direct-to-consumer genetic testing for addiction susceptibility: a premature commercialisation of...

Direct-to-consumer genetic testing for addictionsusceptibility: a premature commercialisation ofdoubtful validity and valueadd_3836 1..6

Rebecca Mathews, Wayne Hall & Adrian CarterThe University of Queensland, UQ Centre for Clinical Research, Queensland, Australia

ABSTRACT

Genetic research on addiction liability and pharmacogenetic research on treatments for addiction have identified somegenetic variants associated with disease risk and treatment. Genetic testing for addiction liability and treatmentresponse has not been used widely in clinical practice because most of the genes identified only modestly predictaddiction risk or treatment response. However, many of these genetic tests have been commercialized prematurely andare available direct to the consumer (DTC). The easy availability of DTC tests for addiction liability and lack ofregulation over their use raises a number of ethical concerns. Of paramount concern is the limited predictive power andclinical utility of these tests. Many DTC testing companies do not provide the consumer with the necessary geneticcounselling to assist them in interpreting and acting on their test results. They may also engage in misleading mar-keting to entice consumers to purchase their products. Consumers’ genetic information may be vulnerable to misuseby third parties, as there are limited standards to protect the privacy of the genetic information. Non-consensual testingand inappropriate testing of minors may also occur. The United States Food and Drug Administration plans to regulateDTC genetic tests. Based on the ethical concerns we discuss below, we believe there is a strong case for regulation of DTCgenetic tests for addiction liability and treatment response. We argue that until this occurs, these tests have morepotential to cause harm than to contribute to improved prevention and treatment of addiction.

Keywords Addiction, advertising, direct to consumer, ethics, genetic testing, prevention, susceptibility, treatment.

Correspondence to: Wayne Hall, The University of Queensland, UQ Centre for Clinical Research, Royal Brisbane and Womens’ Hospital Campus, Brisbane4029 Queensland, Australia. E-mail: [email protected] 2 October 2011; initial review completed 29 December 2011; final version accepted 30 January 2012

INTRODUCTION

A genetic contribution to drug dependence has longbeen suggested by the observation that the disorder‘runs in families’. Twin and adoption studies have sinceconfirmed that the heritability of drug dependenceranges from 30 to 70%, depending on the drug [1]. Thisindicates that susceptibility to addiction is influencedby individual genetic make-up as well as environmentalfactors.

Genomic studies have identified genetic variants thatare associated with drug dependence and dependent indi-viduals’ responses to various treatments (e.g. naltrexoneand acamprosate for alcohol dependence (AD) treatment)(e.g. [2–5]). Optimistic predictions have been made aboutthe potential for these discoveries to personalize medicine[6] and improve the prevention and treatment of sub-stance disorders [7]. As yet, genetic testing has not been

used widely in clinical practice because most of the genesidentified are only modest predictors of addiction risk ortreatment response, and many of these associations havenot been replicated widely (e.g. [2,8–10]). A number ofcommercial companies are none the less offering genetictests direct-to-consumer (DTC) for predicting suscepti-bility to alcohol, nicotine and heroin dependence, physi-ological responses to alcohol consumption (e.g. alcoholflush response, alcohol metabolism) and responses to nal-trexone treatment for alcohol dependence. The tests canbe purchased over the internet, without any referral byor involvement with a health professional, for as little as$200 (e.g. 23andMe).

This paper aims to:• describe the DTC genetic tests that are available for

addiction-related disorders;• review the evidence base for these tests; and• discuss the ethical concerns they raise.

bs_bs_banner

FOR DEBATE doi:10.1111/j.1360-0443.2012.03836.x

© 2012 The Authors, Addiction © 2012 Society for the Study of Addiction Addiction

DTC GENETIC TESTING FORADDICTION: WHAT TESTSARE AVAILABLE?

Table 1 summarizes the DTC genetic tests that are avail-able for addiction-related phenotypes as at January 2011including, who is offering the test and the genetic vari-ants tested. In the next section we analyse briefly thestatus of the research for each of these genetic variants.

