Correlation of Toll-like Receptor (TLR) and PD-1 Pathways in Central Nervous System Lymphoma: A Clinicopathological Study on Outcomes

Introduction:Most patients with central nervous system lymphoma (CNSL) have an inferior response to therapy compared with systemic diffuse large B-cell lymphoma (DLBCL), in part possibly because 95% of CNSL are of activated B-cell (ABC) phenotype. Recent studies have revealed the genetic basis of pathways active in CNSL. Activated anti-apoptotic NFκB and B-cell receptor pathways are key to sustained cancer cell survival. Myeloid differentiation primary-response protein 88 (MYD88) is an adaptor protein that upon stimulation of TLR, mediates downstream activation of the NFκB pathway. The objective of this study is to examine the association of MYD88 mutation and PD-1/PD-L1 expression with clinical outcome and the overlap between PD-1/ PD-L1 expression and MYD88 L265P mutation in CNSL.

Methods:Twenty-two patients with CNSL on whom complete treatment records and tissue were available were included in our study following IRB approval. On a corresponding archived formalin-fixed, paraffin-embedded tissue block from each case, 10 5-um unstained slides for immunohistochemistry and a 10-50 um tissue curl for MYD88 DNA analysis by PCR were prepared. PAX5, CD3, CD68, PD-1 and PD-L1 immunostains were performed and evaluated by light microscopy by two of the authors blinded to the clinical data, including one hematopathologist. Extracted DNA was analyzed using allele-specific PCR designed to detect the common MYD88 gene mutation c.794T>C (p.L265P). DLBCL immunophenotype data, based on Hans algorithm was extracted from the pathology reports. Comparisons between two groups were analyzed using Chi-square and Student t tests. Kaplan Meier method was used to calculate progression-free survival (PFS) and overall survival (OS). Log-rank test was used to determine the differences in survival. Pearson and Spearman's rho coefficients were calculated for correlation analysis. Statistical analysis was performed using SPSS.22 software.

Results:The median age at diagnosis was 67 years (range 28-81). Thirteen patients were male (59%). Twelve patients had ECOG performance status of > 1. Median LDH at diagnosis was 190 U/L. Two patients with CNSL had HIV. Two patients had received prior solid-organ transplant. Seventeen patients received high-dose methotrexate-based therapy, and 9 received radiation. In terms of histopathology, 19 (86%) were of ABC subtype, all were CD20 positive, and 4 were EBV-positive, including both post-transplant cases. The median PFS of all patients was 6.52 months, and the OS was 12 months. Tumor content of the tissue sections was estimated from H&E and PAX-5 stains and showed 30-100% tumor areas, barring 3 cases with <20%. Tumor infiltrating lymphocyte (TIL) content (CD3) was estimated as low (50%), moderate (36%) or high (14%). Tumor associated macrophage (TAM) infiltration (CD68) was estimated as 32% high, 59% moderate and 9% low. 32% of tumors had either PD-L1 or PD-1 expression; all 4 EBV-associated tumors were positive for PD-L1. In 68% of cases, TAMs were positive for PD-L1. TILs co-expressed PD-1 in all cases. MYD88 data was available for 21 patients and 15/21 (71%) were positive. Twenty patients were included in the survival analysis. There was no significant association of MYD88 mutational status or tumor PD-1/PD-L1 expression with PFS or OS. While no correlation was found between the tumor PD1/PD-L1 status and the degree of TAM infiltration, TIL infiltration or the baseline lymphocyte count, there was a negative correlation between MYD88 mutation and tumor PD-1/ PD-L1 (p=0.004;r²=-0.623).

Conclusion: The high prevalence of tumor PD-1/PD-L1 and MYD88 gene mutation are similar to previous reports, however, we found an inverse correlation between MYD88 mutation status and tumor PD-1/PD-L1 expression, not previously reported. This could be due to one pathway being dominant in the individual tumor or to down regulation of checkpoint molecules mediated by MYD88 and needs further exploration. The presence of either PD-1/PD-L1 or MYD88 mutation is associated with a poor prognosis. This study further supports the rational for trials with TLR/NFkB pathway inhibitors and checkpoint inhibitors that penetrate the blood brain barrier.

We acknowledge the Vanderbilt Translational Pathology Shared Resource (TPSR) supported by NCI/NIH Cancer Center Support Grant 2P30 CA068485-14 and the Vanderbilt Mouse Metabolic Phenotyping Center Grant 5U24DK059637-13.