Serum amyloid A as a predictor of coronary fatty plaque progression: a comparative analysis of cardiometabolic biomarkers

Introduction Despite intensive lipid lowering, substantial risk for coronary atherosclerotic progression and events persists, highlighting significant residual cardiovascular risk. Chronic low-grade inflammation drives atherogenesis, plaque progression, and destabilization, even in statin-treated patients. Factors that drive plaque composition may differ from those determining overall plaque burden, yet the magnitude of contribution of inflammation and dyslipidemia to plaque progression remains unclear. This study examined whether inflammatory and lipid biomarkers predict fatty versus total plaque progression. Methods A secondary analysis of 240 statin-treated participants in the Slowing HEART diSease with lifestyle and omega-3 fatty acids (HEARTS) randomized trial evaluated omega-3 fatty acid therapy over 30 months in patients with stable coronary artery disease (CAD). Coronary computed tomographic angiography (CCTA) quantified fatty and total plaque progression, and lipid and inflammatory biomarkers were measured at baseline and follow-up. Biomarkers were standardized per 1 standard deviation for direct comparison. Multivariable linear models were adjusted for age, sex, diabetes, hypertension, baseline plaque burden, and omega-3 index (assessed at 12 months). Incremental explained variance (ΔR²) assessed the added predictive value of each biomarker. Nested models compared serum amyloid A (SAA) with high-sensitivity C-reactive protein (hs-CRP) and assessed independence from lipid burden. Results For fatty plaque progression, SAA had the strongest association (β = 19.9, p < 0.001) and incremental value (ΔR² = 0.08), exceeding total cholesterol (β = 15.4, p = 0.019; ΔR² = 0.025), non-HDL-C (β = 14.6, p = 0.036; ΔR² = 0.020), and LDL-C (β = 13.5, p = 0.056; ΔR² = 0.016); hs-CRP was not significantly associated (β = 10.8, p = 0.070; ΔR² = 0.015) (Table 1). For total plaque progression, lipid measures were stronger predictors compared to inflammatory markers: total cholesterol (β = 16.0, p = 0.004; ΔR² = 0.032), non-HDL-C (β = 13.5, p = 0.026; ΔR² = 0.020), LDL-C (β = 13.1, p = 0.033; ΔR² = 0.019), and lipoprotein (a) (β = 11.4, p = 0.035; ΔR² = 0.018). SAA remained independently associated (β = 10.7, p = 0.008; ΔR² = 0.028), whereas hs-CRP was not (β = 4.75, ΔR² = 0.004, p = 0.36) for total plaque progression. Nested models confirmed SAA improved fit for fatty (ΔR² = 0.08, p < 0.001) and total plaque (ΔR² = 0.028, p = 0.008), while hs-CRP added no independent value. SAA remained significant after adjusting for non-HDL-C (fatty β = 19.02, p < 0.001; total β = 9.81, p = 0.013). Conclusion SAA strongly predicts fatty plaque progression, while lipid measures, particularly total cholesterol, drive total plaque burden. These findings suggest distinct inflammatory and lipid-mediated pathways of residual atherosclerotic risk, with SAA capturing inflammation beyond hs-CRP and independent of lipid levels.