Association of Clonal Hematopoiesis of Indeterminate Potential (CHIP) with Incidence of Venous Thromboembolism Varies By Age: The Atherosclerosis Risk in Communities Study

Importance: Clonal hematopoiesis of indeterminate potential (CHIP) is associated with increased risk of multiple cardiovascular diseases, but its association with venous thromboembolism (VTE) is less clear. CHIP may increase the risk of VTE through several mechanistic pathways, including low-grade inflammation and platelet hyperactivity, endothelial dysfunction, and neutrophil extracellular traps. However, little is known about the association of CHIP with VTE based on human epidemiologic data.

Methods: This study investigated the association between CHIP and incident VTE, and evaluated whether this risk varies by age or CHIP driver gene. The study was conducted using prospective cohorts within the Atherosclerosis Risk in Communities (ARIC) Study, a population-based setting. Participants included two ARIC cohorts without prior history of VTE: a younger cohort (Visit 2 baseline, n=10,765, mean age 57.3 years) and an older cohort (Visit 5 baseline, n=3,980, mean age 75.7 years). Exposures included CHIP (variant allele frequency [VAF] ≥2%) and common gene-specific CHIP subtypes (DNMT3A, TET2, and ASXL1). As the maximum follow-up available for V5 was 8.1 years, we capped the follow-up for both cohorts at 8 years to make the results comparable. The primary outcome was incident VTE, defined as a first diagnosed pulmonary embolism (PE) and/or lower extremity deep vein thrombosis (DVT) during the follow-up. Outcome data was prospectively collected and adjudicated. Hospital records with an ICD code suggestive of VTE were independently reviewed by two physicians using standardized criteria, with differences resolved by discussion. We built Cox proportional hazard risk models, adjusted for age, sex, race (self-report), ARIC center, solid cancer diagnosis, and anticoagulant use at index date. We also performed two sensitivity analyses using the Fine-Gray competing risk model with death ± hematologic malignancies (HM) being the competing factor(s).

Results: After excluding those with prior VTE or HM at the index date, 10,765 participants from V2 with a mean age of 57.3 ± 6.1 and 3,980 participants from V5 with a mean age of 75.7 ± 5.2 were included and subsequently followed for VTE. We identified 1414 CH variants among 1081 individuals (10.0% of the cohort) from V2 and 1225 variants among 985 individuals (24.7% of the cohort) from V5. Of the individuals with CHIP in V2 and V5, 191 (17.7%) and 195 (19.8%) had more than one variant.

In the V2 younger cohort of 1081 individuals, VTE occurred in 12 CHIP carriers (1.1%) vs. 116 non-CHIP participants (1.2%) over a median follow-up of 8 years (HR = 0.86, 95% CI, 0.47-1.56, P = 0.61). adjusted HRs associated with DNMT3A, TET2, and ASXL1 were 0.69 (0.25 - 1.87), P = 0.46; 1.44 (0.53 - 3.91), P = 0.48; and 1.24 (0.17 - 8.89), P = 0.83, respectively.

In contrast, in the V5 older cohort of 985 individuals, VTE occurred in 44 (4.5%) of CHIP carriers vs. 96 (3.2%) of non-CHIP participants over a median follow-up of 7.1 years (HR = 1.45, 95% CI, 1.00-2.10, P = 0.049). The adjusted HRs for DNMT3A, TET2, and ASXL1 in the V5 cohorts were 1.38 (0.85 - 2.24), P = 0.19; 2.21 (1.25 - 3.91), P = 0.007; and 1.34 (0.42 - 4.25), P = 0.62, respectively. In our sensitivity analyses, using the Fine-Gray competing risk model and with either death or incident hematologic malignancies as competing factors, results were unchanged.

Conclusions: This study identifies an age-dependent, gene-specific risk of VTE in individuals with CHIP, with older adults and TET2 mutation carriers being at the highest risk. However, the absolute attributable risk is small and does not warrant routine screening for CHIP in those with VTE.