2.1 Study design and patients

The current open-label, single-arm, multicenter phase II trial was conducted across 19 centers in China.

Eligible patients were aged ≥18 years with histologically proven metastatic or locally recurrent, unresectable and incurable HER2-positive breast cancer, defined as either HER2 immunohistochemical (IHC) 3+ status, HER2 gene amplification confirmed via in situ hybridization [16], or a HER2 gene copy number ≥4 as detected by next-generation sequencing. Patients had not received systemic antitumor therapy in the metastatic phase; however, up to 2 lines of systemic endocrine therapy were allowed. Patients had to have one or more measurable lesions at baseline per Response Evaluation Criteria in Solid Tumors (RECIST) 1.1, with the stipulation that if only one measurable lesion was present, it should not have undergone prior radiotherapy or should show significant progression post-radiotherapy; adequate organ function and an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1.

Patient exclusion was performed for any of the following reasons: untreated active brain or meningeal metastases; cardiotoxic events during previous HER2-targeted therapy; a history of other malignancies; uncontrolled comorbidities, including but not limited to active infections, hypertension and cardiac insufficiency; a history of cardiac systolic dysfunction; a history of drug allergies. Eligibility criteria are detailed in the study protocol.

The current trial was granted ethical approval by the medical ethics review committee of Harbin Medical University Cancer Hospital (No. 2019-180). Informed consent was obtained in written form from all participants.

2.2 Treatment

All participants were administered KN026 (Jiangsu Alphamab Biopharmaceuticals, Suzhou, Jiangsu, China) and docetaxel (No. HJ20140086, Sanofi-Aventis Deutschland GmbH, Frankfurt, Hesse, Germany) in 21-day cycles. KN026 was administered intravenously at 30 mg/kg on day 1 of every cycle. At 2 h after KN026 infusion, the patients received docetaxel intravenously at 75 mg/m². Study treatment was maintained until the occurrence of disease progression per RECIST 1.1 criteria, initiation of a novel antitumor treatment, consent withdrawal by the participant, loss to follow-up, or conclusion of the trial period.

No dose reduction of KN026 was permitted, and its administration would be suspended upon ≥ grade 3 treatment-related toxicity associated with KN026, until the treatment-related toxicity resolves to ≤ grade 1 level. Docetaxel dosage was allowed to be reduced to 55 mg/m² if intolerable. Detailed guidelines for KN026 and docetaxel dose adjustment, as well as treatment discontinuation criteria in the event of adverse reactions, are comprehensively described in the trial protocol.

2.3 Assessments

Patients underwent baseline tumor imaging assessments using either computed tomography or magnetic resonance imaging scans within 21 days prior to KN026 and docetaxel administration. Subsequent imaging assessments were conducted every 6 weeks following treatment initiation. After the initial 48-week period, the frequency of these assessments was adjusted to every 12 weeks, continuing until certain endpoints were reached: disease progression as defined by RECIST 1.1 criteria, initiation of a novel antitumor treatment, consent withdrawal, loss to follow-up, or trial conclusion. In instances where subjects achieved a complete response (CR) or partial response (PR) per RECIST 1.1 criteria, a follow-up imaging assessment was scheduled after 6 weeks (but not earlier than 4 weeks) to confirm the response.

Adverse events (AEs) were followed up at each treatment cycle, at the end of treatment and 30 days after the last dose or initiation of a new antitumor therapy.

2.4 Study endpoints

The primary endpoints of this study were investigator-assessed ORR and duration of response (DOR) per RECIST 1.1 criteria. ORR was the rate of patients achieving either a CR or PR. DOR was the period between initial assessment of CR or PR (whichever was first recorded) and any recorded progressive disease (PD) or death from any cause (whichever occurred first) in patients with CR or PR.

Secondary endpoints comprised PFS, alongside 6-month and 12-month PFS rates, clinical benefit rate (CBR), overall survival (OS), 6-month and 12-month OS rates and AEs. PFS was the time between the first dosing of the study drugs and PD onset or death, while CBR was the percentage of patients showing measurable disease at baseline and exhibiting a CR, PR, or stable disease (SD) as best overall response for a duration of ≥24 weeks. Disease progression was evaluated per RECIST 1.1 criteria. OS was the time from the initial dosing of study drugs to death from any cause. The frequencies and severity of AEs were recorded, encompassing treatment-emergent AEs (TEAEs), treatment-related AEs (TRAEs), serious AEs (SAEs), AEs leading to treatment discontinuation, AEs leading to death, and AEs of special interest. AEs of special interest included all cardiac adverse events and non-infectious pneumonitis, regardless of their relation to KN026, observed during the administration period and up to 30 days after the final dose or start of a new antitumor treatment. The severity of these AEs was assessed using National Cancer Institute Common Terminology Criteria for Adverse Events 5.0.

2.5 Statistical analysis

The sample size calculation, anchored on a 95% confidence interval (CI) for ORR, was assessed by the Clopper-Pearson method. Based on varying ORRs and corresponding 95% CIs, different numbers of patients achieving response were estimated, ranging from 20 patients with an ORR of 40% (95% CI: 26.4%-54.8%) to 37 patients when an ORR of 74% (95% CI: 59.7%-85.4%) was considered. Taking into account those possibilities, the projected enrollment was expected to include 50-60 subjects.

The efficacy analysis set (EAS), which included all subjects administered one or more full or partial doses of KN026 or docetaxel with one or more post-treatment efficacy assessments, was utilized for analyzing tumor efficacy endpoints (ORR, CBR and DOR). The safety set (SS), including all subjects administered one or more full or partial doses of KN026 or docetaxel, was employed for safety analyses. Subgroup analyses of efficacy endpoints were conducted based on baseline characteristics such as visceral metastasis, brain metastasis, HER2 IHC findings, tumor stage, and hormone receptor (HR) status. The 95% CIs of ORR and CBR were determined by the Clopper-Pearson method, while median DOR, PFS, and OS and the respective 95% CIs were derived from Kaplan-Meier curves. Further details on the principles for handling missing data are outlined in the study protocol.

Statistical analysis utilized SAS 9.4 (SAS Institute, Cary, NC, USA). Continuous data are presented as median and range, and categorical data as frequency and percentage.

3.1 Baseline characteristics of patients

From January 13, 2020 to May 27, 2021, 57 patients with HER2-positive recurrent/metastatic breast cancer were enrolled. They were all treated with KN026 combined with docetaxel. By September 15, 2023, treatment was ongoing in 22 patients, while 35 discontinued treatment. The SS comprised 57 patients, while 55 patients were included in the EAS after completing at least one efficacy evaluation (Figure 1).

Baseline clinicodemographic features are shown in Table 1. All patients were women with a median age of 52 years. Most patients were at an advanced clinical stage at baseline, including 52 (91.2%) with stage IV disease. HER2 IHC results were predominantly IHC 3+ (84.2%), and HR positivity was found in 26 patients (45.6%). Visceral metastases were detected in 34 patients (59.6%). Of the 37 patients who previously underwent surgical treatment, 2 (3.5%) had palliative surgery, 32 (56.1%) had curative surgery, and 12 (21.1%) had other types of surgery. Of the 12 patients previously administered radiotherapy, 2 (3.5%) had three-dimensional conformal radiotherapy, and 3 (5.3%) underwent intensity-modulated radiation therapy. Adjuvant therapy was previously administered to 30 (52.6%) patients, and neoadjuvant therapy to 5 (8.8%). Previous endocrine therapy was administered to 12 (12.1%) patients, targeted therapy to 4 (7.0%), and chemotherapy to 31 (54.4%); the most commonly employed chemotherapeutic agents included cyclophosphamide (29 cases, 50.9%), docetaxel (23 cases, 40.4%), and epirubicin (14 cases, 24.6%).

3.2 Tumor response

The median time of exposure to study drugs was 674 (range, 21-1,171) days, and the median duration of treatment was 28 (range, 1-52) weeks. Totally 49 (86.0%) patients in the SS underwent more than six cycles of therapy.

