EBV-associated primary CNS lymphoma occurring after immunosuppression is a distinct immunobiological entity

Abstract

Primary central nervous system lymphoma (PCNSL) is confined to the brain, eyes, and cerebrospinal fluid without evidence of systemic spread. Rarely, PCNSL occurs in the context of immunosuppression (eg, posttransplant lymphoproliferative disorders or HIV [AIDS-related PCNSL]). These cases are poorly characterized, have dismal outcome, and are typically Epstein-Barr virus (EBV)-associated (ie, tissue-positive). We used targeted sequencing and digital multiplex gene expression to compare the genetic landscape and tumor microenvironment (TME) of 91 PCNSL tissues all with diffuse large B-cell lymphoma histology. Forty-seven were EBV tissue-negative: 45 EBV- HIV- PCNSL and 2 EBV- HIV+ PCNSL; and 44 were EBV tissue-positive: 23 EBV+ HIV+ PCNSL and 21 EBV+ HIV- PCNSL. As with prior studies, EBV- HIV- PCNSL had frequent MYD88, CD79B, and PIM1 mutations, and enrichment for the activated B-cell (ABC) cell-of-origin subtype. In contrast, these mutations were absent in all EBV tissue-positive cases and ABC frequency was low. Furthermore, copy number loss in HLA class I/II and antigen-presenting/processing genes were rarely observed, indicating retained antigen presentation. To counter this, EBV+ HIV- PCNSL had a tolerogenic TME with elevated macrophage and immune-checkpoint gene expression, whereas AIDS-related PCNSL had low CD4 gene counts. EBV-associated PCNSL in the immunosuppressed is immunobiologically distinct from EBV- HIV- PCNSL, and, despite expressing an immunogenic virus, retains the ability to present EBV antigens. Results provide a framework for targeted treatment.

Graphical abstract

Figure 1.

Consort diagram providing details of PCNSL subtype and sample testing performed. ACSRB, AIDS Cancer Specimen Resource Bank; AH, Austin Hospital; NNTC, NeuroAIDS National Tissue Collection; PAH, Princess Alexandra Hospital; WH, Westmead Hospital. *One of the 2 EBV⁻ HIV⁺ PCNSL tissues underwent sequencing only, and both had COO. **Brain TK cell-line underwent targeted sequencing only.

Figure 2.

Mutational landscape of PCNSL, according to EBV tissue and HIV serological status. (A) Each column in this plot represents an individual case (with mutation[s] in the displayed genes) of the final sequencing cohort (n = 79), across the 4 tissue subtypes: EBV⁻ HIV⁻ PCNSL, EBV⁻ HIV⁺ PCNSL, EBV⁺ HIV⁺ PCNSL, and EBV⁺ HIV⁻ PCNSL. Mutated genes constitute individual rows and are sorted according to their mutational frequencies of mutated cases as provided on the far right. Mutation types are color coded as indicated in the key; red* brain lymphoma TK cell-line, red** patients without PTLD/iatrogenic immunosuppression. (B) Stacked histograms show the percentage (percentages rounded to whole numbers) of cases with mutations in MYD88, CD79B, PIM1 by EBV-tissue and HIV serological status across the 3 main subtypes. Because EBV⁻ HIV⁺ PCNSL represented only 1 sequenced case, aggregate data are not shown. (C) Number of mutated genes observed using the targeted sequencing panel in EBV⁻ HIV⁻ PCNSL, EBV⁺ HIV⁺ PCNSL, and EBV⁺ HIV⁻ PCNSL, with P values for paired subtypes: *P ≤ .05; **P ≤ .01; ***P ≤ .001; ****P ≤ .0001.

Figure 3.

Molecular COO and copy number alterations in HLA class I/II alleles and antigen presentation/processing genes, separated into the PCNSL subtypes. (A) Stacked histograms show the relationship between the molecular COO classification (using the Lymph2CX assay) across 3 subtypes. Because EBV⁻ HIV⁺ PCNSL represented only 2 COO cases, aggregate data are not shown. UC, unclassified. (B) Each column represents an individual case across the 4 PCNSL subtypes. Copy number altered genes constitute individual rows, with gains and losses shown in gold and blue, respectively. On the far right, percentages indicate the proportion of PCNSL tumors with copy number alterations of the specified gene; red* brain lymphoma TK cell-line, red** patients without PTLD/iatrogenic immunosuppression.

Figure 4.

Combined mutations and copy number loss separated into the 2 EBV-tissue positive PCNSL subtypes and EBV⁻HIV⁻PCNSL. Mutations and/or CN loss were categorized into pathway categories. CN loss is shown for HLA class I/II. For antigen presentation/processing mutations were combined with CN loss. Mutations for immune function, BCR-NF-κB signaling, epigenetic regulation, cell cycle/adhesion and B-cell differentiation. Percentages are shown for each PCNSL subtype, with P values for paired subtypes: *P ≤ .05; **P ≤ .01; ***P ≤ .001; ****P ≤ .0001. There were 42, 20, and 16 cases of EBV⁻ HIV⁻, EBV⁺ HIV⁺, and EBV⁺ HIV⁻ PCNSL, respectively (data for the single EBV⁻ HIV⁻ PCNSL tissue is not shown).

Figure 5.

The tumor microenvironment in the 2 EBV tissue-positive PCNSL subtypes and EBV⁻HIV⁻PCNSL. A targeted gene expression panel was chosen of selected clinically pertinent immune effectors, macrophages, and immune-checkpoint markers. Gene counts are shown for each PCNSL subtype, with P values for paired subtypes: *P ≤ .05; **P ≤ .01; ***P ≤ .001; ****P ≤ .0001. There were 34, 18, and 18 cases of EBV⁻ HIV⁻, EBV⁺ HIV⁺, and EBV⁺ HIV⁻ PCNSL, respectively.

Figure 6.

Pathogenesis of EBV tissue-positive PCNSL in the immunosuppressed, and potential implications for immunotherapy. (A) Mutations in the NF-κB and cell-cycling pathways and deletion of HLA class I/II molecules in EBV⁻ HIV⁻ PCNSL. (B) Viral transformation in EBV tissue-positive PCNSL, with the loss of host immunity providing no selection pressure to prevent presentation of EBV-immunogenic peptides by HLA I/II molecules. (C) EBV-specific viral specific T cells (VST) cross the blood–brain barrier to target the EBV⁺ malignant B cell.