Predictive factors for complete pathologic response in luminal breast cancer: impact of ki67 and HER2 low expression

Abstract

Background:

Complete pathological response to neoadjuvant treatment (NAT) in breast cancer is associated with prolonged survival. Compared to other breast cancer immunophenotypes, luminal tumors are the least chemosensitive with low rates of pathological response within this molecular subtype. Thus, finding predictors of response in this subset remains challenging. The emerging concept of low human epidermal growth factor receptor 2 (HER2) expression has led to a repurpose of the current prognostic system. Little is known about its correlation with response to NAT.

Objectives:

This study aims to evaluate predictors of response in early-stage luminal breast cancer receiving neoadjuvant chemotherapy.

Design:

A total of 252 luminal patients who received NAT were retrospectively assessed in this cohort study.

Methods:

We analyzed the correlation of ki67 and HER2 low expression with the rate of pathologic response. Using ki67 as a continuous variable and applying the receiver operating characteristic curves method.

Results:

We identified that in patients with a ki67 expression level >37%, the probability of having a complete pathological response was 4.80 times higher (odds ratio = 4.80, 95% confidence interval: 1.92–12.04). In Her2-low breast cancer patients, Her2 expression did not correlate with a better response rate.

Conclusion:

In our study, a ki67 expression value greater than 37% constitutes a predictive biomarker of pathological complete response in the subgroup of patients with luminal B tumors and could be considered, therefore, an indicator for treatment decisions in this subgroup.

Keywords

biomarker breast neoplasms ki67 neoadjuvant therapy pathological response

Background

Luminal breast cancer is characterized by positive hormone receptor expression independent of human epidermal growth factor receptor 2 (HER2) and ki67. It is, therefore, a very heterogeneous group, representing 70% of all breast cancers. An increasing number of patients diagnosed in the early stages are amenable to neoadjuvant treatment (NAT), which currently consists of chemotherapy or endocrine therapy. The goals of NAT are tumor downstaging and conversion of unresectable stages to operable disease. Other benefits include achieving higher rates of conservative surgery, monitoring biological response to treatment, and even identifying patients who may benefit from more aggressive adjuvant strategies.^1
–5 Primary endocrine therapy has been limited mainly to less aggressive tumors in postmenopausal patients. On the contrary, preoperative chemotherapy is recommended in hormone receptor-positive (HR+) breast cancer when tumors are unresectable or have high-risk features such as stage ⩾T2, extensive axillary involvement, elevated ki67, low estrogen receptor (ER) expression, or premenopausal status.⁶ Recently, it has been suggested that the assessment of genomic criteria in early-stage breast cancer, based on platforms such as Oncotype DX or MammaPrint, may be a good tool to decide to administer NAT according to the probability of achieving pathologic complete response (pCR).⁷ Although it is already known that there is no benefit in terms of overall survival between neo and adjuvant chemotherapy, there is a growing trend toward neoadjuvant management. However, the increase in overall survival achieved with the recent addition of abemaciclib to adjuvant hormonal treatment in patients with high-risk luminal breast cancer (four or more positive axillary lymph nodes, or between one and three positive axillary lymph nodes and either grade 3 disease or tumor size of 5 cm or larger, or centrally assessed ki67 ⩾20%), whether or not they have received NAT, has represented a new change in treatment strategy to consider.⁸

The pCR is defined as the absence of residual invasive cancer based on the evaluation of the resected breast and all lymph nodes sampled after completing NAT.⁹ In fact, the FDA recognizes pCR as a target for accelerated FDA approval. It has been directly associated with increased overall survival in triple-negative and HER2-positive breast cancer and the adjuvant treatment that the patient will receive after the intervention will be adjusted accordingly. In luminal tumors, however, it is less clear than in the other subtypes and it does not imply a change in the treatment strategy after surgery; all patients receiving hormonal treatment along with abemaciclib if they met the criteria as previously mentioned (pCR is not taken into account). In her2 and triple-negative breast cancer, current neoadjuvant regimens based on chemotherapy combinations have demonstrated pCR rates over 60%. By contrast, luminal her2-negative breast cancer is the subtype with the lowest pCR rates after neoadjuvant chemotherapy. In a meta-analysis published by Cortazar et al.¹⁰ that analyzed samples from 11,955 patients, response rates in luminal HER2-negative breast cancer ranged from 7% to 16%, depending on tumor grade.⁹ Other factors that influence pCR in addition to immunohistochemistry (IHC) have been proposed, such as tumor size, presence of tumor-infiltrating lymphocytes, obesity, genetic mutations, and ki67, among others.¹¹ Due to the low pCR rate achieved in luminal tumors, it has been proposed to guide adjuvant treatment according to the residual cancer burden (RCB). RCB is directly associated with the highest number of relapse events as well as the worst prognosis of patients.¹²

