Crosstalk Between Programmed Death Ligand 1, Ki-67 Labelling Index, and Tumor-Infiltrating Lymphocytes in Invasive Breast Cancer and Clinicopathological Correlations in a Tertiary Care Center in Western India

Background & Objective: Breast cancer is the leading cancer among Indian women and accounts for about 25% of all cancer cases worldwide. The present study aimed to assess Programmed Death Ligand-1 (PD-L1) expression in tumoral cells and tumor-infiltrating lymphocytes (TILs) and evaluate their correlations with the Ki-67 labelling index in invasive breast carcinomas (IBC). Methods: This descriptive observational study was conducted during 2016-2018 and included all diagnosed cases of IBC. The relationships between PD-L1 expression, TILs, hormone receptors, Ki-67, and clinicopathological parameters were studied in IBC. Statistical analysis was performed by SPSS version 23. Results: Out of 114 evaluated cases, 33.33% (N=38) showed PD-L1+ expression in tumor cells and 47.37% (N=54) had PD-L1+ expression in TILs. A high Ki-67 index was observed in 96 cases. Moreover, 49 patients were estrogen receptor (ER)- and 65 were ER+. We observed that 22 of 49 ER- and 49 of 65 ER+ subjects showed PD-L1+ expression, respectively. Conclusion: Our results showed a significant relationship between PD-L1 expression in tumoral cells and TILs, as well as between Ki-67 and TILs. In addition, an inverse correlation was noted between PD-L1 expression and ER. The PD-L1 expression in tumors and TILs and correlation with high Ki-67 may prove the importance of PD-L1 in targeted chemotherapy. An inverse relationship between PD-L1 and ER expression in tumoral cells suggests scope for immunotherapy in ER- IBC. However, further research with more cases is required.


Introduction
Breast cancer is the most common cancer affecting women and accounts for approximately one-fourth of all cancers (1). There is an increasing trend in the incidence of breast cancer, cancer morbidity, and mortality in India. It ranks as the first cancer among all Indian females, with an age-adjusted rate being 25.8 per 100,000 women and a mortality rate of 12.7 per 100,000 women (1,2). Performing core biopsies of palpable breast lumps and enlarged axillary lymph nodes has significantly bypassed the requirement of open surgical biopsy. In addition, assessing the expression of estrogen receptor (ER), progesterone receptor (PgR), and Her-2/neu has facilitated targeted therapy in invasive breast carcinoma

Study Population
All patients diagnosed with IBC were included in the current study. A total of 114 IBC samples were received during January 2016-December 2018 in the Department of Pathology and Laboratory Medicine, All India Institute of Medical Sciences (AIIMS), Jodhpur, Rajasthan, India after fulfilling the inclusion and exclusion criteria. Cases with benign or premalignant breast lesions, cases with a prior history of radiotherapy/chemotherapy for other malignancies, and inadequate samples were excluded from the study. Core biopsies, lumpectomies, quadrantectomies, and modified radical mastectomy (MRM) were received and all the relevant clinical details were collected. Few representative gross images of MRM with photomicrographs of low-grade and high-grade IBC, mucinous carcinoma, and lobular breast carcinoma are shown in Figure 1.  Semiquantitative analysis of PD-L1, ER, and PgR was completed, and the Allred score was given as per intensity and proportion staining (5,6). The membranous expression of PD-L1 was noted. The Ki-67 nuclear positivity was noted, and percentage was calculated in the tumoral cells according to the recommendations of the International Ki-67 Breast Cancer Working Group (7). Her-2/neu was assessed based on ASCO/CAP guidelines, 2018 (8). The TILs were assessed in Hematoxylin and Eosin (H&E) stained slides according to the recommendations of the International TILs Working Group, 2014 (9). The PD-L1 expression in the tumoral cells and TILs and total effect on both was studied and correlated with factors, such as age, hormone receptor status, and lymph node metastases. A similar section of the tissue was chosen for scoring the percentage of TILs and applying IHC.

Statistical Analysis
The data were entered in Excel and were analyzed by SPSS version 23 (IBM Co., Armonk, NY, USA). Chisquared and Fisher's exact tests were used to analyze the expression of PD-L1 in tumors and TILs and evaluate its possible association with the existing biomarkers ER, PgR, and Her-2/neu. The TILs were quantified and compared with the Ki-67 labelling index.

Characteristics of Patients and Tumors
Out of 114 cases, 112 were women and 2 were men. We observed that 28 patients had received pre-op neoadjuvant chemotherapy (NACT), whereas 86 cases did not receive any NACT (summarized in Table 1). Most of the patients with IBC were in the age group of 51-60 years (28.07%) (Tables 1 and 2).

