Genetic Testing and Pharmacogenomics: Issues for Determining the Impact to Healthcare Delivery and Costs

, , , ,
The American Journal of Managed Care, July 2004 - Part 1, Volume 10, Issue 7 Pt 1

Objectives: To determine the potential impact of genetic testingand pharmacogenomics on healthcare delivery and costs.

Study Design: Literature review.

Methods: We examined 3 examples: (1) BRCA1/2 testing forbreast cancer risk, (2) HER2/neu overexpression testing to guidedrug treatment in women with breast cancer, and (3) CYP2C9 testingbefore the use of the anticoagulant warfarin. We discussedeach genetic testing example from the perspective of the patient,provider, insurer, industry, government, and society.

Results: The expanded use of genetic information offers manypotential clinical benefits, but also many economic challenges.One of those challenges will be managing the impact of genetictesting on healthcare delivery and costs.

Conclusions: Systematic, evidence-based technology assessmentsand economic evaluations will have to be used to guide theincorporation of genomics into clinical practice. More researchalso will be needed to assess patient preferences and willingness topay for genomic technologies; how providers can assess and usegenomic technologies; and how the industry, insurers, and governmentcan best balance the relevant costs and benefits.

(Am J Manag Care. 2004;10:425-432)

TheNew England Journal of Medicine

Many have asserted that the wealth of knowledgeoffered by the human genome portends a timeof rapid change in medicine.1 However, genomictechnologies will influence not only health outcomesbut also the delivery and cost of healthcare; and in thisera of increasing concern about healthcare costs, it willbe impossible to consider the implications of genomicmedicine without also considering the economic implications.2-5 Harold Varmus, MD, former director of theNational Institutes of Health, noted in an editorial in that a key questionis: "How much will the expanded use of genetic informationfurther escalate the cost of healthcare, and who willpay for it?"4 Although many studies have addressedissues related to the privacy and fair use of genetic information,2 there has been surprisingly little discussion ofthe implications of genetic testing and pharmacogenomicsfor healthcare delivery and costs.6-9 Ultimately,these issues will substantially influence how patients andproviders adopt and utilize genomic technologies.

This paper reviews the current literature to assessthe potential impact of genetic testing and pharmacogenomicson healthcare delivery and costs. Our objectivesare: (1) to provide an overview of the key issues fromthe perspectives of patients, providers, insurers, industry,government, and society; (2) to discuss examplesfrom the empirical literature; and (3) to point out areasfor future research. In particular, we highlight how keyconsiderations are evolving as genetic testing is beingdeveloped for common conditions and drug-prescribingdecisions, instead of being used only to predict futurerisk of disease in small, relatively high-risk populations.

The Figure illustrates 2 applications of genomics: (1)the use of genetic testing to predict disease risk and (2)the use of genetic information to guide drug developmentand prescribing ("pharmacogenetics/pharmacogenomics").This review examines both of theseapplications (also called "genetic testing" or "genomicmedicine" for brevity), but it does not examine the useof genomics to develop gene-based therapies.

The majority of growth in genetic testing todayinvolves finding tests that aid in diagnosing diseases,predicting future disease risk, or predicting drugresponse.3,10-14 The number of genetic tests that areexpected to move from the research setting to marketwill require providers without specialty training ingenetics to become "practitioners of genomic medicine."3,7 For example, Roche announced the launch of agenetic test to predict how individuals will react to a"large range of drugs."15 Therefore, the translation ofbasic science findings in genomics into clinical practicemust become a major priority.16,17

It is critical that we begin now to understand theimpact of genetic testing–both positive and negative–on healthcare delivery and costs, even though thesenew technologies have generally not been widelyadopted in healthcare systems. With concerns over risinghealthcare costs, the impact of genetic testing onhealthcare delivery and costs will need to be consideredin assessing its potential added value before suchtechnologies are disseminated. Analyses can informfuture decision making and highlight the circumstancesunder which genetic testing strategies couldprove to be either a poor or a good use of scarce healthcareresources.

STAKEHOLDER PERSPECTIVES

Below we examine the conflicts and challenges tothe relevant US stakeholders in implementing genetictesting. We provide illustrations throughout using 3types of genetic testing that offer interesting contrastsand comparisons:

