Precision medicine has long promised individualized treatments based on genetic profiles, but a fundamental challenge has been distinguishing between genes that cause disease versus those that determine how well someone responds to therapy. This distinction matters because the same genetic variant might simultaneously predispose someone to hypertension while also affecting how their blood pressure responds to medication—creating confounded treatment decisions.
Researchers developed a novel graphical modeling approach using electronic health records from 211,845 individuals, analyzing over 8.4 million genetic variants alongside 1.4 million blood pressure measurements and prescription records. Their framework mathematically separates three distinct genetic effects: disease susceptibility, medication selection patterns, and actual treatment response magnitude. The analysis identified previously unknown pharmacogenetic variants in genes SLC35F2, PKD1, and KCNIP4 that specifically influence how patients respond to angiotensin receptor blockers, independent of their underlying hypertension risk.
This methodological advance addresses a critical gap in pharmacogenomics research, where most studies conflate genetic effects on disease with genetic effects on treatment. The approach revealed that genetic influences on blood pressure are predominantly established before age 50, but identified 127 additional genetic loci that specifically affect blood pressure changes in later life. For longevity-focused adults, this suggests genetic testing could eventually guide not just disease risk assessment, but optimize medication selection and dosing strategies. However, the study's reliance on observational data and focus on blood pressure limits immediate clinical applications. The framework represents a significant step toward truly personalized medicine, though translation to clinical practice will require validation across diverse populations and additional therapeutic areas.