PD-L1 Expression in Tumoral Cells and TILs
A statistically significant relationship was noted between PD-L1 expression in the tumoral cells and TILs (Table 3). Our results showed that 33.33% of patients (38 cases) had PD-L1 expression in the tumoral cells and 47.37% (54 cases) had PD-L1 expression in TILs. The PD-L1 helps in tumor escape, implying that without chemotherapeutic intervention, the tumor has the potential to bypass the immune system, progress, and become poorly differentiated. There was a significant association between TILs on H&E and the Ki-67 labelling index (N=114) ( Table 4). Therefore, it was suggested that the more proliferative activity a tumor exhibits, the higher the number of TILs to restrain it. In addition, the present study indicated 47.37% cases with PD-L1 expression in TILs implying a purported greater benefit by targeted chemotherapy and in suppressing the tumor cell population.

IHC for ER, PgR, Her-2/neu, and PD-L1
Results of IHC for ER and PgR were assessed as positive/negative based on the Allred scoring. Quantification of TILs (H&E), membranous staining of PD-L1 IHC in tumors, and TILs are shown in Figure 2. The ER and PD-L1 expression had a significant inverse relationship (Table 5). This finding suggested the possibility of using immunotherapy for PD-L1 in ER-cases. No statistically significant association was noted between PD-L1 expression, PgR, Her-2/neu, and clinicopathological parameters, namely age, clinical features, menstrual history, tumor grade, tumor size, pTNM, lymph node metastases, and extranodal extension. However, PD-L1 expression had a statistically significant correlation with triple-negative breast cancer (TNBC), but the results could not be validated due to the small sample size (N=24).

Discussion
Breast cancer detection and treatment have evolved a lot in the last two decades. Cancer therapy has undergone a revolution due to the development of immune checkpoint inhibitors. The CTLA-4 and PD-1 are two negative regulatory proteins that share a 33% amino acid match. They suppress the co-stimulatory molecule by either binding to it or its own ligand and downregulate and eventually suppress the immune system (10)(11)(12). The PD-1 is a B7 family negative regulatory molecule on T, B, and myeloid cells which facilitates immune tolerance by causing cellular anergy following binding one of its ligands PD-L1 (11). The PD-L1 and PD-L2 are also expressed in other nonlymphoid organs (11,13). Consequently, targeted therapies against CTLA-4 or PD-1/PD-L1 signaling pathways can rejuvenate the anti-tumor response and generate a good clinical response (14). The present study described the role of PD-L1 in carcinogenesis. Our findings (Tables 2, 3 (Table 5), except that we did not use cell lines (20). Ghebeh et al. proposed an inverse association between these two parameters. As ER+ breast carcinoma responds better to chemotherapy, they postulated that ER-status and PD-L1+ expression in the tumor may lead to tumor evasion and may be considered a poor prognostic indicator for patients (21). The present investigation indicated that the expression of PD-L1 in the tumoral cells did not have a significant correlation with PgR and Her-2/neu. This was in contrast to the results of Ghebeh et al., who demonstrated that PD-L1 expression in the tumor was significantly correlated with PgR-and Her-2/neu+ breast cancers (21 (25,26).
However, Yuan et al. noted a discordant expression of PD-L1 in primary breast cancer and paired axillary lymph nodes (23). Ming Li et al. noted that the expression of PD-L1 in the lymph node metastases was higher than in paired breast cancer and hence, they suggested that the expression of PD-L1 protein was better analyzed in the lymph nodes (27). Alves et al. investigated PD-L1 expression in primary breast cancer and the lymph nodes and suggested PD-L1 expression in the lymph node metastases, but unrelated to primary breast cancer clinicopathological features (28). Therefore, large-scale studies of PD-L1 expression in primary cancer as well as paired lymph nodes were suggested. Targeted chemotherapy with PD-1 and PD-L1 inhibitors, pembrolizumab, and atezolizumab has widened the horizon in breast cancer

Conclusion
The PD-L1 expression in IBC could be a suitable marker for immunotherapy. The PD-L1 expression has a direct relationship with the Ki-67 labelling index and an inverse relationship with ER expression in the tumoral cells. As a result, we suggest that IBC cases that are ER-can benefit from immunotherapy. The TNBC cases have a worse prognosis, while PD-L1 targeted therapy may prove beneficial even in such cases. Furthermore, TILs had a statistically significant association with PD-L1 in the tumoral cells, suggesting the immune capacity of the host against the rapidly increasing the tumor burden. Large-scale studies are required to further establish PD-L1 marker expression in IBC and its benefit for targeted treatment in the cohort of IBC cases which are ER-.