Prognostic significance of tumor infiltrating lymphocytes with CTLA-4, PD-1, and LAG-3 expression in breast carcinomas

Objective: Invasive breast carcinoma is the most frequent malignancy in women and it is also the 4th cause of cancer-related deaths worldwide. Therefore, determining the course of the disease is extremely important. Routine histopathologic reporting of breast carcinomas includes histological grade, stage, histological and molecular subtype so as to predict the clinical course and determine a properly personalized treatment. Based on current genetic studies and emerging molecular classification, new biomarkers that are both expressed in tumor cells and tumor microenvironment and have a potential role in prognosis and treatment, have been identified, and targeted therapies have been suggested. The assessment of immune infiltrate in tumors, most commonly referred to as tumor infiltrating lymphocytes (TILs), is also gaining importance in the current quest for optimal biomarkers to select patients with the highest likelihood of responding to immunotherapeutic agents. Cytotoxic T lymphocyte antigen 4 (CTLA-4) that is one of the immune checkpoint molecules, has a possible poor prognostic value in breast carcinomas. High CTLA-4 positive lymphocyte density was significantly correlated with good prognosis only when tumor CTLA-4 expression was low. Programmed death protein 1 (PD-1), as a member of B7/CD28 co-regulatory factor family, interracts with its ligands -programmed death ligand 1 (PD-L1) and 2 (PD-L2)- which leads to supression of anti-tumor T cell activity and immune escape of tumor cells. PD-1 expression in TILs -there were some conflicting studies- was associated with better outcome in breast carcinomas. Even some studies showed that PD-1 positivity in TILs was an independent prognostic factor. Lymphocyte activation gene 3 (LAG-3) represents another example of a new immune checkpoint target. LAG-3 inhibitors were intended as a treatment for autoimmune diseases but its clinical development has more recently been re-focused on its use as anti-cancer therapeutics. In some studies, LAG-3 had significantly improved breast cancer-specific survival. In our study, we aimed to determine the relationship between expression of CTLA-4, PD-1 and LAG-3 and prognosis in a group of host defense factor (HDF) positive breast carcinomas and tried to predict prognosis using these biomarkers. Materials and Methods: Archive records between January 2009-December 2013 were retrieved from Istanbul University Istanbul Faculty of Medicine Department of General Surgery and Istanbul University Oncology Institute. The records were compared to computer-based archive records in Department of Pathology, Istanbul Faculty of Medicine. HDF-positive breast carcinoma cases were chosen among these records to constitute the study group. Tumor blocks of 238 patients were available and 62 tissue microarray blocks were prepared. Immunohistochemical analyses were performed on these blocks using antibodies against CTLA-4, PD-1 and LAG-3. Clinicopathological characteristics of the study group were retrieved from archive records. Some patient records were completed over telephone calls. Information on the patients' medical history, tumor stage, recurrence or metastasis, treatment regimens and survival data were updated. Results: Average follow-up time was 75.44 (7-130, median: 79) months. A total of 36 (15.1%) patients died of disease, whereas 200 (84%) patients were alive. Forty patients (16.8%) showed progression, either as local recurrence or systemic metastasis. The 10-year-local recurrence free survival (LRFS) was 94.4%; 5-year-disease free survival (DFS) and 10-year-DFS were 83.3% and 79.3%; 5-year-overall survival (OS) and 10-year-OS were 90.1% and 79.2%, respectively. LRFS, DFS and disease-spesific survival (DSS) rates were lower in multifocal tumors compared to unifocal tumors. LRFS, DFS, and DSS were also lower in cases with lymph node metastases compared to those without. TNM-stage were correlated with survival and survival rates decreased as TNM-stages increased. Although high TIL expression was significantly associated with hormone receptor negativity, high proliferation rate and histological grade 3 tumors, the increase in TILs was also associated with longer survival. Multivariate analysis confirmed that high TIL score was an independent prognostic marker for LRFS and DFS. HDF scoring system, that is based on ordinal scale (1+, 2+, 3+), was an independent prognostic marker for only DFS. No correlation was shown between CTLA-4 expression in tumor cells and demographic, clinical and histopathological features. Patients with CTLA-4 immunoexpression in immune cells exhibited significantly less lymphovascular invasion. Tumors with high proliferation rate showed significantly more CTLA-4 positivity in immune cells. LRFS, DFS and DSS rates were lower in cases with high rates of CTLA-4 positivity in TILs that was shown to be an independent factor in predicting DFS and DSS, in multivariate analysis. PD-1 positivity in TILs were more likely to appear in estrogen receptor-negative tumors, tumors with higher Ki-67 index, and histological grade 3 tumors. Cases with strong PD-1 immunolabeling in TILs showed statistically longer DSS. LAG-3 positivity in TILs were inversely correlated with lymphovascular invasion but positively correlated with tumor size. LAG-3 positive TILs were significantly associated with negative prognostic factors: progesterone receptor-negative tumors, high proliferation, and histological grade 3 tumors. DFS and DSS rates were higher in cases that had ≥20% lymphocytes expressing LAG-3. Correlation between percentage of TILs and percentages of CTLA-4, LAG-3 and PD-1 were investigated and statistically significant relationship was shown. Also these immunomarkers were significantly associated with each other. Conclusion: For the percentage of TILs predicts survival better than HDF scoring system, the assessment of TILs is preferable for evaluation of the immune microenvironment and should be a routine parameter in pathology reports with regard to breast carcinoma. Based on the results of our study, we may suggest that high rates of CTLA-4 positivity in TILs adversely affects rates of LRFS, DFS, and DSS, while no correlation exists between CTLA-4 expression in tumor cells and survival rates. We think that further studies are needed to confirm our result about strong PD-1 immunolabeling in TILs is related to longer DSS, due to well-known variation in the strength of immunostaining depending on the method used. We also suggest that high expression of LAG-3 (≥20% lymphocytes) is related to high rates of DFS and DSS. Furthermore, expressions of CTLA-4, LAG-3 and PD-1 are strongly associated with each other and also with the percentage of TILs. Finally, these three immune markers can not only predict prognosis but may also be used in determination of proper treatment. Therefore, immunohistochemical analysis of CTLA-4, LAG-3, and PD-1 can be conceivable, notedly in tumors with high levels of TILs.