American Heart Association, Circulation, Suppl_1(147), 2023
DOI: 10.1161/circ.147.suppl_1.27
Full text: Unavailable
Background: Polygenic risk scores (PRS) have shown promise in complementing existing clinical risk factors and improving early diagnosis of cardiovascular disease. Recently, several studies have developed PRS for blood pressure traits; however, few have examined to what extent these PRS predict antihypertensive (AHT) efficacy. Hypothesis: We hypothesize that applying a systolic blood pressure (SBP) PRS developed and trained in multi-ancestral observational studies will predict response to chlorthalidone (CHL), as well as non-response to AHT treatment (apparent treatment resistant hypertension, aTRH), among African Americans (AA). Methods: We applied an optimized multi-ancestry PRS generated in a pooled Trans-Omics for Precision Medicine (TOPMed) cohort of >21,000 adults for SBP (PRS SBP ) to 4,297 AA Genetics of Hypertension Associated Treatment (GenHAT) participants randomized to CHL as part of the Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) with genome wide association study data. We then used linear regression to test the association of PRS SBP quintiles with CHL response (6 month BP - baseline BP) adjusting for age, sex, genetic ancestry, and baseline SBP. For the analysis of aTRH, cases were defined as individuals (a) treated with 3 different AHT classes, with average BP ≥140/90 mmHg at year 3 follow-up or (b) ≥4 AHT classes regardless of BP (cases, n=286). Treated controls were defined as individuals with BP <140/90 mmHg and on trial drug monotherapy at year 3 (n=2,391). Logistic regression models adjusting for age, sex, and genetic ancestry were used to calculate the odds ratio for aTRH comparing the highest and lowest quintile (Q5 and Q1, respectively). Results: The PRS SBP was associated with reduced SBP response in Q5 (-5.01 [-6.15, -3.87] mmHg) versus Q1 (-8.24 [-9.39, -7.09] mmHg) and Q2 (-7.25 [-8.39, -6.11] mmHg). In regard to aTRH, Q5 was nominally associated with higher odds of aTRH compared to Q1 (1.48 [1.00, 2.20]). Conclusions: We found that a general SBP PRS was associated with BP response in AAs from GenHAT. On average, a greater SBP reduction was found in participants at low genetic risk (Q1 or Q2) compared to the high risk (Q5) taking the same AHT. Similarly, when compared to participants in the bottom 20% of the distribution, those in the top 20% had increased odds of aTRH. Additional work using the PRS SBP in other ancestral populations, as well as developing novel AHT class treatment response PRS, is warranted and in progress to determine whether we can identify individuals who would benefit the most from specific AHT classes.