Chromosome 9p21 genetic variation explains 13% of cardiovascular disease incidence but does not improve risk prediction




To evaluate the proportion of cardiovascular disease (CVD) incidence that is explained by genetic variation at chromosome 9p21 and to test whether such variation adds incremental information with regard to CVD prediction, beyond traditional risk factors.

Design, setting and participants

rs4977574 on chromosome 9p21 was genotyped in 24 777 subjects from the Malmö Diet and Cancer study who were free from CVD prior to the baseline examination. Association between genotype and incident CVD (= 2668) during a median follow-up of 11.7 years was evaluated in multivariate Cox proportional hazard models. Analyses were performed in quartiles of baseline age, and linear trends in effect size across age groups were estimated in logistic regression models.


In additive models, chromosome 9p21 significantly predicted CVD in the entire population (hazard ratio 1.17 per G allele, 95% confidence interval 1.11–1.23, < 0.001). Effect estimates increased from the highest (Q4) to the lowest quartile (Q1) of baseline age, but this trend was not significant. The overall population attributable risk conferred by chromosome 9p21 in fully adjusted models was 13%, ranging from 17% in Q1 to 11% in Q4. Addition of chromosome 9p21 to traditional risk factors only marginally improved predictive accuracy.


The high population attributable risk conferred by chromosome 9p21 suggests that future interventions interfering with downstream mechanisms of the genetic variation may affect CVD incidence over a broad range of ages. However, variation of chromosome 9p21 alone does not add clinically meaningful information in terms of CVD prediction beyond traditional risk factors at any age.