Neoadjuvant anti-HER2 antibody-drug conjugate SHR-A1811 in Hormone Receptor-positive, HER2-low Breast Cancer: A Phase 2 Clinical Trial (HELEN-015)

Abstract

As an anti-HER2 antibody-drug conjugate, SHR-A1811 has exhibited significant antitumor activity in patients with HER2-low advanced breast cancer. This trial aims to explore the efficacy and safety profile of SHR-A1811 as a neoadjuvant therapy for hormone receptor-positive, HER2-low (HR + /HER2-low) breast cancer. We conducted a single-arm, phase 2 study enrolling women aged 18 to 70 years with untreated HR + /HER2-low invasive breast cancer (cT2-3, cN0-3) and a Ki-67 proliferation index exceeding 14%. Eligible patients had an Eastern Cooperative Oncology Group performance status of 0 or 1, and received intravenous SHR-A1811 at a dose of 6.4 mg/kg every three weeks for eight cycles. The primary endpoint was the objective response rate, assessed in all patients who received at least one treatment cycle. Secondary endpoints included pathological complete response (pCR, residual cancer burden (RCB) 0–1 rate, event-free survival (EFS), disease-free survival (DFS), and distant disease-free survival (DDFS). In total, 66 of 101 screened patients (65.3%) were enrolled, with 65 included in the efficacy analysis and 66 in safety analysis. The objective response rate was 81·5% (95% confidence interval (CI), 70·0-90·1), exceeding the protocol-specified threshold of 66·0%. Pathological complete response (ypT0/Tis ypN0) was observed in 2 patients (3.1%; 95% CI, 0.4-10.8), while 6 patients (9.2%; 95% CI, 3.5-19.6) achieved a residual cancer burden class 0-1. Treatment-related grade ≥3 adverse events occurred in 37.9% of patients. The most common grade ≥3 adverse events were neutropenia (27.3%), leukopenia (16.7%), and anemia (13.6%). No treatment-related deaths or interstitial lung disease were reported. In summary, SHR-A1811 monotherapy achieved favorable response rates with manageable toxicity, suggesting a potential alternative neoadjuvant option for HR + /HER2-low breast cancer. Further randomized trials may be necessary to assess its clinical value more comprehensively. Follow-up continues for survival. ClinicalTrials.gov identifier: NCT05911958

Anti-HER2 antibody-drug conjugates (ADCs) have been approved for patients with Hormone Receptor positive (HR + ), HER2-low metastatic breast cancer. Here, the authors present a phase II interventional clinical trial evaluating the anti-HER2 ADC SHR-A1811 as neoadjuvant monotherapy in patients with HR + HER2-low breast cancer (HELEN-015).

Introduction

Hormone receptor-positive (HR+), HER2-negative (HR+/HER2−) breast cancers comprise a substantial portion of breast cancer cases, with approximately 65% exhibiting low HER2 expression (HER2-low)^1,2, defined by an immunohistochemical (IHC) score of 1+ or 2+ without ERBB2 amplification³. This subgroup generally receives limited benefit from conventional HER2-targeted therapies⁴. Similar to other HR+/HER2− cancers, current standard neoadjuvant and adjuvant treatment primarily involves chemotherapy and endocrine therapy. Although these neoadjuvant approaches yield modest rates of pathologic complete response (pCR), their clinical utility extends beyond pCR by facilitating therapeutic sensitivity assessments, offering an opportunity to downstage locally advanced disease for resection, promoting breast-conserving surgery, and allowing de-escalation of axillary interventions in selected cases⁵.

Chemotherapy remains the preferred neoadjuvant treatment for HR + /HER2- breast cancer, while neoadjuvant endocrine therapy may be considered for low-risk postmenopausal patients. Current chemotherapy regimens yield objective response rates (ORRs) of 47–66%, yet pCR rates are typically below 10%⁵. These multi-agent regimens are often associated with significant toxicity^6,7. Notably, HER2-low tumors show poorer responses compared to HER2 IHC 0 (HER2-zero) cases². The clinical adoption of neoadjuvant endocrine therapy remains limited due to unresolved questions regarding the optimal treatment duration and response assessment⁸. Recent advancements in anti-HER2 antibody-drug conjugates (ADCs) have provided new therapeutic options for patients with HER2-low breast cancer. Trastuzumab deruxtecan (T-DXd) has shown significant efficacy in metastatic HR+, HER2-low (HR+/HER2-low) disease⁹. Following FDA approval for advanced HER2-low breast cancer, the reported incidence of treatment-related interstitial lung disease (ILD) at 12.1%, including 0.8% fatal cases, underscores the need for careful risk stratification and management strategies in curative settings¹⁰.

SHR-A1811 is an investigational HER2-targeting ADC with a drug-to-antibody ratio of 5.7. Its optimized design incorporates a chiral cyclopropyl group for enhanced plasma stability (<1% payload release over 21 days). Its payload SHR169265 exhibits approximately 1000-fold greater in vitro potency versus conventional chemotherapeutics. In a clinical study involving heavily pretreated metastatic HER2-low breast cancer patients, SHR-A1811 demonstrated an ORR of 60.4% with an acceptable safety profile. The incidence of ILD was only 2.6%¹¹. However, the efficacy and safety of SHR-A1811 in the neoadjuvant setting for HR+/HER2-low breast cancer remain undefined.

We conducted this phase 2 trial to evaluate the efficacy and safety of SHR-A1811 as a neoadjuvant monotherapy for HR+/HER2-low breast cancer. By concentrating on this specific population, the study aims to offer valuable insights into the potential of SHR-A1811 and inform future research. This report details the primary endpoint of ORR, along with safety and other secondary outcomes.

Results

Between July 12, 2023, and March 29, 2024, a total of 101 Asian women were screened, with 66 meeting the eligibility criteria and enrolled in the study (the primary reason for screen failure was unmet laboratory inclusion criteria). All 66 patients received at least one treatment cycle. One patient withdrew consent during cycle one and, due to the lack of effective efficacy evaluation, was excluded from the efficacy analysis but included in safety assessments. Consequently, 65 patients were included in the efficacy analysis. Of these, 62 underwent surgery and 58 completed all eight treatment cycles. All 66 patients were included in the safety analysis (Fig. 1).

Fig. 1. CONSORT flow diagram of patient enrollment and disposition.

Of 101 patients screened, 66 were enrolled and received treatment (safety set); 1 withdrew before efficacy evaluation, leaving 65 in the efficacy set.

The cohort had a median age of 49 years (interquartile range [IQR], 43–54 years). Most patients presented with node-positive disease (57.6%) and were classified as stage II (86.4%). HER2 IHC 2+ expression was observed in 74.2% of cases, while ER positivity of ≥50% was documented in 95.5% of the participants (Table 1).

Table 1.

Baseline characteristics

	Patients (N = 66)
Median age, years	49 (43–54)
Age, years
≤50	40 (60.6%)
>50	26 (39.4%)
Women	66 (100%)
Asian	66 (100%)
Eastern Cooperative Oncology Group performance status
0	61 (92.4)
1	5 (7.5)
Menopausal status
Premenopausal	44 (66.7)
Postmenopausal	22 (33.3)
Tumor stage
T2	62 (93.9)
T3	4 (6.1)
Lymph node status
Negative	22 (33.3)
Positive	38 (57.6)
Stage
IIA	21 (31.8)
IIB	36 (54.5)
IIIA	9 (13.6)
Grade
Ⅰ/Ⅱ	46 (70.8)
Ⅲ	19 (29.2)
HER2 status
IHC 1+	17 (25.8)
IHC 2+	49 (74.2)
ER positivity
<50%	3 (4.5)
≥50%	63 (95.5)
PR positivity
<50%	18 (27.3)
≥50%	48 (72.7)
Ki-67
<30%	14 (21.2)
≥30%	52 (78.8)
TILs
<10%	47 (72.3)
≥10%	18 (27.7)

Data are median (IQR) or n (%). IHC, immunohistochemistry; TILs, tumor-infiltrating lymphocytes.

Efficacy

Per the study protocol’s Simon two-stage design, 35 patients were enrolled in the first stage, with 26 achieving an objective response, exceeding the prespecified threshold of 18 required for study continuation. The ORR was 81.5% (53/65; 95% CI, 70.0–90.1) in the efficacy analysis set (n = 65), consisting of one complete response (1.5%) and 52 partial responses (80.0%). (Fig. 2 and Table 2). Pathological complete response was observed in 2 patients (3.1%; 95% CI, 0.4–10.8), while 6 patients (9.2%; 95% CI, 3.5–19.6) achieved a residual cancer burden (RCB) class 0–1. In a post-hoc analysis, the breast-conserving surgery rate was 32.3% (21/65; 95% CI, 21.4–45.0) (Table 2).