ki67 is a nuclear protein of MKI67 whose expression varies over the course of the cell cycle. It is not expressed in the G0 phase and reaches its highest levels during mitosis. To date, ki67 is the only marker that is cell division specific and is not expressed in cells with DNA in repair.^13,14 Therefore, it is strongly correlated with the rate of cell proliferation. In luminal cancer, ki67 is the most important discriminator between luminal A and luminal B tumors. Previous studies support that cell proliferation, identified via ki67 expression levels, may be used as a prognostic and predictive biomarker in breast cancer. The Preoperative Endocrine Prognostic Index, which also includes the ki67 value, has also been considered to have prognostic value in this subtype. However, its use as a predictor of response to neoadjuvant chemotherapy has not been standardized.^14
–18 Some studies suggest that a ki67 cutoff point of 50% may be correlated with a higher pCR after NAT in patients with luminal tumors.¹⁹ Also, the ki67 decrease allows differentiation between endocrine-sensitive and endocrine-resistant tumors in luminal HER2-negative breast cancer treated with primary endocrine therapy.^20,21 In addition, the ki67 cutoff used in these studies was established by arbitrary criteria, making it difficult to set up a predictive ki67 threshold value for the classification of responses to NAT for the different tumor subtypes in clinical practice.²²

The emergence of the concept of her2-low expression has changed the previous paradigm of HER2-positive or negative disease. Her2 expression is considered low when the tumor shows expression by IHC with a score of 1 or 2 with negative SISH. Her2 low tumors are estimated to be 50%–55% of breast cancer and are more likely to be hormone receptor-positive compared to Her2-negative and Her2-positive tumors (64% vs 37% and 50%, respectively). This concept is the result of the advent of conjugated antibodies against Her2-positive metastatic disease where their predictive role of response has been demonstrated but also in Her2-low metastatic disease. Although their applicability in Her2-low localized disease has not yet been demonstrated, they have addressed the point of her2-low expression as a prognostic and predictive factor for response. In fact, it has been described as having different response rates to NAT according to her2-low and negative expression. Regarding pCR, several retrospective analyses have attempted to study whether there is an association between her2-low expression and pCR, without consistent results among them.^23,24

Objectives

The objective of this study was to analyze and identify predictive factors of response to NAT in breast cancer patients in a large series from a referral hospital with an oncology population of more than 1 million inhabitants. We have established the relevant role played by the different clinical biomarkers in breast cancer to identify factors associated with higher rates of pCR after NAT such as ki67 and her2 expression level.

Design

Study population

This was a descriptive retrospective cohort study that included 252 patients who received NAT for localized or locally advanced luminal breast cancer from January 2016 to December 2020, and who subsequently underwent breast and/or axilla surgery at Virgen del Rocío University Hospital (HUVR, acronym in Spanish) (Seville, Spain). It is a comprehensive sample where all analyzable cases were taken into account. The inclusion criteria were as follows: patient diagnosed with Luminal B breast cancer who will receive NAT with available ki67 IHC determination and pathological response data and age over 18 years. Patients who participated in another clinical trial or progressed during NAT and if they had a contraindication to receiving standard treatment were excluded. Patients signed a biobank informed consent for the handling of biological samples.

This research was carried out in accordance with Regulation (EU) 2016/679 on the protection of natural people with respect to the processing of personal data and the free circulation of these data and Organic Law 3/2018 of December 5, Protection of Personal Data and Guarantee of Digital Rights. All information containing personal identification was removed from the dataset before analysis. The study was approved by the Ethics Coordinating Committee of Biomedical Research of Andalusia (Code: 0920-N-20).

Clinical data, tumor characteristics, treatment received, type of surgery, and follow-up data were obtained from electronic health records: computerized clinical history (Diraya) and pharmacy records (HUVR) specific for oncological treatments in Farmis-Oncofarm 3.0^® (V.11.38), in which the systemic treatments received by patients are collected (Table 1).

Table 1.

Clinicopathological characteristics of patients.

Characteristics	N = 252
Age at diagnosis (years)
Mean (range)	51 (22–80)
Menopausal status
Premenopausal	126 (50%)
Postmenopausal	126 (50%)
ER status
Positive	246 (97.6%)
Negative	6 (2.4%)
PR status
Positive	223 (88.5%)
Negative	29 (11.5%)
Proliferation index (ki67%)
<20%	78 (31.0%)
⩾20%–49%	142 (56.3%)
⩾50%	32 (12.7%)
Luminal subtype
Luminal A like	78 (31%)
Luminal B like	174 (69%)
Histologic grade
Grade 1	34 (13.5%)
Grade 2	151 (59.9%)
Grade 3	30 (14.7%)
Unknown	37 (14.7%)
Clinical tumor size (prior to NAT)
cT0/cTis	3 (1.2%)
cT1	23 (9.1%)
cT2	143 (56.7%)
cT3	55 (21.8%)
cT4	28 (11.1%)
Clinical nodal status (prior to NAT)
cN0	86 (34.1%)
cN+	165 (65.5%)
cNx	1 (0.4%)
Clinical stage (prior to NAT)
Stage I	9 (3.6%)
Stage II	159 (63.1%)
Stage III	84 (33.3%)
Pathological tumor size
ypT0	32 (12.7%)
ypTis	4 (1.6%)
ypT1	129 (51.2%)
ypT2	73 (29%)
ypT3	11 (4.4%)
ypT4	3 (1.2%)
Pathological nodal status
ypN0sn	104 (41.3%)
ypN1sn	86 (34.1%)
ypN2sn	48 (19%)
ypN3sn	14 (5.6%)
Pathological stage
y0	27 (10.7%)
yI	76 (30.2%)
yII	82 (32.5%)
yIII	67 (26.6%)
Type of breast and axillary surgery
Breast-conserving surgery and SLNB	74 (29.4%)
Breast-conserving surgery and ALND	55 (21.8%)
Mastectomy and SLNB	39 (15.5%)
Mastectomy and ALND	84 (33.3%)
Histopathologic type
Invasive ductal carcinoma no specific type	208 (82.5%)
Invasive lobular carcinoma	27 (10.7%)
Others	17 (6.8%)

