Background
CHIP (clonal hematopoiesis of indeterminate potential) mutations (hereafter ‘CHIP’) are acquired somatic DNA mutations that are associated with increased risk of cancer, mortality, and vascular disease. People with treated and suppressed HIV (PWH) have higher frequencies of CHIP than people without HIV (PWoH), but the underlying mechanisms linking CHIP and clinical events remains unclear among PWH or PWoH.
Methods
We sampled 1000 PWH on ART from the U.S Military HIV Natural History Study using a case-cohort design: 200 PWH with incident vascular events (VE) and 800 PWH without VE. CHIP (e.g., DNMT3A, TET2, ASXL1) were quantified by PCR from PBMCs and cytokines were measured from plasma (MesoScale). We analyzed two composite clinical outcomes: atherogenic VE (AVE: coronary artery disease, myocardial infarction, cerebrovascular accident, or peripheral arterial disease) and VE (deep vein thrombosis, pulmonary embolus, or AVE). Cox proportional hazard models were performed to estimate the risk of AVE/VE, including covariates for traditional vascular risk factors and HIV disease progression (Fig 1). Linear mixed effects models were fit to estimate 2-fold changes in cytokines associated with CHIP.
Results
The median baseline age was 39, and 92% of participants were male. 56 of 975 sequenced participants (5.7%) had detectable CHIP (3 participants had >1 CHIP mutation). The most common CHIP variant was in DNMT3A. CHIP was associated with incident AVE (HRAVE=2.24, 95%CI=1.19-4.24) and VE (HRVE=2.21, 95%CI=1.24-3.91), even after adjustment for hyperlipidemia and hypertension (aHRAVE=1.81, 95%CI=0.99-3.31; aHRVE=1.82, 95%CI=1.07-3.10). However, this effect was largely driven by the middle age tertile and did not meet statistical significance after age adjustment (Figure 1A). This may be due to lower incidence of CHIP and VE in the lowest age tertile and competing risks in the highest tertile. CHIP was associated with several cytokines (IL-6, IP-10, MCP-3, MIP-3a) of which IP-10 remained statistically significant after adjustment for age (Fig 1B).
Conclusions
CHIP mutations were associated with incident VE/AVE in PWH on ART. These data are the first to highlight circulating immune pathways potentially involved in CHIP-mediated pathogenesis, especially in PWH (e.g., IP-10 is a critical proinflammatory factor responding to IFN-γ in HIV disease). Further studies are needed to compare the risk of CHIP in PWH vs. PWoH and determine whether the identified cytokine pathways are unique to PWH.