Cancer Research • 2024-04-07

Genomic distinctions of HIV- and EBV-associated DLBCL in a diverse African cohort

Authors

Katherine Antel¹, Aleksander Nesmelov², Meng-Xuan Wu¹, Charlotte Slome³, Dharshnee Chetty³, Raymond Kriel³, Ekaterina Postovalova², Robert Redd¹, Martin Aryee¹, Delphine Armand¹, Lauren McGrath¹, Nina Xiong¹, Jenna Oosthuizen³, Karryn Brown³, Andrey Tyshevich², Svetlana Podsvirova², Felix Frenkel², Natalia Kuzkina², Jessica Scher², Cagdas Tazearslan², Ksenia Zornikova², Nikita Kotlov², Aleksander Bagaev², Nathan Fowler², Arnold Freedman¹, Scott Rodig4, Estelle Verburgh³, Mark A. Murakami¹
  1. Dana-Farber Cancer Institute, Boston, MA, US
  2. BostonGene, Corp., Waltham, MA, US
  3. University of Cape Town, Cape Town, South Africa, Brigham and Women's Hospital, Boston, MA, US

Abstract

Lymphoma, a leading cancer cause of death among people living with HIV (PLHIV) globally, poses a pressing public health challenge in Sub-Saharan Africa (SSA), the epicentre of the HIV epidemic. Despite its prevalence, the genetic underpinnings of lymphoma in Black Africans and PLHIV remains poorly understood, contributing to regional disparities. This study aims to elucidate the impact of HIV status and EBV on the genetic landscape, molecular subtypes and survival outcomes of diffuse large B-cell lymphoma (DLBCL) in ethnically diverse SSA patients.
We established a cohort of 117 newly diagnosed DBLCL cases (55 HIV+, 21 EBV+, 62 Black Africans), with histopathologic and clinical annotation from a public teaching hospital in Cape Town, South Africa (2010-22). Tumor specimens underwent histologic review, whole exome sequencing (WES) and bulk whole transcriptome sequencing. Germline variants were filtered by paired normal WES (when available) and normal references augmented by supervised machine learning in tumor-only specimens. Aberrant somatic hypermutation (ASHM) was assessed in curated genes within paired samples. Microsatellite instability (MSI) was assessed via MSI-sensor.
Canonical genetic aberrations reproduced in this cohort included recurrent SNVs in TP53, CARD11, TET2, SOCS1, MYD88, and KMT2D with differences in relative proportion compared to the NCI consortium cohort being more attributable to HIV or EBV status than ethnicity. HIV+ cases harbored frequent mutations in TET2, BRAF, and KRAS, with fewer in NFKB pathway genes including CARD11, IKBK, and BTK. ABC-type HIV+ cases manifested higher expression of genes in the JAK-STAT pathway. EBV+ cases exhibited distinct genomic features including amplifications in NOTCH1, CARD11, NFKBIE, MYC, CD274, and JAK2, as well as frequent gain-of-function (GOF) mutations in the NOTCH1 PEST domain (40% vs 3.8% q=0.04). Tumor mutation burden was consistent across ethnic and virological strata (median mutations 139-167, p>0.05). Notably, ASHM was not identified as a predominant mutational signature in either HIV-positive or negative cases, and 4 HIV+ DLBCL cases exhibited MSI, corroborated by protein-affecting mutations in mismatch repair genes. The 2-year overall survival (OS) was less than 40%, markedly inferior to published Western cohorts, even when controlled for clinical prognostic factors and treatment regimen.
This study highlights genetic and biological distinctions in DLBCL due to HIV and EBV. Key findings include the absence of ASHM correlates of pathogenic driver mutations, challenging a prevailing hypothesis of HIV-associated lymphomagenesis, and the identification of MSI in HIV+DLBCL. The dismal survival of this cohort and distinct genomic features, especially in EBV+DLBCL, highlight the urgency of further biological interrogation and inclusive clinical trials for these demonstrably underserved populations.
Read full publication Download