Nicotine dependence: variants in the neuronalnicotinic receptor gene cluster (CHRNA3, CHRNA5and CHRNB4)

The associations between risk of nicotine dependence insmokers and variants in the CHRNA3, CHRNA5 andCHRNB4 gene cluster that encodes the nicotinic acetyl-choline receptors have been well replicated (see [11]).Nicotine binds to these receptors within 60 seconds ofinhaling cigarette smoke and produces its rewardingeffects indirectly. Smokers with this gene cluster are 1.4times more likely to develop nicotine dependence thanthose without [12]. This modest association has onlybeen shown in studies conducted in European samples,however [11]. One of the variants does not exist inAfrican populations, and is present in only one-third ofEast Asian people [11]. 23andMe acknowledge on theirwebpage that the test for nicotine dependence is onlyapplicable to Europeans, but deCODE does not.

Significantly, the test provides information only aboutrisk of nicotine dependence in current smokers. It doesnot provide any information about the risk of nicotinedependence in non-smokers (e.g. smoking initiation andabstinence). The clinical utility of such a test is question-able, given that smoking carries significant negativehealth consequences regardless of whether the smokerdevelops nicotine dependence. This test is of limited valueto the smoker whose money could be better spent invest-ing in smoking cessation assistance.

Alcohol flush response: aldehydedehydrogenase (ALDH2*2)

Three companies test for alcohol flush response byscreening for the ALDH2*2 allele. This allele producesan inactive ALDH2 enzyme that allows acetaldehyde (aby-product of alcohol metabolism) to accumulate in thebody producing facial flushing, sweating and headache.In more severe cases, it can cause cardiovascular col-lapse, arrhythmias, unconsciousness and convulsions.These aversive symptoms protect against developing AD.Carriers of two copies of this allele (homozygotes) have a10 times lower risk of AD, but are rare [2]. Heterozygoteswho carry one copy of ALDH2*2 have a fivefold reductionin risk of AD [2].

While ALDH2*2 appears to be associated with signifi-cantly lower risk of AD, it occurs mainly in East Asians(about 30% prevalence), with cases in European andAfrican populations being rare. From the informationthat is publicly available, none of the DTC genetic testingcompanies appear to inform prospective consumers ofthis.

The clinical and economic utility of this test withadults is also doubtful. First, in western cultures wherealcohol use is ubiquitous, people positive for ALDH2*2will not need a genetic test to tell them that small quan-tities of alcohol make them flush. Secondly, those whotest negative for ALDH2*2 may be under the misappre-hension that they can drink without harm or risk ofaddiction. This is of particular concern because of evi-dence that moderate drinkers with ALDH2*2 have anincreased risk of developing oesophageal cancer [13].

AD: Taq1A polymorphism of dopamine2 receptor (DRD2)

Variants in the DRD2 mediate the rewarding effects ofalcohol, as well as many other drugs of addiction [5].The Taq1A polymorphism of DRD2 has been linked most

Table 1 Genetic tests offered direct to the consumer for addiction related disorders/phenotypes by company.

23andMe Decode Gene planetBiomarkerpharmaceuticals

My geneprofile Lumigenix

Alcohol dependence Taq1a SNP of DRD2 Yes† Yes†

Alcohol flush response ALDH2*2 ALDH2*2 ALDH2*2 Yes†

Alcohol metabolism Yes†

Alcohol intoxication Yes†

Naltrexone response OPRM1 (Asp40)Nicotine dependence CHRNA3, CHRNA5

and CHRNB4CHRNA3, CHRNA5

and CHRNB4Yes† Yes† Yes†

Heroin addiction OPRM1 (Asp40)

Table adapted from [44]. †but gene not specified. ALDH: aldehyde dehydrogenase; CHRN: neuronal nicotinic receptor genes; DRD2: dopamine 2 receptor;OPRM1: m-opioid type 1 receptor; SNP: single nucleotide polymorphism.

2 Rebecca Mathews et al.


consistently to AD. Meta-analyses suggest that people withTaq1A are 1.3 times more likely to develop AD than thosewithout [14,15]. However, recent evidence suggests thatTaq1A maps onto an adjacent gene (ANKK1) [9], prompt-ing questions about whether Taq1A increases AD risk oris a marker of a region where multiple alleles involved inAD risk are located. Meta-analyses suggest that publica-tion bias may have influenced initial optimism about thestrength and significance of the association, which variesbetween different populations [14]. The results of the testfor AD using this single nucleotide polymorphism (SNP)should therefore be interpreted with caution.

