Genetic Profiling Reveals the Distinctions Among MTX-Associated DLBCL, EBV-Positive Mucocutaneous Ulcer, and EBV + DLBCL

Abstract

The WHO recently changed the outline of immunodeficiency/dysregulation (IDD)-associated lymphoproliferative disorders (LPDs)/lymphomas from underlying IDD settings to an overarching framework and accommodates commonalities in histology, the involvement of various oncogenic viruses, and specific clinical/therapeutic consequences. A mutational analysis has been performed on post-transplantation and HIV-positive lymphomas, but not on other iatrogenic immunodeficiency (OII)-associated LPDs mainly caused by methotrexate (MTX) to treat rheumatoid arthritis. We herein conducted next-generation sequencing (NGS) to examine the genetic spectrum along with a fluorescence in situ hybridization analysis of 9p24.1 and PD-L1 expression in 37 MTX-associated diffuse large B-cell lymphoma (DLBCL) cases, 17 Epstein-Barr virus (EBV)-positive mucocutaneous ulcer (EBVMCU) cases, and 26 EBV-positive DLBCL (EBV + DLBCL) cases. Targeted NGS identified 177 mutations. The mutation frequency was significantly higher in EBV^- MTX-DLBCL than in EBV-positive LPDs/lymphomas (EBVMCU, EBV⁺ MTX-DLBCL, and EBV + DLBCL). Regrowth or resistance to spontaneous regression after MTX withdrawal was more likely in EBV^- MTX-DLBCL than in EBV⁺ MTX-DLBCL. Therefore, accumulated gene mutations, sustained by the restored immune status in EBV^- MTX-DLBCL, may affect clinical outcomes after MTX discontinuation. Several unique genetic findings were obtained for each category. Fewer TET2/DNMT3A and CD58 mutations in OII-LPD/lymphomas (EBVMCU and MTX-DLBCL) than in EBV + DLBCL indicate that clonal hematopoiesis and an immune evasion-related background contributed less to lymphomagenesis in OII-LPDs. MYD88^L265P/CD79B^Y196 mutations were only detected in EBV^- MTX-DLBCL. SOCS1 mutations were significantly more frequent in EBV-positive LPD/lymphoma categories, irrespective of the immune status, than in EBV^- MTX-DLBCL. These results reveal distinct genetic features among MTX-DLBCL (EBV+/-), EBVMCU, and EBV + DLBCL.

FIGURE 1

Progression‐free survival in all cases or cases that had discontinued MTX in each entity or category after LPD onset. (A) Kaplan–Meier curves of PFS for all cases analyzed in this study. (B) Kaplan–Meier curves of PFS for EBVMCU and MTX‐DLBCL that had discontinued MTX after LPD onset. (C) Kaplan–Meier curves of PFS for EBVMCU and MTX‐DLBCL (EBV^+/−) that had discontinued MTX after LPD onset.

FIGURE 2

Mutational profile in each subtype of lymphomas/LPDs. (A) All non‐synonymous mutations detected are depicted for each sample category (n = 80), and genes are sorted by the gene pathway. Bar graphs at the top depict the number of called mutations per sample. Mutations are color‐coded. Age, sex, the EBV status, cell of origin (COO), primary immune disease (PID), and MTX use for each case are indicated at the bottom. EBVMCU, Epstein–Barr virus (EBV) –positive mucocutaneous ulcer; EBV + DLBCL, EBV‐positive diffuse large B‐cell lymphoma; MTX‐DLBCL, methotrexate‐associated DLBCL. (B) The average variant number per case in each category is shown. EBVMCU (yellow), EBV⁻ MTX‐DLBCL (ivory), EBV⁺ MTX‐DLBCL (orange), and EBV + DLBCL (blue). (C) The mutation frequency classified by biological pathways in each category is shown.

FIGURE 3

Genetic alterations in 9p24.1 and a PD‐L1 expression analysis of each subtype of lymphomas/LPDs. Summary of 9p24.1 alterations and H‐scores. Bar graphs show the percentages of tumor cells with 9p24.1 disomy (light gray), polysomy (gray), copy gain (light orange), and amplification (brown), plotted on the left y‐axis. The H‐score of tumor cells is shown as a red circle on the column for each case, plotted on the right side y‐axis.

FIGURE 4

Clinical courses and clinical outcomes of MTX‐DLBCL. (A) The clinical courses of MTX‐DLBCL (EBV⁻, light orange diamond; EBV⁺, brown triangle). Among the 37 cases (−, 17; +, 20) with MTX‐DLBCL, watchful waiting after the withdrawal of MTX was performed for 25 cases (−, 12; +, 13). Among them, 12 cases (−, 2; +, 10) had spontaneous regression (SR), whereas 12 (−, 10; +, 2) had non‐SR, including non‐regression, relapse, and/or regrowth. (B) Kaplan–Meier curves of the PFS of 12 SR cases and 12 non‐SR cases according to the status of EBV infection (top left), TP53 mutation (top right), PIM1 mutation (middle left), CREBBP/EP300 mutation (middle right), MYD88/CD79B mutation (bottom left), and pathway mutation (bottom right). The x‐axis presents the time from MTX discontinuation (month) after the diagnosis of MTX‐DLBCL.

FIGURE 5

Comparison of mutational profiles between MTX‐DLBCL and PT‐DLBCL. The twenty genes that were commonly analyzed in the present study and in a previous study by Menter et al. with respect to the various cases: EBV⁻ MTX‐DLBCL (light orange), EBV⁻ PT‐DLBCL (light green), EBV⁺ MTX‐DLBCL (brown), and EBV⁺ PT‐DLBCL (green). The mutation frequency is plotted on the y‐axis.

FIGURE 6

Overview of Genetic and Immunophenotypic Features in IDD‐LBCL and EBV + DLBCL. The classification framework is shown in the upper part and the genetic and immunophenotypic features for each disease category are presented in the lower part. CH, Clonal hematopoiesis; LBCL, large B‐cell lymphoma; LPDs, lymphoproliferative disorders; OII, other iatrogenic immunodeficiency; MTX, methotrexate.