Abstract P096: Association of Cystatin C Gene Cluster Variants with Cystatin C levels, Chronic Kidney Disease, and Risk of iIncident Cardiovascular Disease and Mortality

Abstract

Background: Accounting for genetic variation in cystatin C production may improve glomerular filtration rate estimation (eGFR) and the classification of chronic kidney disease (CKD). As the majority of known SNPs influencing cystatin C tend to result in lower circulating levels, we hypothesized that accounting for genetic variation would result in higher eGFR calculated from cystatin C level (eGFRcys) and a lower prevalence of CKD. Furthermore, we hypothesized that adjusting eGFRcys for the genetic component would strengthen the associated cardiovascular and mortality risk estimates. Methods: Four prospective cohorts were included: the Atherosclerosis Risk in Communities (ARIC), Cardiovascular Health (CHS), Framingham Heart (FHS), and Health ABC (HABC) studies. We developed a cystatin C genotype score using 7 SNPs within the cystatin gene cluster on chromosome 20, which accounted for 8% of the variability in eGFRcys in FHS. We compared the association of the cystatin C genotype score with cystatin and creatinine-based eGFR and CKD (eGFR <60ml/min/1.73m2). We also tested whether eGFRcys, adjusted for the genotype score, resulted in a change in the prevalence of CKD. Finally, we performed longitudinal analyses of the genotype score, and both adjusted and unadjusted eGFRcys, with mortality and cardiovascular events. Results were combined using inverse-variance meta-analysis. Results: A total of 14,647 participants were included. A higher cystatin C genotype score was associated with lower cystatin C levels (beta-coefficient −0.45; p=4.0E−103) and higher eGFRcys (beta-coefficient 0.53; p=1.0E−95), and was not associated with creatinine-based eGFR (p=0.4). The absolute prevalence of CKDcys was 10.7%; after adjustment for the genotype score, the absolute prevalence of CKDcys was 8.9%, a difference of 1.8%. Adjustment for the genotype score had no impact on the association between eGFRcys with CVD events (HR 0.38 [95% C.I. 0.33–0.44] per unit of log(eGFRcys)) and mortality (HR 0.29 [95% C.I. 0.26–0.32] per unit of log(eGFRcys)). Conclusion: In general, adjusting for the genetic contribution of cystatin C levels resulted in higher estimated GFR and marginally fewer CKD cases. This indicates the presence of a genetic bias in circulating cystatin C levels; however, this does not appear to be of clinical importance with respect to cardiovascular and fatal outcomes.