J Clin Oncol 28: by American Society of Clinical Oncology INTRODUCTION

Transcription

1 VOLUME 28 NUMBER 20 JULY JOURNAL OF CLINICAL ONCOLOGY O R I G I N A L R E P O R T Comparison of Anticancer Drug Coverage Decisions in the United States and United Kingdom: Does the Evidence Support the Rhetoric? Anne Mason, Michael Drummond, Scott Ramsey, Jonathan Campbell, and Dennis Raisch See accompanying editorial on page 3212 From the Centre for Health Economics, University of York, York, United Kingdom; Fred Hutchinson Cancer Research Center, Seattle, WA; and the University of New Mexico College of Pharmacy and Department of Veterans Affairs Cooperative Studies Program Pharmacy, Albuquerque, NM. Submitted September 28, 2009; accepted March 26, 2010; published online ahead of print at on May 24, Authors disclosures of potential conflicts of interest and author contributions are found at the end of this article. Corresponding author: Anne Mason, MA, Centre for Health Economics, Alcuin A Block, University of York, York, YO10 5DD, United Kingdom; arm10@york.ac.uk by American Society of Clinical Oncology X/10/ /$20.00 DOI: /JCO A B S T R A C T Purpose In contrast to the United States, several European countries have health technology assessment programs for drugs, many of which assess cost effectiveness. Coverage decisions that consider cost effectiveness may lead to restrictions in access. Methods For a purposive sample of five decision-making bodies, we analyzed US and United Kingdom coverage decisions on all anticancer drugs approved by the US Food and Drug Administration (FDA) from 2004 to Data sources for the timing and outcome of licensing and coverage decisions included published and unpublished documentation, Web sites, and personal communication. Results The FDA approved 59 anticancer drugs over the study period, of which 46 were also approved by the European Medicines Agency. In the United States, 100% of drugs were covered, mostly without restriction. However, the United Kingdom bodies made positive coverage decisions for less than half of licensed drugs (National Institute for Health and Clinical Excellence [NICE]: 39%; Scottish Medicines Consortium [SMC]: 43%). Whereas the Centers for Medicare and Medicaid Services (CMS) and the Department of Veterans Affairs (VA) covered all 59 drugs from the FDA license date, delays were evident for some Regence Group decisions that were informed by cost effectiveness (median, 0 days; semi-interquartile range [SIQR], 122 days; n 22). Relative to the European Medicines Agency license date, median time to coverage was 783 days (SIQR, 170 days) for NICE and 231 days (SIQR, 129 days) for the SMC. Conclusion Anticancer drug coverage decisions that consider cost effectiveness are associated with greater restrictions and slower time to coverage. However, this approach may represent an explicit alternative to rationing achieved through the use of patient copayments. J Clin Oncol 28: by American Society of Clinical Oncology INTRODUCTION In the United States, evidence-based health technology assessment (HTA) programs, which examine short-term and long-term consequences of the application of health care technology, have been embraced as a method to improve the value of new drug prescribing. 1,2 The US Centers for Medicare and Medicaid Services (CMS) typically include all US Food and Drug Administration (FDA) approved medications on their formulary, without a formal HTA process. Private health plans make formulary decisions individually, and the process varies widely; decisions are sometimes based on HTA, but this is the exception rather than the rule. 3 In contrast, several European countries go beyond assessing clinical efficacy (benefits achievable under optimal conditions), effectiveness (benefits achievable under real-life conditions), and safety to include assessments of cost effectiveness, the formal approach to estimating value for money. One of the most widely discussed is the National Institute for Health and Clinical Excellence (NICE) in the United Kingdom. Restricting access to new pharmaceuticals is controversial, particularly for life-threatening diseases such as cancer. 4 Anticancer therapies provide an important test case of the nature of restrictions on access imposed as a result of HTA and the implications for the United States of the broader adoption of this approach by American Society of Clinical Oncology

