Recommendations from the EGAPP Working Group: Genomic profiling to assess cardiovascular risk to improve cardiovascular health

The Summary of Recommendations: The Evaluation of Genomic Applications in Practice and Prevention Working Group (EWG) found insufficient evidence to recommend testing for the 9p21 genetic variant or 57 other variants in 28 genes (listed in Table 1) to assess risk for cardiovascular disease (CVD) in the general population, specifically heart disease and stroke. The EWG found that the magnitude of net health benefit from use of any of these tests alone or in combination is negligible. The EWG discourages clinical use unless further evidence supports improved clinical outcomes. Based on the available evidence, the overall certainty of net health benefit is deemed "Low." Rationale: It has been suggested that an improvement in CVD risk classification (adjusting intermediate risk of CVD into high-or low-risk categories) might lead to management changes (e. g., earlier initiation or higher rates of medical interventions, or targeted recommendations for behavioral change) that improve CVD outcomes. In the absence of direct evidence to support this possibility, this review sought indirect evidence aimed at documenting the extent to which genomic profiling alters CVD risk estimation, alone and in combination with traditional risk factors, and the extent to which risk reclassification improves health outcomes. Analytic validity: Assay-related evidence on available genomic profiling tests was deemed inadequate. However, based on existing technologies that have been or may be used and on data from two of the companies performing such testing, the analytic sensitivity and specificity of tests for individual gene variants might be at least satisfactory. Clinical validity: Twenty-nine gene candidates were evaluated, with 58 different gene variant/disease associations. Evidence on clinical validity was rated inadequate for 34 of these associations (59%) and adequate for 23 (40%). Inadequate grades were based on limited evidence, poor replication, existence of possible biases, or combinations of these factors. For heart disease (25 combined associations) and stroke (13 combined associations), profiling provided areas under the receiver operator characteristics curve of 66% and 57%, respectively. Only the association of 9p21 variants with heart disease had convincing evidence of a per-allele odds ratio of between 1.2 and 1.3; this was the highest effect size for any variant/disease combination with at least adequate evidence. Although the 9p21 association seems to be independent of traditional risk factors, there is adequate evidence that the improvement in risk prediction is, at best, small. Clinical utility: Clinical utility was not formally evaluated in any of the studies reported to date, including for 9p21. As a result, no evidence was available on the balance of benefits and harms. Also, there was no direct evidence available to assess the health benefits and harms of adding these markers to traditional risk factors (e. g., Framingham Risk Score). However, the estimated additional benefit from adding genomic markers to traditional risk factors was found to be negligible. Contextual Issues: Prevention of CVD is a public health priority. Improvements in outcomes associated with genomic profiling could have important impacts. Traditional risk factors such as those used in the Framingham Risk Scores have an advantage in clinical screening and risk assessment strategies because they measure the actual targets for therapy (e. g., lipid levels and blood pressure). To add value, genomic testing should lead to better outcomes than those achievable by assesment and treatment of traditional risk factors alone. Some issues important for clinical utility remain unknown, such as the biological mechanism underlying the most convincing marker's (9p21) association with CVD; the level of risk that changes intervention; whether long-term disease outcomes will improve; how individuals ordering direct to consumer tests will understand/ respond to test results and interact with the health care system; and whether direct to consumer testing will motivate behavior change or amplify potential harms. Genet Med 2010:12(12):839-843.