Clonal hematopoiesis of indeterminate potential (CHIP) and cardiovascular toxicity from androgen deprivation therapy in men with prostate cancer.

Abstract

312 Background: Androgen deprivation therapy (ADT) with agonists of the gonadotropin releasing hormone (GnRH) receptor (e.g. leuprolide) is associated with coronary artery disease and major adverse cardiovascular (CV) events. CHIP, an age-related expanded somatic blood cell clone without other hematologic abnormalities, is associated with inflammation and accelerated atherosclerosis. Whether CHIP is associated with adverse CV outcomes in patients receiving ADT is unknown. Methods: Twenty-five men with localized prostate cancer pursuing radiotherapy plus ≥ 6 months of leuprolide were enrolled on a prospective biorepository. Using pre-treatment blood samples, whole exome sequencing libraries were generated; identified variants were filtered to retain <1% MAF variants found amongst the three most common CHIP genes ( DNMT3A, TET2, ASXL1). Pre-treatment and 6-month blood samples from each patient subsequently were analyzed using a commercially available, validated 4-protein risk score that is predictive of 4-year major adverse CV events (Prevencio Inc HART CVE, Seattle WA). Multivariable logistic regression measured the association of CHIP with increased CV risk after initiation of leuprolide, defined as a clinically significant rise in 4-protein risk score at six months compared with baseline. Results: Median age was 70 years (IQR 65-76), 10 men (40%) were Black, and 4 men (16%) had a history of myocardial infarction. CHIP gene variants were found in 7 men (28%). Amongst the men with CHIP variants, 6 (85.7%) were Black. Amongst the entire cohort, 11/25 (44%) men had a clinically significant rise in CV risk score after six months of ADT. These men, compared with those who did not experience a rise in CV risk, were more often Black (72.7% versus 14.3%), on anti-hypertensive medication (72.7% versus 14.3%), and had higher baseline LDL cholesterol level (median 98 versus 81 mg/dL). On univariate analysis, harboring ≥1 CHIP variant was associated with a significantly increased odds of having an elevated CV risk score after 6 months of ADT. On multivariable analysis, after adjusting for race, anti-hypertensive medication use, and baseline LDL cholesterol, harboring ≥1 CHIP variant was associated with increased odds of having an elevated CV risk score after 6 months of ADT (adjusted OR 1.44, 95% CI 0.88-1.83), which did not meet statistical significance but limited by small sample size. Conclusions: Six months of GnRH-agonist leuprolide resulted in a rise in CV risk score in 44% of this cohort. CHIP variants, identified in 28% of this cohort, was associated with increased odds of experiencing a rise in CV risk score. Notably, Black men were more likely to harbor CHIP and experience increased CV risk after 6 months of leuprolide, respectively. Larger studies are underway to determine if CHIP variants may serve as a biomarker of ADT-induced CV risk, especially in older Black men.