In the EAS comprising 55 patients, 3 patients achieved CR and 39 achieved PR, culminating in an ORR of 76.4% (95% CI, 63.0%-86.8%). The CBR was 85.5% (95% CI, 73.3%-93.5%) (Table 2). Subgroup analyses of ORR stratified by baseline visceral and brain metastases, HER2 IHC findings, clinical stage, and HR status revealed that the therapeutic benefits extended across most subgroups, maintaining a consistent trend with the total population (Supplementary Figure S1).

In addition, the median DOR was not reached (95% CI, 20.7 months-not reached) in the 42 responding patients (Table 2). DOR rates at 12, 24 and 30 months were 85.2%, 59.3% and 51.6%, respectively (Figure 2A).

3.3 Survival

Median follow-up was 31.1 months (range: 0.8-41.4 months), and median PFS was 27.7 months (95% CI, 18.0 months-not reached) (Figure 2B). PFS rates at 6, 12, 24 and 30 months were 87.4% (95% CI, 75.4%-93.8%), 76.3% (95% CI, 62.6%-85.5%), 54.2% (95% CI, 39.7%-66.6%), and 45.2% (95% CI, 30.9%-58.4%), respectively (Figure 2B). Subgroup analyses of PFS stratified by baseline visceral and brain metastases, HER2 IHC findings, clinical stage, and HR status indicated survival benefits across most groups, consistent with the overall cohort (Supplementary Figure S2).

Median OS was not reached, and OS rates at 6, 12, 24, and 30 months were 98.2% (95% CI, 88.2%-99.8%), 93.0% (95% CI, 82.4%-97.3%), 84.1% (95% CI, 71.7%-91.4%), and 78.5% (95% CI, 65.2%-87.2%), respectively (Figure 2C).

3.4 Safety profiles

All patients experienced TEAEs. Grade ≥3 TEAEs were recorded in 36 patients (63.2%). SAEs were documented in 12 patients (21.1%), of whom 10 (17.5%) had an SAE of grade ≥3 (Supplementary Table S1). Eleven patients died due to disease progression. No deaths were attributed to KN026 or docetaxel. A fatal AE of unknown cause was reported in 1 patient who received one dose of KN026. In this patient, KN026 administration was stopped at 12 min with 162.8 mg (planned dose: 1,860 mg) due to grade 3 allergic reactions, which were resolved on the same day at discharge. The patient was withdrawn from the study without the use of docetaxel and ten days later developed grade 4 sepsis. The patient died more than 23 days after KN026 administration, and this death was considered not related to either KN026 or docetaxel.

The commonest TEAEs (occurring in ≥20% of patients) were white blood cell count decrease (63.2%), neutrophil count decrease (61.4%) and diarrhea (36.8%). The commonest grade ≥3 TEAEs (≥2%) included neutrophil count decrease (40.4%), white blood cell count decrease (28.1%) and hypokalemia (8.8%) (Table 3).

Fifteen patients (26.3%) had TEAEs leading to KN026 dose interruption and 3 (5.3%) had TEAEs leading to permanent KN026 discontinuation, including one case of grade 3 type I hypersensitivity related to KN026, one grade 3 hypersensitivity related to KN026, and one grade 3 pericardial effusion not related to KN026 and possibly related to disease progression. In addition, 6 patients (10.5%) had docetaxel dose reduction due to a TEAE. Nine patients (15.8%) had TEAEs leading to docetaxel dose interruption, and 2 (3.5%) had TEAEs resulting in permanent discontinuation of docetaxel (Supplementary Table S1).

AEs of special interest were recorded in 5 patients (8.8%). Of these, 3 (5.3%) had grade 2 pneumonitis, and 2 (3.5%) had left ventricular ejection fraction (LVEF) decrease ≥15% from the baseline absolute value, both of which were grade 2 (Supplementary Table S1). There were no reports of symptomatic left ventricular systolic dysfunction or LVEF reduction ≥10% from the baseline absolute value and LVEF <50.