In this randomized, double-blind, parallel-controlled, multicenter, phase III equivalence clinical trial, patients were recruited from 52 medical centers in China. Eligible participants were aged 18 to 75 years, with histologically confirmed HER2-positive breast cancer at accredited local laboratories. The study included patients with early (T2-3, N0-1, M0) or locally advanced (T2-3, N2 or N3, M0; T4, any N, M0) stage breast cancer, based on the 8th edition of the American Joint Committee on Cancer (AJCC) staging system for breast cancer [22], and with confirmed HER2-positivity and ER/PR negative statuses. Other inclusion criteria included an Eastern Cooperative Oncology Group performance status of 0 to 1 and a baseline left ventricular ejection fraction (LVEF) ≥55%, measured by echocardiography (first choice) or multigated acquisition scan. The main exclusion criteria included stage IV or bilateral breast cancer, prior antineoplastic treatment or radiation therapy for any malignancy, and pregnancy or lactation. The full inclusion and exclusion criteria are detailed in the methods section in the Supplementary Material.

The study received ethical approval from independent ethics committees for each center and adhered to the Declaration of Helsinki, Good Clinical Practice guidelines, and all applicable regulatory requirements. The study protocol, including subsequent modifications, was approved by ethics committees, and written informed consent was obtained from all participants.

Following eligibility confirmation, patients were centrally and randomly (1:1) assigned to one of two treatments: QL1209 (Qilu Pharmaceutical, Jinan, China) combined with trastuzumab (Roche, Basel, Switzerland) and docetaxel (Qilu Pharmaceutical, Jinan, China) for the QL1209 group, or pertuzumab (Roche, Basel, Switzerland) plus trastuzumab (Roche, Basel, Switzerland) and docetaxel (Qilu Pharmaceutical, Jinan, China) for the pertuzumab group. The allocation schedule was generated using central stratified randomization, stratified by disease stage (early-stage vs. locally advanced-stage, per the 8th AJCC staging system). As the study was double blinded, patients, investigators, study site personnel, review committee, and the sponsor’s study team were masked to treatment allocation.

This study was initially designed to investigate the equivalence of QL1209 to reference pertuzumab in efficacy and safety according to the 1.1 and 2.0 version of protocol. As the study progressed, adjuvant treatment was added post-surgery in the updated version 3.0 protocol, in accordance with the Guidelines for clinical trials of biosimilars for pertuzumab injection, released on April 21, 2021, by the Center for Drug Evaluation NMPA in China [23].

During the neoadjuvant period, patients underwent treatment with QL1209 or pertuzumab, combined with trastuzumab and docetaxel for 4 cycles (every 3 weeks). QL1209 or reference pertuzumab received an 840 mg loading dose in cycle 1, followed by 420 mg in cycles 2-4. Trastuzumab was administered at a loading dose of 8 mg/kg in cycle 1, followed by 6 mg/kg in cycles 2–4. Docetaxel was administered at 75 mg/m² immediately after QL1209 or reference pertuzumab in cycles 1–4. Dose adjustments were permitted at the investigator’s discretion, with discontinuation of QL1209/pertuzumab recommended if reference trastuzumab was discontinued. Surgery was performed within 2 weeks after completing the 4-cycle neoadjuvant treatment.

In the subsequent neoadjuvant period, patients received FEC chemotherapy (fluorouracil 500-600 mg/m², epirubicin 90-120 mg/m², cyclophosphamide 500–600 mg/m²) in cycles 5-7, along with QL1209 and trastuzumab at the previous doses in cycles 8–20. Treatment continued until the completion of 20 cycles or until an event occurred, such as disease recurrence, loss to follow-up, death, withdrawal of informed consent, switch to alternative tumor treatment, or intolerable toxicity, whichever occurred first. The end-of-treatment or treatment-discontinuation visit was scheduled 28 days after the final administration of the study drug, marking the conclusion of the study.

Laboratory parameters, including hematology and serum chemistries, physical examination, 12-lead electrocardiogram, Eastern Cooperative Oncology Group Performance Status (ECOG PS), and vital signs were assessed at each cycle. Routine tumor response, evaluated through clinical breast examination, was conducted every 2 cycles during the initial 4 treatment cycles, followed by assessments every 4 cycles from cycles 8 to 20, in accordance with Response Evaluation Criteria in Solid Tumors version 1.1 (RECIST v.1.1). A preoperative assessment, encompassing a physical examination, mammogram, and ultrasound (if deemed necessary according to local practice), was performed. The independent review committee (IRC) and investigators (INV) evaluated surgical specimens. Pathologic response was determined by local histopathology assessment of surgical breast specimens and lymph node tissues post neoadjuvant therapy, according to the 8th edition AJCC ypTNM staging system.

