Clonal Hematopoiesis and Cardiovascular Outcomes in the Elderly

Background Clonal hematopoiesis of indeterminate potential (CHIP) is an emerging aging-related risk factor for cardiovascular disease (CVD). However, previous studies suggest that CHIP’s relevance to CVD may diminish with advancing age. Purpose To test the association of CHIP (subtypes) with coronary heart disease (CHD) and other common CVD outcomes in an elderly population. Methods Participants in the Women’s Health Initiative (WHI) Long Life Study (LLS) completed study assessments approximately 15 years after enrollment in the main WHI study. Individuals with prevalent hematologic malignancy were excluded. CHIP status was ascertained using single-molecule molecular inversion probe sequencing (median coverage depth 4,609x) of peripheral blood DNA. Given the heterogeneity of CVD associations among major CHIP drivers and stronger associations previously observed for TET2 CHIP, the co-primary exposures were (1) composite CHIP and (2) TET2 CHIP, with CHIP defined as a variant allele fraction of ≥2%. DNMT3A, non-DNTM3A, ASXL1, and JAK2 CHIP were examined as secondary exposures. Individuals without CHIP constituted the reference group. The primary outcome was incident CHD. Secondary outcomes were incident heart failure with preserved (HFpEF) and reduced ejection fraction (HFrEF), ischemic stroke, and cardiovascular death. Outcomes were ascertained using standardized physician review, classification, and adjudication. Cox models tested associations between CHIP and CVD with follow-up starting at the LLS visit, adjusting for age, race, body mass index, diabetes, total cholesterol, high-density lipoprotein cholesterol, systolic blood pressure, cholesterol-lowering and antihypertensive medication use, and smoking status. Results Our analytic cohort included 6,572 women (median [interquartile range] age 80 [73-85] years, 2,123 [32.3%] with any CHIP at the LLS visit) followed up for a median of 10.2 (7.1-10.8) years. TET2 CHIP was independently associated with incident CHD (HR, 1.39 [95% CI, 1.07-1.81]; P=0.015). In secondary analyses, TET2 CHIP was also associated with HFpEF (HR, 1.50 [95% CI, 1.07-2.12]; P=0.02), ASXL1 CHIP with HFrEF (HR, 3.81 [95% CI, 1.74-8.33]; P<0.001), and JAK2 CHIP with ischemic stroke (HR, 2.68 [95% CI, 1.26-5.71]; P=0.01) and cardiovascular death (HR, 2.69 [95% CI, 1.70-4.26]; P<0.001). No significant associations were observed with composite or DNMT3A CHIP (Table 1). Findings were consistent in sensitivity analyses incorporating non-cardiovascular death as a competing risk and, for secondary outcomes, in analyses further adjusted for prevalent CHD. Conclusion In this cohort of elderly women, key subtypes of CHIP (TET2, ASXL1, and JAK2) were independently associated with incident CVD, with evidence of heterogeneous associations across CHIP driver mutations. These findings validate previous observations in younger cohorts and suggest that CHIP continues to be associated with cardiovascular health well into later life.Associations between CHIP and CVD