Response to naltrexone treatment of alcoholdependence: m-opioid type 1 receptor (OPRM1)

23andMe offers a genetic test that purports to predict anindividual’s response to naltrexone treatment for AD. Theuse of genetic information to target medical treatmentis called pharmacogenetics. OPRM1 mediates AD indivi-duals’ response to treatment with naltrexone; a drug thatblocks opioid signalling and blunts the euphoria associ-ated with alcohol consumption (e.g. [16]). Individualswith the Asp40 allele of OPRM1 are three times morelikely to respond to naltrexone therapy for AD than thosehomozygous for the Asn40 allele [3,17,18]. Such testshave the potential to significantly improve the effective-ness of naltrexone treatment of AD, given that itis only marginally effective in the AD population [19].Because naltrexone is a prescription drug, there is a goodcase for restricting such pharmacogenetic tests to use inclinical contexts. However, there are few guidelines toaddress the ethical issues raised by the application ofsuch tests [20].

Heroin dependence: OPRM1

Asp40 has also been claimed to predict the risk of devel-oping heroin dependence, but results are conflicting [21].Asp40 was more prevalent among heroin-dependentpeople in two studies (odds ratios of 2.9) [22,23], but intwo other studies it was more prevalent in people withoutheroin dependence [24,25]. As the majority of studies ofAsp40 show no association with heroin dependence [21],it is premature for 23andMe to be offering tests for thisvariant. The association between Asp40 and naltrexonetherapy for AD appears to be more robust, but caution isstill needed in interpreting test results because other envi-ronmental or epigenetic factors may affect whether thisgenotype predicts better treatment outcomes.

Ideally, consumers should be told about the popula-tion prevalence of the genes that are tested. This is par-ticularly pertinent for these tests because Asp40 is muchmore common in people of Asian descent (about 48%)than in Americans of Caucasian (15–18%) or African

American descent (5%). An African American personcontemplating a genetic test for naltrexone response orsusceptibility to heroin dependence may decide not to betested because the gene has limited predictive power, andthe predictive variant only occurs in 5% of such people.

ETHICAL CONCERNS ABOUT DTCGENETIC TESTING FOR ADDICTION

Predictive power, reliability, validity and clinical utility

The genes tested for by DTC companies have limitedpredictive power for addiction liability. With the exceptionof the gene for alcohol flushing, the effect sizes for eachof the above-mentioned genes are small, the associationshave not been well replicated and the prevalence of someof the alleles are very rare in certain cultural groups,limiting their power to predict addiction liability. In thecase of nicotine dependence, there is evidence thatgenetic risk information does not improve upon predic-tions based on family history [26]. The validity of DTCtests is undermined further by evidence showing that thesame tests produced contradictory disease risk estimateswhen administered by two different companies, and insome cases tests results were inconsistent with thepatients’ actual disease status [27].

The ethical defensibility of these DTC tests is broughtfurther into question by the fact that they do not give theconsumer clinically actionable information. In practice,someone who tested positive for a gene associated withheightened risk of AD would be advised to drink inmoderation, good advice regardless of genetic risk. Simi-larly, recommending that individuals quit smoking andnot use heroin are public health messages that should begiven irrespective of genetic risk. Pharmacogenetic testsfor response to naltrexone provide more clinically usefulinformation, although such tests would be best adminis-tered by the physician prescribing naltrexone and coun-selling should be given about the meaning of test results.

Genetic counselling and test interpretation

A clinically useful genetic test would yield actionableinformation to prevent disease and improve health, e.g.by undertaking a pharmacological, behavioural or psy-chological intervention. None of the DTC tests for depen-dence liability meet these criteria. Most are providedwith uninterpreted genetic risk information that, at best,advises consumers that their risk of developing depen-dence is average, or slightly above or below average. Mostcompanies do not offer genetic counselling either beforeor after testing (Navigenics is the exception, but it doesnot offer genetic testing for addiction-related pheno-types). It is unlikely that the average consumer would beable to understand their genetic tests results. A recent

Direct-to-consumer genetic testing for addiction liability 3


survey of potential DTC genetic test consumers foundthat most did not understand their test results and 78%indicated a need to consult a general practitioner (GP) orphysician to help interpret them [28].

