Abstract Body (Do not enter title and authors here):
Background:
While polygenic risk scores (PRS) and social determinants of health (SDOH) are established predictors of coronary artery disease (CAD), interactions between PRS and individual SDOH remain largely underexplored.

Research Question:
Do income, education, material deprivation, and social isolation modify the effect of PRS on incident CAD?

Methods:
We studied 409,877 participants aged 40-69 years from the UK Biobank free from CAD at baseline and with available genotyping and SDOH data. A CAD PRS consisting of 1,296,172 variants (GPSmult, PGS PGP000466) was calculated. Four SDOH were examined: income (defined as self-reported household income before tax), education (highest educational qualification), area-level material deprivation (quantified through the Townsend Deprivation Index), and social isolation (defined as living alone). Cox proportional hazards models were used to examine the associations between each SDOH and incident CAD. Interactions between PRS and each SDOH were assessed to understand whether social factors modify the effect of PRS on CAD risk.

Results:
Among 409,877 participants (mean age 56.0 ± 8.1 years, 53.5% female), 12,244 incident CAD events occurred over a median follow-up time of 13.7 years. Each standard deviation (SD) increase in PRS was associated with a 1.88-fold increase in incident CAD risk (95% CI: 1.84, 1.91). Lower income (Hazard Ratio (HR) 1.48, 95% CI: 1.33, 1.64), lower educational attainment (HR 1.34, 95% CI: 1.27, 1.41), and higher TDI (HR 1.18, 95% CI: 1.11, 1.25) were all independently associated with greater incident CAD risk. Living alone was not significantly associated with CAD risk (HR 1.04, 95% CI: 0.99, 1.09). The absolute CAD risk difference between the high and low income groups was 2.62% in the highest PRS decile, compared to only 0.77% in the lowest decile. Similarly, the absolute risk difference between college graduates and those without a high school diploma decreased from 1.40% in the highest PRS decile to 0.68% in the lowest. (Figure). Multiplicative interactions between PRS and both income (P_interaction= 4.9×10^-5) and education (P_interaction= 6.1×10^-6) were statistically significant.

Conclusion:
These findings suggest a complex interaction between genetic risk and SDOH in relation to CAD risk, particularly for income and education. Integrating these insights into predictive models could enable more accurate risk stratification and inform targeted interventions for those at greatest risk.