Introduction: The PD-1 / PD-L1 interaction is a crucial immune checkpoint in cancer, which has been targeted by monoclonal antibodies currently used in clinical trials. The genetic basis and transcriptional regulation of PD-L1 has been previously investigated in ALK+ and ALK- anaplastic large cell lymphomas (ALCL) by our group (Atsaves et al, Leukemia 2017, (7):1633-1637), however, the regulatory mechanisms of PD-1/PD-L1 immune checkpoint in other T-cell lymphomas are not yet known. PD-L1 expression in neoplastic cells has been associated with worse clinical outcome in other hematologic malignancies including Hodgkin lymphoma, diffuse large B-cell lymphoma (DLBCL) and plasma cell myeloma, however, the clinical significance of PD-L1 expression in neoplastic cells and in the immune cells of the tumor microenvironment in PTCLs has not been studied to date. In this study, we investigated PD-L1 gene copy number alterations using a genome-wide approach and diagnostic biopsies. We also assessed for the expression patterns of PD-L1 in the neoplastic cells and histiocytes in PTCL tumors as well as for their possible associations with clinical outcome.

Methods: The PTCL study group included 64 patients (median age 62 years; 61% males, various histologic types), diagnosed and treated at Karolinska University Hospital (Sweden). All tissue samples were obtained prior to therapy. PD-L1 expression was assessed by immunohistochemistry performed on tissue microarrays or full tissue sections using a PD-L1 monoclonal antibody (SP263, Ventana). At least 500 lymphoma cells were counted to calculate the percentage of PDL1-positive tumor cells. PD-L1 expression in histiocytes was scored as "high" or "low" (arbitrary) based on the number of positive infiltrating histiocytes. Overall survival (OS) defined as time from diagnosis to death or last follow-up was chosen as the endpoint for outcome. Survival analyses were performed using the Kaplan-Meier method (log-rank test) and Cox hazards models. In addition, a pilot series of 24 PTCLs were analysed by Oncoscan (Affymetrics) for PD-L1 gene copy number alterations and the results were confirmed by digital q-RT PCR at the mRNA level. Genomic DNA and total RNA were extracted from diagnostic formalin-fixed, paraffin-embedded (FFPE) biopsies using standard protocols. Associations between PD-L1 expression and PTCL type were statistically analysed with Fisher's exact, chi-square, Mann-Whitney and Kruskal-Wallis tests. Survival analysis was performed using the Kaplan-Meier method (log rank test) and Cox proportional hazards model.

Results: Whole genome analysis using Oncoscan (Affymetrics) showed low-level chromosomal gains in 2 (8.3%) PTCLs, which were confirmed by digital q-RT PCR that demonstrated slightly increased PD-L1 mRNA levels (PD-L1/reference gene <2). No correlation of the gene copy number findings with PD-L1 protein expression assessed by immunohistochemistry was seen. Expression of PD-L1 was observed in 92% ALK+ ALCL, 57% ALK- ALCL, 38% PTCL, NOS and 18% angioimmunoblastic T-cell lymphomas (AILT) and less frequently in other PTCL types (p=0.0019). After a median follow up of 66 months, PD-L1 expression by the lymphoma cells was associated with superior OS in the entire study group and in the subgroup of PTCL other than ALK+ ALCL (p<0.0001, logrank). By contrast, high PD-L1 expression in histiocytes was associated with inferior OS (p<0.0001, logrank). The PTCL patients with PD-L1-positive tumor cells and PD-L1-low expression in histiocytes showed a 5-year OS 71% as compared to 24% for the PTCLs with PD-L1-negative tumor cells and PD-L1-high expression in histiocytes (p=0.0002, logrank) and this difference remained highly significant in the subgroup of PTCL other than ALK+ ALCL (p=0.0005, logrank). Both the PD-L1 negativity in lymphoma cells (p=0.0007) and PD-L1-high expression in histiocytes (p=0.04) were independent adverse prognostic factors in the multivariable analysis (Cox model).

Conclusions: Copy number alterations of the PD-L1 gene (9p24.1 locus) are uncommon in PTCL subtypes and therefore, PD-L1 gene expression is likely regulated at the transcriptional of post-translational levels in T-cell lymphomas. PD-L1 negativity in the lymphoma cells and high PD-L1 expression in histiocytes are strong adverse prognostic factors in PTCL, however, larger series are needed to confirm these associations in each histologic type of PTCL.

Disclosures

Österborg:Pharmacyclics: Research Funding; Janssen: Research Funding; Gilead: Consultancy, Research Funding; Abbvie: Research Funding; Beigene: Research Funding.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution