Pathologic Complete Response to Neoadjuvant Chemotherapy in HER2-Low Versus HER2-Zero Breast Cancer: A Retrospective Cohort Study

Introduction

Breast cancer remains the most prevalent malignancy among women worldwide, with approximately 2.3 million new cases diagnosed annually [1-3]. Although incidence increases with age, nearly 7% of cases occur in women younger than 40 years, underscoring the biological and clinical heterogeneity of the disease across age groups [4]. Despite major advances in screening, systemic therapy, and multidisciplinary management, breast cancer continues to account for substantial morbidity and mortality globally, particularly in countries undergoing economic and demographic transition [5-7].

Contemporary breast cancer management relies on molecular profiling, especially the status of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2). These markers are essential for current subtype classification and guiding therapy decisions [4, 8, 9]. HER2positive tumors, characterized by HER2 protein overexpression or gene amplification (IHC 3+ or IHC 2+/ISH+), account for about 20–30% of cases and are known for their aggressive nature. Yet, prognosis has improved significantly with HER2-targeted therapies like trastuzumab and pertuzumab [10-15]. Conversely, roughly 80% of breast cancers lack HER2 amplification and are classified as HER2-negative, thus limiting the use of HER2targeted treatments in these cases [16].

Recent refinements in HER2 assessment have led to the recognition of a HER2-low subgroup, defined as tumors with immunohistochemistry (IHC) 1+ or IHC 2+ staining in the absence of ERBB2 gene amplification [16-22]. HER2-low tumors account for nearly half of breast cancers previously classified as HER2-negative and have gained clinical relevance following the demonstrated efficacy of antibody-drug conjugates, particularly trastuzumab deruxtecan, in metastatic disease [19]. Beyond therapeutic implications, accumulating evidence suggests that HER2-low tumors may represent a biologically distinct subset rather than a purely technical classification.

HER2-low breast cancers are consistently more likely to be hormone receptor-positive and tend to fall into luminal molecular subtypes. They also show lower proliferative activity and reduced Ki-67 expression compared to HER2-zero tumors [16, 29, 35-

37]. These biological traits are known to provide some resistance to cytotoxic chemotherapy and correlate with lower rates of pathological complete response (pCR) in neoadjuvant treatments [32, 39-41]. Consequently, HER2-low status might indicate a tumor phenotype driven mainly by hormone signaling, with decreased chemosensitivity and unique therapeutic vulnerabilities, regardless of eligibility for HER2-targeted therapies.

Neoadjuvant chemotherapy (NACT) is frequently used to treat patients with locally advanced and high-risk early-stage breast cancer, aiming to shrink tumors and assess the treatment response in vivo [22, 23]. A complete response at the pathological level after NACT is strongly associated with improved long-term outcomes, especially for aggressive subtypes like triple-negative breast cancer [24]. However, dependable markers for chemotherapy response in HER2-negative breast cancer are limited, and the role of low HER2 expression in affecting chemosensitivity remains uncertain [25].

Previous research on how HER2-low status influences the response to neoadjuvant chemotherapy and survival has produced mixed results. Some studies report notably lower pCR rates in HER2-low tumors compared to HER2-zero tumors, while others find no significant differences in response or long-term outcomes [17, 26]. Many of these analyses are hampered by small sample sizes, diverse patient groups, inconsistent HER2 testing methods, or limited subgroup analysis, which fuels ongoing uncertainty about the predictive and prognostic significance of HER2-low expression in early-stage disease [16, 20].

Against this background, the present study aimed to evaluate whether HER2-low expression, compared with HER2-zero status, is associated with pathological complete response to neoadjuvant chemotherapy in patients with stage II–III HER2-negative breast cancer. Secondary objectives included comparison of disease-free survival (DFS) and overall survival (OS) between these groups, with exploratory analyses within the HER2-low subgroup. By anchoring HER2-low expression within a biologically plausible framework of hormone receptor enrichment and reduced proliferation, this study seeks to clarify its role as a predictive construct in the neoadjuvant setting and to inform future strategies for treatment stratification in HER2-negative breast cancer.

Methods

Study Design and Setting

This was a retrospective cohort study utilizing clinical and pathological data from patients treated at the Breast Cancer Research Center of the Motahari Clinic between 2016 and 2021, with longitudinal follow-up for survival outcomes. This study aimed to evaluate the influence of HER2 expression on the pathological response to neoadjuvant chemotherapy and subsequent survival outcomes in patients with HER2-negative breast cancer.

Study Population and Sample Size

The study population consisted of female patients aged 18–75 years with histologically confirmed invasive breast carcinoma who received neoadjuvant chemotherapy followed by definitive breast surgery at the Breast Cancer Research Center of the Motahari Clinic, Shiraz, Iran, between January 2016 and December 2021. This retrospective cohort study with longitudinal follow-up was designed to evaluate pathological response and survival outcomes.

Patients were eligible if they had stage II–III HER2-negative breast cancer, defined according to contemporaneous ASCO/CAP guidelines based on immunohistochemistry (IHC) and in situ hybridization (ISH) testing [27, 28]. Tumors were classified as HER2-zero (IHC score 0) or HER2-low (IHC 1+ or IHC 2+ with negative ISH), consistent with emerging consensus definitions within the HER2-negative spectrum [16, 29, 38]. HER2 assessment was performed locally at the time of diagnosis, without centralized pathology review. Since identifying HER2-zero versus HER2-low tumors can vary between observers, especially at low expression levels, there is a potential for misclassification bias that cannot be entirely ruled out [29, 38].

Prespecified exclusion criteria were consistently applied to all potentially eligible patients. Patients were excluded if they had inflammatory breast cancer, bilateral synchronous breast cancer, or a history of invasive malignancy (except non-melanoma skin cancer or in situ cervical carcinoma) due to their unique biology, treatment strategies, and prognosis. Additionally, patients who had prior systemic therapy for breast cancer, did not complete the planned neoadjuvant chemotherapy, or did not undergo definitive breast surgery were excluded because pathological complete response could not be reliably assessed in these cases. Patients with missing, indeterminate, or unavailable HER2 or hormone receptor (ER/PR) status were also excluded from the primary analysis.

Only patients with comprehensive clinicopathologic data and available surgical pathology after neoadjuvant chemotherapy were included in the final analysis. Pathological complete response was evaluated solely in patients who underwent surgery after neoadjuvant treatment.

This study’s sample size comprises all eligible patients treated during the study period with complete data, without a formal prestudy size calculation. The final count of 934 patients represents the entire source population meeting eligibility criteria within the specified timeframe, enhancing statistical power and reducing selection bias typical in retrospective cohorts.

Patients received standard neoadjuvant chemotherapy following institutional protocols and current international guidelines for early-stage and locally advanced breast cancer [22, 23]. The neoadjuvant treatment mainly involved anthracycline- and taxane-based regimens, either sequentially or combined, which is the standard approach for HER2-negative breast cancer [8, 22]. Typical regimens included doxorubicin plus cyclophosphamide followed by a taxane, or upfront taxane-based treatments, with the specific choice based on tumor features, patient performance status, and physician judgment.

Chemotherapy was administered at standard doses and schedules. Dose-dense regimens were used selectively based on clinical judgment and patient tolerance. The duration of neoadjuvant chemotherapy typically ranged from 4 to 8 cycles, depending on the regimen and clinical response. No patients received HER2targeted therapy, consistent with HER2-negative disease status.

Treatment selection was not based on HER2-low versus HER2zero status, since HER2-low did not affect neoadjuvant treatment choices during the study period. As a result, there was some variation in chemotherapy regimens across the cohort; however, all patients received contemporary cytotoxic chemotherapy regarded as standard care for HER2-negative breast cancer. Considering this real-world treatment setting, the type of chemotherapy regimen was not used as a stratification variable in the main analyses.

Data Collection

Data, including demographic information (age and menopausal status), tumor characteristics (size, histological type, grade, and hormone receptor status), treatment details (neoadjuvant chemotherapy regimens and radiotherapy), and pathological outcomes, were extracted from the center’s electronic and paper medical records. HER2 expression was assessed by experienced pathologists according to contemporaneous ASCO/CAP criteria.. Pathological complete response (pCR) was defined as the absence of invasive cancer in both the breast tissue and lymph nodes following neoadjuvant chemotherapy (ypT0/Tis ypN0), allowing for the presence of ductal carcinoma in situ. Radiotherapy data included chest wall, axillary, and intraoperative radiotherapy.

Outcome Measures

Pathological complete response (pCR) was defined as the absence of residual invasive disease in the breast and axillary lymph nodes at surgery following neoadjuvant chemotherapy (ypT0/Tis ypN0).

For survival analyses, disease-free survival (DFS) and overall survival (OS) were calculated from the date of histopathologic diagnosis. DFS was defined as the time to first documented local, regional, or distant recurrence, or death from any cause. OS was defined as the time to death from any cause. Recurrence and survival events were ascertained through institutional medical records. Patients without events were censored at the date of last clinical follow-up.

Statistical Analysis

Descriptive statistics summarized clinicopathologic features. Continuous variables were compared with independent-sample t tests or ANOVA, while categorical variables were analyzed using chi-square or Fisher’s exact tests, as appropriate.

Pathological complete response (pCR) was assessed through both univariable and multivariable logistic regression models. A priori covariates included age at diagnosis, hormone receptor status (ER/PR), menopausal status, tumor size, histologic grade, lymphovascular invasion, and HER2 status (HER2-low vs HER2zero), based on known clinical relevance and prior literature. Chemotherapy regimen type was not included as a covariate due to heterogeneity and lack of protocolized assignment. Interaction terms between HER2 status and hormone receptor status were explored in prespecified analyses. Only complete cases were included in regression analyses.

Survival outcomes were evaluated with the Kaplan–Meier method and compared using log-rank and Wilcoxon tests. No multivariable Cox models were conducted. Subgroup analyses, such as those based on menopausal status, were planned as exploratory. Due to their exploratory purpose and the small number of main hypotheses, no formal adjustment for multiple comparisons was made; hence, subgroup results were interpreted cautiously.

Subgroup analyses were prespecified as exploratory and interpreted descriptively.

All statistical analyses were performed using IBM SPSS Statistics, version 26, with a two-sided significance threshold of p<0.05.

Ethical Considerations

The study protocol was approved by the Institutional Review

Board of the Motahari Clinic (Approval number IR.SUM.MED. REC.1403.600). Given the retrospective nature and anonymization of the patient data, the need for informed consent was waived. All procedures complied with the Declaration of Helsinki and local data protection regulations.

Results

Patient Demographics and Tumor Characteristics:

The overall cohort consisted of 934 patients with invasive breast cancer who received neoadjuvant chemotherapy. The mean age at diagnosis was 46.81 years (SD: 10.88), with a 95% confidence interval of 46.02 to 47.60 years . When stratified by HER2 status, patients with HER2 1+/2+ tumors were slightly older (mean age 47.36 years, SD: 11.01) than those with HER2-Zero tumors (mean age: 45.10 years, SD: 10.61). This age difference was not statistically significant (ANOVA, p= 0.077). The average tumor size in the entire population was 2.34 cm (SD: 1.97). Patients with HER2 1+/2+ had a slightly larger mean tumor size of 2.40 cm compared to 2.24 cm in HER2-negative patients, but this difference was not statistically significant (p=0.245). Most tumors were invasive ductal carcinomas: 81.2% in HER2-zero and 85.6% in HER2-low groups (p=0.078). Grade 2 was most common in both HER2-zero (45%) and HER2-low (46.7%) groups. HER2low tumors exhibited higher rates of vascular and perineural invasion (p=0.006) and a greater proportion of in situ components (40.3% vs. 33.8%, p=0.045). There was no significant difference in tumor necrosis presence between the groups (p=0.573) (Table 1). 

Table 1: Clinicopathological features of patients stratified by HER2-zero(Negative) and HER2-low breast cancer treated with neoadjuvant chemotherapy.

Figure 1: Pathologic complete response rates according to HER2 status following neoadjuvant chemotherapy.

Histopathologic Characteristics across HER2 Groups 

The distribution of tumor grades did not differ significantly between the HER2-zero and HER2-low groups (χ²=2.190, df=3, p=0.534). Grade 2 was the most common in both groups, with 45.0% in HER2-zero and 46.7% in HER2-low. Logistic regression, using grade 1 as the reference, showed no significant differences in the odds of higher grades in either group: for HER2-zero, the odds ratios for grades 2 and 3 were 0.523 (p=0.311) and 0.349 (p=0.059), respectively; for HER2-low, the odds ratios were 1.621 (p=0.539) for grade 2 and 1.939 (p=0.926) for grade 3. Overall, these results indicate no substantial difference in histologic grade distribution based on HER2 status.

A notable difference was found in invasion patterns across HER2 groups (χ²=13.309, df=3, p=0.004). The HER2-low group showed a greater percentage of vascular invasion (31.6% vs 25.5%) and dual lymphovascular and perineural invasion (16.1% vs 11.3%) compared to the HER2-zero group. Logistic regression using “no invasion” as the baseline confirmed that both vascular and dual invasion were significantly linked to HER2 expression (p=0.001 for the relevant contrasts). These findings suggest that HER2-low tumors have a higher invasive potential, especially for vascular and dual invasion.

Pathological Complete Response (pCR)

The overall pathological complete response (pCR) rate was 16.0%. When divided by HER2 status, HER2-negative patients had a notably higher pCR of 19.6%, compared to 13.5% in the HER2 1+/2+ group (Chi-square = 6.06, p = 0.014) (Table 2). This significant difference highlights the potential clinical variation in chemotherapy effectiveness based on HER2 expression, even among those without HER2 amplification (Figure 1). 

Table 2: Crosstabulation of Response by HER2 Group: Association between HER2 status and pathologic complete response following neoadjuvant chemotherapy.

Hormone Receptor Subgroup Analysis

Exploratory subgroup analyses were performed to assess pathological complete response (pCR) according to hormone receptor (ER/ PR) status within HER2-zero and HER2-low tumors. In both HER2 groups, ER/PR-negative tumors demonstrated higher pCR rates compared with ER/PR-positive tumors (Table 3). Among ER/PR-negative tumors, pCR rates were similar between HER2-zero and HER2-low groups. Among ER/PR-positive tumors, pCR rates were numerically lower in the HER2-low group compared with the HER2-zero group; however, this difference did not reach statistical significance and should be interpreted with caution.

Table 3: Pathologic complete response rates stratified by hormone receptor status and HER2 expression.

Survival Outcomes

In this study, we compared disease-free survival (DFS) and overall survival (OS) between HER2-zero and HER2-low tumors over five years. The average DFS was 51.20 months (95% CI, 48.81–53.58) for HER2-zero patients and 49.66 months (95% CI, 47.76–51.57) for HER2-low patients. Kaplan–Meier analysis showed no significant difference (log-rank χ²=0.881, p=0.348). The five-year DFS rates were 74% for the HER2-zero group and 71% for the HER2-low group, with overlapping confidence intervals (69%–78% vs. 66%–75%), suggesting no clear separation of survival curves (Table 4, Figure 2).

Similarly, no statistically significant difference in OS was observed between the HER2-zero and HER2-low groups (log-rank χ²=0.147, p=0.702). Although HER2-low tumors demonstrated numerically lower survival estimates, these differences were not statistically significant. Overall, univariable Kaplan–Meier analyses showed no evidence of an association between HER2 status and DFS or OS within the 5-year observation period (Table 4).

Table 4: Five-year disease-free survival and overall survival estimates according to HER2 status.

Figure 2: Mean survival-time estimates of (A) DFS ; (B) OS stratified by HER2 status.

Menopausal Status and Response

Among non-menopausal breast cancer patients, a significant link was found between HER2 expression (HER2=0 vs. HER2-low) and treatment response. Chi-square analysis showed a p-value of 0.024, while Fisher’s exact test confirmed this with p=0.016, suggesting HER2-low tumors may be less responsive to treatment in this group. Specifically, 60.7% of non-responders had HER2-low tumors compared to 39.3% with HER2=0, highlighting a clear response difference. Conversely, in menopausal patients, this association was not significant (chi-square p=0.288 and Fisher’s exact p=0.179), indicating that HER2 status did not significantly influence treatment response in this subgroup.