroup analyses in terms of PD-L1 status and stromal TILs. Baseline tumor samples from 26 and 18 patients were available for PD-L1 and TIL assessment, respectively. PD-L1 status was associated with neither ORR nor PFS (CPS ≥1 vs. <1: 75% vs. 60%, P = 0.968; 18.4 vs. 16.0 months, P = 0.482). Patients with more stromal TILs (≥10%) had a trend toward higher ORR and longer median PFS than those with fewer TILs (<10%), although the differences were not statistically significant (54.5% vs. 14.3%, P = 0.550; 13.8 vs. 3.5 months, P = 0.216).

Among the 36 patients, the median duration of treatment with KN026 and KN046 was 24.2 weeks (range: 6.0–104.9) and 23.9 weeks (range: 5.9–104.7), respectively. Treatment-related adverse events (TRAE) occurred in 34 (94.4%) patients, and the most common TRAEs were infusion-related reactions (44.4%), aspartate aminotransferase (AST) increased (36.1%), alanine aminotransferase (ALT) increased (30.6%), and diarrhea (30.6%) (Table 3). Grade 3–4 TRAEs occurred in 10 (27.8%) patients, and the most common TRAEs were ejection fraction decreased (5.6%) and γ-glutamyl transferase increased (5.6%). Of the two patients who experienced a grade 3–4 ejection fraction decreased, one had a decrease in the left ventricular ejection fraction (LVEF) to 51% (565 days after receiving the initial study treatment), which represented a 22% decline from baseline, and the LVEF had recovered to the baseline level after study treatment discontinuation; another patient had a decrease in LVEF from 54% to 23% (126 days after receiving the initial study treatment), and the LVEF had recovered to 49% after study treatment discontinuation. No patient experienced a TRAE leading to death.

A total of four (11.1%) patients had TRAEs leading to discontinuation of KN026, including peripheral edema in one patient, myocardial ischemia in one patient, infusion-related hypersensitivity reaction in one patient, and myocarditis, cardiac failure, cardiac and respiratory arrest, upper respiratory tract infection, and hypoxic–ischemic encephalopathy in one patient (Supplementary Table S3). There were seven (19.4%) patients who experienced TRAEs leading to KN046 discontinuation, including phosphocreatine kinase increased (one patient), adrenal insufficiency (one patient), peripheral ischemia (one patient), immune-mediated pneumonitis (one patient), infusion-related hypersensitivity reaction (one patient), diabetic ketoacidosis (one patient), and myocarditis, cardiac failure, cardiac and respiratory arrest, upper respiratory tract infection, and hypoxic–ischemic encephalopathy in another patient (Supplementary Table S3). In addition, one (2.8%) patient experienced a serious AE related to KN026 (Table 4). Six (16.7%) patients experienced a serious AE related to KN046 (Table 4).

Sixteen (44.4%) patients had immune-related AEs (irAE), and the most common ones (incidence rate ≥5%) were pruritus (22.2%), hypothyroidism (19.4%), hyperthyroidism (16.7%), and pulmonary inflammation (5.6%). Among them, two patients had grade 3 irAEs, including adrenal insufficiency in one patient and diabetic neuropathy, diabetic angiopathy, and diabetic ketoacidosis in another patient. No patient experienced grade 4–5 irAEs (Supplementary Table S4). Thirty-two (88.9%) patients had anti-HER2–related AEs, and the most common ones (incidence rate ≥10%) were pruritus (27.8%), infusion-related reaction (27.8%), diarrhea (27.8%), rash (22.2%), AST increased (16.7%), ALT increased (16.7%), weight loss (13.9%), lymphocyte count decreased (11.1%), neutrophil count decreased (11.1%), and abnormal hepatic function (11.1%). Among them, five patients had grade 3 AEs, including ALT increased in one patient, lymphocyte count decreased in one patient, neutrophil count decreased in one patient, γ-glutamyl transferase increased in two patients, and upper respiratory tract infection in one patient. No patient experienced grade 4–5 anti-HER2–related AEs (Supplementary Table S5).

This phase II, multicenter study demonstrated that the chemo-free combination of KN026 and KN046 as second- or later-line therapy for patients with advanced HER2-positive breast cancer resulted in an encouraging ORR of 47.2% (95% CI, 30.4–64.5) and a median PFS of 5.6 months (95% CI, 4.1–13.8), with a consistent safety profile. To the best of our knowledge, this is the first trial to assess the efficacy and safety of triple blockade of HER2 and PD-L1/cytotoxic T lymphocyte–associated protein treatment in patients with trastuzumab-resistant, HER2-positive advanced breast cancer.