DTC tests that offer no genetic counselling and limitedinterpretation of results are likely to create additionalwork for GPs and physicians who will be asked to inter-pret and clinically manage these test results. This is prob-lematic for two reasons. First, it creates additional costsand pressure on an already stretched health system. Sec-ondly, many clinicians lack the knowledge, skills and edu-cation to interpret genetic test results accurately [29].Even if they were able to interpret the test result accu-rately, they may feel pressured to provide a therapeuticintervention to prevent the development of a diseasewhen the patient’s level of risk is average.

Discrimination and privacy

DTC genetic tests raise many of the same privacy issues astests administered in clinical and research contexts. Theresults of DTC genetic tests may be misused by thirdparties, such as health insurance companies and employ-ers to justify discrimination, or by the criminal justiceand legal systems to link individuals to crime [30]. DTCcompanies may be legally required to communicate ordisclose genetic information to these third parties undersubpoena.

Some countries have introduced legal safeguardsto minimize such discrimination. For instance, in theUnited States, the 2009 Genetic Information Non-Discrimination Act protects asymptomatic individualswho have a genetic mutation against discrimination byhealth insurers and employers. The Act prohibits healthinsurers from using genetic information or requestingtesting in setting insurance premiums or employers fromusing it in decisions about hiring, promoting or firing anemployee [31]. However, it does not cover life, long-termcare or disability insurance; nor is it applicable to peoplewho have clinical symptoms of a genetically baseddisease. Consequently, the potential for genetic discrimi-nation still exists. The legislation has also yet to be testedin the courts, so it is not clear what level of protection itoffers. However, because alleles for addiction susceptibil-ity are nowhere near as predictive as those for single genedisorders (e.g. Huntington’s disease), discrimination onthe basis of these alleles is unlikely.

Proponents of DTC genetic testing argue that it offersmore privacy protection than genetic testing ordered viaa medical professional because test results will not beincluded in the person’s medical record that may besubject to subpoena (e.g. [29]). This claim is contes-table. The security and privacy of any online transac-tions, including the purchasing of genetic tests, is not

guaranteed, with hacking and identity theft possibilities[32]. Furthermore, the companies who offer DTC testingare not subject to the privacy restrictions of the HealthInsurance Portability and Accountability Act (HIPAA),although most claim to be ‘HIPAA-compliant’. [33]. IfDTC companies are not being HIPAA-compliant, con-sumers may not have rights to access their genetic infor-mation or to request corrections to it. Their geneticinformation may also be vulnerable to misuse by associ-ates of DTC companies, who may be able to access geneticdatabases but are not contractually required to protecttheir privacy [34].

Informed consent

It is more difficult to ensure informed consent to DTCtesting than to clinical genetic tests, because there is nohealth professional acting as an intermediary to autho-rize and support the collection of DNA. Accordingly,there is less quality control in the collection of specimens.An individual may also take a DNA specimen fromanother person (e.g. through hair strands, a used cottontip, cigarette butts) and send it to a DTC company withoutthat person’s consent. This is referred to as non-consensual, or surreptitious, testing. It is unclear howcommon it is [35]. None the less, it could be amelioratedby having some enforceable minimum standards forobtaining informed consent (i.e. the person being testedwould have to sign consent forms and provide signedcopies of identification) and the samples (e.g. by compa-nies refusing to analyse specimens other than a vial ofsaliva or a cheek swab). The United Kingdom has madenon-consensual analysis of DNA an offence under theHuman Tissues Act [36].

Testing of minors

The consensus in published guidelines is that genetictesting of asymptomatic children should be postponeduntil they are able to make an informed choice about suchtesting [37,38]. For DTC genetic tests, there is no consis-tent policy on testing minors. A recent analysis of DTCgenetic companies’ websites revealed four differentapproaches to testing minors [38]. One group of compa-nies provided no information at all about testing minors.A second group allowed genetic testing of minors, ifauthorized or requested by their parents/guardians (e.g.deCODE and 23andMe). A third group made clear thattheir website is not ‘directed to minors’, but do not explic-itly refuse to test minors if parents request it (e.g. Biomar-ker Pharmaceuticals). A fourth group states that peoplemust be 18 years or older to be tested (e.g. Navigenics).

