TY - JOUR
AU - Zehnbauer,, Barbara
AB - Direct-to-consumer (DTC)2 genetics testing is a commercial method for providing DNA testing for genetic variations that are associated with >1000 diseases and traits, as requested by an individual. People seeking insight about personal genetic traits and the risks of disease hope that this information will encourage changes in diet, exercise, behaviors (smoking, alcohol consumption, overeating, and so forth) or will indicate early detection that might improve health and decrease disease. Typically, the customer requests these DTC services via the company's Web site, collects either a saliva or buccal swab sample at home, and submits it for genotyping and analysis. This general, nondiagnostic screening does not require a physician or medical indication, and the individual bears the costs. DTC genetics testing has drawn criticism from many genetics professionals and scrutiny from US government agencies [the General Accountability Office (1) and the Food and Drug Administration (2)]. Concerns abound about the clinical utility of the genetic associations, the lack of medical oversight, the need for quality practices, protection of patient privacy and integrity of data files, and the misunderstanding of risk assessments as indications of disease. In addition to the regulatory quandary and the public policy outcry, clinical laboratories are also asking how these services stack up to more accepted methods. The public wants genetic information, whether it is marketed as medical information or complies with accepted regulatory standards. Many DTC firms use genome-wide association studies, in which hundreds of thousands of genetic markers [single-nucleotide polymorphisms (SNPs)] are detected. Disease associations are interpreted on the basis of comprehensive literature reviews and advanced data comparisons of marker frequencies in populations and the diseases in question. The strength or value of the interpreted associated risk depends on the published scientific research correlating a particular genetic variant(s) with a specific disease or condition. The majority of peer-reviewed data has focused on Caucasian populations of European ancestry with limited relevance for extrapolating to genetic associations for an African, Asian, or Hispanic individual, although companies are increasingly augmenting their databases with new studies of marker frequencies in these ethnic groups. Each DTC service uses a different set of genetic markers as indicators of disease risk and compares these markers for affected and unaffected populations. Alleles that occur more commonly in affected populations and exhibit odds ratios >1 are labeled risk alleles. The greater the allele frequency in patients with disease relative to the unaffected controls, the higher the odds ratio associated with the risk allele. Alleles that occur less frequently in patients with disease have odds ratios <1 and might confer protection against the disease. An essential limitation that is frequently misunderstood is that a disease-risk estimate measures the risk relative to the general population or the probability that an individual will develop disease; it is not equivalent to prediction or diagnosis of disease. In this issue of Clinical Chemistry, Imai et al. (3) present data from the SNP-genotyping analyses of a single individual's sample by 3 commercial DTC firms and 1 DNA-testing service. The concordance rates among all of the services for this single sample were >99.6%. This result is consistent with the 99.7% concordance documented in a previous comparative study (4) for 2 of the same companies. The authors conclude that these leading companies have comparable analytical quality in their methods, although an analysis of additional samples would have strengthened this assertion. Concordance rates among separate firms using the same microarray technology (Illumina) were >99.9%, indicating similar performance characteristics. Therefore, the relative detection rates of the genetic variants are high and similar to the values reported by these firms. Imai et al. also compared the relative-risk assessments for several clinical conditions that were predicted by each DTC company's proprietary data-analysis program. Similar to the results of a previous study (4), there are variations in relative disease risk assigned for this individual, with some assessments showing agreement and others with divergent conclusions. For 6 of 8 different conditions, all of the DTC services gave the same qualitative findings of either increased risk or protective effects from disease. Both the number of SNPs and the identity of the SNPs used in the risk calculations varied with the DTC service, and more markers did not necessarily predict a higher risk. Certain diseases have better prediction agreement than others (4), and strong-effect markers tend to produce more consensus predictions. For example, there was good agreement among the data from all 3 DTC services for both celiac disease and Crohn disease, 2 conditions with well-characterized strong-effect markers and pronounced odds ratios. The magnitude of the relative-risk factors over all the conditions surveyed in this report varied between 0.3 and 1.8, which is slightly different than the “modest” range of 0.5–1.5 observed across the 5 individuals tested in the study of Ng et al. (4). Two discrepancies in disease risk estimation were detected for this individual among the 3 DTC strategies: a decreased risk for atrial fibrillation was noted by only deCODEme, whereas a decreased risk for rheumatoid arthritis was reported only by Navigenics. The Navigenics testing strategy has apparently changed since the report of Ng et al. (4). In 2009, Navigenics was using the Affymetrix 6.0 technology (currently used by Expression Analysis). In the study of Imai et al., the Navigenics genotyping used an Applied Biosystems TaqMan® technology that is limited to about 120 SNPs. A specific explanation for this switch was absent (3), but perhaps Navigenics chose to focus on the more predictive, strong-effect SNPs (4) after recognizing that the previous unique markers did not contribute significantly to overall disease risk prediction. The difference in the predictive value of the revised assay is