2 Comparison of Anticancer Drug Coverage Decisions This article considers anticancer drugs approved by the FDA from 2004 through The objectives of this study were to quantify and compare the restrictions on access (or lack thereof) to new anticancer drugs imposed in the United States and United Kingdom and to understand the degree to which HTA and evaluations of cost effectiveness played a role in the decisions that were made. METHODS Study Settings National Health Service (United Kingdom). Covering approximately 54 million people in England and Wales, NICE issues guidance to the National Health Service (NHS) on the use of selected health technologies, including pharmaceuticals. Its remit is to consider clinical and cost effectiveness, and its assessments typically include a systematic review of the clinical evidence and an economic model. NICE decisions are guided by a willingness-to-pay threshold of 20,000 to 30,000 per quality-adjusted life-year gained. 5 There are two types of appraisal multiple technology appraisals (MTAs) and single technology appraisals (STAs). MTAs cover at least one technology, and the process is typically approximately 14 months but may be longer. Both new and existing technologies are appraised. 6 STAs cover only one technology, and NICE aspires to provide recommendations within 13 weeks of license. 7 The Scottish Medicines Consortium (SMC), which covers approximately 5 million people in Scotland, operates a technology appraisal program similar to NICE s STAs. The main difference is that guidance is issued on all new drugs. Therefore, it represents a more comprehensive approach, and we consider it here because it increases the number of drugs for which United Kingdom access data are available. CMS. In the United States, the CMS maintains the Medicare and Medicaid programs. Medicare provides coverage for health care items and services for more than 43 million beneficiaries and is the primary provider of health insurance for persons age 65 years or older and all persons with endstage kidney disease. Medicaid is available only to certain low-income individuals and families who fit into an eligibility group that is defined by federal and state laws. CMS is required to cover all reasonable and necessary services. 8 Regarding coverage of anticancer drugs, decisions are in part determined by federal regulations, which stipulate that anticancer drugs must be covered for indications approved by the FDA. 9 New drugs are generally approved for coverage from the date of their FDA approval; thus the coverage decision is effectively one to default to the licensing date. Under a policy in place since 1993, drugs that are used for health conditions that are not FDA approved (off-label use) are covered by Medicare and Medicaid if they are included in specified compendia. 10 Medicare can also pay for noncompendia off-label use in the context of clinical trials. Department of Veterans Affairs. The Veterans Affairs (VA) provides coverage for former members of armed forces. Each year, more than 4 million patients across the United States receive care through more than 150 VA medical centers. Drugs are approved through a national formulary committee, with representatives from all regions. The committee is supported by the Pharmacy Benefits Management group, which reviews and assesses all drugs being considered for inclusion on the national formulary. The Pharmacy Benefits Management group often develops monographs and prescribing criteria for anticancer drugs to assure their appropriate use. Only a few new anticancer drugs are on the formulary, but the nonformulary approval process makes virtually all anticancer drugs available to patients for whom treatment is indicated. The nonformulary process entails completion of forms documenting and justifying the treatment for each individual patient. After chief of service departmental approval, the use of nonformulary drugs is reviewed by the local and regional VA formulary committees. The committees may determine that certain restrictions or controls are necessary. New anticancer drugs are rarely disapproved unless the drug is being used outside of its labeled indication without sufficient justification from the medical literature. Cost or cost effectiveness is secondary to effectiveness in the assessment or control of anticancer therapies. The Regence Group. The Regence Group is a network of Blue Cross Blue Shield affiliated plans offering commercial health insurance to residents in four states in the Pacific Northwest, with approximately 3 million enrollees. Regence maintains a formulary and has an internal HTA group that reviews new drugs following FDA approval, including published and unpublished evidence. On request from Regence, pharmaceutical manufacturers typically provide evidence to Regence s HTA group in the form of a summary dossier that includes reports from clinical trials, unpublished studies, and economic evaluations. Formal evaluations of cost effectiveness are occasionally provided in the dossiers. The HTA group grades each separate study supporting the product into four categories ranging from useful to not useful. On the basis of these assessments, the HTA group presents its evidence summaries and recommendations regarding formulary status to a standing Pharmacy and Therapeutics (P&T) committee comprised of clinicians