Genetic testing for addiction liability during childhoodis difficult to justify, given the poor predictive value ofthese tests. We would recommend that DTC genetic tests



not be offered to people under the age of 16 years. UntilDTC genetic tests are regulated, this standard will bedifficult to enforce.

Misrepresentation

The promises made by different companies about theutility of DTC genetic testing may vary. Not all DTCtesting companies promise the same personalized health-care services from their genetic tests [39], with some pro-moting them only as a product for consumers to satisfytheir curiosity. A major problem with DTC genetic testsbeing promoted for clinical use is that there is often asignificant gap between the claims made by providersabout their clinical usefulness and the research evidenceabout this [27]. This is misleading marketing. Many com-panies’ websites, for example, include disclaimers thattest results are not diagnostic and should not be consid-ered as medical advice, but a Government AccountabilityOffice (GOA) audit found that company marketing mate-rials, and company representatives, often highlighted theclinical implications of the tests [40]. Moreover, it foundthat 10 of the 15 DTC genetic testing companies auditedengaged in ‘some sort of fraudulent, deceptive or other-wise questionable marketing practices’.

In the United States, federal law prohibits companiesfrom using unfair, deceptive or fraudulent trade practices,or making false or misleading advertising claims [41]. Intheory, this law prohibits false genetic-testing claims, butin practice fair-trade commission regulators may lack theknowledge that enables them to assess when the claimsmade by DTC genetic testing are misleading. This suggeststhe need for DTC genetic testing companies to be regu-lated and for their tests to be subject to pre-market review.

The case for regulation

There is a need for a regulatory framework to betterprotect consumers of DTC genetic tests. This frameworkwould need to consider: the clinical validity and utility ofthe tests, the quality of genetic counselling offered to con-sumers, protecting consumer privacy and minors, mini-mizing risks of non-consensual testing and preventingmisleading claims. Researchers, doctors, lawyers, bioet-hicists and policy makers have advocated for the regula-tion of DTC genetic tests for some time [42], and multiplead-hoc task forces have examined the issues that need tobe addressed by regulation [43]. In spite of this, no con-crete action was taken on regulation until 22 July 2010,when the US Food and Drug Administration announcedits plans to regulate DTC genetic tests [42]. Thisannouncement coincided with the US GAO review of DTCtests, which concluded that DTC test results were mis-leading and had limited clinical value [40].

CONCLUSION

While different DTC genetic testing companies maypromise different outcomes from their tests, in generalgenetic tests have been commercialized prematurely withlittle oversight, regulation or respect for scientific rigour[43]. There is often no intervening period in which testresults are replicated and clinically tested before beingcommercialized. In the case of genetic tests for addictionliability and treatment response, tests are commerciallyavailable that have failed to demonstrate clinical validityand utility, and which test for variants that are rare insome population groups. This information is not beingcommunicated to consumers, who may have the expec-tation that these tests are clinically relevant. Conse-quently, they will be unable to make well-informedchoices about whether to undergo these tests or interpretthe information the tests can reliably provide them with.Moreover, because genetic counsellors and medical pro-fessionals are not typically involved in the disclosure ofDTC genetic test results, there is the potential for consum-ers to misinterpret their test results. This, in turn, maycause harm by making them unnecessarily anxious andprompt them to pursue unnecessary medical interven-tions, or mislead them into believing that they canconsume alcohol, heroin or nicotine with a low risk ofdeveloping dependence. Until DTC tests for addiction areregulated, their potential for harm is likely to exceed theirlimited potential to prevent addictive behaviour or reducethe harms of drug use.

References

1. Agrawal A., Lynskey M. T. Are there genetic influences onaddiction: evidence from family, adoption and twin studies.Addiction 2008; 103: 1069–81.

2. Luczak S. E., Glatt S. J., Wall T. L. Meta-analyses of ALDH2and ADH1B with alcohol dependence in Asians. PsycholBull 2006; 132: 607–21.

3. Anton R. F., Oroszi G., O’Malley S., Couper D., Swift R.,Pettinati H. et al. An evaluation of mu-opioid receptor(OPRM1) as a predictor of naltrexone response in the treat-ment of alcohol dependence: results from the CombinedPharmacotherapies and Behavioral Interventions forAlcohol Dependence (COMBINE) study. Arch Gen Psychiatry2008; 65: 135–44.

