Epigenome-Wide Association Studies of Proteasome Inhibitor-Related Cardiotoxicity in Patients with Multiple Myeloma

<b>Background/Objectives</b>: Carfilzomib (CFZ) and bortezomib (BTZ) are proteasome inhibitors used as the first-line therapy for relapsed or refractory multiple myeloma (MM) but are associated with cardiovascular adverse events (CVAEs). This study aims to identify differentially methylated positions (DMPs) and regions (DMRs), and enriched pathways associated with CVAEs related to CFZ or BTZ-based treatment. <b>Methods</b>: Baseline germline DNA methylation profiles from 79 MM patients (49 on CFZ and 30 on BTZ) in the Prospective Study of Cardiac Events During Proteasome Inhibitor Therapy (PROTECT) were analyzed. Epigenome-wide analyses were performed within each group, followed by meta-analyses to identify signals common to CVAEs associated with both medicines. <b>Results</b>: Four DMPs were significantly associated with CFZ-CVAEs, including cg15144237 within <i>ENSG00000224400</i> (<i>p</i> = 9.45 × 10^-10), cg00927646 within <i>TBX3</i> (<i>p</i> = 9.78 × 10^-8), and cg10965131 within <i>WDR86</i> (<i>p</i> = 1.00 × 10^-7). One DMR was identified in the <i>FAM166B</i> region (<i>p</i> = 5.46 × 10^-7). There was no evidence of any DMPs in BTZ-treated patients, however two DMPs and one DMR reached a suggestive level of significance (<i>p</i> < 1.00 × 10^-5): cg09666417 in <i>DNAJC18</i> (<i>p</i> = 3.41 × 10^-7) and cg12987761 in <i>USP18</i> (<i>p</i> = 5.00 × 10^-7), and a DMR mapped to the <i>WDR86/WDR86-AS1</i> region (<i>p</i> = 8.11 × 10^-8). Meta-analysis did not find any significant DMPs, with the top CpG being cg17933807 in <i>GNL2</i> (<i>p</i> = 7.38 × 10^-5). Pathway enrichment analyses identified peroxisome, MAPK, Rap1, adherens junction, phospholipase D, autophagy, and aldosterone-related pathways to be implicated in CVAEs. <b>Conclusions</b>: Our study identified distinct DMPs, DMRs, and pathways enrichment associated with CVAE, suggesting epigenetic contributors to CVAEs and supporting the need for larger validation studies.