and other experts not otherwise connected with the company. The P&T committee approves, amends, or rejects Regence s HTA-based recommendations regarding formulary status. Informal input from a standing oncology physician expert panel is sought in cases where the evidence is considered uncertain. Medications that are not put on the formulary may still be covered but usually at a higher copayment. Similar to most private health insurers, drugs that are delivered intravenously in hospital or physician office settings are covered under medical benefit. Drugs that are taken in pill or tablet form or through subcutaneous injection by patients at home are covered under a separate pharmacy benefit. During the years of this study, Regence s HTA process applied only to drugs supplied under the pharmacy benefit. Drugs covered under Regence s medical benefit have historically been covered from the date they are licensed by the FDA. Coverage of drugs under the pharmacy benefit generally occurs some time after FDA approval as a result of time involved for internal review and presentation to the P&T committee for consideration. Identification of FDA-Licensed Anticancer Drugs We included drugs in our sample if they were licensed by the FDA from January 2004 through December 2008 and were specifically for the treatment or prevention of cancer (ie, drugs for treating symptoms such as pain or anemia were excluded). Data sources included the FDA ( Drugs/default.htm), EmaxHealth ( and CenterWatch ( Web sites. Drugs used for different indications were analyzed separately. For the subgroup analysis, we classed drugs as biologic, chemotherapy, or hormonal. Identifying Coverage Data for Anticancer Drugs Data sources for coverage decisions included published and unpublished documentation, Web sites, and personal communications. The end point for coverage decisions was December 31, Our data source for all of NICE s appraisals (MTAs and STAs) of the anticancer drugs in our sample was the NICE Web site ( org.uk/). We obtained marketing authorization dates and licensed therapeutic indications from several Web sites, including those of the European Medicines Agency, Medicines and Healthcare Products Regulatory Agency, and Electronic Medicines Compendium. We identified missing data by contacting the company holding the marketing authorization. We classified NICE s recommendations into the following three categories: as per license (ie, no restrictions), restricted (ie, limited to some patients or indications), and no routine NHS use (ie, total ban). We documented the reasons for and type of restrictions. We also reviewed NICE s appraisals in progress. At the time of analysis, some appraisals had reached the stage of preliminary or final guidance but were still ongoing because of stakeholder comments or appeals. Others were on NICE s to-do list. These details were important because one of the claims about HTA is that it denies access to life-saving therapy merely because of the time taken to conduct the assessments. 11 For the analysis of time to coverage, we included appraisals in progress because NICE provides dates of when by American Society of Clinical Oncology 3235

3 Mason et al Anticancer drugs FDA approved (N = 59) licensed in Europe (EMEA) (n = 13) Covered by Covered by Covered by CMS VA Regence Approved in Europe (EMEA) (n = 46) Formulary (n = 14) Non formulary (n = 41) Medical benefit coverage NA Formulary (n = 14) Non formulary (n = 8) Medical benefit coverage (n = 37) Appraised by NICE (n = 23) Accepted (n = 8) Restricted (n = 10) appraised by NICE (n = 23) NICE in progress (n = 10) NICE proposed (n = 1) Appraised by SMC (n = 42) Accepted (n = 7) Restricted (n = 13) appraised by SMC (n = 4) Fig 1. Anticancer drugs approved by the US Food and Drug Administration from 2004 to end 2008: overview of decisions made by selected United Kingdom and US bodies. FDA, US Food and Drug Administration; EMEA, European Medicines Agency; CMS, Centers for Medicare and Medicaid Services; VA, Veterans Affairs; NICE, National Institute for Health and Clinical Excellence; SMC, Scottish Medicines Consortium. No routine NICE No routine use terminated use (n = 5) (n = 1) (n = 22) assessed (n = 11) guidance is expected. However, this approach may underestimate the actual time taken for NICE to make a coverage decision because it does not account for delays caused by appeals. From the SMC Web site ( we extracted data on the drug name, indication, type of submission, decision date, and recommendation (accepted for use, accepted for restricted use, or not recommended). Where the SMC had reviewed a drug for a particular indication more than once, details of the first and last decisions were recorded. When assessing time to coverage, we used the first decision for the assessment, since we required a measure of the time taken to undertake the initial HTA. However, we