ALND, axillary lymph node dissection; ER, estrogen receptor; NAT, neoadjuvant treatment; PR, progesterone receptor; SLNB, sentinel lymph node biopsy.

The analysis of ER, PR, her2, and ki67 was performed by IHC.^25,26 The evaluation was performed by two pathologists specialized in breast cancer by conventional optical microscopic visualization following the American Society of Clinical Oncology/College of American Pathologist guidelines for HER2 expression assessment.^27,28 HER2-low-positive breast cancer was defined as tumors with HER2 IHC score 1+ or score 2+ with negative fluorescence in situ hybridization (FISH)/in situ hybridization (ISH), and HER2-zero group with an IHC score of 0. To determine ki67 expression by IHC, an anti-ki67-(30-9) monoclonal antibody (Ventana-Roche) was used.²⁹ ki67 was done on core biopsies to baseline samples. The average percentage of staining in neoplastic cells of a complete section of the tumor was assessed, following recommendations of the international group “International ki67 in Breast Cancer Working Group.” To convert the ki67 result provided as an interval to an absolute number to simplify its analysis, it was decided to homogenize considering the average value of the interval.

Classification of pathological response to neoadjuvant therapy

The pathological response to NAT was determined by analyzing the surgical tumor specimen after completing NAT. The RCB system, proposed by Symmans et al.,³⁰ is divided into four categories. A complete pathological response, pCR or RCB-0, was defined as no evidence of invasive residual tumor in the breast and absence of disease in the axillary lymph nodes (ypT0 and ypN0). Patients with residual carcinoma in situ and no evidence of invasive disease were also included in this category (ypTis and ypN0). The few cases with lymphatic vascular invasion as the only finding, without evidence of invasive residual tumor in the breast or in axillary lymph nodes (ypT0, ypN0), were excluded from the pCR category.

Methods

Statistical analysis

Descriptive statistics were used to analyze the general characteristics of the patients (age and menopausal status) and tumor histopathological characteristics: HR and HER2 status, ki67 proliferation index, tumor grade, clinical stage (cTNM), pathological stage (ypTNM), and pathological response of the tumor. The report of this study conforms to the STROBE statement and is attached as a Supplemental File. To determine whether there was an association between the categorical variable ki67 and pCR, we used the Pearson Chi-square test; subsequently, we assessed different associations between ki67 subsets and pCR ranges using the Z test with Bonferroni correction. In addition, for the continuous variable ki67, we used receiver operating characteristic curves (ROC) to establish cutoff points for pCR risk estimates were also analyzed. The best cutoff point for ki67 was determined according to Youden’s J statistic, this cutoff point determines the highest sensitivity and specificity and divides the study population (luminal B breast cancer patients and premenopausal and postmenopausal patients) into two groups and predicts the ability of ki67 discriminating between patients with and without pCR to NAT.

Correlations between HER2 status (HER2-low-positive vs HER2-zero) and pCR were evaluated using Pearson’s Chi-square test (two-sided, p < 0.05).

All statistical analyses were performed with SPSS 28.0 (Statistical Package for the Social Sciences, IBM, USA).

Results

Patient characteristics and treatment

In total, 2845 patients were diagnosed with breast cancer at Virgen del Rocío Hospital from January 2016 to December 2020. A total of 556 patients received NAT, of which 252 patients with a diagnosis of luminal breast cancer (45.3%) were selected. Seventy-eight patients (31%) had luminal A phenotype, while 174 (69%) had luminal B. Median age was 50 years at diagnosis. 48.92% of patients had a postmenopausal status. Other patient characteristics are described in Table 1.

At diagnosis, 9 patients (3.6%) were in the initial stage (stage I), 159 patients (63.1%) were in stage II, and 84 patients (33.3%) were in the locally advanced stage (stage III) (Table 1). In total, 230 patients (91.3%) received an anthracycline regimen followed by taxanes; for 3 patients (1.2%), carboplatin was added with the taxane, 9 patients (3.6%) received cyclin-dependent kinase 4 and 6 (CDK4/6) inhibitors and 10 patients (3.9%) only received endocrine therapy.

