Introduction

Breast cancer is the most common malignancy among woman and its incidence is increasing worldwide and in the Indian subcontinent [1]. Various options are available for the treatment of breast cancer which include surgery, chemotherapy, radiotherapy, hormonal therapy, and targeted therapy. Chemotherapy is ubiquitously used in breast cancer both in neoadjuvant and adjuvant settings. Neoadjuvant chemotherapy is necessary for local and systemic tumor control and it improves the resectability of the tumor by downsizing it [2]. However, about 20% of patients subjected to neoadjuvant chemotherapy do not show response [3, 4]. To predict the response of the tumor for neoadjuvant chemotherapy, several factors have been proposed such as ER (estrogen receptor), PR (progesterone receptor), Her-2/neu receptor, and Ki-67 index. But, independently none of the above factors could accurately predict chemotherapy response. Hence, there is ongoing research for analyzing other factors predicting chemotherapy response in patients of breast carcinoma.

E-cadherin is one such protein being studied for its role in predicting response to neoadjuvant chemotherapy. Cadherins are Ca2+-dependent trans-membrane proteins that are connected to other proteins like actin through other cytoplasmic proteins. E-cadherin loss has been implicated in the epithelial-mesenchymal transition in several malignancies [5]. Selective loss of E-cadherin protein can lead to epithelial de-differentiation and invasiveness in human carcinomas [6, 7]. An inverse relationship has been documented between expression of E-cadherin and invasiveness of tumors [7].

Among the histological types of breast carcinoma, loss of E-cadherin was associated with majority of the invasive lobular carcinomas when compared with invasive ductal carcinomas and is frequently used for differentiating the histological varieties [8]. The results from various studies were contradictory about the association of E-cadherin with tumor size; nodal status; ER, PR, and Her-2/neu receptor positivity; and overall survival. But most of the studies concluded that E-cadherin has limited role as a prognostic factor but it is useful as a phenotype marker [8,9,10,11,12,13,14]. However, there have been promising results from studies comparing E-cadherin with response to chemotherapy in triple-negative breast cancers [15, 16]. These studies are also complimented with studies correlating other cadherins with response to chemotherapy [17]. This has been the basis of our interest to conduct this study with the aim of comparing E-cadherin expression with response to neoadjuvant chemotherapy in breast carcinoma.

Patients and Methods

Institutional ethics committee approval for the study was taken prior to the commencement as it involved human participants. All patients were enrolled in the study after taking written informed consent.

Study Design and Population

A prospective single-center study was conducted in our institution which is a tertiary care and academic center located in India from November 2016 to March 2018. Thirty-three samples of paraffin blocks from breast cancer core biopsy and mastectomy specimens have been included in this study on accrual. All female patients who presented to the surgical department of our hospital during the study period satisfying the inclusion and exclusion criteria and who consented were enrolled in the study. All patients belonged to a single cohort as per the protocol and were followed up prospectively until the end of neoadjuvant chemotherapy and surgery. The average follow-up period was 10 weeks.

Inclusion Criteria

  • Female patients with diagnosis of invasive carcinoma of breast (both ductal and lobular) on core needle biopsy who were planned for neoadjuvant chemotherapy.

Exclusion Criteria

  • Patients with metastasis at presentation.

  • Patients with recurrent disease.

  • Patients who were lost to follow-up during neoadjuvant chemotherapy.

  • Patients in who were unfit for chemotherapy.

Procedure

Female patients who presented to the surgical department with history of symptoms and signs of breast malignancy were evaluated by triple assessment and the size of the lump was measured clinically and radiologically. Patients with a diagnosis of breast carcinoma on core needle biopsy for whom neoadjuvant chemotherapy was planned were invited to enroll in the study. Indications of chemotherapy were:

  • All patients with locally advanced breast carcinoma

  • Patients with early breast carcinoma to downsize the tumor to make it amenable for breast conservation surgery.

All patients in the study were administered 3 cycles of CAF (cyclophosphamide, adriamycin, and 5-flurouracil) as 3-weekly regimen. Size of the lump was monitored every 3 weeks and final size was documented at the end of neoadjuvant chemotherapy. The response was assessed as per the RECIST (Response Evaluation Criteria in Solid Tumors) criteria 1.1 and patients graded into clinical complete responder, partial responder, and non-responder (stable or progressive disease) [18]. E-cadherin expression is compared with neoadjuvant chemotherapy response, age, tumor size, lymph node status, tumor grade, and ER, PR and Her-2/neu receptor status. After the surgery, postoperative specimen was subjected to histopathological examination and pathological complete response if present was documented.

Immunohistochemistry for E-cadherin

The immunohistochemical analysis was performed on formalin-fixed, paraffin-embedded tissue sections. The tissue sections were initially deparaffined in xylene and rehydrated using graded alcohols. The sections were then put in microwave along with buffer solution to retrieve the antigens using heat-induced epitope retrieval technique. Endogenous peroxidase activity was reduced by using hydrogen peroxide solution. Then, the sections were incubated with primary antibody towards E-cadherin for 90 min. We used mouse monoclonal antibody E-cadherin (clone: EP6, Biogenex, USA). The slides were washed and reincubated with secondary antibody and then followed by streptavidin-biotin-peroxidase complex. After washing, brown color reaction was developed using diaminobenzidine as the chromogen, counter stained with haematoxylin, and then observed. All steps were carried out at room temperature. Normal breast tissue sections were used as controls on which all steps were carried out except for the addition of primary antibody. Staining was considered to be positive when the pattern was membranous or membranous and cytoplasmic.

The intensity of E-cadherin was graded as 0 (no color), 1+ (light yellow), 2+ (claybank), 3+ (sepia), and 4+ (dark brown). The proportion of cells stained was graded as 0 (absent), 1+ (≤ 30% of cells stained), 2+ (31–60% of cells stained), and 3+ (> 60% of cells stained) (Fig. 1a, b, and c).

Fig. 1
figure 1

Microscopic images showing various intensities and proportions of E-cadherin staining× 400. a IHC staining of E-cadherin, intensity 4+ and proportion 3+, × 400. b IHC staining showing E-cadherin, intensity 3+ and proportion 3+, × 400. c IHC staining of E-cadherin, intensity 1+ and proportion 1+, × 400

Full size image

Statistical Analysis

The data acquired was analyzed using Statistical Package for Social Sciences (SPSS) version 21.0 using Kruskal-Wallis’s test and Fisher’s exact test.

Results

Out of 33 female patients, 32 patients had invasive ductal carcinoma on histology. Only 1 patient had invasive lobular carcinoma on histology which stained negative for E-cadherin. There was no sample regression due to loss of follow-up.

Mean age of patients was 54.82 years with range of 25 to 80 years. Out of 33 patients, 3.03% patients were below 25 years of age, 39.39% of patients were in the age range of 26–50 years of age, 45.45% of patients were in the age range of 51–75 years, and 12.21% of patients were above the age of 75 years (Table 1). Six out of 33 patients were triple negative constituting 18.1% of subjects. Out of 33 subjects, 24.2% patients belonged to stage IIB, 46.4% patients belonged to stage IIIA, 20.2% patients belonged to stage IIIB, and 9.2% patients belonged to stage IIIC.

Table 1 Age groupwise frequency distribution of patients
Full size table

Out of 33 patients, 3.0% patients showed no reactivity of E-cadherin, 12.1% were 1+ positive, 9.1% were 2+ positive, 36.4% were 3+ positive, and 39.4% were 4+ positive for E-cadherin expression (Table 2).

Table 2 Frequency distribution of E-cadherin intensity and proportion
Full size table

Out of 33 patients, 3.0% patients were negative, 9.1% showed E-cadherin proportion of 1+, 15.2% showed E-cadherin proportion of 2+, and 72.7% showed E-cadherin proportion of 3+ (Table 2).