Adverse events (AEs) were monitored until hospital discharge post-surgery or 28 days after the last adjuvant treatment. AE severity was coded using MedDRA v25.0 and graded according to the National Cancer Institute common terminology criteria version 4.03 (NCI-CTCAE v4.03). Cardiac function, including LVEF and left ventricular systolic dysfunction (LVSD), was assessed using echocardiograms or multiple-gated acquisition scans at specified intervals. Symptomatic LVSD was defined as a symptomatic decrease in LVEF or explicit/highly likely cardiac-related death, while asymptomatic significant reduction in LVEF was defined as a decrease of ≥10% from baseline to an LVEF of ≤50%. Blood samples for PK and immunogenicity analysis were collected before each QL1209 or reference pertuzumab infusion at cycles 1-5, 10, 15, and the end of treatment/discontinuation visits.

The primary endpoint was the total pathologic complete response (tpCR) rate assessed by IRC, defined as the absence of invasive tumor cells in the breast and ipsilateral axillary lymph nodes upon microscopic examination following primary tumor excision (ypT0/is, ypN0). Secondary endpoints included tpCR rate assessed by INV, breast pathological complete response rate (bpCR) based on IRC and INV, objective response rate (ORR), event-free survival (EFS), disease-free survival (DFS), safety, PK, and immunogenicity. The bpCR rate was defined as the absence of invasive tumor cells in the breast upon microscopic examination following primary tumor excision (ypT0/is). ORR was the percentage of patients with the best overall response of complete response (CR) or partial response (PR) according to RECIST v.1.1. EFS was the time from randomization to the first occurrence of disease progression, recurrence, or death from any cause, while DFS was the time from surgery to disease recurrence or death from any cause.

PK was assessed by minimum serum concentration (Ctrough). Immunogenicity analysis involves detecting antidrug antibodies (ADAs) and neutralization antibodies (NAbs). Treatment-induced ADA positivity was defined as ADA-positive after treatment in patients who were ADA-negative at baseline or had ADA-positive titers at baseline with an increase in ADA concentration of ≥4-fold post-baseline.

A sample size of 512 patients was planned to achieve 80% power in detecting equivalence at a predefined margin, with a significance level determined by two one-sided tests (α = 0.025), assuming a dropout rate of 10% and an estimated 50% of patients achieving a tpCR. The sample size was calculated using PASS software (NCSS, LLC, Kaysville, UT). The equivalence margin was determined through a meta-analysis of efficacy data in the ER/PR-negative subgroup from the NeoSphere and PEONY studies [3, 10]. The pooled relative risk (RR) of tpCR for pertuzumab plus trastuzumab and docetaxel versus trastuzumab plus docetaxel was 2.11 (70% confidence interval (CI), 1.74-2.57). The chosen equivalence margin of 0.76 to 1.32 aimed to preserve at least half of the efficacy observed in these studies [23].

Analyses of the primary endpoint (tpCR) and secondary endpoints (bpCR, ORR, EFS, and DFS) were conducted in the full analysis set (FAS). A 2-sided 90% Wald CI for the ratio of tpCR rate was calculated using logarithmic transformation without covariate adjustment. Subjects with concomitant events were categorized as non-responders, and those with missing assessment results were not imputed. Equivalence was affirmed if the CI entirely fell within the range of 0.76 to 1.32. Time-to-event endpoints (EFS and DFS) were estimated using the Kaplan-Meier method, with 95% CIs, and inter-group comparisons were performed via the stratified log-rank test. RR and associated 90% CIs were assessed using a stratified Cox proportional-hazards model.

To ensure the robustness and avoid or minimize potential bias, we analyzed the primary endpoint in per-protocol set (PPS; including patients in the FAS who had completed primary efficacy evaluations, except those who had a major protocol deviation or did not receive 4 doses of study drug at least), performed supplementary analyses of the primary endpoint in patients with centrally confirmed HER2-positive and ER/PR-negative, conducted subgroup analyses according to stratification factor, tipping-point analyses. More detailed statistical analyses can be seen in the methods section in the Supplementary Material. Statistical analyses were performed using SAS software (SAS Institute) version 9.4 or later.