Fig. 2. Waterfall plot for the best change in breast target lesions from baseline.

Dashed lines at 20% and –30% denote thresholds for progressive disease and partial response, respectively, according to the Response Evaluation Criteria in Solid Tumors, version 1.1. * Two patients achieved pathological complete response (ypT0/is ypN0). Source data are provided as a Source Data file.

Table 2.

Clinical and pathological responses and surgical approaches in the efficacy analysis population

	Efficacy analysis population (N = 65)
Best response
Complete response	1 (1.5%)
Partial response	52 (80.0%)
Stable disease	12 (18.5%)
Progression disease	0 (0.0%)
Objective response	53 (81.5%, 70.0–90.1)
Residual cancer burden
0	2 (3.1%)
1	4 (6.2%)
2	33 (50.8%)
3	23 (35.4%)
No record	3 (4.6%)
Residual cancer burden (0-1) rate	6 (9.2%, 3.5–19.0)
Breast surgery
Breast-conserving surgery	21 (32.3%)
Mastectomy*	41 (63.1%)
No surgery	3 (4.6%)

Data are n (%) or n (%; 95% CI), unless otherwise stated. CI, confidence interval. *Mastectomy includes immediate breast reconstruction surgery.

Post-hoc subgroup analyses of ORRs are presented in Table 3. Patients with HER2 IHC 1+ tumors demonstrated an ORR of 94.1% (16/17; 95% CI, 71.3–99.9), while those with IHC 2+ expression showed an ORR of 77.1% (37/48; 95% CI, 62.7–88.0). Among 53 evaluable patients with paired baseline and surgical tumor samples, Ki67 expression decreased by a mean of 35.6% (95% CI, 30.4–40.7; P = 4.46 × 10⁻¹⁹) following treatment (Fig. 3).

Table 3.

Subgroup analysis of objective response rates

	Objective response, No./Total No.	Objective response, %(95%CI)
Overall	53/65	81.5 (70.0–90.1)
Age, years
≤50	35/39	89.7 (75.6–96.6)
>50	18/26	69.2 (48.2–85.4)
Menopausal status
Premenopausal	39/43	90.7 (77.9–97.4)
Postmenopausal	14/22	63.6 (40.7–82.8)
Tumor stage
T2	49/61	80.3 (68.2–89.4)
T3	4/4	100.0 (39.8–100.0)
Lymph node status
Negative	17/21	81.0 (58.1–94.6)
Positive	36/44	81.8 (68.5–90.9)
Stage
IIA	16/20	80.0 (56.3–94.3)
IIB	28/36	77.8 (60.9–89.9)
IIIA	9/9	100.0 (66.4–100.0)
Grade
Ⅰ/Ⅱ	36/46	78.3 (63.6–88.9)
Ⅲ	17/19	89.5 (66.9–98.7)
HER2 status
IHC 1+	16/17	94.1 (71.3–99.9)
IHC 2+	37/48	77.1 (62.7–88.0)
ER positivity
<50%	3/3	100.0 (29.2–100.0)
≥50%	50/62	80.7 (68.6–89.6)
PR positivity
<50%	16/18	88.9 (65.3--98.4)
≥50%	37/47	78.7 (64.3--89.2)
Ki-67 level at baseline
<30%	10/14	71.4 (41.9–91.6)
≥30%	43/51	84.3 (71.4–93.0)
TILs
<10%	37/47	78.7 (64.3–89.2)
≥10%	16/18	88.9 (65.3–98.4)

CI, confidence interval; IHC, immunohistochemistry; TILs, tumor-infiltrating lymphocytes.

Fig. 3. Changes in Ki-67 expression levels from baseline to surgery.

A Paired Ki-67 scores at baseline and at surgery for 53 patients. Lines connect pre- and post-treatment values from the same individual. (B) Box plot of Ki-67 scores at baseline and surgery (n = 53). Center line represents the median; box bounds indicate the 25th and 75th percentiles; whiskers represent minima and maxima. A two-sided paired t-test was used for statistical analysis. The exact P value = 4.46 × 10⁻¹⁹. Source data are provided as a Source Data file.

In the post-hoc analysis, the radiographic assessment after 2 cycles of chemotherapy did not reach a statistically significant association with postoperative RCB (P = 0.138). In contrast, assessments after 4 (P= 0.012), 6 (P = 0.004), and 8 (P = 0.022) cycles were significantly associated with RCB categories (Supplementary Fig. S1). No statistically significant difference in ORR was observed among the 35 PAM50-subtyped cases (P = 0.570). Notably, an ORR was achieved in all 6 cases of the Normal-like subtype (Supplementary Fig. S2).