  1. Genetic testing for the purpose of predicting futuredisease risk based on an inherited mutation in anindividual (ie, germline mutation). We use theexample of testing for BRCA1/2 mutations,which are associated with a higher risk ofbreast and ovarian cancer. It has been estimatedthat 5% to 10% of breast cancers have aninherited component (including BRCA mutationsas well as others), and that approximately60% to 80% of women with BRCA mutationswill develop breast cancer.18 BRCA testing(BRACAnalysis® test)is relatively expensive (>$2000 for counselingand testing19), and theprimary available interventionsrequire surgery (prophylactic removalof the breasts and/or ovaries).
  2. Genetic testing for the purpose of prescribingdrug therapy based on genetic variation of thedisease such as acquired mutations in tumors (ie,somatic mutation). We use the example of testingfor overexpression of HER2/neu oncogenes,which are genetic alterations in specific cell typesthat determine prognosis and the potential forresponse to drug therapy, specifically trastuzumab(Herceptin®, Genentech, South San Francisco,Calif) in the case of breast cancer. It has been estimatedthat 25% to 30% of breast cancer casesexhibit overexpression of HER2/neu,20 and womenwho are treated with Herceptin have demonstrateda 25% improved median survival time.21 Thetest for HER2/neu is relatively inexpensive, but anannual course of Herceptin therapy has been estimatedto cost $27 000 to $81 000.22
  3. Genetic testing for the purpose of prescribingdrug therapy based on genetic variation of theindividual such as variant alleles in drug-metabolizingenzymes. We use the example of testingfor variation in CYP2C9 before the use of theanticoagulant warfarin, which requires carefuldosing and patient follow-up because of bleedingrisk and significant interpatient variability indrug response. Approximately 20% to 30% ofCaucasian patients receiving long-term warfarintherapy have at least 1 variant allele, which hasbeen shown to be associated with an increasedrisk of overanticoagulation and bleedingevents.23-25 A genetic test for warfarin is not yet inroutine clinical use, but because warfarin is prescribedto more than 1 million patients in theUnited States annually, the potential cost impactis large. Table 1 summarizes these examples andstakeholder perspectives, which are discussed indetail below.

Patients

Patients are the ultimate consumers of healthcaretechnology, and many are aware of or interested ingenetic testing. According to a nationally representativesurvey, 41% of individuals in the United States haveheard that genetic tests can determine whether a personis at greater risk of developing cancer,26 and 79% ofrespondents in a population-based survey stated thatthey would take a genetic test to predict whether theywill develop Alzheimer's disease.27

Will patients actually agree to genetic testing or individualizeddrug therapies based on genetics? Most of thepatient preference research to date has been conductedwithin research settings, focusing on testing for rare,high-penetrance mutations such as BRCA1/2. Thesestudies have generally shown that uptake of genetictesting is much less than interest in testing because ofconcerns about cost, lack of effective treatment optionsfor those testing positive, privacy and discriminationconcerns, limited predictive value, and negative impacton quality of life.28

Will patients pay out-of-pocket or higher prices forgenetic testing or individualized drug therapies? Thereis little evidence on this issue. One poll found that twothirds of respondents said they would pay extra for a"genetically customized drug that you knew would workfor you."29 Anecdotal evidence suggests that patientsoften pay out-of-pocket for genetic testing for risk susceptibilitybecause of concerns about confidentiality orlack of coverage by insurance companies, although thisis unlikely to be the case for testing for drug therapies.

One indication that demand for genetic technologiescould soon affect clinical practice is the increase indirect-to-consumer advertising of genetic testing.30-32For example, Genelex advertises testing for 3 importantdrug-metabolizing enzymes (CYP2D6, CYP2C19, andCYP2C9) at $250 per test or $650 for all 3 tests(http://www.healthanddna.com, accessed July 31, 2003);and Myriad Genetics advertises testing for inheritedcancer risk including BRCA1/2 (http://www.myriad.com, accessed July 31, 2003). As with other tests,33there is widespread debate over the wisdom of marketingtests directly to consumers.30,34

Overall, the clinical and economic impact of genetictesting will be larger if direct-to-consumer advertisingand insurance coverage cause patients to embracegenetic testing for common diseases and for individualtargeting of drug therapies. Research can help us understandthese preferences,35 particularly in situationswhere the information reveals modest increases in riskfor common diseases or adverse drug effects.Ultimately, if consumers come to consider genetic testsin the same way they perceive, for example, a cholesterolscreening test, demand for these tests will be drivenless by concern about the risks of their use than bytheir perceived benefits and costs. The examples ofBRCA1/2 and HER2/neu testing illustrate how patientdemand is likely to differ for genetic testing to predictrisk versus drug prescribing. In a nationally representativesurvey of the US adult population, only 1% reportedhaving had any genetic test for cancer risk.26 For example,1 study found that only one quarter of high-riskwomen actually pursued BRCA testing after counseling.36 In contrast, there was strong patient demand andadvocacy for access to Herceptin and for fast Food andDrug Administration (FDA) approval of the therapy andtesting.37 Thus, it appears likely that there will be strongpatient demand for genetic testing when it is directlyrelated to key treatment decisions.

