M. H. Shahin,1 D. M. Rotroff,2 Y. Gong,1 T. Langaee,1 C. W. McDonough,1 A. L. Beitelshees,3 T. J. Garrett,4 A. B. Chapman,5 J. G. Gums,1 S. T. Turner,6 A. Motsinger-Reif,2 R. F. Frye,1 S. E. Scherer,7 W. Sadee,8 O. Fiehn,9 R. M. Cooper-DeHoff,1 R. Kaddurah-Daouk,10 J. A. Johnson1; 1Department of Pharmacotherapy and Translational Research, College of Pharmacy, University of Florida, Gainesville, FL, 2Bioinformatics Research Center, North Carolina State University, Raleigh, NC, 3Department of Medicine, University of Maryland, Baltimore, MD, 4Department of Pathology, Immunology, and Laboratory Medicine, College of Medicine, University of Florida, Gainesville, FL, 5Department of Medicine, Emory University, Atlanta, GA, 6College of Medicine, Mayo Clinic, Rochester, MN, 7Human Genome Sequencing Center, Baylor College of Medicine, Houston, TX, 8Program in Pharmacogenomics, Department of Pharmacology, The Ohio State University, Columbus, OH, 9Genome Center, University of California at Davis, Davis, CA, 10Department of Psychiatry and Behavioral Sciences, Duke University, Durham, NC

BACKGROUND: Hydrochlorothiazide (HCTZ) is among the most commonly prescribed antihypertensives in the US, yet less than 50% of HCTZ treated patients achieve blood pressure (BP) control. In this study, we used a genomics-metabolomics integrative approach to identify novel biomarkers of HCTZ BP response.
METHODS: This study included 228 White hypertensive PEAR participants with BP determined at baseline and after 9 weeks of HCTZ treatment. Genome-wide genotyping was determined via Illumina Omni 1M Quad Chip. Untargeted metabolomics analysis was performed on fasting plasma samples using a GC TOF MS platform. Pathway analysis was conducted to integrate the BP GWAS signals at P < 1x10-4 with the metabolomics findings. Gene expression was also tested using RNA-Seq in extreme BP responders (25 responders and 25 non-responders).
RESULTS: Metabolomics analysis revealed 212 known metabolites, of which 13 were significantly associated with systolic and diastolic BP response (FDR < .05). Integrative pathway analysis identified metabolites in the platelet activation pathway (p = .009) and Rho Kinase 1 gene (ROCK1) as a potential factor influencing HCTZ BP response. ROCK1 rs8085654 variant carriers had a poor BP response vs. non carriers (∆SBP/∆DBP: -5.8/-2.7 vs. -10.7/-5.9 mmHg, respectively, ∆SBP p = 5x10-5 and ∆DBP p = 9x10-4). Additionally, ROCK1 baseline expression levels were significantly different between HCTZ BP responders vs. non responders (23.7±5.8 vs 20.0±3.2 FPKM, respectively, p = .01).
CONCLUSION: These results align with recent animal studies showing ROCK1 contribution to increased BP. Moreover, this study highlights the strength of using different omics to identify novel biomarkers of drug response, and suggests that ROCK1 might be an important determinant of HCTZ BP response.