By Robert Flaumenhaft, MD, PhD

2009-01-01

Dr. Flaumenhaft indicated no relevant conflicts of interest.

Bezemer ID, Bare LA, Doggen CJ,  et al. Gene variants associated with deep vein thrombosis. JAMA. 2008;299:1306-14.

Both acquired and genetic factors contribute to deep-vein thrombosis (DVT). It is estimated that the contribution of genetic factors is ~60 percent. DVT is a complex multigenic disorder; while several heritable risk factors have been identified, many remain to be determined. The earliest identified inherited abnormalities associated with DVT included several that were relatively uncommon but imparted a substantial increase in risk of DVT (e.g., heterozygosity for protein C, protein S, or antithrombin III deficiency). Subsequently, genetic variants that are more common, but demonstrate more moderate increases in DVT risk, were identified (e.g., factor V Leiden, prothrombin gene mutation). Although the odds ratio associated with these more common gene variants is smaller, they carry a greater risk to the overall population (e.g., a greater population-attributable risk) because of their high prevalence. Can we further improve our understanding of the heritable contributions to DVT by identifying genetic variants that are even more common, but difficult to detect because of their subtle phenotype?

Bezemer, et al. have identified several such gene variants by studying single nucleotide polymorphisms (SNPs) in large populations of patients with DVT. SNPs are variants in the genetic code that are relatively common in the population (frequency ≥1%) and occur at a rate of approximately 1 per 1,000 base pairs. Bezemer, et al. screened for 19,682 SNPs located in 10,887 genes (e.g., nearly half the genome) using SNPs selected for their potential to affect gene function or expression. They evaluated 3,155 patients with DVT and 5,087 controls in three distinct groups, enabling confirmation of results. The objective was to identify SNPs that occurred at an increased frequency in patients with DVT compared with controls. 

Three SNPs that are consistently and significantly associated with DVT were found in genes for antithrombin (SERPINC1), glycoprotein VI (GP6), and a cytochrome P450 family gene (CYP4V2). That variants in antithrombin III, an essential natural anticoagulant, and glycoprotein VI, a platelet receptor required for normal thrombus formation, could contribute to DVT is pathophysiologically plausible. CYP4V2 may be linked to a nearby causal gene, such as F11 (factor XI). SERPINC1, GP6, and CYP4V2 variants were associated with relatively weak odds ratios, ranging from 1.15 for the GP6 variant to 1.29 for the SERPINC1 variant. However, the frequencies of alleles associated with DVT were high, ranging from 0.10 for the SERPINC1 variant to 0.84 for the GP6 variant. When considering such a common allele over an entire population, even these relatively weak odds ratios become important. For example, the population-attributable risk associated with the GP6 variant is similar to that of factor V Leiden.

The studies of Bezemer, et al. demonstrate a means whereby genome-wide association studies can assist in elucidating the many subtle inheritable factors that contribute to DVT. By identifying high-frequency gene variants, such as those in antithrombin and glycoprotein VI, these studies detect small but common DVT risks. These newly identified common risk factors for DVT will need to be verified as thrombophilic in non-northwestern European populations. Once confirmed, results such as those described in this study could be combined with those of other genome-wide association studies and gene-centric genotyping studies to provide the basis for a profiling screen, consisting of a battery of SNPs, to assess DVT risk. Ultimately, the results of such individual profiling could guide clinical decisions in thrombosis management.

back to top