Demand for Evidence Drives Research Process

The expansion of government reimbursement for prescription drugs under the Medicare drug benefit is increasing interest in obtaining more comparative information on drugs and medical products. In addition to the basic safety and efficacy data produced in pre-approval clinical trials, payers want evidence on how a new drug compares to existing therapies and whether it is cost effective. Recognizing that prescription drug coverage for seniors would drive up utilization and spending, the Medicare Modernization Act of 2003 supports more independent research on comparative drug effectiveness; the aim is to limit reimbursement by the Center for Medicare and Medicaid Services (CMS) to those products that demonstrate clinical advantage over available medicines and to reduce inappropriate spending.

Jill Wechsler

Coordinating research

This demand for comparative effectiveness information to gain coverage by public and private payers is affecting biomedical R&D. There has been considerable talk about CMS playing a more visible role in the FDA's drug approval process to accelerate patient access to some new therapies. However, there are concerns that demanding more comparative and effectiveness information from pre-approval testing programs could make drug development even longer and more costly.

Even though regulators agree with industry that a formal coordinated review process is not advisable, manufacturers of drugs and medical devices are consulting more often with CMS early in the development process. Sponsors want to avoid surprise last-minute information requests, such as additional safety data for certain population subgroups. FDA also is working more closely with CMS in preparing guidance on certain research issues, such as the use of pharmacogenomic tests with personalized medicines. CMS and FDA are crafting a memorandum of understanding that permits the two agencies to share proprietary information if needed to accelerate reimbursement decisions.

Conditional coverage

Moreover, CMS' Coverage with Evidence Development (CED) program is shaping drug development and utilization decisions more directly. This controversial initiative authorizes Medicare reimbursement for a drug or medical procedure provided that the sponsor collects additional effectiveness information through clinical trials, patient registries or other assessments. While CMS officials paint this approach as expanding payment for "promising but not persuasive" uses of drugs and medical products, pharma companies are wary that Medicare will routinely require postapproval studies as a condition for new drug reimbursement.

The CED program also has raised ethical questions about protecting the rights of elderly patients who may be pressured to provide personal treatment information in order to obtain coverage for recommended care. Last year, CMS decided to cover four colorectal cancer therapies for off-label uses, but only for patients enrolled in certain clinical trials sponsored by the National Cancer Institute. CMS also proposed to greatly expand the use of PET (positron emission tomography) imaging for diagnosis and staging of a broader range of cancers for those patients who submit data to a PET imaging registry. Although the PET coverage plan seeks to better understand how imaging affects treatment decisions, the program was placed on hold for several months when concerns arose that it could coerce seniors into providing confidential treatment information in order to gain access to scanning services.

Such criticism has prompted CMS officials to revise its draft guidance on the CED program, and a new document has been promised this year. Meanwhile, CMS and the radiologists developing the PET registry devised an alternative process for collecting treatment data on nonconsenting patients. The initiative also avoids extensive informed consent procedures and oversight by IRBs at some 800 participating facilities.

Push for registries

Less controversial is another CED registry to track the effectiveness of implantable cardioverter defibrillators (ICDs) in preventing fatal heart attacks. CMS decided to cover ICD use more broadly in January 2005, and required patients receiving these implants to enroll in the National Cardiovascular Data Registry developed by the American College of Cardiology and the Heart Rhythm Society; it is gathering data on all patients (not just Medicare beneficiaries) receiving ICDs at 1300 hospitals. With more than 125,000 ICDs implanted annually in the United States, the registry expects to collect considerable information.

CMS anticipates that the ICD registry will be the "poster child" for its CED policy, commented CMS special advisor Steven Pearson at a May conference on evidence-based medicine sponsored by Avalere Health. Registration of patients at the time of service should ensure that treatment is targeted to individuals most likely to benefit; data review should point out key differences in outcomes for different providers, devices, drugs, and patients; and all this information should be useful in developing pay-for-performance programs that reward physicians able to provide higher quality care.

These initiatives reflect increased interest in using patient registries to obtain more outcomes information on innovative treatments. This approach may gain wider acceptance with the development of a reference guide on patient registries by the Effective Healthcare Program at the Agency for Healthcare Research and Quality (AHRQ). The agency has posted on its Web site an outline to this guide on establishing, maintaining, and evaluating registries that collect information on patient outcomes following medical treatment, and a final version from consultant Outcome Sciences is due by year-end. The guide will provide case studies on effective trial registries and will address many of the patient protection issues that surfaced related to the PET scan registry.

Comparing effectiveness

While sponsors feel pressure to develop more outcomes information, government-sponsored comparative effectiveness reviews of widely used drug classes are emerging from Evidence-based Practice Centers (EPCs) established by AHRQ's Effective Health Program. A May report comparing drugs to treat anemia in cancer patients, for example, finds little differences in safety and effectiveness. Another review similarly concluded that most second-generation antidepressants have few differences in main effects and adverse events.

States and consumers also seek more evidence to support the use of new medical products. The Drug Effectiveness Review Program (DERP) administered by an Oregon evidence-based policy center provides pharmacy clinical reviews to 15 state Medicaid programs and other organizations. Such information helps states compile preferred drug lists for Medicaid and other state pharmacy programs.

In the international arena, the National Institute for Health and Clinical Excellence (NICE) in the United Kingdom has made cost-effectiveness analysis an acceptable part of the evaluation process for drugs, particularly those therapies that raise coverage questions due to high cost or uncertain effectiveness, pointed out NICE Chief Executive Andrew Dillon at the May conference. This institute conducts technical appraisals of new drugs and medical procedures that consider both clinical and economic issues; final recommendations are for routine or selective drug use or use only in a research setting. A recent high-profile assessment recommends use of aromatese inhibitors to treat beast cancer. NICE also is expected to issue a guidance this summer on appropriate use of Herceptin for early-stage breast cancer. Because the UK health system has a fixed budget, Dillon explained that a local health service that wants to pay for a very costly drug has to cut back somewhere else.

Off the label

One unintended consequence of expanded payer support for more information on real-world drug effectiveness may be to discourage manufacturers from conducting post-approval studies needed to gain FDA approval of new indications. FDA Deputy Commissioner Scott Gottlieb pointed out that if government and private health plans decide to cover an unapproved drug based on their own analysis, sponsors have less incentive to underwrite studies needed to put a new indication on the product label.

This development would only increase off-label drug use, which often is inappropriate, according to another study sponsored partly by AHRQ. An article in the May 8, 2006, issue of the Archives of Internal Medicine finds that about 21% of drug use in 2001—some 150 million prescriptions—was for off-label uses; most alarming is that 73% of those off-label uses had "little or no scientific support."

Industry and FDA officials contend that much off-label prescribing is appropriate. Physicians are highly protective of their right to prescribe drugs off label, particularly for children and cancer patients. FDA does not conduct any specific assessment of off-label uses and limits the practice only by challenging industry-sponsored promotions. But requirements could stiffen if policymakers agree that there is a need for more surveillance and further analysis of the clinical and economic implications of such "non-evidence-based prescribing."

Jill Wechsler is the Washington editor of Applied Clinical Trials, (301) 656-4634 jwechsler@advanstar.com