Safety

In the safety population of 66 patients, 97.0% experienced at least one treatment-related adverse event. The most frequently reported events were nausea (89.4%), asthenia (81.8%), and vomiting (75.8%). Grade ≥3 adverse events occurred in 37.9% of patients, most commonly neutropenia (27.3%), leukopenia (16.7%), and anemia (13.6%) (Table 4). Among the initial cohort of 22 patients who did not receive primary pegylated recombinant human granulocyte colony-stimulating factor (PEG-rhG-CSF) prophylaxis, grade ≥3 hematologic toxicities were prevalent, with neutropenia in 77.3%, anemia in 27.3%, and thrombocytopenia in 18.2%. In contrast, among 44 patients receiving PEG-rhG-CSF, the incidence of severe hematologic toxicities was markedly reduced, with only 2.3% experiencing neutropenia, 6.8% with anemia, and no cases of thrombocytopenia.

Table 4.

Adverse events

Event, No. (%)	N = 66
	Any Grade	Grade 1	Grade 2	Grade 3	Grade 4
All	64 (97.0)	13 (19.7)	26 (39.4)	12 (18.2)	13 (19.7)
Nausea	59 (89.4)	58 (87.9)	1 (1.5)	0	0
Asthenia	54 (81.8)	54 (81.8)	0	0	0
Vomiting	50 (75.8)	32 (48.5)	16 (24.2)	2 (3.0)	0
Diarrhea	36 (54.5)	21 (31.8)	10 (15.2)	5 (7.6)	0
Anemia	28 (42.4)	8 (12.1)	11 (16.7)	9 (13.6)	0
Leukopenia	27 (40.9)	1 (1.5)	15 (22.7)	7 (10.6)	4 (6.1)
Neutropenia	26 (39.4)	0	8 (12.1)	5 (7.6)	13 (19.7)
Weight decreased	25 (37.9)	12 (18.2)	13 (19.7)	0	0
Alanine aminotransferase increased	22 (33.3)	18 (27.3)	4 (6.1)	0	0
Aspartate aminotransferase increased	19 (28.8)	18 (27.3)	1 (1.5)	0	0
Hepatic function abnormal	19 (28.8)	15 (22.7)	3 (4.5)	1 (1.5)	0
Constipation	16 (24.2)	12 (18.2)	4 (6.1)	0	0
Platelet count decreased	13 (19.7)	6 (9.1)	3 (4.5)	2 (3.0)	2 (3.0)
Alopecia	12(18.2)	8(12.1)	4(6.1%)	0	0
Rash	9 (13.6)	9 (13.6)	0	0	0
Gamma-glutamyl transferase increased	8 (12.1)	3 (4.5)	5 (7.6)	0	0
Lymphocyte count decreased	7 (10.6)	2 (3.0)	2 (3.0)	2 (3.0)	1 (1.5)
Abdominal pain	7 (10.6)	7 (10.6)	0	0	0
Stomatitis	6 (9.1)	5 (7.6)	1 (1.5)	0	0
Abdominal distension	6 (9.1)	5 (7.6)	1 (1.5)	0	0
Headache	6 (9.1)	6 (9.1)	0	0	0
Blood alkaline phosphatase increased	4 (6.1)	4 (6.1)	0	0	0
Dizziness	4 (6.1)	4 (6.1)	0	0	0
Influenza like illness	3 (4.5)	3 (4.5)	0	0	0
Flatulence	3 (4.5)	2 (3.0)	1 (1.5)	0	0
Hypoalbuminaemia	3 (4.5)	1 (1.5)	2 (3.0)	0	0
Hypokalaemia	3 (4.5)	1 (1.5)	1 (1.5)	1 (1.5)	0
Palpitations	3 (4.5)	3 (4.5)	0	0	0
Decreased appetite	3 (4.5)	3 (4.5)	0	0	0
Pyrexia	2 (3.0)	0	2 (3.0)	0	0
Taste disorder	2 (3.0)	1 (1.5)	1 (1.5)	0	0
Myalgia	2 (3.0)	2 (3.0)	0	0	0
Platelet count increased	2 (3.0)	2 (3.0)	0	0	0
Anal fissure	2 (3.0)	2 (3.0)	0	0	0
Abdominal discomfort	2 (3.0)	1 (1.5)	1 (1.5)	0	0

Data are all grade 1 or grade 2 adverse events (n [%]) that occurred in ≥3% of patients and all grade 3 and grade 4 adverse events. Patients might have had more than one adverse event. One patient died from a non-treatment-related accidental cause.

