Mantle Cell Lymphoma Is Associated with a Variable Number of PD-1-Positive Tumor Infiltrating T-Cells

Background: Immune checkpoint blockade agents have been developed for the treatment of solid and hematological tumors. Programmed cell death protein-1 (PD-1), one of the co-receptors of T-cell, is expressed mainly on the cell surface of activated T-cells and inhibits the anti-tumor immunity. Activated T-cells are the main inducers of soluble interleukin-2 receptor (sIL-2R), which is widely used in care of lymphoma in Japan. We recently reported that patients with mantle cell lymphoma (MCL) showed higher levels of sIL-2R compared to those with other histological subtypes (Tomita N, et al. Ann Hematol 2015). In this study, we explored the application of anti-PD-1 antibody for patients with MCL by investigating the presence of PD-1 on tumor-infiltrating T-cells (TITs). Methods: This study was approved by the St. Marianna University School of Medicine Ethics Committee. The subjects of this study were 17 patients with MCL who were diagnosed at St. Marianna University Hospital between 2005 and 2015. The sIL-2R levels were tested in 16 patients (normal range: 124-466 U/mL) at the time of diagnostic biopsy. Diagnostic specimensfrom the 17 patients were examined. Five reactive (non-lymphoma) lymph node specimens from other patients were also investigated. Various immunohistochemical staining techniques including anti-PD-1 antibody staining were performed. CD3-positive cells were defined as TITs. Two hematopathologists (N.N. and H.I.) reviewed all the specimens. Results: Biopsies of MCL were performed at initial diagnosis in 15 patients and at relapse in 2 patients. All the 17 patients were men. The median age at biopsy was 65 years (range: 50-83 years). The biopsied sites were the lymph node in 11 patients, Waldeyer's ring in 3, bone marrow in 2, and colon in 1. The median sIL-2R level was 3,610 U/mL (range: 553-41,600 U/mL). The results of the immunohistochemical staining of tumor cells of the 17 specimens were as follows: 0% (0/17 specimens) were positive for CD3, 76% (13/17) were positive for CD5, 0% (0/17) were positive for CD10, 100% (17/17) were positive for CD20, 67% (8/12) were positive for CD43, 100% (16/16) were positive for BCL2, 0% (0/7) were positive for BCL6, 100% (17/17) were positive for cyclin D1, and 100% (17/17) were positive for SOX11. The median MIB-1 labeling index was 10% (range: <5%-80%). The median number of TITs varied from 100 to 1,000 with a median of 233 in a high-power field (HPF). The median number of tumor-infiltrating PD-1-positive T-cells varied from 20 to 200 with a median of 70 in a HPF. The percentage of PD-1-positive T-cells in TITs in each patient ranged from 10% to 80% with a median of 30%. The sIL-2R levels did not correlate with the number of TITs or tumor-infiltrating PD-1-positive T-cells. In reactive lymph node specimens, PD-1-positive T-cells were found mainly in the germinal centers and rarely in inter-follicular areas. Conclusion: The number ofinfiltrating PD-1-positive T-cells were varied, which may be targets of anti-PD-1 antibody.