4. Ooteman W., Naassila M., Koeter M. W., Verheul R.,Schippers G. M., Houchi H. et al. Predicting the effect ofnaltrexone and acamprosate in alcohol-dependent patientsusing genetic indicators. Addict Biol 2009; 14: 328–37.

5. Le Foll B., Gallo A., Le Strat Y., Lu L., Gorwood P. Genetics ofdopamine receptors and drug addiction: a comprehensivereview. Behav Pharmacol 2009; 20: 1–17.

6. Collins F. S., Green E. D., Guttmacher A., Guyer M. S. Avision for the future of genomics research. Nature 2003;422: 835–47.

7. Uhl G. R., Grow R. W. The burden of complex genetics inbrain disorders. Arch Gen Psychiatry 2004; 61: 223–9.

Direct-to-consumer genetic testing for addiction liability 5


8. Edenberg H. J. The collaborative study on the geneticsof alcoholism: an update. Alcohol Res Health 2002; 26:214–8.

9. Gelernter J., Kranzler H. R. Genetics of alcohol dependence.Hum Genet 2009; 126: 91–9.

10. Kohnke M. D. Approach to the genetics of alcoholism: areview based on pathophysiology. Biochem Pharmacol 2008;75: 160–77.

11. Greenbaum L., Lerer B. Differential contribution of geneticvariation in multiple brain nicotinic cholinergic receptors tonicotine dependence: recent progress and emerging openquestions. Mol Psychiatry 2009; 14: 912–45.

12. Thorgeirsson T. E., Geller F., Sulem P., Rafnar T., Wiste A.,Magnusson K. P. et al. A variant associated with nicotinedependence, lung cancer and peripheral arterial disease.Nature 2008; 452: 638–42.

13. Lee C. H., Lee J. M., Wu D. C., Goan Y. G., Chou S. H., WuI. C. et al. Carcinogenetic impact of ADH1B and ALDH2genes on squamous cell carcinoma risk of the esophaguswith regard to the consumption of alcohol, tobacco andbetel quid. Int J Cancer 2008; 122: 1347–56.

14. Munafo M. R., Matheson I. J., Flint J. Association of theDRD2 gene Taq1A polymorphism and alcoholism: a meta-analysis of case-control studies and evidence of publicationbias. Mol Psychiatry 2007; 12: 454–61.

15. Smith L., Watson M., Gates S., Ball D., Foxcroft D.Meta-analysis of the association of the Taq1A poly-morphism with the risk of alcohol dependency: a HuGEgene-disease association review. Am J Epidemiol 2008; 167:125–38.

16. Ray L. A., Hutchison K. E. Effects of naltrexone on alcoholsensitivity and genetic moderators of medication response:a double-blind placebo-controlled study. Arch Gen Psy-chiatry 2007; 64: 1069–77.

17. Oslin D. W., Berrettini W. H., O’Brien C. P. Targeting treat-ments for alcohol dependence: the pharmacogenetics ofnaltrexone. Addict Biol 2006; 11: 397–403.

18. Oroszi G., Anton R. F., O’Malley S., Swift R., Pettinati H.,Couper D. et al. OPRM1 Asn40Asp predicts response tonaltrexone treatment: a haplotype-based approach. AlcoholClin Exp Res 2009; 33: 383–93.

19. Anton R. F., O’Malley S. S., Ciraulo D. A. et al. Combinedpharmacotherapies and behavioral interventions foralcohol dependence: the COMBINE study: a randomizedcontrolled trial. JAMA 2006; 295: 2003–17.

20. Hoop J. G., Lapid M. I., Paulson R. M., Roberts L. W. Clinicaland ethical considerations in pharmacogenetic testing:views of physicians in 3 ‘early adopting’ departments ofpsychiatry. J Clin Psychiatry 2010; 71: 745–53.

21. Zhang H., Luo X., Kranzler H. R., Lappalainen J., Yang B. Z.,Krupitsky E. et al. Association between two mu-opioidreceptor gene (OPRM1) haplotype blocks and drug oralcohol dependence. Hum Mol Genet 2006; 15: 807–19.