Serious adverse events were observed in 9 patients (13.6%), with 8 cases considered treatment-related (12.1%). The predominant toxicities were hematologic (9.1%) and gastrointestinal (3.0%). There was one death (1.5%), which was deemed unrelated to SHR-A1811. Dose reductions were required in 11 patients (16.7%), mainly due to hematologic (6.1%) and gastrointestinal (10.6%) adverse events. Treatment discontinuation occurred in 7 patients (10.6%), primarily driven by adverse events (6.1%, including hematologic and gastrointestinal toxicities), voluntary withdrawal (3.0%), and one unrelated death. The patient experienced acute viral fever at home during the intertreatment period after the second cycle of therapy and suffered rapid clinical deterioration, passing away on the way to the hospital. No formal laboratory confirmatory tests were performed due to the acute death. The study’s independent Adverse Event Adjudication Committee determined this death was attributable to severe acute viral infection with rapid progression, with no causal relationship to SHR-A1811 treatment) (1.5%).

Discussion

To the best of our knowledge, the HELEN-015 study is the first to explore the third-generation HER2-directed ADC SHR-A1811 as a neoadjuvant therapy for HR+/HER2-low breast cancer, demonstrating an ORR of 81.5% and a manageable toxicity profile. Importantly, there were no cases of ILD, and neutropenia was effectively managed through primary prophylaxis for myelosuppression. While survival data require longer follow-up, this study provides a rationale for future large-scale randomized controlled trials.

For most HR+/HER2− breast cancer patients, chemotherapy remains the preferred neoadjuvant treatment. A meta-analysis indicated that anthracycline- and taxane-based neoadjuvant chemotherapy achieved an ORR of 52% (range: 47–66%) based on imaging assessments². Notably, our study excluded HER2-zero patients, as HER2-low tumors are known to exhibit lower chemosensitivity compared to HER2-zero counterparts. In this context, the observed ORR of 81.5% in HR+/HER2-low early breast cancer is noteworthy. Although the pCR rate was relatively low at 3.1%, this aligns with the intrinsic biology of HR+/HER2- tumors and corroborates previous findings^12–14. Select HR+/HER2- populations, such as those with histological grade 3 tumors or estrogen receptor expression below 10%, can achieve pCR rates of around 15% following neoadjuvant chemotherapy. The incorporation of immunotherapy may further elevate pCR rates above 20% in these patients^15,16. However, it is clear that pCR may not be an appropriate endpoint for assessing treatment efficacy in all HR+/HER2- breast cancer patients, particularly within the HER2-low subgroup. Future investigations into the use of SHR-A1811, either alone or in combination with immunotherapy, could provide valuable insights for this selective HR+/HER2- population. Furthermore, our study recorded a breast-conserving surgery rate of 32.3%, significantly surpassing the national average of 22% in China¹⁷, which is especially noteworthy given that we focused on the HR+/HER2-low subgroup, typically associated with poorer responses to neoadjuvant therapy.

Neoadjuvant endocrine therapy continues to be actively investigated in combination strategies, primarily due to its favorable safety profile^18–24. However, the TALENT trial demonstrated that adding endocrine therapy to T-DXd did not provide additional benefit in patients with HR+/HER2-low breast cancer, with ORRs of 58% compared to 68% observed with T-DXd monotherapy²⁵. Notably, SHR-A1811 possesses distinct pharmacological attributes relative to T-DXd, including structural modifications, an optimized drug-to-antibody ratio, and enhanced payload characteristics such as improved membrane permeability¹¹. Our data indicate that SHR-A1811 monotherapy achieved an ORR of 81.5%, compared to 68% reported in the TALENT study for T-DXd. However, it is important to note that the TALENT data represent only a subset of the enrolled patient population, and the complete dataset has not yet been published. Therefore, cross-trial comparisons should be approached with caution, as differences in patient selection, treatment protocols, and assessment criteria may influence outcomes. Nonetheless, the encouraging activity observed with SHR-A1811 in HR+/HER2-low breast cancer underscores the necessity for further preclinical and clinical investigations to evaluate its potential synergistic effects when combined with endocrine therapy. Ideally, such studies should be conducted through prospective, randomized controlled trials involving well-matched patient cohorts.

