To the Editor: Genomic studies of cardiovascular disease that are based on samples obtained from patients have identified variants associated with the risk of disease that do not always predict the onset of disease shown in prospective studies. 1, 2 Dewey et al.(March 24 issue) 3 report lower levels of triglycerides and a lower risk of coronary artery disease among carriers of E40K and other inactivating mutations in ANGPTL4 than among noncarriers. Also, in the same issue of the Journal, Stitziel et al. report that in a cross-sectional study, low-frequency coding-sequence variants were associated with a reduced risk of coronary disease. 4

We carried out an independent, large-scale study and confirmed the role of these genetic variants in incident coronary heart disease. Our study involved data that was collected prospectively from seven cohorts (the Age, Gene/Environment Susceptibility–Reykjavik Study; the Atherosclerosis Risk in Communities Study; the Cardiovascular Health Study; the Framingham Heart Study; the Genetic Studies of Atherosclerosis Risk; the Rotterdam Study; and the Women’s Genome Health Study). A total of 5427 of 49,518 persons of European ancestry in these studies had incident coronary heart disease. 5 We investigated 35,904 autosomal exome-array variants with a minor-allele frequency of greater than 1% for association, with adjustment for age, sex, and population substructure. Significant associations (Bonferroni-corrected P< 1.39 E-06) are summarized in Table S1 in the Supplementary Appendix, available with the full text of this letter at NEJM. org. We confirm previously reported6 associations with noncoding common variants at 9p21 and PHACTR1, and we also found an association between a low-frequency variant in ANGPTL4 and a reduced risk of incident coronary heart disease.