22. Bart G., Heilig M., LaForge K. S., Pollak L., Leal S. M., Ott J.et al. Substantial attributable risk related to a functionalmu-opioid receptor gene polymorphism in association withheroin addiction in central Sweden. Mol Psychiatry 2004; 9:547–9.

23. Szeto C. Y., Tang N. L., Lee D. T., Stadlin A. Associationbetween mu opioid receptor gene polymorphisms andChinese heroin addicts. Neuroreport 2001; 12: 1103–6.

24. Tan E. C., Tan C. H., Karupathivan U., Yap E. P. Mu opioid

receptor gene polymorphisms and heroin dependence inAsian populations. Neuroreport 2003; 14: 569–72.

25. Schinka J. A., Town T., Abdullah L., Crawford F. C., OrdoricaP. I., Francis E. et al. A functional polymorphism within themu-opioid receptor gene and risk for abuse of alcohol andother substances. Mol Psychiatry 2002; 7: 224–8.

26. Gartner C. E., Barendregt J. J., Hall W. D. Multiple genetictests for susceptibility to smoking do not outperform simplefamily history. Addiction 2009; 104: 118–26.

27. Kuehn B. M. Inconsistent results, inaccurate claims plaguedirect-to-consumer gene tests. JAMA 2010; 304: 1313–5.

28. McGuire A. L., Diaz C. M., Wang T., Hilsenbeck S. G. Socialnetworkers’ attitudes toward direct-to-consumer personalgenome testing. Am J Bioeth 2009; 9: 3–10.

29. Bender L., Silverman L. M., Dinulos M. B., Nickel J., Grody W.W. Direct-to-consumer genotyping: are we ready for a bravenew world? Clin Chem 2010; 56: 1056–60.

30. Coors M. E., Raymond K. M. Substance use disorder geneticresearch: investigators and participants grapple with theethical issues. Psychiatr Genet 2009; 19: 83–90.

31. Appelbaum P. S. Genetic discrimination in mental disorders:the impact of the genetic information non-discriminationact. Psychiatr Serv 2010; 61: 338–40.

32. Synovate. United States federal trade commission—identitytheft survey report; 2003.

33. Wolfberg A. J. Genes on the Web—direct-to-consumermarketing of genetic testing. N Engl J Med 2006; 355:543–5.

34. Centre for Disease Control. HIPAA privacy rule and publichealth. Morbidity and Mortality Weekly Report. Atlanta:Centre for Disease Control; 2003, p. 1–12.

35. Genetics and Public Policy Centre. Surreptitious DNAtesting. Washington, DC: Johns Hopkins University; 2009.

36. Tamir S. Direct-to-consumer genetic testing: ethical–legalperspectives and practical considerations. Med Law Rev2010; 18: 213–38.

37. Borry P., Stultiens L., Nys H., Cassiman J. J., Dierickx K.Presymptomatic and predictive genetic testing in minors: asystematic review of guidelines and position papers. ClinGenet 2006; 70: 374–81.

38. Borry P., Howard H. C., Senecal K., Avard D. Health-relateddirect-to-consumer genetic testing: a review of companies’policies with regard to genetic testing in minors. Fam Cancer2010; 9: 51–9.

39. Arribas-Ayllon M., Sarangi S., Clarke A. Promissory strate-gies of personalisation in the commericalisation of genomicknowledge. Commun Med 2011; 8: 53–66.

40. United States Government Accountability Office. Direct-to-Consumer Genetic Tests: Misleading Test Results Are FurtherComplicated by Deceptive Marketing and Other QuestionablePractices. Washington, DC: United States GovernmentAccountability Office; 2010.

41. Hudson K., Javitt G., Burke W., Byers P. ASHG Statement* ondirect-to-consumer genetic testing in the United States.Obstet Gynecol 2007; 110: 1392–5.

42. Brower V. FDA to regulate direct-to-consumer genetic tests.J Natl Cancer Inst 2010; 102: 1610–2, 1617.

43. Hogarth S. Myths, misconceptions and myopia: searchingfor clarity in the debate about the regulation of consumergenetics. Public Health Genomics 2010; 13: 322–6.

44. Genetics and Public Policy Centre. DTC genetic testing com-panies. Washington, DC: Johns Hopkins University; 2010.



direct-to-consumer genetic testing for addiction susceptibility: a premature commercialisation of...

Documents