There was no significant association between tumor size, lymph node positivity, tumor grade, ER positivity, PR positivity, Her-2/neu receptor positivity, and E-cadherin intensity (Table 3). There was significant association between clinical response to neoadjuvant chemotherapy and E-cadherin intensity with p value of < 0.001 (Table 3). There was significant association between E-cadherin intensity and pathological complete response with p value of 0.001 (Table 3).

Table 3 Association between E-cadherin intensity and parameters
Full size table

There was no significant association between tumor size, lymph node positivity, tumor grade, ER positivity, PR positivity, Her-2/neu receptor positivity, and E-cadherin proportion (Table 4). There was significant association between clinical response to neoadjuvant chemotherapy and E-cadherin intensity with p value of 0.003 (Table 4). There was significant association between E-cadherin intensity and pathological complete response with p value of 0.008 (Table 4).

Table 4 Association between E-cadherin proportion and parameters
Full size table

Discussion

E-cadherin is a glycoprotein belonging to the class of calcium-dependent cell adhesion molecules [19]. It is coded by CDH1 gene located at locus 16q22.1 and is a tumor suppressor gene. It plays a very important role in cell-cell adhesion and tissue morphogenesis [20]. The protein is also connected to the intracellular cytoskeleton and also plays a role in various signaling pathways [21]. Loss of E-cadherin is associated with cancer progression in various malignancies [22]. This has prompted research regarding the role of E-cadherin as a prognostic marker in carcinoma breast.

The association of E-cadherin with various prognostic variables is contradictory in various studies. In a study by Singhai R et al., E-cadherin was not associated with any variables like tumor size, lymph node status, and ER, PR, and Her-2/neu receptor status [8]. In a study by Parker C et al., there was significant association between E-cadherin and tumor grade and ER positivity [21]. In a study by Fulga V et al., there was positive association between E-cadherin expression and lymph node metastasis and ER positivity [23]. In a study by Younis LK et al., there was significant correlation between E-cadherin expression and lymph node negative status [9]. In a study by Qureshi HS et al., there was no correlation between E-cadherin and established prognostic variables [24]. In a study by Chintamani et al., there has been positive correlation between E-cadherin expression and ER positivity status and inverse relation with p53 gene mutation [25]. In the present study, there was no significant association between E-cadherin expression and studied prognostic variables similar to the previous studies. However, the association between E-cadherin intensity and ER positivity was close but not significant.

Epithelial to mesenchymal transformation will result in the formation of fibroblastoid cells which are less sensitive to chemotherapy [26]. Epithelial to mesenchymal transformation will also result in upregulation of N-cadherin which is also postulated to have a role in cell invasion [27]. In various studies, it has been proved that lobular carcinomas respond less to chemotherapy and this is attributed to loss E-cadherin [28, 29]. Loss of E-cadherin was associated with decreased pathological complete response after neoadjuvant chemotherapy and poor prognosis after adjuvant therapy in patients of triple-negative breast cancer [15, 16, 30]. In the present study, the significant association proved between E-cadherin intensity and proportion and response to chemotherapy can be correlated with previous studies on patients with lobular breast carcinoma and triple-negative breast cancers. The prognostic value of other cadherins like T-cadherin was also studied and negative expression was correlated with better response to chemotherapy [17].

The limitation of the present study was small sample size and single-center population. Other prognostic predictors like Ki-67 and p53 were not studied simultaneously. However, this study stimulates further research in this regard due to exiting results.

Conclusion

Immunohistochemically detected E-cadherin effectively predicts response to chemotherapy in breast cancer subjects as seen in this study of rather smaller sample of 33 patients. E-cadherin negative cases should be advised early surgery as the role of neoadjuvant chemotherapy is limited in this setting. A multicenter study with a larger sample size and multiple prognostic indicators would help to conclude the above results.