Providers

Providers face complex questions regarding the costs,benefits, and risks of genetic-testing technologies (seeSuther and Goodson38 for a review). First, they face thechallenge of keeping abreast of what tests are available;their accuracy, predictive validity, and cost; and whichpatients are most appropriate for testing. Second, theymust consider whether the test offers true advantagesover the current standard of care. For example, therehas been debate about whether genetic tests are superiorto ferritin screening for detecting individuals at riskfor adverse effects of hemochromatosis because of thelow penetrance of the relevant mutations.39 A third issueis how the test might change patient management. Asnoted above, many genetic tests are available to identifyrisks for diseases for which there is no treatment. Giventhe rising costs of healthcare, health insurers may limitpatients' and providers' interest in getting this informationbecause of the need to limit the use of tests withuncertain clinical benefit.40 Fourth, clinicians mustdecide the appropriate level of counseling and informedconsent for genetic tests–and who will provide and payfor it. Most providers are not trained to provide geneticcounseling,41 and insurers may not reimburse for counselingoutside of genetics-specialist settings. Finally,some genetic variants may be quite uncommon in certainracial or ethnic groups.42 Clinicians will face the difficulttask of balancing the arguments against testing inpopulations where prevalence of the variant is quite lowagainst the practical and ethical issues of identifyingindividuals from these populations and informing themthat testing is not indicated.

These challenges are likely to differ for genetic testingto predict risk of disease versus genetic testing toguide drug prescribing. Genetic testing to predict highdisease risk in currently healthy persons (eg, BRCA testing) often requires providers to have extensiveknowledge of genetic counseling. A nationally representativesurvey found that, although 31% of physicians hadordered or provided cancer susceptibility tests in thepast 12 months (including, but not limited to, testing forBRCA), the majority of these physicians (87%) hadreferred these patients elsewhere rather than providingthe testing themselves and 50% of physicians overall feltunqualified to recommend cancer susceptibility testing.43 In contrast, testing for specific drug targets (eg,HER2/neu testing) involves clinical decisions that arepart of treating a person with a medical condition.However, future use of genetic testing to target drugsbased on individuals' genetic variation may be morecomplex, for example, when determining the appropriateuse of genetic testing for drugs with complex metabolicpathways.

Industry

The availability of genetic tests and targeted drugtherapies will depend on the economic incentives forpharmaceutical, diagnostic, and genomics companies todevelop and market them. From the pharmaceutical-industryperspective, genomic technologies offer aninteresting contrast in opportunities for drug developmentversus threats to market share.

On the one hand, there are economic incentives forindustry to develop genomic-based drug therapies.Genomics will likely provide a multitude of new drugtargets, enable the development of drugs that avoidproblematic genetic variants in drug-metabolizingenzymes, and increase the development of preventiveinterventions for patients identified as being at higherrisk for future disease (eg, use of cyclooxygenase-2inhibitors to reduce risk of familial colon cancer44).Furthermore, if pharmacogenomics leads to drugtherapies with greater efficacy and safety, these therapiesare likely to have a higher price that reflects thisadded value. Genetic profiling of patients enrolled inPhase I and II clinical trials may allow for smallerPhase III clinical trials that have a greater chance ofsuccess, resulting in decreased development time andcost, and shorter FDA approval times. Additionally,genomics offers the opportunity to rescue drugs thathave not been approved because of side effects or lackof efficacy by targeting them to patients with specificgenetic profiles.7,45,46

There also are economic disincentives for the pharmaceuticalindustry to develop genomic-based drugtherapies. Of most concern, targeted drugs may lead todecreased market share and fewer blockbuster drugs.For example, genetic testing may help us decide whichpatients will not respond well to specific types of antihypertensives,cholesterol-lowering medications, orantidepressants.47,48 If the relevant mutations are prevalent,large numbers of patients may be steered towardspecific drugs. It also is not clear whether the currentmarket structure will provide the necessary incentivesfor the development of drugs targeted toward patientswith rare genetic variants, such as many racial minoritypopulations.45,46,49 However, even a "niche" drug, ifused as a presymptomatic preventive measure for thoseat increased genetic risk, may prove profitable.10

A major shift as a result of the availability of geneticinformation is that drug/diagnostic combinations arebecoming more important; thus, the diagnostics industrywill play a critical role in the implementation ofgenetics.50 (We thank an anonymous reviewer for thispoint.) The shift toward combination products also portendsgreater integration of the diagnostics and pharmaceuticalindustries, which have historically operatedand been regulated relatively independently.51,52