Response to NAT

Twenty-seven patients (10.71%) reached a pCR, 24 (9.52%) presented minimal RCB (RCB-I), 110 (43.62%) moderate RCB (RCB-II), and 69 patients (27.38%) extensive RCB (BCR-III). The pathological response was not quantifiable in 22 cases (8.7%; Table 2). The percentage of patients with luminal A and luminal B tumors who achieved a pCR was 3.8% and 13.8%, respectively. The percentage of patients with RCB-III tumors was higher in the luminal A subtype (34.6%) than in the luminal B subtype (24.1%).

Table 2.

Pathologic response (RCB) to neoadjuvant treatment.

Variable	pCR	RCB-I	RCB-II	RCB-III	Not calculable	Total
Subtype
Luminal A like	3 (3.8%)	4 (5.1%)	34 (43.6%)	27 (34.6%)	10 (12.8%)	78
Luminal B like	24 (13.8%)	20 (11.5%)	76 (43.7%)	42 (24.1%)	12 (6.9%)	174
ki67%
<50%	17 (7.7%)	15 (6.8%)	102 (46.4%)	65 (29.5%)	21 (9.5%)	220
⩾50%	10 (31.3%)	9 (28.1%)	8 (25%)	4 (12.5%)	1 (3.1%)	32
ki67%
ki67 ⩽37%	11 (5.7%)	15 (7.8%)	89 (46.4%)	59 (30.7%)	18 (9.4%)	192
ki67 >37%	16 (26.7%)	9 (15.0%)	21(35.0%)	10 (16.7%)	4 (6.7%)	60
Total	27 (10.7%)	24 (9.5%)	110 (43.7%)	69 (27.4%)	22 (8.7%)	252
	pCR	No pCR				Total
ki67%
ki67 ⩽37%	11 (5.7%)	181 (94.3%)				192
ki67 >37%	16 (26.7%)	44 (73.3%)				60
Total	27 (10.7%)	225 (89.3%)				252

pCR, pathologic complete response; RCB, residual cancer burden.

Correlation between response to NAT and ki67 expression

Analyzing ki67 as a continuous variable and applying ROC curves, ki67 expression and pCR were significantly correlated. When the analysis was performed considering only patients with luminal B breast cancer (N = 174), the results revealed a correlation between this subtype and the ki67 expression index. The cutoff point was ki67 ⩽ 37%, and the area under the ROC curve was 0.684 (confidence interval (CI): 0.560–0.808), with a sensitivity of 79.2% and a specificity of 70.7% (Figure 1(a)). For luminal B tumors with ki67 ⩽ 37%, the pCR after NAT was 7%; in comparison, for those with ki67 >37%, the pCR was 26.7% (p = 0.0001) (Figure 1(b)). For these patients, the probability of having a pCR was 4.80 times higher for tumors with ki67 >37% than with ki67 ⩽37% (p = 0.0001, odds ratio (OR) = 4.80, 95% CI: 1.92–12.04; Table 2).

Figure 1.

(a) ROC curve of ki67 expression with respect to pCR. (b) Representative diagram of pCR with respect to ki67 expression.

In premenopausal patients (n = 126), the cutoff point was ki67 ⩽37% and the area under the ROC curve was 0.690 (CI: 0.537–0.842; p = 0.025) with a sensitivity of 71.4% and a specificity of 68, while in postmenopausal patients (n = 89), the cutoff point was ki67 ⩽ 42%. In these patients, the area under the ROC curve was 0.668 (CI: 0.464–0.872; p = 0.086) with a sensitivity of 60% and a specificity of 77.3%.

Reduction in post-neoadjuvant tumor stage based on the baseline clinical stage

Subsequently, we evaluated the changes between the clinical and pathological stages after neoadjuvant chemotherapy (Table 3). In 48.82% (n = 124) of patients, the stage decreased after NAT and 10.71% of these (n = 27) presented pCR. However, in 40% (n = 101), the stage did not change and even 10.71% of patients presented disease progression. Of the patients with luminal immunophenotype and ki67 greater than 37% (N = 60), 71.66% (N = 43) presented a decrease in staging (26.66% achieved pCR). In 21.66% (n = 13), the stage did not change and 6% of patients showed progression (n = 4).

Table 3.

Downstaging of status according to baseline clinical stage (ki67 >37%).

Clinical stage	Pathological stage				Total (N)
Clinical stage	y0	yI	yII	yIII	Total (N)
Stage I	1 (50%)	1 (50%)	0 (0%)	0 (0%)	2
Stage II	11 (26.8%)	20 (48.8%)	6 (14.6%)	4 (9.8%)	41
Stage III	4 (23.5%)	2 (11.8%)	5 (29.4%)	6 (35.3%)	17
	16	23	11	10	60

PCR rates according to the HER2-low status

A total of 158 (66.38%) out of 238 tumors were HER2-low-positive (58 of them were IHC1+ and 100 ICH2+), and 80 tumors (33.61%) were Her2-Zero. In 14 out of the total 252 tumors analyzed (5.6%), Her2 status was not known. No differences in pCR rate were seen in luminal Her2-low (12%) versus HER2-zero tumors (7.5%) (p = 0.282). We could not find any differences between IHC1+ (12.1%) and IHC2+ tumors (12%) either.

Discussion

To date, the main applications of ki67 in early HR-positive HER2-negative breast cancer are as follows: differentiation between luminal A and luminal B tumors, prognostic value (with increased ki67 values related to worse prognosis), indicator of response to neoadjuvant endocrine therapy, and determination of adjuvant treatment.³¹ The International ki67 in Breast Cancer Working Group established a ki67 index cutoff of ⩽5% or ⩾30% to estimate prognosis. However, there is still a large difference between 5% and 30%, and a large percentage of patients would fall within this range. The 2013 update of the St. Gallen Consensus Statement redefined >20% as the new threshold for substratification of luminal subtypes based on the work of panel of experts³²who highlighted the relevance of this cutoff for predicting survival outcomes in luminal breast cancer. However, the 2015 St. Gallen consensus subsequently abandoned a specific cutoff point for ki67 due to analytical and preanalytical barriers to standardized assessment.³³ Biologically, cell proliferation is a phenomenon that correlates with chemosensitivity and is considered a prognostic factor. Breast cancers with high levels of proliferation prior to NAT can be expected to have a more aggressive clinical course but also a better response to chemotherapy. On the contrary, high levels of ki67 after chemotherapy, as well as a lack of reduction in ki67, are associated with a worse prognosis. Although the use of ki67 expression is clinical, its association with a higher pCR rate and greater overall survival in ER-positive tumors in values above 20% has been demonstrated by numerous studies. In our study, for luminal breast cancer, the basal ki67 index greater than 37% was significantly correlated with an increase in pCR after NAT.

Alba et al. proposed that ki67 >50% was correlated with greater efficacy of neoadjuvant chemotherapy for breast cancer (N = 262), especially in patients with tumors with ER expression and HER2 expression. However, pCR was reported in 11%.¹⁹ Our study included a cohort of patients, and we identified a cutoff value that allowed classifying pCR after NAT based on the quantification of ki67 as a continuous variable. This approach further allowed us to robustly define a ki67 cutoff value for use in clinical practice. That is, the analysis performed by Alba et al. was carried out with categorical values of ki67 (in a range of 10 expression levels), and in our study, as a continuous variable, the classification of the response to NAT was better defined with respect to the ki67 proliferative index. Notably, the series of patients studied show differences regarding breast cancer subtypes, which could have influenced the results. In our study, the ki67 result has been simplified into an absolute number to simplify its analysis, although it is usually provided as an interval given the interobserver variability. This may lead to differences between the results of different studies.

A recent study published by Ma et al.³⁴ proposes a ki67 cutoff point of 22.5% as a prognostic factor in patients with HR+/HER2− breast cancer, with a worse prognosis being associated with a greater risk of distant relapse in patients who presented a ki67 greater than the proposed value. However, this study did not include patients who received NAT so it does not present results of pCR and its influence on overall survival or relapse-free interval. This may influence the difference in cutoff point with our study.

A meta-analysis of 53 studies proposed that high ki67 expression is associated with a higher pCR, with an OR of 3.10 (2.52–3.81) regardless of the tumor subtype; however, they reported that this association was only valid in patients who had received anthracyclines in combination with taxanes or anthracyclines in monotherapy.³⁵ Other meta-analyses have already highlighted the importance of ki67 expression, relating it to pCR; however, they did not establish a significant cutoff point.^4,5

While pCR in breast cancer has been associated with better prognosis in terms of disease-free survival and overall survival, the results of the Minckwitz analyses suggest that this relation was stronger in patients with more aggressive cancer subtypes: triple-negative breast cancer and in those with HER2-positive/hormone-receptor-negative breast cancer.³⁶ However, in the GeparTrio study, in which three groups of patients were distinguished according to ki67 index level (0–15% vs 15.1%–35% vs >35.1%). The low ki67 group had a comparable outcome to the PCR group, while the high ki67 group had a significantly higher risk of recurrence and death compared to the low or intermediate ki67 group. Taken together, the post-therapy ki67 index level may provide additional prognostic information in ER-positive breast cancer in which PCR has limited prognostic value.³¹

ki67 expression has also been investigated after NAT (not frequent in routine practice), as performed by Cabrera-Galeana et al.³⁷ In this study, there was a decrease in ki67 expression (on average 10%) in surgical specimens after completing NAT; this decrease was related to a higher pCR as well as a greater disease-free interval. Overall survival increased compared with that for patients in whom there was no change or an increase in ki67 expression (increased risk of recurrence and death with hazard ratios of 3.39 (1.8–6.37) and 7.03 (2.6–18.7), respectively). Recently, the role of ki67 in adjuvant treatment settings is increasingly controversial given the latest advances in the field because adding CDK4/6 inhibitors such as abemaciclib or ribociclib to hormonal therapy provides survival benefit regardless the ki67 value.^38,39

