Gene testing stakes a claim in the health benefits marketplace

The application of pharmacogenomics-the science of how genetic variations influence drug response-is new to the field of health benefit management, and its potential to improve care and reduce costs is significant. Medications are at the front line of treating chronic and complex diseases and, yet, a medication on the market may be effective in only 50% of the people who take it, although effectiveness may vary with the disease category.¹ Genomic tests can often indicate whether a patient is likely to benefit from the use of certain medications, guide clinicians to the correct medication dosage, and help to improve safety and effectiveness in the use of medications. This can lead to less waste and higher-quality care.

Pharmacogenomics is building toward a tipping point in the movement to personalized medicine. Not a week goes by without mention of pharmacogenomics in major medical journals, as the knowledge gained since mapping the human genome in 2003 becomes increasingly pertinent to clinical practice. A recent Medco study found that 1 of every 4 patients of the 55 million in the Medco prescription claims database uses a medication for which FDA has included genetic information in the label.² New biomarkers for older medicines will increase this trend toward pairing medications with diagnostics. Drug pipelines are becoming more interconnected, with the use of companion genomic tests to guide clinicians in prescribing medications. Some companies have estimated that between 10% and 20% of drugs in development will be tested along with an associated genomic diagnostic test during the next several years, especially in the field of oncology.³ About 50% of cancer drugs in the pipeline through 2013 are for oral administration, and 85% of them will include genetic biomarkers.⁴ Our knowledge about genes and medicine is growing rapidly and more drugs are increasingly likely to be paired with diagnostic, genomic tests.⁵

While pharmacogenomic testing is gaining traction, there is still some confusion over genomic tests that predict the response to therapy as opposed to genetic tests that predict a predisposition to disease. Direct-to-consumer disease-predictive tests are heavily marketed, although most have no proven clinical value.⁶ By contrast, pharmacogenomic tests have been validated by peer-reviewed research, and their worth in improving drug safety and efficacy is now reflected in many drug labels.

MEASURING THE BENEFITS FOR HEALTH PLAN SPONSORS

Benefit plan administrators understand the value of pharmacogenomics as they strive to improve health outcomes for their members and reduce total plan costs. ⁷ There are obstacles to adoption, but the foundation of research being laid may enable broader use of these tests as their predictive and economic value becomes established.

For health plan sponsors, pharmacogenomic tests may appear to introduce a new cost. But this cost can be offset by savings realized through avoidance of hospitalizations, reduction of waste, greater safety and efficacy in the use of medicines, decrease in drug spending, and avoidance of costs due to disability and rehabilitation. The value needs to be measured by weighing the cost of a test against the benefits to be gained from the clinical and health economic standpoints. Studies currently in progress are expected to show that pharmacogenomics makes economic sense. However, many employers and plan sponsors already appreciate the value in the science and do not believe in the need to wait for further cost-benefit evidence before applying pharmacogenomics to the care of their members. Medco currently has more than 220 clients who now include gene testing in pharmacy benefit plans covering approximately 8 million members.⁸ A 2008 Medco survey showed that larger employers are adopting this coverage faster than small and mid-sized enterprises.9 About 17% of employers who took part in a recent survey of Medco clients incorporated genetic testing into their programs in 2009.⁹ In the same survey, 62% of respondents believe that personalized medicine should be a routine part of healthcare, because it has the potential to reduce the waste caused by trial-and-error treatments.

MIXED SIGNALS ON THE FEDERAL FRONT

The various agencies of the federal government, the country’s largest healthcare payer, are not coordinated in their approach to personalized medicine.¹⁰ FDA is supportive of personalized medicine and continues to work toward changing drug labels to adapt to the science. However, in May 2009, the Centers for Medicare & Medicaid Services (CMS), which oversees the $444 billion (fiscal 2010) Medicare program, limited reimbursement of genetic tests related to the metabolism of warfarin to “coverage with evidence development” ie, to those tests used in clinical studies that will gather evidence of improvements in patient health.^10,11 This followed FDA’s revision of the warfarin label to include a recommendation that patients should receive genetic testing to prevent adverse drug events commonly associated with warfarin treatment.Warfarin, the widely used blood thinner, is notoriously difficult to dose properly. According to results from a joint study by Medco and Mayo Clinic, only 29.2% of patients were likely to experience a “normal response to warfarin” and another 12.2% of patients didn’t need a dose change but frequent monitoring was recommended.¹² The balance of patients were recommended to change the dosage, 25.4% warranting an increase and 33.2% decreasing the dose. Event rates for bleeding or thromboembolism range as high as 16.5% and 25% during the first 6 months of warfarin treatment in usual care settings.^13,14 Subsequently, in January 2010, FDA further revised the warfarin label to include specific dosing recommendations based on patients’ genotypes.¹⁵

SAVINGS THROUGH TESTING

CMS has expressed a willingness to reconsider expanded coverage as further evidence emerges. ¹⁶ The Medco/Mayo Clinic joint study reported a roughly 30% drop in hospitalization rates for patients who received gene tests as part of warfarin therapy initiation.¹² With data from this and other studies, CMS will have the opportunity to re-evaluate its judgment. The decision is especially important for seniors covered by Medicare, who are those most likely to be taking warfarin and other anticoagulants.

Pharmacogenomic science is still in its early stages, and new connections are continually being drawn between genes and drugs. As these connections become stronger, greater efficiencies are likely to emerge. A geneticist recently quoted in The New York Times affirmed that technology is driving the costs of genetic testing down 10-fold every year.¹⁷ While the cost of an entire-genome sequence today is just under $50,000, computing companies and biotechnology firms are competing to cut the cost to less than $1,000. Single cytochrome 450 (CYP450) enzyme tests, which can indicate how drugs are likely to be metabolized by the liver, cost a small fraction of a full genome sequence and can be applied to broad categories of drugs, including antidepressants, pain relievers, cancer treatments, and gastrointestinal medications, among others.¹⁸ For example, the CYP2D6 liver enzyme affects the metabolism of tamoxifen, as well as of beta blockers, antidepressants, and antipsychotic drugs, among other widely used therapies. (For more details visit the Indiana University Department of Medicine chart at: http://medicine.iupui.edu/clinpharm/ddis/table.asp.) The ability to apply results from a single test to cover a variety of medications further strengthens the economic case for pharmacogenomic testing.

RESEARCH AND CURRENT APPLICATION OF PHARMACOGENOMICS

The genetically controlled liver enzyme system called CYP450 affects the body’s response to about a quarter of all drugs, providing the basis of much of the current research connecting genes and medications. For example, results of studies have shown that up to 10% of women do not adequately metabolize the breast cancer drug tamoxifen and thus incur a heightened risk of breast cancer recurrence.^19,20 If clinicians can identify such patients with a genetic test, they may consider other therapies offering a greater potential for preventing recurrence.^21,22Cancer drugs not associated with CYP450 enzymes are increasingly paired with other genetic tests. Patients using imatinib for the treatment of chronic myeloid leukemia are advised to be tested for the Philadelphia chromosome-a biomarker for the disease-every 3 to 6 months after the start of drug therapy to assess the response to treatment.²² Screening for the gene that causes overexpression of the tumor growth factor HER2, a target for trastuzumab, can indicate whether a patient with breast cancer is likely to respond to treatment with this drug.²³

Human immunodeficiency virus (HIV) has emerged as a condition that warrants gene testing as part of a medication protocol. For example, the antiretroviral drugs abacavir and maraviroc illustrate how gene or biomarker tests may influence safety and efficacy. Before receiving abacavir, patients should have a gene test for HLA B*5701 to identify those at risk for a hypersensitivity reaction to the drug, which can cause serious and sometimes fatal complications. Physicians prescribing maraviroc should test for a biomarker that can indicate whether the patient’s HIV type is susceptible to the drug. Maraviroc prevents the virus from replicating by binding specifically to a T-cell receptor known as CCR5. Patients with HIV strains that bind to CXCR4 receptors do not benefit from maraviroc treatment.

Studies are looking more deeply into the application of pharmacogenomic tests in larger trials for a broader range of drugs, while also considering the costs of gene tests and, therefore, the overall cost benefit. Universities and drug manufacturers have been aggressively researching the connections between drug classes and biomarkers. The evidence provided by these studies influences managed care organizations in clinical decisions and rules for coverage, as well as regulators who may request drug label changes as a result of the research. Some publications, such as the bi-monthly Pharmacogenomics Journal, specialize in this field. The subject has also gained steady coverage in general medical journals. For example, New England Journal of Medicine cited “pharmacogenomics” 53 times and “pharmacogenetics” 51 times between 2000 and 2009, with some overlap between articles. Search results for “pharmacogenomics” in the Journal of the American Medical Association disclosed mentions in 66 articles between 1998 and 2009, and 127 mentions of “pharmacogenetics” from 1962 to the same end date.²⁴

As a pharmacy benefit manager, Medco has conducted pharmacogenomic research to help health-plan sponsors make better-informed decisions about gene testing and the use of medications. These decisions are largely driven by analyses of a database of pharmacy and medical claims. The database allows insights into medications prescribed, other drugs that patients received concomitantly, the results of any gene tests performed, and the overall outcomes for patients. Beyond the warfarin study with the Mayo Clinic, ongoing Medco research includes the Genotype-guided Comparison of Clopidogrel and Prasugrel Outcomes study (GeCCO), a head-to-head comparative effectiveness study of clopidogrel versus prasugrel.²⁵ Clopidogrel, an antiplatelet agent metabolized by the CYP2C19 enzyme pathway, is the third largest-selling drug in the United States. ²⁶ It will lose patent protection in 2012. Prasugrel, an anticoagulant from the same drug class as clopidogrel but metabolized by different enzymes, entered the market in 2009. The GeCCO study will enroll more than 14,000 patients to compare outcomes in patients treated with prasugrel with those genetically identified as normal metabolizers of clopidogrel. Additionally, the study will identify patients whose disease does not respond well to clopidogrel and who may benefit from prasugrel on the basis of their genetic profiles.

Medco is also conducting genetic studies with LabCorp on tamoxifen, with Celera Corporation on statins, and with FDA on incorporating genetic testing into medical practice. ^27–30 With the American Medical Association (AMA), Medco conducted a survey of more than 10,000 physicians to determine their attitudes toward gene testing. Although 98% of respondents believed genetics affect drug response, only 10% considered themselves well enough informed about pharmacogenomic testing to put it into practice.³¹

The results of this survey point to a gap between physicians’ perceived value of pharmacogenomic testing and the information physicians have acquired to incorporate such testing into the delivery of care. In addition, the results indicated that the doctors who believed they are well informed about pharmacogenomic tests are twice as likely to order such tests for their patients.³¹ Clearly, clinician education initiatives will be central to encouraging the adoption of these important clinical tools. It also may help both clinicians and patients to know that the tests are often as simple as taking a cheek swab or a saliva collection to obtain a DNA sample; for most of these tests, blood does not need to be drawn.

CONCLUSION

Today we are on the ground floor of this emerging medical innovation. The economic and clinical benefits are becoming increasingly clear, especially as the cost of gene testing declines. When patient populations are better defined by pharmacogenomics, drugs in development are less likely to fail in the late stage of a costly clinical trial. The development of trastuzumab in the 1990s illustrates how gene tests turned a drug of indeterminate value into a real advance in treatment, which now routinely incorporates a test for a HER2 gene.²³ Many products now in development are following the path of pairing drugs with diagnostics. Pharmacogenomic testing also offers opportunities to improve the safety and efficacy of older drugs, as shown by the examples of warfarin and tamoxifen. A growing number of healthcare payers see pharmacogenomic tests as tools that will lead to more precise treatments. The outcome will be safer and more-effective medications for members, leading to improved health outcomes.

A critical next step is broader pharmacogenomic education to equip clinicians to apply the science with confidence in the clinical setting. Surprisingly, the Medco/AMA survey revealed that the small number of physicians now ordering pharmacogenomic tests tended to be out of medical school for 15 to 29 years.³¹ The potential conclusion to be drawn is that younger physicians, more recently out of medical school, are less confident than their elders in using these advances in technology. With medication therapies clearly heading in the direction of personalized medicine, medical schools, training curricula for other health professions, and continuing education programs for clinicians need to do a better job of incorporating pharmacogenomics. The science is racing ahead. Health benefits administrators recognize that greater understanding of pharmacogenomics can help keep patients out of the hospital, improve the likelihood that they will get well, and allow clinicians to practice better medicine.

Dr Barlow is vice president of personalized medicine business solutions at Medco Health Solutions, Franklin Lakes, NJ.

REFERENCES

• Spear BB, Heath-Chiozzi M, Huff J. Clinical application of pharmacogenetics. Trends Mol Med. 2001;7:201–204.
• Frueh FW, Amur S, Mummaneni P, et al. Pharmacogenomic biomarker information in drug labels approved by the United States Food and Drug Administration: prevalence of related drug use. Pharmacotherapy. 2008;28:992–998.
• The Royal Society. Personalised medicines: hopes and realities. September 2005, page 18. Available at: http://royalsociety.org/Personalised-medicines-hopes-and-realities/. Accessed June 14, 2010.
• Medco Health Solutions, Inc. Frueh F. Personalized medicine: health plan strategies to improve outcomes and control Rx costs. Slide 37. AIS audioconference aired on February 26, 2009. Available at: http://www.aishealth.com/Products/C9P05_022609.html. Accessed June 14, 2010.
• Davis RL, Khoury ML. A public health approach to pharmacogenomics and gene-based diagnostic tests. Pharmacogenomics. 2006;7:331–337. Available at: http://www.cdc.gov/genomics/population/file/print/pharmacogenomics.pdf. Accessed June 14, 2010.
• Offit K. Genomic profiles for disease risk: predictive or premature? JAMA. 2008;299:1353–1355.
• Personalized Medicine Coalition. The Case for Personalized Medicine. 2nd ed. Washington, DC: Personalized Medicine Coalition; 2009. Available at: http://www.personalizedmedicinecoalition.org/communications/TheCaseforPersonalizedMedicine_5_5_09.pdf. Accessed June 14, 2010.
• Carey J. Personalized medicine spurred by Medco’s gene testing (Update 2). Bloomberg Businessweek online, March 11, 2010. Available at http://businessweek.com/news/2010-03-11/personalized -medicine-spurred-by-medco-s-cost-saving-gene-tests.html.
• Medco Health Solutions, Inc. 9 leading trends in Rx plan management: Findings from a national peer study [third annual independent national study on prescription benefit management trends]. 2009;20–21.
• Department of Health & Human Services. FY 2011 Budget in Brief. Washington, DC; Department of Health & Human Services; 2010. Available at: http://www.hhs.gov/asrt/ob/docbudget/2011budgetinbrief.pdf. Accessed July 9, 2010.
• Centers for Medicare & Medicaid Services. Available at: http://www.cms.hhs.gov/ContractorLearningResources/downloads/JA6715.pdf. Accessed July 9, 2010.
• Epstein RS, Moyer TP, Aubert RE, et al. Results from the MM-WES (Medco-Mayo Warfarin Effectiveness Study). J Am Coll Cardio., published online March 30, 2010. Available at: http://content.onlinejacc.org/cgi/reprint/55/25/2804.pdf. Accessed June 15, 2010.
• Willey VJ, Bullano MF, Hauch O, et al. Management patterns and outcomes of patients with venous thromboembolism in the usual community practice setting. Clin Ther. 2004;26:1149–1159.
• Beyth RJ, Quinn L, Landefeld CS. A multicomponent intervention to prevent major bleeding complications in older patients receiving warfarin: a randomized, controlled trial. Ann Intern Med. 2000;133:687–695.
• Coumadin [package insert]. Princeton, NJ: Bristol Myers Squibb; 2010.
• Centers for Medicare & Medicaid Services. Pharmacogenomic Testing for Warfarin Response – JA6715, January 8, 2010. Available at:http://www.cms.hhs.gov/ContractorLearningResources/downloads/JA6715.pdf. Accessed July 12, 2010.
• Markoff J. I.B.M. joins pursuit of $1,000 personal genome. The New York Times. October 6, 2009.
• Indiana University Department of Medicine, Division of Clinical Pharmacology. P450 drug interaction table. Available at: http://medicine.iupui.edu/clinpharm/ddis/table.asp. Accessed June 14, 2010.
• Goetz MP, Knox SK, Suman VJ, et al. The impact of cytochrome P450 2D6 metabolism in women receiving adjuvant tamoxifen. Breast Cancer Res Treat. 2007;101:113–121.
• Frasor J, Chang E, Komm B, et al. Gene expression preferentially regulated by tamoxifen in breast cancer cells and correlations with clinical outcome. Cancer Res. 2006; 66(14): 7334–7340. Available at: http://cancerres.aacrjournals.org/cgi/content/abstract/66/14/7334. Accessed June 15, 2010.
• Jin Y, Desta Z, Stearns V, et al. CYP2D6 genotype, antidepressant use, and tamoxifen metabolism during adjuvant breast cancer treatment. J Natl Cancer Inst. 2005;97:30–39.
• Radich JP, Zelenetz AD, Chan WC, et al. NCCN task force report: molecular markers in leukemias and lymphomas. JNCCN. 2009;7[suppl 4]: S7.
• Harries M, Smith I. The development and clinical use of trastuzumab (Herceptin). Endocr Relat Cancer. 2002;9:75–85.
• Results from search of http://content.nejm.org and http://jama.ama-assn.org/ as of Feb. 12, 2010.
• Medco Health Solutions, Inc. Genotype-guided comparison of clopidogrel and prasugrel outcomes study (GeCCO). Study in progress.
• IMS Health Data. 2009 U.S. sales and prescription information. Available at: http://www.imshealth.com/deployedfiles/imshealth/Global/Content/StaticFile/Top_Line_Data/Top%2015%20Products%20by%20U.S.Sales.pdf. Accessed June 15, 2010.
• Medco, LabCorp strike strategic agreement for research on personalized medicine and pharmacogenomics” [press release]. Medco Health Solutions; October 22, 2007.
• Medco launches collaboration to study emerging field of personalized medicine [press release]. Medco Health Solutions; December. 4, 2006.
• Celera and Medco enter research collaboration to evaluate effect of KIF6 testing on compliance with statin therapy in patients with cardiovascular disease [press release]. Medco Health Solutions; September 22, 2009.
• Medco announces research collaboration with FDA focused personalized medicine [press release]. Medco Health Solutions; August 18, 2008.
• Stanek EJ, Sanders CL, Teagarden JR, et al. National pharmacogenomics physician survey: who are adopters of pharmacogenomics among U.S. physicians? Poster presented at: The 59th annual meeting of the American Society for Human Genetics; October 22, 2009; Honolulu, Hawaii. Abstract 2256. Available at: http://www.ashg.org/2009meeting/pdf/poster_4up.pdf. Accessed July 15, 2010.