The combination of KN026 and KN046 exhibited favorable antitumor activity, with an ORR of 47.2%, a CBR of 50.0%, and a median PFS of 5.6 months in this trial. These results surpassed those from a prior phase Ib to II trial of pembrolizumab plus trastuzumab for trastuzumab-resistant, HER2-positive advanced breast cancer, which showed an ORR of 11.5% and a median PFS of 2.7 months (10). The antitumor activity of KN026 plus KN046 observed in this study was notably superior to that seen in a previous phase Ib study of durvalumab plus trastuzumab in patients with HER2-positive metastatic breast cancer previously treated with chemotherapy and anti-HER2 antibodies, in which no patients achieved a response (PR/CR) and the median PFS was only 1.4 months (11). It is important to note that most patients enrolled in this study had a high tumor burden (median sum of diameters of target lesions of 54.6 cm, with 41.7% having liver metastasis) and were heavily pretreated (a median of two previous lines of anti-HER2 therapy after relapse or metastasis, with 63.9% having received more than three lines of anticancer treatment in the advanced setting). Interestingly, the two patients who achieved persistent CR (lasting more than 18 months) had progressed after prior treatment with trastuzumab and a taxane and anti-HER2 tyrosine kinase inhibitor drugs, such as lapatinib or pyrotinib. In addition, patients with PD-L1–negative tumors also benefited from the chemo-free combination therapy in this study. Furthermore, three patients with brain metastases were enrolled in the current study, which had an ORR of 33.3% and a disease control rate of 66.7% in these patients with brain metastases. Although only 3 of the total 36 patients had brain metastases, we could see the positive sign of the chemo-free combination of KN046 and KN026 in treating patients with HER2-positive breast cancer with brain metastases. Further studies on more patients with brain metastases are needed to explore the clinical benefit of this combinational therapy in these special populations.

The mechanism of the antitumor activity of KN026 involves increasing tumor cell binding, enhancing blockade of ligand-dependent and -independent tumor growth, as well as elevating target avidity, thereby promoting increased HER2 receptor internalization (15). A preclinical study demonstrated that KN046 showed higher affinity for PD-L1 and mediated the depletion of regulatory T cells in the tumor microenvironment, which could enhance the antitumor immune response (19). However, the specific underlying mechanisms of the KN026 and KN046 combination remain unclear and warrant further investigation.

The safety profile of KN026 plus KN046 in this trial was consistent with the previously known toxicity profiles of each drug. The prior reported all-grade TRAEs and ≥grade 3 TRAEs of KN026 monotherapy (at doses of 5, 10, 20, and 30 mg/kg) occurred in 88.9% and 6.3% of patients with breast cancer, respectively, and those of KN046 monotherapy (at doses of 1, 3, and 5 mg/kg) occurred in 83.0% and 14.0% of patients with multiple solid tumors, respectively, although there was a lack of AE data for KN046 monotherapy in patients with breast cancer previously (16, 18). The all-grade AEs of KN046 (5 mg/kg) combined with KN026 (30 mg/kg) in our study occurred in 94.4%, and the ≥grade 3 TRAEs occurred in 27.8%, which was higher than those rates of AEs reported in the above two prior phase I studies with monotherapy, and this might be explained by the combination of these two drugs with higher dosage. However, all the patients who experienced grade 3–4 TRAEs had recovered after interventions, and there were no new signals of side effects related to combination treatment in this study. The most common ≥grade 3 TRAEs were ejection fraction decreased (5.6%) and γ-glutamyl transferase increased (5.6%). All the patients who experienced a grade 3–4 TRAE had recovered after interventions. In an early-phase trial, which explored KN026 monotherapy in treating HER2-positive metastatic breast cancer, higher rates of grade 3 or more ALT increased (22.2%) and AST increased (22.2%) were reported but a slightly lower incidence of grade 3 ejection fraction decreased (1.6%) compared with the rates observed in this study.

This study was limited by its single-arm design and small sample size. Large randomized controlled trials are needed to definitively demonstrate the role of this novel chemo-free combination treatment in patients with advanced HER2-positive breast cancer who were resistant to prior trastuzumab-based therapies. Additionally, the median PFS of KN026 plus KN046 was relatively short in the current trial, and thus further study should focus on identifying specific patient subgroups that may benefit more from this chemo-free combination therapy.

In summary, this phase II, multicenter study demonstrated that the chemo-free treatment of KN026 plus KN046 as second- or later-line therapy in patients with advanced HER2-positive breast cancer had a favorable efficacy with a comparative safety profile. Further randomized controlled trials are required to confirm these findings.

L. Shen reports grants and other support from BeiGene, Ltd., outside the submitted work. No disclosures were reported by the other authors.

J. Liu: Conceptualization, resources, data curation, formal analysis, funding acquisition, validation, visualization, methodology, writing–original draft, project administration. C. Song: Conceptualization, resources, data curation, supervision, validation, investigation, project administration, writing–review and editing. Y. Yang: Data curation, formal analysis, supervision, methodology. Xiangcai Wang: Resources, data curation, supervision, validation, investigation, project administration. M. Ni: Resources, data curation, formal analysis, validation, investigation, project administration. Xujuan Wang: Resources, data curation, formal analysis, validation, investigation, project administration. L. Chen: Resources, data curation, validation, investigation, project administration. H. Yang: Resources, data curation, validation, investigation, project administration. R. Zhao: Resources, data curation, validation, project administration. T. Xu: Resources, data curation, formal analysis, visualization. L. Shen: Conceptualization, resources, data curation, supervision, funding acquisition, methodology, project administration, writing–review and editing.

We thank the patients who participated in the trial, the members of the data monitoring committee, the members of the trial steering committee, the staff members who assisted with the trial at each site, and the team that supported the trial. We thank Prof. Wen Jiang from MD Anderson Cancer Center for assistance in English language editing. This study was supported by grants from the Natural Science Foundation of China (82072906) and Guangdong Science and Technology Department (2022A1515012238).