The pharmacodynamic effects of SHR-A1811 were evidenced by a significant mean reduction of 35.0% in Ki-67 expression, indicating potent antiproliferative activity²⁶. The observed dose-response relationship in ORR across treatment cycles (increasing from 43.1% at cycle 2 to 81.5% at cycle 8) and the significant association between radiographic assessments at cycles 4–8 with RCB categories suggest that sustained treatment exposure is essential for achieving deeper pathological responses. These findings align with the mechanism of ADCs and highlight the importance of adequate treatment duration for optimal tumor response²⁷.

Subgroup analyses revealed no positive correlation between HER2 immunohistochemistry status and response to SHR-A1811. Objective response rates were 94.1% in HER2 1+ and 77.1% in HER2 2+ patients. This aligns with findings from DESTINY Breast04, where T-DXd showed comparable progression-free survival, at 10.3 versus 10.1 months, across HER2 low subgroups⁹. These observations challenge the conventional HER2 expression response paradigm, suggesting that antibody drug conjugate efficacy in HER2-low disease may depend more on payload properties such as bystander effect and membrane permeability than on absolute HER2 levels²⁸. The current immunohistochemistry classification system appears insufficient for predicting antibody drug conjugate response, underscoring the need for novel biomarkers, including HER2 trafficking dynamics or tumor microenvironment features, to optimize patient selection^29,30.

The safety data for SHR-A1811 were consistent with prior reports of anti-HER2 antibody drug conjugates^11,31. A critical intervention in this study was the use of primary PEG-rhG-CSF prophylaxis in 44 patients, which dramatically reduced both the incidence and severity of neutropenia. Grade ≥3 adverse events decreased from 77.3% to 2.3%. A reduction was also observed in the occurrence of anemia and thrombocytopenia. Importantly, no cases of ILD or treatment-related deaths were observed. SHR-A1811 exhibits a unique toxicity profile likely attributable to its proprietary linker-payload chemistry. Available evidence suggests that the reduced risk of ILD may be associated with structural optimizations that decrease pulmonary payload release, while the increased myelosuppression could be related to the same payload’s enhanced membrane permeability and potent activity against dividing bone marrow cells³². Although the incidence of diarrhea was relatively high, most cases were low-grade and manageable with standard supportive care. While these preliminary results suggest a more favorable safety profile for SHR-A1811, longer follow-up and larger studies are needed to fully assess its long-term risks.

Several limitations of this study warrant consideration. First, the single-arm design precludes definitive efficacy comparisons between SHR-A1811 and established standard therapies. Second, the modest sample size may affect the generalizability of these findings. To address these limitations, we are initiating a larger randomized controlled trial to validate these results and determine the optimal role of SHR-A1811 in the neoadjuvant management of HR+/HER2-low breast cancer. Finally, a longer follow-up is required to fully assess survival outcomes and treatment durability.

In conclusion, this phase 2 trial demonstrates that SHR-A1811 exhibits promising efficacy and a favorable safety profile as a neoadjuvant therapy for patients with HR+/HER2-low breast cancer. These preliminary findings should be confirmed through rigorously designed randomized controlled trials, which should also investigate potential combination strategies. Additional insights into the long-term therapeutic benefits of SHR-A1811 will be provided through ongoing follow-up analyses from this cohort.

Methods

Study oversight

The HELEN-015 study is a single-arm, phase II exploratory trial conducted at Henan Cancer Hospital in Zhengzhou, China. The trial was designed and conducted in accordance with the Declaration of Helsinki and the Guidelines for Good Clinical Practice, and received approval from the Ethics Committee of Henan Cancer Hospital (ID: 2023-198-001). All participants provided written informed consent, and the study protocol is available in the supplementary materials.

Patients

Eligible participants were treatment-naïve female patients aged 18–70 years. All patients had histologically confirmed HR+/HER2-low invasive breast cancer with clinical stage cT2-3 and cN0-3. Sex was determined based on official identity documentation. HER2-low status is defined as IHC results of 1+ or IHC 2+ without ERBB2 amplification on in situ hybridization (ISH), centrally confirmed, while HR positivity is determined by estrogen receptor (ER) and/or progesterone receptor (PR) expression of ≥1%. HER2 IHC testing was performed at Henan Cancer Hospital using the clinical-grade Ventana PATHWAY 4B5 assay (Roche, Switzerland). For cases with HER2 IHC (2+), reflex ISH testing was conducted using APPBIO probes (Guangzhou Anbiping Biotechnology Co., Ltd., China). Participants are required to have an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1. Key inclusion criteria also mandate adequate hematologic function, characterized by an absolute neutrophil count (ANC) of ≥1.5 × 10⁹ cells/L, hemoglobin (Hb) levels of ≥90 g/L, and platelets ≥100 × 10⁹ cells/L; hepatic function defined by total bilirubin, aspartate aminotransferase (AST), and alanine aminotransferase (ALT) levels ≤1.5 times the upper limit of normal (ULN); renal function with a creatinine clearance of ≥60 mL/min; and cardiac function measured by a left ventricular ejection fraction (LVEF) of ≥50% via echocardiography.

Exclusion criteria included bilateral or metastatic disease, prior anticancer therapies, active secondary malignancies (except for non-melanoma skin cancers), known drug contraindications, history of clinically significant pulmonary diseases (including but not limited to interstitial pneumonia, pneumonia, pulmonary fibrosis, and radiation pneumonitis), and pregnancy status confirmed by β-human chorionic gonadotropin (β-hCG) testing. Full eligibility criteria are listed in the Supplementary Information.

Trial design and treatment

Eligible patients in the HELEN-015 trial received SHR-A1811 as neoadjuvant therapy. The drug was administered intravenously at a dose of 6.4 mg/kg on day 1, with the initial infusion lasting 90 ± 10 min. If no infusion-related adverse reactions occurred, subsequent infusions were shortened to approximately 30 min. Treatment was scheduled every three weeks for a total of eight cycles, amounting to an overall treatment duration of 24 weeks.

Adverse events were monitored at each treatment and follow-up visit. In instances of severe toxicities, discontinuation and dose reductions were mandated, with adjustments made in increments of 0.8 mg/kg. If a participant remained intolerant to treatment following two consecutive dose reductions, the investigator would evaluate the risk-to-benefit ratio to determine whether to further reduce the dose or discontinue treatment entirely. Neoadjuvant therapy could also be discontinued in cases of disease progression, withdrawal of informed consent, or study termination, whichever occurred first. Delays in SHR-A1811 administration were not permitted to exceed 42 days (equivalent to two cycles). For patients who discontinued treatment due to disease progression, alternative neoadjuvant therapies or surgical options were determined based on the investigator’s judgment. Primary prophylaxis for gastrointestinal toxicities was recommended as part of the treatment plan. After the enrollment of 22 patients, PEG-rhG-CSF was administered as a primary prophylactic measure to mitigate myelosuppression and improve treatment tolerance.

During the screening phase, primary tumors and regional lymph nodes were evaluated using contrast-enhanced magnetic resonance imaging (MRI), ultrasound, mammography, and physical examination. Subsequently, ultrasound and physical exams were conducted prior to each treatment cycle, while contrast-enhanced MRI was performed every two cycles. Preoperative assessments employed the same four modalities for a comprehensive evaluation. Treatment efficacy was primarily determined by contrast-enhanced MRI results, in accordance with the Response Evaluation Criteria in Solid Tumors (RECIST 1.1)³³. Additional investigations, including chest radiographs, liver imaging, and bone scans, were conducted as clinically indicated to rule out metastatic disease. Routine laboratory assessments were performed at baseline, before each treatment cycle, and prior to surgery. Radiological and pathological assessments were performed by independent radiologists and pathologists, respectively, with blinding to patient treatment and outcome data to ensure objectivity.

Surgery was scheduled to occur 3–6 weeks after the final dose of treatment. During the adjuvant phase, the decision to administer adjuvant chemotherapy was determined by the investigators. Radiotherapy and endocrine therapy were provided in accordance with established clinical treatment guidelines. A five-year follow-up protocol was implemented to collect data on disease recurrence and disease-free survival rates, with follow-up visits scheduled every three months during the first year and every six months for the subsequent four years.

Assessments

The ORR was evaluated by an independent radiologist using contrast-enhanced MRI, following the RECIST 1.1 criteria³³. For patients who did not undergo surgery, the last evaluable assessment was used for analysis. Upon completion of neoadjuvant therapy, the pCR was determined by an independent pathologist according to the American Joint Committee on Cancer (AJCC) staging criteria (ypT0/Tis ypN0)³⁴. Patients who did not undergo surgery were classified as not achieving pCR. Event-free survival (EFS) was defined as the time from enrollment to the first preoperative disease progression, postoperative recurrence, or death from any cause. Disease-free survival (DFS) was defined as the time from surgery to the first postoperative recurrence or death from any cause. Distant disease-free survival (DDFS) was measured from surgery to the first occurrence of a distant recurrence. Safety was assessed in accordance with National Cancer Institute Common Terminology Criteria for Adverse Events (NCI CTCAE) v5.0³⁵, covering the incidence and severity of all adverse events.