uncertain, because the same person's sample was not tested with both Navigenics' approaches (Affymetrix 6.0 and TaqMan methods). Data that were previously discordant between 2 of these firms (23andMe and Navigenics) for some of the same disorders (4) were not discordant for the single individual genotyped in the study of Imai et al. Whether this discrepancy was due to changes in the SNP analyses, data interpretation, or a specific feature of the genotype of this individual was not explored. These authors attribute most of the differences in disease risk assessment across the DTC services to the inclusion of different SNPs in the separate platforms. Relative-risk calculations and interpretive algorithms for the significance of each pattern of SNPs are also unique to each DTC company's approach. Finally, the ethnic populations that constitute the reference database for these SNP genotypes are not standardized across these companies. The authors investigated this variable by changing the ethnicity of the same individual's submission information to help direct a more appropriate consideration of the significance and relevance of the genetic variations. This approach is limited by the availability of data about disease risk in alternative ethnic groups within each company's scientific genome-association database. The same genetic marker patterns may yield very different disease risks when European ancestry is changed to African ancestry, as Imai et al. found for rheumatoid arthritis and colorectal cancer. The analytical details are challenging for most medical professionals, even those who are aware of this testing. A CDC review of 1880 DocStyles survey respondents showed that 42% were aware of DTC genetic testing, and 52% of these respondents said they would be somewhat or very likely to use the DTC testing data to influence their clinical care decisions. Many might be unaware that although the analytical validity of the available tests is high, it does not equate to strong or obvious clinical validity or utility (5). Companies that provide genotyping and data-analysis services variably describe their respective services and the intended use of the information they provide. 23andMe offers genotyping services “for health, disease, and ancestry,” but they do not provide medical advice to their customers (6). Whether consumers can distinguish the differences between medical advice and personalized disease risk information is not clear. deCODEme promotes their services as a “genetic health scan,” justifying that “early detection of [disease] risk allows you to prevent a disease,” although the outcome data to support these claims of preventive value are limited (7). Navigenics presents “health focused genetic testing and analysis” with the aim to “empower you with genetic insights … to improve your health” and “control your genetic information” (8). This firm includes board-certified genetic counselors to work with physicians and patients to “understand test results,” “review … diagnostic, preventive, and early detection options,” and “facilitate a physician's informed decision-making.” This description might approximate the role that clinical diagnostics laboratories and genetic counseling services provide. DTC genetic testing has drawn customers, as well as criticism from genetics professionals and scrutiny from regulatory bodies, but the lure persists for the general public curious about their future health predicted from genome-wide association studies. It remains to be determined whether strict regulatory intervention or industry self-regulation, as proposed by the UK Human Genetics Commission (9), is the most effective method to focus on the safety of consumer interests. 2 Nonstandard abbreviations: DTC direct-to-consumer SNP single-nucleotide polymorphism. " Author Contributions:All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article. " Authors' Disclosures or Potential Conflicts of Interest:No authors declared any potential conflicts of interest. " Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, or preparation or approval of manuscript. References 1. U.S. Government Accountability Office . Direct-to-consumer genetic tests: misleading test results are further complicated by deceptive marketing and other questionable practices . http://www.gao.gov/products/GAO-10-847T (Accessed December 2010). 2. U.S. Food and Drug Administration . Medical devices. Letters to industry (medical device manufacturers, direct-to-consumer genetic testing) . http://www.fda.gov/MedicalDevices/ResourcesforYou/Industry/ucm111104.htm (Accessed December 2010). 3. Imai K , Kricka LJ, Fortina P. Concordance study of 3 direct-to-consumer genetic-testing services . Clin Chem 2011 ; 57 : 518 – 521 . Google Scholar Crossref Search ADS PubMed WorldCat 4. Ng PC , Murray SS, Levy S, Venter JC. An agenda for personalized medicine . Nature 2009 ; 461 : 724 – 6 . Google Scholar Crossref Search ADS PubMed WorldCat 5. Kolor K , Liu T, St. Pierre J, Khoury MJ. Health care provider and consumer awareness, perceptions, and use of direct-to-consumer personal genomic tests, United States, 2008 . Genet Med 2009 ; 11 : 595 . Google Scholar Crossref Search ADS PubMed WorldCat 6. 23andMe . For physicians. Open letter to the medical community . https://www.23andme.com/for/physicians (Accessed December 14, 2010). 7. deCODEme . Confront risk early, prolong your health. Minimizing the impact of your inherited risk . http://www.decodeme.com/genes-and-health (Accessed December 2010). 8. Navigenics . For Physicians. Our genetic counseling team . http://www.navigenics.com/visitor/for_physicians/our_services/genetic_counseling (Accessed December 2010). 9. Human Genetics Commission . A Common Framework of Principles for direct-to-consumer genetic testing services . http://www.hgc.gov.uk/Client/document.asp?DocId=280&CAtegoryId=10 (Accessed December 2010). © 2011 The American Association for Clinical Chemistry This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Direct-to-Consumer Genetics Testing—Fair Comparisons?
JF - Clinical Chemistry
DO - 10.1373/clinchem.2010.160085
DA - 2011-03-01
UR - https://www.deepdyve.com/lp/oxford-university-press/direct-to-consumer-genetics-testing-fair-comparisons-LzdBrMo4cs
SP - 369
VL - 57
IS - 3
DP - DeepDyve
ER -