Focusing on HER2-low luminal tumors, the articles published on the pCR rate do not find differences depending on the level of HER2 expression. Denkert et al. published a pooled analysis of patients with HER2-negative (HER2-low and HER2-zero) who participated in four clinical trials with standard chemotherapy at a neoadjuvant setting (except GeparX clinical trial that added denosumab). In this analysis, a statistically significant difference (p = 0.024) was identified between the pCR rate of HER2-low luminal tumors (23.6%) and the pCR rate of HER2-zero luminal tumors (17.5%). However, there was no difference between HER2-low IHC 1+ patients and HER2-low IHC 2+/ISH non-amplified patients.²³

Other published studies, such as de Moura et al.,²⁴ did not find significant differences between the pCR rate of HER2-low or HER2-zero luminal tumors with standard chemotherapy in a neoadjuvant context. In this, the pCR rate was 13% versus 9.5%, respectively (p = 0.27). These results are in line with the results of our study where the pCR rate was 12% versus 7.5% (p = 0.282) in luminal HER2-low and HER2-zero tumors. Other studies obtained the same results, finding no statistically significant difference in pCR between HER2-low and HER2-zero patients, regardless of the level of ER expression.^40,41

In the HELENA study,⁴² the role of HER2 expression as a predictor of pCR was evaluated. No difference in pCR rate was found between HER2-low IHC 1+ and HER2-low IHC 2+/ISH non-amplified tumors. However, it was not described whether the tumors that reached pCR expressed hormone receptors.

Clinical trials with targeted treatments on NAT such as antibody drug conjugated (ADCs) in her2-low breast cancer are ongoing. In the future, we will be able to find out whether targeted treatment for this tumor subtype obtains better pCR rates than with standard chemotherapy.

Our study has some limitations. It was conducted at a single center, and the reproducibility of ki67 expression measurements was not assessed; however, the probable bias through interobserver variability was likely low. Importantly, evaluations were performed in a standardized manner by expert pathologists following international recommendations.

Conclusion

The role of biomarkers in predicting response to NAT in luminal breast cancer is being actively investigated. Biomarkers such as the ki67 proliferation index and hormone receptors expression or her2 amplification can provide important information about the aggressiveness of the tumor and the probability of response to NAT. Luminal B tumors, which usually have a higher ki67 index, tend to be more sensitive to NAT with chemotherapy, and the result of our cohort of ki67 37% can be used as a reference to predict the behavior of the disease and anticipate pCR when advising patients on chemotherapy treatment. AUC of 0.7 is acceptable and useful in this clinical context, indicating that the model has significant but moderate discriminative ability. The model can capture certain important patterns in the data, distinguishing between cases with complete and incomplete pathological responses, which is a valuable advancement in this field. However, we propose, as future work directions, the improvement of the standardization of the marker estimation process, support for new computerized methods or artificial intelligence, the establishment of universal reference points, and the search for new predictive biomarkers in this group of patients. An example of this is the her2-low status in the neoadjuvant setting. Currently, her2-low cannot be considered a biomarker of response with approved treatments, and its role will have to be evaluated when adding targeted drugs in this scenario. In the future, the development of more precise and reproducible biomarkers could allow better patient stratification and more effective customization of NAT in luminal breast cancer.

Supplemental Material

sj-docx-1-tam-10.1177_17588359241309169 – Supplemental material for Predictive factors for complete pathologic response in luminal breast cancer: impact of ki67 and HER2 low expression

Supplemental material, sj-docx-1-tam-10.1177_17588359241309169 for Predictive factors for complete pathologic response in luminal breast cancer: impact of ki67 and HER2 low expression by Isabel Miras, Ana Gil, Marta Benavent, María Ángeles Castilla, Begoña Vieites, María Ángeles Dominguez-Cejudo, Sonia Molina-Pinelo, Lina Alfaro, Javier Frutos, Manuel Ruiz-Borrego, Alejandro Falcón, Mónica Cejuela and Javier Salvador-Bofill in Therapeutic Advances in Medical Oncology

Footnotes

Acknowledgements

The authors acknowledge support from the Andalusia-Roche Network Mixed Alliance in Precision Medical Oncology, resources of which are composed of financial contribution from Andalusian Public Health System through grant entitled “ALIANZA MIXTA EN RED ANDALUCÍA—ROCHE EN ONCOLOGÍA MÉDICA DE PRECISIÓN (INVESTIGACIÓN BÁSICA/TRASLACIONAL)” financed by Regional Ministry of Health and Families (PIP-0044-2020) through a competitive public call for proposals and by Roche Farma S. A. with private funds. Ana Gil and Sonia Molina-Pinelo were funded by the Regional Ministry of Health and Families (RB-0003-2019 and RC-0004-2020, respectively).

Declarations

ORCID iDs

Isabel Miras

Sonia Molina-Pinelo

Supplemental material

Supplemental material for this article is available online.

References

Berruti

Amoroso

Gallo

, et al. Pathologic complete response as a potential surrogate for the clinical outcome in patients with breast cancer after neoadjuvant therapy: a metaregression of 29 randomized prospective studies. J Clin Oncol 2014; 32: 3883–3891.

Hayashi

Yagata

Tsugawa

, et al. Response and prognosis of docetaxel and cyclophosphamide as neoadjuvant chemotherapy in ER (+) HER2(−) breast cancer: a prospective phase II study. Clin Breast Cancer 2020; 20: 462–468.

Van der Hage

van de Velde

Julien

, et al. Preoperative chemotherapy in primary operable breast cancer: results from the European Organization for Research and Treatment of Cancer trial 10902. J Clin Oncol 2001; 19: 4224–4237.

Mauri

Pavlidis

Ioannidis

JP.

Neoadjuvant versus adjuvant systemic treatment in breast cancer: a meta-analysis. J Natl Cancer Inst 2005; 97: 188–194.

Tao

Chen

Zhang

, et al. Ki-67 labeling index is a predictive marker for a pathological complete response to neoadjuvant chemotherapy in breast cancer. Medicine (Baltimore) 2017; 96(51): e9384.

Ayala de la Peña

Antolín Novoa

Gavilá Gregori

, et al. SEOM-GEICAM-SOLTI clinical guidelines for early-stage breast cancer (2022). Clin Transl Oncol 2023; 25(9): 2647–2664.

Freeman

Shubeck

Howard

, et al. Evaluation of multigene assays as predictors for response to neoadjuvant chemotherapy in early-stage breast cancer patients. NPJ Breast Cancer 2023; 9: 33.

Johnston

SRD

Toi

O’Shaughnessy

, et al. Abemaciclib plus endocrine therapy for hormone receptor-positive, HER2-negative, node-positive, high-risk early breast cancer (monarchE): results from a preplanned interim analysis of a randomised, open-label, phase 3 trial. Lancet Oncol 2023; 24: 77–90.

Von Minckwitz

Untch

Blohmer

J-U

, et al. Definition and impact of pathologic complete response on prognosis after neoadjuvant chemotherapy in various intrinsic breast cancer subtypes. J Clin Oncol 2012; 30: 1796–1804.

10.

Cortazar

Zhang

Untch

, et al. Pathological complete response and long-term clinical benefit in breast cancer: the CTNeoBC pooled analysis. Lancet 2014; 384(9938): 164–172.

11.

Alamoodi

Factors affecting pathological complete response in locally advanced breast cancer cases receiving neoadjuvant therapy: a comprehensive literature review. Eur J Breast Health 2023; 20(1): 8–14.

12.

Yau

Osdoit

van der Noordaa

, et al. Residual cancer burden after neoadjuvant chemotherapy and long-term survival outcomes in breast cancer: a multicentre pooled analysis of 5161 patients. Lancet Oncol 2022; 23: 149–160.

13.

Azambuja

Cardoso

de Castro

, et al. ki67 as prognostic marker in early breast cancer: a meta-analysis of published studies involving 12,155 patients. Br J Cancer 2007; 96: 1504–1513.

14.

Kreipe

Harbeck

Christgen

Clinical validity and clinical utility of ki67 in early breast cancer. Ther Adv Med Oncol 2022; 14: 17588359221122725.

15.

Von Minckwitz

Untch

Nuesch

, et al. Impact of treatment characteristics on response of different breast cancer phenotypes: pooled analysis of the German neo-adjuvant chemotherapy trials. Breast Cancer Res Treat 2011; 125: 145–156.

16.

Hirata

Shimizu

Yonemori

, et al. Change in the hormone receptor status following administration of neoadjuvant chemotherapy and its impact on the long-term outcome in patients with primary breast cancer. Br J Cancer 2009; 101: 1529–1536.

17.

Faneyte

Schrama

Peterse

, et al. Breast cancer response to neoadjuvant chemotherapy: predictive markers and relation with outcome. Br J Cancer 2003; 88: 406–412.

18.

Berruti

Generali

Kaufmann

, et al. International expert consensus on primary systemic therapy in the management of early breast cancer: highlights of the fourth symposium on primary systemic therapy in the management of operable breast cancer, Cremona, Italy (2010). J Natl Cancer Inst Monogr 2011; 2011: 147–151.

19.

Alba

Lluch

Ribelles

, et al. High proliferation predicts pathological complete response to neoadjuvant chemotherapy in early breast cancer. Oncologist 2016; 21(2): 150–155.

20.

Dowsett

Smith

Ebbs

, et al. Short-term changes in Ki-67 during neoadjuvant treatment of primary breast cancer with anastrozole or tamoxifen alone or combined correlate with recurrence-free survival. Clin Cancer Res 2005; 11: 951s–958s.

21.

Ellis

Tao

Luo

, et al. Outcome prediction for estrogen receptor-positive breast cancer based on postneoadjuvant endocrine therapy tumor characteristics. J Natl Cancer Inst 2008; 100: 1380–1388.

22.

Luporsi

André

Spyratos

, et al. Ki-67: level of evidence and methodological considerations for its role in the clinical management of breast cancer: analytical and critical review. Breast Cancer Res Treat 2012; 132: 895–915.

23.

Denkert

Seither

Schneeweiss

, et al. Clinical and molecular characteristics of HER2-low-positive breast cancer: pooled analysis of individual patient data from four prospective, neoadjuvant clinical trials. Lancet Oncol 2021; 22(8): 1151–1161.

24.

de Moura Leite

Cesca

Tavares

, et al. HER2-low status and response to neoadjuvant chemotherapy in HER2 negative early breast cancer. Breast Cancer Res Treat 2021; 190(1): 155–163.

25.

Allison

Hammond

MEH

Dowsett

, et al. Estrogen and progesterone receptor testing in breast cancer: ASCO/CAP guideline update. Arch Pathol Lab Med 2020; 144(5): 545–563.

26.

Dowsett

Nielsen

A’Hern

, et al. Assessment of ki67 in breast cancer: recommendations from the international ki67 in breast cancer working group. J Natl Cancer Inst 2011; 103(22): 1656–1664.

27.

Wolff

Hammond

Hicks

, et al. Recommendations for human epidermal growth factor receptor 2 testing in breast cancer: American Society of Clinical Oncology/College of American Pathologists clinical practice guideline update. J Clin Oncol 2013; 31(31): 3997–4013.

28.

Wolff

Hammond

MEH

Allison

, et al. Human epidermal growth factor receptor 2 testing in breast cancer: American Society of Clinical Oncology/College of American Pathologists clinical practice guideline focused update. J Clin Oncol 2018; 36(20): 2105–2122.

29.

Polley

Leung

McShane

, et al. An international ki67 reproducibility study. J Natl Cancer Inst 2013; 105(24): 1897–1906.

30.

Symmans

Peintinger

Hatzis

, et al. Measurement of residual breast cancer burden to predict survival after neoadjuvant chemotherapy. J Clin Oncol 2007; 25(28): 4414–4422.

31.

Denkert

Loibl

Müller

, et al. ki67 levels as predictive and prognostic parameters in pretherapeutic breast cancer core biopsies: a translational investigation in the neoadjuvant GeparTrio trial. Ann Oncol 2013; 24(11): 2786–2793.

32.

Goldhirsch

Winer

Coates

, et al.; Panel Members. Personalizing the treatment of women with early breast cancer: highlights of the St Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2013. Ann Oncol 2013; 24(9): 2206–2223.

33.

Vasconcelos

Hussainzada

Berger

, et al. The St Gallen surrogate classification for breast cancer subtypes successfully predicts tumor presenting features, nodal involvement, recurrence patterns and disease free survival. Breast 2016; 29: 181–185.

34.

Liu

She

, et al. The role of Ki-67 in HR+/HER2− breast cancer: a real-world study of 956 patients. Breast Cancer (Dove Med Press) 2024; 16: 117–126.

35.

Chen

Han

, et al. The predictive value of Ki-67 before neoadjuvant chemotherapy for breast cancer: a systematic review and meta-analysis. Future Oncol 2017; 13(9): 843–857.

36.

von Minckwitz

Blohmer

Costa

, et al. Response-guided neoadjuvant chemotherapy for breast cancer. J Clin Oncol 2013; 31(29): 3623–3630.

37.

Cabrera-Galeana

Muñoz-Montaño

Lara-Medina

, et al. ki67 changes identify worse outcomes in residual breast cancer tumors after neoadjuvant chemotherapy. Oncologist 2018; 23(6): 670–678.

38.

Harbeck

Rastogi

Martin

, et al.; MonarchE Committee Members. Adjuvant abemaciclib combined with endocrine therapy for high-risk early breast cancer: updated efficacy and Ki-67 analysis from the monarchE study. Ann Oncol 2021; 32(12): 1571–1581.

39.

Ergun

Dogan

Ucar

, et al. Efficacy of adjuvant CDK4/6 inhibitors in hormone receptor-positive breast cancer: a systematic review and meta-analysis. Expert Opin Pharmacother 2023; 24(17): 1901–1909.

40.

Reinert

Sartori

Souza

, et al. Abstract PS4-22: prevalence of HER2-low and HER2-zero subgroups and correlation with response to neoadjuvant chemotherapy (NACT) in patients with HER2-negative breast cancer. Cancer Res 2021; 81(4S): PS4–PS22.

41.

Guan

Shi

Zheng

, et al. Real-world data on pathological response and survival outcomes after neoadjuvant chemotherapy in HER2-low breast cancer patients. Clin Breast Cancer 2024; 24(5): 463–472.e2.

42.

Cherifi

Da Silva

Johnson

, et al. HELENA: HER2-Low as a prEdictive factor of response to Neoadjuvant chemotherapy in eArly breast cancer. BMC Cancer 2022; 22(1): 1081.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB