Clonal Hematopoiesis in Older Women with Extremely Skewed, Non-Random X-Inactivation Is Associated with Previous Cardiovascular and Cerebrovascular Events

Age-related clonal hematopoiesis (CH) is a common condition that is associated with an increased risk of hematologic malignancies (HM) and cardiovascular disease (CVD). The majority of candidate driver mutations occur in epigenetic regulatory genes such as ASXL1, DNMT3A, and TET2 genes. However, a significant proportion of older people harbor clonal hematopoiesis without candidate driver mutations. In older women, clonal hematopoiesis can also be detected by the human androgen receptor A gene (HUMARA) assay regardless of the presence or absence of candidate driver mutation(s). The HUMARA assay evaluates non-random X-inactivation (NRX-I) as a marker for clonal hematopoiesis. The purpose of this study is to evaluate the association between NRX-I and cardiovascular risk factors and other health correlates.

We screened for NRX-I in 904 women ages 65 and older participating in an ongoing, prospective cohort study in Oregon (Women Engaged in Advancing Health Research [WEAR] study). This approach was used to enhance the investigation of changes in CH that occur over time that may prove to be impactful for health outcomes. We examined the HUMARA results and any association with previous health history using a cross-sectional study design. Analysis of variance and logistic regression analyses were used to examine the relationship between HUMARA assay results and baseline CVD risk, controlling for age and race. HUMARA assay results were quantified and 3 groups were established: random X-inactivation, NRX-I, and extremely skewed NRX-I.

No significant differences were detected between groups with respect to age, race, education, reproductive health indices or body mass index. With respect to cardiovascular risk factors, women with extreme NRX-I were more likely to be lifetime non-smokers compared to women without NRX-I and those with NRX-I with less skewing (P: 0.049), though no difference was seen in the proportion of current smokers (P: 0.213) and the total pack years smoked by those with a smoking history (P: 0.846).The proportion of subjects reporting statin or other cholesterol lowering medication did significantly differ between groups, with 24.3% women without NRX-I reporting statin use, versus 30.4% in women with NRX-I, and 36.7% in women with NRX-I extreme skewing (P: 0.005). Post-hoc pairwise tests showed significant differences between women with no NRX-I and women with NRX-I extreme skewing (P: 0.004). Cardiovascular event history differed significantly between the three groups of interest. 5.6% of women without NRX-I reported a history of transient ischemic attack (TIA), cerebral vascular accident (CVA), or myocardial infarction (MI) versus 7.0% in the NRX-I group and 11.0% in the NRX-I with extreme skewing group (P: 0.05). Post-hoc pairwise tests showed significant differences between women with no NRX-I and women with NRX-I extreme skewing (P: 0.043). The adjusted odds of TIA, CVA, or MI more than doubled (OR: 2.15, 95%CI: 1.14-3.42) among women with extremely skewed HUMARA results compared to women with NRX-I, controlling for age, high cholesterol, smoking history, hypertension, and type II diabetes. Cholesterol was also found to be independently associated with TIA, CVA, and MI. Women with hypercholesterolemia were 1.84 times as likely to report a history of TIA, CVA, or MI compared to women without hypercholesterolemia (95% CI: 1.06, 3.20).

In summary, prediction of major adverse cardiac events is based on the presence of traditional risk factors including high blood pressure, high cholesterol, uncontrolled diabetes, smoking, and family history. Yet there is significant residual risk; many will still die from CVD without these risk factors. NRX-I, in addition to enriching for mutations known to confer CVD and HM risk; may be a marker for additional and unique health risks. An association between NRX-I, high cholesterol, and history of TIA, CVA, and MI, supports the current biological hypothesis that clonal hematopoiesis, perhaps irrespective of the cause or underlying driver mutation, is a driver for CVD.