Endpoints

The primary endpoint was ORR, defined as the proportion of patients achieving complete response or partial response after neoadjuvant therapy but prior to definitive surgery in the efficacy analysis (all randomly assigned patients receiving ≥1 treatment dose with ≥1 evaluable post-baseline contrast-enhanced MRI) population. Secondary endpoints included pCR (ypT0/is ypN0, indicating no residual invasive carcinoma in breast or sampled axillary nodes), RCB 0-1 rate, EFS, DFS and DDFS in the efficacy analysis population. Safety analyses were done in all patients receiving at least one dose of study treatment. As the survival data remain immature, these time-to-event endpoints will be reported after extended follow-up.

Statistical analyses

This phase 2 trial employed a minimax Simon’s two-stage optimal design to evaluate treatment efficacy. Based on historical data³⁶, we established a null hypothesis ORR of 50% versus an alternative hypothesis ORR of 66%, with a one-sided alpha level of 0.05 and 80% statistical power. The study design required evaluation of 35 patients in the first stage, with progression to full accrual contingent upon observing 18 or more objective responses among these initial participants. The trial would demonstrate final treatment efficacy if 37 or more responses were observed among the planned 60 evaluable patients. To account for a potential 10% dropout, the total target enrollment was set at 66 participants.

Continuous variables were summarized as medians with IQR, and categorical variables as frequencies with percentages. The primary and secondary efficacy endpoints (ORR, pCR, and RCB 0-1 rates) were analyzed using the Clopper-Pearson method to calculate exact 95% CIs.

In post-hoc analyses, breast-conserving surgery rates were evaluated with exact 95% CIs calculated using the Clopper-Pearson method. Chi-square tests were employed to evaluate the differences in ORR across PAM50 subtypes. The correlation between the degree of radiological regression and RCB categories was assessed using Spearman’s rank correlation coefficient. The difference in the mean percentage of Ki-67 expressing tumor cells between baseline and surgical samples was analyzed using the paired t-test. All Statistical analyses were conducted using SPSS software (version 27.0). All statistical tests were two-sided with α = 0.05.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Author contributions

All authors had access to the data used to prepare the manuscript and participated in the writing, critical review and editing of the manuscript. The first draft was written by the first and senior author. D.J., J.Z., X.S., Z.D.L., J.W., M.J., Y.W., and Z.Z.L. conceptualized, designed and planned the study. X.S., Z.D.L., J.Q., X.C., M.Y., C.Z., L.L., and Z.Z.L. did the acquisition of the data. D.J., J.Z., X.S., Z.D.L. and Z.Z.L. analyzed the data. D.J., J.Z., and Z.Z.L. interpreted the results. D.J., J.Z. and Z.Z.L. drafted the manuscript. D.J., J.Z., X.S., and Z.Z.L. did the reviewing or revising of the manuscript for important intellectual content. X.S., Z.D.L., J.Q., X.C., M.Y., C.Z., L.L., and Z.Z.L. provided the study materials/patients. All authors decided to submit the manuscript for publication. All authors approved the final draft of the manuscript.

Peer review

Peer review information

Nature Communications thanks Christos Vaklavas and other anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.

Funding

This work was supported by an unrestricted research grant and study drug supply from Jiangsu Hengrui Pharmaceuticals. Additional funding was provided by the Training Program for Young and Middle-aged Health and Technology Innovation Leaders in Henan Province (Grant No. YXKC2022025), the National Natural Science Foundation of China (Grant No. 82372712), and the Science and Technology Development Research Center of the National Health Commission of China (Grant No. WKZX2023CX070001).

Data availability

The patient-level data generated in this study have been deposited in Figshare https://doi.org/10.6084/m9.figshare.29068586³⁷. Restricted access to the dataset is available solely for academic research due to patient privacy regulations. Researchers who submit data access requests to the corresponding author will receive a formal reply within 14 working days after application submission. Once data access is approved, the authorized dataset will remain accessible to the requester for a continuous period of 12 months. The processed study data are provided in the Source Data file. Source data are provided with this paper.

Code availability

No custom code or mathematical algorithm was used in the HELEN-015 study.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s41467-026-77041-x.