In contrast to the mixed incentives for pharmaceuticalfirms, the diagnostics industry has a potentiallyclearer incentive to use genetic information.53 Geneticinformation is likely to increase the market for diagnostics,as genetic tests will be required for risk predictionand to stratify patients regardless of whether theyreceive a drug or intervention.54,55 Indeed, it has beensuggested that diagnostic firms may profit sooner fromgenomics than pharmaceutical firms.54 Still, the availabilityof tests will depend on reimbursement rates,which to date have varied.55,56 Moreover, the regulatoryclimate as it relates to the diagnostic industry is uncertain.The regulation of diagnostics has historically beenless comprehensive than the regulation of drugs, but theuse of genetic information may require changes in theprevailing regulatory framework.57

Herceptin and the associated tests offer an illustrationof the issues for both the pharmaceutical and diagnosticsindustries. Herceptin, which is considered to bea major success story for the biotech industry, illustratesthe promise of finding novel therapies for certaingenetic variations in a disease state. Herceptin also illustratesthe potential for diagnostic firms, as 2 availabletests are listed on the label, thus conferring a monopolyto those firms.58

Insurers

Health insurers play a large role in defining patientand provider access to new medical technologies throughtheir coverage policies and clinical guidelines.59,60

Limited data suggest many genetic tests are not coveredor coverage policies have not been determined. Forexample, 84% of insurers reported having never made adecision to cover BRCA testing, 12% had received aninquiry and decided not to cover testing, and 4% hadmade a decision to cover the test.59 Another surveyfound that half of plans had no coverage policy for prophylacticmastectomy or oophorectomy followingBRCA1/2 testing, and an additional one quarter of plansdid not cover such surgery.61

Genetic tests can be expensive; for example,BRCA1/2 full sequencing costs $2975 (http://www.myriadtests.com/doc/Myriad_Patient_Auth_Form.pdf.Accessed June 15, 2004), and the downstream coststhat follow from testing can be much more costly thanthe test itself (eg, prophylactic surgery). In addition,although the cost of genetic testing is expected to continueto drop, health insurers may face large outlays ifinitial testing results in extensive testing of family members.Because most genetic tests will become availablewell in advance of evidence that using these tests willfavorably affect patient outcomes, insurers will face thedilemma of whether to bear the economic burden oftesting while waiting for definitive studies to be conducted.45 Although cost is only 1 factor in determiningcoverage policies, insurers have reported concerns thatthe costs of covering genetic testing will be too high.59Furthermore, many patients change plans frequently;thus, individual health plans are unlikely to realize thefull benefits of preventive interventions.

Another determinant of the impact of genetic testingwill be clinical guidelines. For example, many managedcare organizations use drug formularies to guide drugutilization.62 However, targeting drugs based on genetictest results will increase the complexity of formulariesand require that formularies consider relative trade-offs.Payers also will have to determine whether to requiregenetic testing before initiating certain drug therapies.63From the insurer perspective, testing for risk predictionversus drug prescribing provides a study in contrasts.Insurance companies have not universally covered testingfor risk prediction (eg, BRCA testing); in contrast,because HER2/neu testing and appropriate treatmentnow are considered the standard of care, insurers coverboth testing and Herceptin. It remains to be seenwhether insurers will cover broader genetic testing.Testing for relatively rare, highly penetrant gene mutationsis expensive on an individual level but modest ona health plan level. In contrast, testing for more commonmutations and for targeting drug therapies is relativelyless expensive on an individual basis, but willapply to a much larger population; thus, the total costsmay be substantial, although these costs may be balancedby improvements in health and savings in othermedical costs. Therefore, it will be critical that insurershave the cost and outcome data necessary to evaluategenomic technologies.59

Government

Government and regulatory agencies have a role indetermining (1) the safety and efficacy of genomic technologies;(2) reimbursement policies for public healthinsurance programs; and (3) relevant laws, regulations,and guidelines.64,65 The FDA will face a number of issuesregarding the approval of genetic tests, including towhat extent genetic data may be required in the drugapproval process, whether there will be further reviewof previously approved drugs as relevant genetic databecome available, whether testing may be requiredbefore drug therapy is initiated or after initiation, thecomarketing and labeling of tests and drugs, and patentprotection for drugs developed for small, geneticallydefined populations through the Orphan Drug Act.9,66-69A long-standing issue that will gain prominence asmore genetic tests are developed is the current statusof regulatory review of genetic tests. Although theFDA oversees the safety and efficacy of new drugs anddiagnostics in the United States, most genetic tests arenot regulated by the FDA because they are considered"clinical services" rather than marketed kits.57,70Laboratory testing is regulated under the ClinicalLaboratory Improvement Amendments (CLIA); however,CLIA does not examine clinical validity or utility(http://www.cms.hhs.gov/clia).51,71

Reimbursement policies are determined for publicinsurance enrollees through the Centers for Medicare &Medicaid Services. The government also has a role indeveloping guidelines for population screening5 and inprotecting individuals from discrimination based ongenetics for purposes of health insurance.72

Overall, any changes in the regulatory structure relevantto genetic tests will be key events in the future ofgenomic medicine. Government may have a role in encouragingprivate and increasing public investment ingenomic technologies if current levels are not sociallyoptimal.45 This is particularly true because commercialtesting has advanced rapidly in the United States, causingconcerns about the effects on accessibility, price,and quality.73,74 Because of this commercialization,industry and economic incentives will play a relativelylarger role in determining the overall impact of genomictechnologies unless there is government intervention.73

BRCA testing in particular illustrates the controversiesover the commercialization of genetic testing andthus where the government may play an important role.The patenting of human genes has been very controversial,and the case of BRCA has been no exception. Thepatent for the BRACAnalysis® test is held by MyriadGenetics, which has been criticized for monopolizing theBRCA test by requiring that full sequencing testing bedone only at its laboratory and by pricing the test higherthan if were more readily available. The concern isthat these restrictions will reduce access and discouragefurther research.75

Societal

The perspectives discussed above ultimately contributeto an overall societal perspective. From this perspective,the goal is to maximize the benefits fromgenetic testing while minimizing the risks and costs forall stakeholders. Two formal, systematic approaches toweighing the costs, benefits, and risks of genetic testingare cost-effectiveness and cost-benefit analysis.However, in a systematic review of theliterature, we found that there are currentlyfew economic analyses of genetictesting, particularly for drug therapies,and that the studies cover a limited numberof genetic mutations and diseases (ie,the majority of analyses of genetic testingfor risk prediction are concerned with prenatalscreening such as that for cystic fibrosis; K. A. Phillips, S. L. VanBebber, J. Sakowski, unpublished data, 2004).

before

Although the lack of cost-effectiveness evaluations ofgenetic testing undoubtedly reflects the currently limiteduse of these technologies, it is important to systematicallyevaluate their likely costs and benefits they are widely implemented. We have identified keyfactors that are likely to determine the cost-effectivenessof genetic testing, and they provide a researchagenda for future analyses (Table 2; see also references10 and 76).

Surprisingly, there has been only 1 published, comprehensivecost-effectiveness analysis of testing forBRCA. This analysis found that screening high-riskwomen (Ashkenazi Jews) for BRCA1/2 is relatively cost-effectiveonly if all women who test positive undergo prophylacticsurgery.77 We found only 1 cost-effectivenessanalysis of testing for HER2/neu and treatment withHerceptin.78 Yet in other countries that require cost-effectivenessanalyses for new pharmaceuticals (eg,Australia, the United Kingdom), Herceptin was notimmediately covered because of the high cost of the drugtherapy and the limited impact on survival (therapy hasbeen estimated to increase median survival times from20 to 25 months21).79

CONCLUSION

The expanded use of genetic information offersmany potential clinical benefits but also many economicchallenges. One of those challenges will be managingthe impact of genetic testing on healthcaredelivery and costs. Thus, systematic, evidence-basedtechnology assessments and economic evaluations willhave to be used to guide the incorporation of genomicsinto clinical practice. Future research also will be neededto assess patient preferences and willingness to payfor genomic technologies; how providers can assess anduse genomic technologies; and how the industry, insurers,and government can best balance the relevantcosts and benefits.

Acknowledgments

We are grateful for comments on earlier versions from WylieBurke, PhD, MD, Charles Epstein, MD, Neil Holtzman, MD, MPH,Kathy Giacomini, PhD, Robert Hiatt, MD, PhD, Jennifer Haas,MD, MSPH, Alan Guttmacher, MD, and colleagues at theUniversity of California, San Francisco.

From the University of California, San Francisco, Calif (KAP, DLV, SLB, JS); and theFred Hutchinson Cancer Research Center, Seattle, Wash (SDR).

This work was partially supported by funding to Dr Phillips from the National CancerInstitute (grant R01 CA81130), the National Institute for Allergies and Infectious Diseases(grant R01 AI43744), and the Agency for Healthcare Research and Policy (grants P01HS10771 and P01 HS10856).

Address correspondence to: Kathryn A. Phillips, PhD, Professor of Health Economicsand Health Services Research, and Director, Program in Pharmacogenomics andPopulation Screening, School of Pharmacy, Institute for Health Policy Studies, and UCSFComprehensive Cancer Center, University of California, San Francisco, 3333 California St#420, Box 0613, San Francisco, CA 94143. E-mail: kathryn@itsa.ucsf.edu.

Nature

1. Collins FS, Green ED, Guttmacher AE, Guyer MS. A vision for the future ofgenomics research. . 2003;422:835-847.

N Engl J Med.

2. Collins FS. Medical and societal consequences of the human genome project[Shattuck lecture]. 1999;341(1):28-37.

JAMA.

3. Collins FS, McKusick VA. Implications of the human genome project for medicalscience. 2001;285(5):540-544.

N Engl J Med

4. Varmus H. Getting ready for gene-based medicine. .2002;347(19):1526-1527.

N Engl J Med.

5. Khoury MJ, McCabe LL, McCabe ER. Population screening in the age ofgenomic medicine. 2003;348(1):50-58.

Inquiry

6. Swartz K. The human genome and medical care in the new century. .2000;37:3-6.

Health Serv Res

7. Phillips KA, Veenstra DL, Sadee W. Implications of the genetics revolution forhealth services research: pharmacogenomics and improvements in drug therapy.. 2000;35(5):1-12.

JAMA

8. Phillips KA, Veenstra DL, Oren E, Lee JK, Sadee W. Potential role of pharmacogenomicsin reducing adverse drug reactions: a systematic review. .2001;286(18):2270-2279.

Health Aff (Millwood)

9. Robertson JA, Brody B, Buchanan A, Kahn J, McPherson E. Pharmacogeneticchallenges for the health care system. . 2002;21(4):155-167.

Pharmaceutical-Sciences

10. Veenstra DL, Higashi MK, Phillips KA. Assessing the cost-effectiveness ofpharmacogenomics. . September 14, 2000;2(3):article 29.Available at: www.aapspharmsci.org/view.asp?art=ps020329. Accessed June 16,2004.

N Engl J Med

11. Holtzman N, Marteau T. Will genetics revolutionize medicine? .2000;343(2):141-144.

Nat Genet

12. Weiss KM, Terwilliger JD. How many diseases does it take to map a genewith SNPs? . October 2000;26(2):151-157.

Clinical Utility of Pharmacogenetics and Pharmacogenomics

13. Holtzman N. .New York, NY: Wiley-Liss; 2003.

N Engl J Med.

14. Evans WE, McLeod HL. Pharmacogenomics–drug disposition, drug targets,and side effects. 2003;348(6):538-549.

The Wall StreetJournal

15. Fuhrmans V. Roche test promises to tailor drugs to patients. . June 25, 2003:A3.

Intern Med J

16. Ward RL, Hawkins NJ. Checking the scoreboard: the impact of cancer geneticson clinical practice. . 2001;31:249-253.

17. National Human Genome Research Institute. Beyond the Beginning: TheFuture of Genomics; December 12-14, 2001; Warrenton, Va. Meeting contentavailable at: http://www.genome.gov/10001649. Accessed June 16, 2004.

Am J Hum Genet

18. Easton DF, Ford D, Bishop DT. Breast and ovarian cancer incidence inBRCA1-mutation carriers. Breast Cancer Linkage Consortium. .1995;56(1):265-271.

Cancer Epidemiol Biomarkers Prev

19. Lawrence WF, Peshkin BN, Liang W, Isaacs C, Lerman C, Mandelblatt JS.Cost of genetic counseling and testing for BRCA1 and BRCA2 breast cancer susceptibilitymutations. . 2001;10(5):475-481.

N Engl J Med.

20. Slamon DJ, Leyland-Jones B, Shak S, et al. Use of chemotherapy plus a monoclonalantibody against HER2 for metastatic breast cancer that overexpressesHER2. 2001;344(11):783-792.

Cancer Res

21. Baselga J, Norton L, Albanell J, et al. Recombinant humanized anti-HER2antibody (Herceptin) enhances the antitumor activity of paclitaxel and doxorubicinagainst HER2/Neu overexpressing human breast cancer xenografts. .1998;58:2825-2831.

New Drug Review, Caremark New Drug Quick-Facts

22. Caremark Inc. Biotechnology breakthrough in breast cancer wins FDAapproval. . 1998;1(11).

JAMA

23. Higashi MK, Veenstra DL, Kondo LM, et al. Association between CYP2C9genetic variants and anticoagulation-related outcomes during warfarin therapy.. 2002;287(13):1690-1699.

Pharmacogenomics J

24. Wadelius M, Sorlin K, Wallerman O, et al. Warfarin sensitivity related toCYP2C9, CYP3A5, ABCB1 (MDR1) and other factors. . 2004;4(1):40-48.

J Thromb Thrombolysis

25. Linder MW, Looney S, Adams 3rd JE, et al. Warfarin dose adjustments basedon CYP2C9 genetic polymorphisms. . 2002;14(3):227-232.

26. National Center for Health Statistics. Sample adult person section–publicuse (p 744). 2000 National Health Interview Survey. Available at: ftp://ftp.cdc.gov/pub/Health_Statistics/NCHS/Dataset_Documentation/NHIS/2000/samaDult.pdf.Accessed June 15, 2004.

Health Aff (Millwood)

27. Neumann PJ, Hammitt JK, Mueller C, et al. Public attitudes about genetictesting for Alzheimer's disease. . 2001;20(5):252-264.

JAMA.

28. Lerman C, Hughes C, Trock BJ, et al. Genetic testing in families with hereditarynonpolyposis colon cancer. 1999;281(17):1618-1622.

Time

29. Gorman C. Drugs by design. . January 11, 1999:1-3.

JAMA.

30. Gollust SE, Hull SC, Wilfond BS. Limitations of direct-to-consumer advertisingfor clinical genetic testing. 2002; 288(14):1762-1767.

The Wall Street Journal

31. Pearson H. At-home DNA tests are here. . June 25,2002:D6.

N Engl J Med.

32. Rosenthal MB, Berndt ER, Donohue JM, Frank RG, Epstein AM. Promotion ofprescription drugs to consumers. 2002;346(7):498-505.

N Engl J Med

33. Phillips KA, Flatt SJ, Morrison KR, Coates TJ. Potential use of home HIV testing.. 1995;332(19):1308-1310.

34. Zitner A. Firms sell gene tests directly to public. Los Angeles Times. August11, 2002:A1.

Health Serv Res

35. Phillips KA, Johnson FR, Maddala T. Measuring what people value: attitude orpreference surveys. . 2002;37(6):1659-1679.

Cancer

36. Lee S, Bernhardt BA, Helzsouer KJ. Utilization of BRCA1/2 genetic testing inthe clinical setting. report from a single institution. . 2002;94:1876-1885.

USA Today

37. Friend T. Dying Patients Plead for Unproven Cancer Drugs. . May16, 2001:D.09.

Genet Med

38. Suther S, Goodson P. Barriers to the provision of genetic services by primarycare physicians: a systematic review of the literature. . 2003;5(2):70-76.

Best Pract ResClin Haematol

39. Burke W, Reyes M, Imperatore G. Hereditary haemochromatosis: a realisticapproach to prevention of iron overload disease in the population. . June 2002;15(2):315-328.

Hosp Pract (Off Ed)

40. Holtzman NA. Bringing genetic tests into the clinic. .January 15, 1998;33(1):107-112, 115-106, 121-102 passim.

ANA Publ

41. Scanlon C, Fibison W. Managing genetic information: implications for nursingpractice. . February 1995;(NP-102 2. 5m):i-iv; 1-50.

N Engl J Med.

42. Weinshilboum R. Inheritance and drug response. 2003;348(6):529-537.

Cancer Epidemiol Biomarkers Prev

43. Wideroff L, Freedman A, Olson L, et al. Physician use of cancer susceptibilitytests in the US: results of a national survey. .2003;12:295-303.

N Engl J Med

44. Steinbach G, Lynch PM, Phillips RK, et al. The effect of celecoxib, a cyclooxygenase-2 inhibitor, in familial adenomatous polyposis. .2000;342(26):1946-1952.

Value Health

45. Danzon P, Towse A. The economics of gene therapy and of pharmacogenetics.. January-February 2002;5(1):5-13.

Pharmacogenomics J

46. Lindpaintner K. Pharmacogenetics and the future of medical practice: conceptualconsiderations. . 2001;1(1):23-26.

JAMA.

47. Psaty BM, Smith NL, Heckbert SR, et al. Diuretic therapy, the alpha-adducingene variant, and the risk of myocardial infarction or stroke in persons with treatedhypertension. 2002;287(13):1680-1689.

N Engl JMed

48. Kuivenhoven JA, Jukema JW, Zwinderman AH, et al. The role of a commonvariant of the cholesteryl ester transfer protein gene in the progression of coronaryatherosclerosis. The Regression Growth Evaluation Statin Study Group. . 1998;338(2):86-93.

Encyclopedia of Genetics,Genomics, Proteomics and Bioinformatics

49. Veenstra DL. The clinical and economic implications of pharmacogenomics.In: Jorde LB, Little PFR, Dunn MJ, Subramaniam S, eds. . Chichester, England: John Wiley &Sons, Ltd; 2004.

Clin Chem

50. Linder MW, Prough RA, Valdes R Jr. Pharmacogenetics: a laboratory tool foroptimizing therapeutic efficiency. . 1997;43(2):254-266.

Pharmacogenomics: Social, Ethical, and ClinicalDimensions

51. Feigal DW, Gutman S. Drug development, regulation, and genetically guidedtherapy. In: Rothstein MA, ed. . Hoboken, NJ: John Wiley & Sons, Inc; 2003:99-108.

Food Drug Law J

52. Huang A. FDA regulation of genetic testing: institutional reluctance and publicguardianship. . 1998;53(3):555-591.

53. Cambridge Healthtech Institute. How will the diagnostic company businessmodel evolve to incorporate pharmacogenomics: an interview with Steven Littleof DxS Ltd. 2003. Available at: http://www.healthtech.com/newsarticles/issue37_2.asp. Accessed December 9, 2003.

Nat Biotechnol

54. Persidis A. Pharmacogenomics and diagnostics. .1998;16(suppl):20-21.

Pharmacogenomics

55. Tsongalis GJ. A reality check for molecular diagnostics in clinical practice.. 2003;4(5):667-668.

Arch PatholLab Med

56. Ross JS. Financial determinants of outcomes in molecular testing. . 1999;123:1071-1075.

Nat Med

57. Getting a grip on genetic testing [editorial]. . 2003;9(2):147.

58. Genentech. Full prescribing information: Herceptin® (trastuzumab). Availableat: http://www.gene.com/gene/products/information/oncology/herceptin/insert.jsp.Accessed November 7, 2003.

Int J Technol Assess HealthCare

59. Schoonmaker MM, Bernhardt BA, Holtzman NA. Factors influencing healthinsurers' decisions to cover new genetic technologies. . 2000;16(1):178-189.

Reducing Breast Cancer Risk in Women

60. Hiatt RA. Ethics and economics of genetic testing for breast cancer. In: Stoll B,ed. . Dordecht, the Netherlands: KluwerAcademic Press; 1994:55-68.

Ann SurgOncol

61. Kuerer HM, Hwang ES, Anthony JP, et al. Current national health insurancecoverage policies for breast and ovarian cancer prophylactic surgery. . 2000;7(5):325-332.

J Manag Care Pharm

62. Sullivan SD, Lyles A, Luce B, Grigar J. AMCP guidance for submission of clinicaland economic evaluation data to support formulary listing in US health plansand pharmacy benefits management organizations. . 2001:272-282.

Am J Manag Care

63. Higashi MK, Veenstra DL. Managed care in the genomics era: assessing thecost-effectiveness of genetic tests. . 2003;9(7):493-500.

BMJ

64. Holtzman NA, Shapiro D. Genetic testing and public policy. . 1998;316(7134):852-856.

Science

65. Holtzman NA, Murphy PD, Watson MS, Barr PA. Predictive genetic testing:from basic research to clinical practice. . 1997;278:602-605.

Pharmacogenomics J

66. Lesko LJ, Woodcock J. Pharmacogenomic-guided drug development: regulatoryperspective. . 2002;2(1):20-24.

Pharmacogenomics

67. Salerno RA, Lesko LJ. Pharmacogenomics in drug development and regulatorydecision-making: the genomic data submission (GDS) proposal.. 2004;5(1):25-30.

Pharmacogenomics

68. Eisenberg RS. Will pharmacogenomics alter the role of patents in drug development?. 2003;3(5):571-574.

Pharmacogenomics

69. Issa AM. Pharmacogenomic profiling in postmarketing surveillance: prospectsand challenges. . 2003;4(5):647-655.

Science

70. Holtzman NA. Are genetic tests adequately regulated? [editorial]. .1999;286(5439):409.

71. Feigal DW. FDA's review of genetic tests. Paper presented at: Second Meetingof the Secretary's Advisory Committee on Genetics, Health, and Society; October22, 2003; Washington DC.

Science

72. Rothenberg KH, Terry SF. Human genetics. Before it's too late–addressingfear of genetic information. . 2002;297(5579):196-197.

Soc Sci Med

73. Kaufert PA. Health policy and the new genetics. . 2000;51(6):821-829.

Enhancing the Oversight of Genetic Tests: Recommendations of the SACGT

74. .Bethesda, Md: Secretary's Advisory Committee on Genetic Testing, NationalInstitutes of Health; July 2000.

New Scientist

75. Westphal SP. Your money or your life. . 2002;175(2351):28-33.

J Periodontol

76. Higashi MK, Veenstra DL, del Aguila M, Hujoel P. The cost-effectiveness ofinterleukin-1 genetic testing for periodontal disease. .2002;73(12):1474-1484.

J Clin Oncol

77. Grann V, Whang W, Jacobson JS, Heitjan DF, Antman KH, Neugut AI.Benefits and costs of screening Ashkenazi Jewish women for BRCA1 and BRCA2.. 1999;17(2):494-500.

J Clin Oncol

78. Elkin EB, Weinstein MC, Winer EP, Kuntz KM, Schnitt SJ, Weeks JC. HER-2testing and trastuzumab therapy for metastatic breast cancer: a cost-effectivenessanalysis. . 2004;22(5):854-863.

Reuters Health

79. Bitomsky M. Cost cited in rejecting Herceptin for advanced breast cancer inAustralia. . October 1, 2001. Available at:www.cancerpage.com/cancernews/cancernews3406.htm. Accessed June 24, 2003.