used the last decision for the assessment of the final coverage decision, on the grounds that the final outcome was the one of importance to patients. To ensure the SMC and NICE classifications were consistent, SMC restrictions include only cases where eligible patients were unable to access the drug. Therefore, if the SMC designated a drug as accepted for restricted use because only specialists could prescribe the drug, we reclassed this drug as accepted for use. For CMS, we obtained time-to-coverage data based on marketing authorization dates from the US Food and Drug Administration Web site. In the VA, most new anticancer drugs were only available on a nonformulary basis. We obtained details of nonformulary anticancer drug procedures from the VHA (Veterans Health Administration) Formulary Management Process in the VHA Handbook. 12 We also reviewed copies of all drug monographs and prescribing criteria for the anticancer drugs included in this review. The VA restricts all anticancer drug prescribing to oncology specialists. Based on confidential documents from the Regence P&T committee, we obtained the dates of formulary approval and formulary classification (ie, preferred, nonpreferred) for drugs administered in the outpatient setting. The date of approval by the P&T committee is used as the initial date of coverage. In cases where drugs were reviewed under the medical benefit, the date of US Food and Drug Administration licensing is the date of formulary approval. Data Analysis Data were analyzed in Microsoft Office Excel 2003 and StatsDirect statistical software version RESULTS From 2004 through 2008, the US Food and Drug Administration approved 59 anticancer drugs (details are in Online Table 1). For the 46 drugs approved by both agencies, the US Food and Drug Administration date preceded the European Medicines Agency date by 3.6 months on average (median difference: 109 days; semiinterquartile range (SIQR): 171 days). Figure 1 provides an overview of decisions taken by the licensing and regulatory bodies. Figure 2 shows how regulatory bodies varied in the time taken to make coverage decisions. The CMS and the VA covered all 59 drugs from the US Food and Drug Administration license date. For anticancer drugs covered under medical benefit, the Regence coverage decision also matched the US Food and Drug Administration license date. However, time-to-coverage for drugs covered under pharmacy benefit ranged from zero days (ie, identical to the US Food and Drug Administration license date) to 771 days. Relative to the European license date, the median time for NICE coverage decisions was 25.7 months (783 days; SIQR: 170 days). Subgroup analysis of NICE decisions indicated that the single technology (STA) process had not significantly shortened the time-to-coverage compared with MTAs (Mann-Whitney U test: P.783). The SMC made decisions significantly more quickly than NICE, typically taking 7.6 months (231 days; SIQR: 129 days; Mann Whitney U test: P.001). When time-to-coverage was analyzed by drug class, some variations were evident. For Regence, NICE and the SMC, time-tocoverage was shortest for hormonal drugs. Whereas Regence coverage by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

4 Comparison of Anticancer Drug Coverage Decisions FDA to CMS FDA to VA FDA to Regence EMEA to NICE (all) EMEA to NICE (STA) EMEA to NICE (MTA) EMEA to SMC ,000 1,500 min -[ lower quartile - median - upper quartile ]- max Time to Coverage From License Date (days) Fig 2. Time to coverage from license date. FDA, US Food and Drug Administration; CMS, Centers for Medicare and Medicaid Services; VA, Veterans Affairs; NICE, National Institute for Health and Clinical Excellence; EMEA, European Medicines Agency; STA, Single Technology Appraisal; MTA, Multiple Technology Appraisal; SMC, Scottish Medicines Consortium. decisions on biologics were significantly slower than for chemotherapy drugs, the reverse was true for SMC decisions. NICE decisions on biologics and chemotherapy drugs were not significantly different (Mann Whitney U test: P.265). Figure 3 shows how coverage decisions differed between the United Kingdom and US bodies. In the US, all 59 anticancer drugs were covered by all the decision-making bodies, with some drugs subject to partial restrictions. For example, Regence imposed a requirement for prior authorization on 31% of the drugs, all of which were covered under pharmacy benefit. The VA also used prior authorization requirements to restrict access (69% of drugs). 12 In the United Kingdom, only 46 anticancer drugs were licensed for use. NICE made positive recommendations for 39% of the 46 drugs, of which 22% were subject to restrictions. For the (final) SMC decision the corresponding percentages were 43% and 28%. The comparatively low rate of positive approvals by NICE partly reflects the CMS VA Regence NICE SMC Unrestricted Partial restriction Total restriction No decision Coverage Restrictions (%) Fig 3. Coverage restrictions for eligible anticancer drugs, US Food and Drug Administration (FDA) approved 2004 to During the study period, the FDA approved 59 anticancer drugs, of which 46 were licensed in Europe and so eligible for United Kingdom (National Institute for Health and Clinical Excellence [NICE] and Scottish Medicines Consortium [SMC]) coverage decisions. CMS, Centers for Medicare and Medicaid Services; VA, Veterans Affairs. high proportion of drugs without a NICE decision (23/46), some of which were in progress at the time of the analysis (Fig 1). Differences in the distribution of coverage decisions between the five bodies were statistically significant at the 5% level ( 2 : P.001). DISCUSSION To our knowledge, this is the first study comparing access to anticancer drugs in the US with that in a jurisdiction applying HTA. Our study confirms the widely-held belief among Americans that the introduction of economic evaluation as part of health technology assessments can lead to restrictions in access to anticancer drugs. At one end of the spectrum, CMS made all the drugs available from the time of their licensing by the US Food and Drug Administration. At the other end of the spectrum, NICE issued guidance on 23 drugs, recommending that five of them should not be available on the NHS, that 10 should be restricted to certain patient sub-groups and that eight should be available as per license. A limitation of our study is that the selection of decisionmaking bodies was not random, but constrained by data availability, particularly from private payers. Instead, our sample was purposive, addressing a spectrum of managed care (Regence), government payer (CMS) and government provider (NHS, VA). Moreover, the US bodies varied in their use of HTA and of costeffectiveness evidence. Although CMS decisions were not based on HTA evidence, both Regence and the VA employed evidencebased HTA programs in the decision-making process. These decision-makers restricted access by the use of prior authorization requirements and, in the case of Regence, who also considered cost-effectiveness for drugs covered under pharmacy benefit, delays in coverage decisions were evident. Nonetheless, no US decision-maker restricted access completely, whereas NICE often imposed restrictions as result of uncertainties about the clinical data, as well as because of a lack of cost-effectiveness. 14 At first sight, the United Kingdom s restrictions might be considered unacceptable by many in the US. However, US patients face varying copayment levels for almost all branded drugs, which can be 100% for nonformulary drugs. Because newer oncology medicines can cost the patient thousands of dollars each month, these copays can represent substantial barriers to access. Recently, some private health plans have dramatically increased coinsurance rates for branded anticancer drugs covered under the medical benefit, or have moved medical benefit drugs to a pharmacy benefit, allowing them to increase copayments to as much as 50% of the plan s purchase price. 15 This trend is likely to continue, as many of the newer drugs are oral compounds. In the United Kingdom, if the drug is covered, the patient faces no copay. If it is not covered, patients can still access the drug but must pay the manufacturer s price. 16 In the US, the ultimate assessment of value for money paid out of pocket is made by patients, presumably based on the guidance offered by their physician, instead of on independent information from a body like NICE. However, in life-threatening conditions such as cancer, evidence suggests that patient decisionmaking is unlikely to be well-informed. 17 Although pharmacy departments of many US health plans now have analysts with expertise in cost-effectiveness, restructuring formulary decision-making processes to include cost-effectiveness would by American Society of Clinical Oncology 3237

5 Mason et al require significant changes in long-held practices. To date, no health insurer has been willing to alter their decision procedures in this way. 18 The recent Federal initiative on comparative effectiveness research is an important step on the road toward HTA, but it is not clear if this will influence health plan drug formulary decisions. A common criticism of the United Kingdom experience is the delay that the application of HTA introduces. On average, NICE s coverage decisions were made just over 2 years from the license date. withstanding recent changes to accelerate the NICE appraisal process, 19 such delays are likely to be unacceptable in the US. As a solution, researchers and policy makers have proposed coverage with evidence development arrangements for drugs, whereby the drug s formulary listing would be made dependent on evidence of clinical benefits being gathered once the drug is being used. This approach has been used by CMS for devices 20 and has also been used for some anticancer drugs by NICE. However, we are not aware of its use for drugs by private US payers. Oberlander observed that the fate of health care reform in 2009 rests largely on reformers ability to solve the financing puzzle. 21 If cost-effectiveness criteria were to be formally applied in the US, it is likely that the threshold of maximum willingness-to-pay used by decision-makers would be higher than that applied by NICE. In the literature, a figure of $100,000 per QALY is quoted as the threshold for cost-effectiveness in the US. 22 With such a threshold, it is probable that some of the drugs restricted by NICE would have been deemed costeffective in the US, although many more would still be restricted than is currently the case. AUTHORS DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a U are those for which no compensation was received; those relationships marked with a C were compensated. For a detailed description of the disclosure categories, or for more information about ASCO s conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors. Employment or Leadership Position: None Consultant or Advisory Role: Scott Ramsey, Regence Blue Shield (C) Stock Ownership: None Honoraria: None Research Funding: None Expert Testimony: None Other Remuneration: None AUTHOR CONTRIBUTIONS Conception and design: Michael Drummond, Scott Ramsey Collection and assembly of data: Anne Mason, Michael Drummond, Scott Ramsey, Jonathan Campbell, Dennis Raisch Data analysis and interpretation: Anne Mason, Michael Drummond, Scott Ramsey, Jonathan Campbell, Dennis Raisch Manuscript writing: Anne Mason, Michael Drummond, Scott Ramsey, Jonathan Campbell, Dennis Raisch Final approval of manuscript: Anne Mason, Michael Drummond, Scott Ramsey, Jonathan Campbell, Dennis Raisch REFERENCES 1. Sullivan SD, Watkins J, Sweet B, et al: Health technology assessment in health-care decisions in the United States. Value Health 12:S39-S44, 2009 (suppl 2) 2. Fry RN, Avey SG, Sullivan SD: The Academy of Managed Care Pharmacy Format for Formulary Submissions: An evolving standard a Foundation for Managed Care Pharmacy Task Force report. Value Health 6: , Watkins JB, Minshall ME, Sullivan SD: Application of economic analyses in U.S. managed care formulary decisions: A private payer s experience. J Manag Care Pharm 12: , McGreal C: In the US things are getting nasty over NICE. BMJ 338:b2058, National Institute for Health and Clinical Excellence: Guide to the Methods of Technology Appraisal. London, United Kingdom, National Institute for Health and Clinical Excellence, Barham L: Single technology appraisals by NICE: Are they delivering faster guidance to the NHS? Pharmacoeconomics 26: , National Institute for Health and Clinical Excellence: Guide to the Single Technology Appraisal (STA) Process. London, United Kingdom, National Institute for Health and Clinical Excellence, Neumann PJ, Rosen AB, Weinstein MC: Medicare and cost-effectiveness analysis. N Engl J Med 353: , Bach PB: Limits on Medicare s ability to control rising spending on cancer drugs. N Engl J Med 360: , Abernethy AP, Raman G, Balk EM, et al: Systematic review: Reliability of compendia methods for off-label oncology indications. Ann Intern Med 150: , House of Commons Health Committee: National Institute for Health and Clinical Excellence. London, United Kingdom, The Stationery Office Limited, Department of Veterans Affairs: VHA Formulary Management Process. VHA HANDBOOK Washington, DC, Veterans Health Administration, 2009, pp StatsDirect: StatsDirect Statistical Software (version 2.6.6). Cheshire, United Kingdom, StatsDirect Ltd, Mason AR, Drummond MF: Public funding of new cancer drugs: Is NICE getting nastier? Eur J Cancer 45: , Lee TH, Emanuel EJ: Tier 4 drugs and the fraying of the social compact. N Engl J Med 359: , Richards M: Improving access to medicines for NHS patients. A report for the Secretary of State for Health by Professor Mike Richards CBE, November London, United Kingdom, Department of Health, Stacey D, Samant R, Bennett C: Decision making in oncology: A review of patient decision aids to support patient participation. CA Cancer J Clin 58: , Neumann PJ: Why don t Americans use costeffectiveness analysis? Am J Manag Care 10: , Darzi A: High quality care for all: NHS next stage review final report. London, United Kingdom, Department of Health, Tunis SR, Pearson SD: Coverage options for promising technologies: Medicare s coverage with evidence development. Health Aff 25: , Oberlander J: Picking the right poison: Options for funding health care reform. N Engl J Med 360: , Ubel PA, Hirth RA, Chernew ME, et al: What is the price of life and why doesn t it increase at the rate of inflation? Arch Intern Med 163: , by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY