Realigning Incentives for Developing and Pricing New Anticancer Treatments

The Centers for Medicare & Medicaid Services (CMS) proposed in late March that Medicare cover sipuleucel-T, a new therapy for patients with terminal prostate cancer. Three independently conducted randomized controlled trials found that sipuleucel-T extends survival by 3 to 5 months. But the price is $93 000 for a course of therapy.¹ It is probably pure coincidence that the announcement occurred as Congress was locked in intense negotiations over the budget, but few recent developments better symbolize the challenges facing lawmakers who want to rein in Medicare spending.

Sipuleucel-T is only one of an increasing number of costly and controversial cancer treatments. Previously, bevacizumab and other genetically engineered chemotherapeutics attracted attention from insurers. These drugs, like sipuleucel-T, extend life by months, not years, and are costly. Between 1995 and 2005, lifetime medical costs for Medicare patients who received a diagnosis of metastatic colorectal cancer and who received chemotherapy increased from $63 000 to $100 000 (2006 dollars).² More drugs are forthcoming; in late March, the Food and Drug Administration (FDA) approved a new treatment for late-stage melanoma, ipilimumab, that extends survival by 4 to 5 months. The cost is $120 000.³

Manufacturers of new chemotherapeutics face few constraints on their pricing power. Oncologists make money from office-administered chemotherapeutics, and most insurers, such as Medicare, cover FDA-approved drugs. Patient cost-sharing provides little incentive to economize on care for privately insured patients who have exceeded their deductibles and for Medicare patients with supplemental coverage.

Standard economic models do not adequately explain pricing patterns for new chemotherapeutics. Why should the price of sipuleucel-T be so much greater than the price of bevacizumab (a chemotherapeutic originally approved for metastatic colorectal cancer) when the benefits are similar? It seems as if social norms, rather than market forces, dictate pricing decisions. For instance, if one company introduces a drug that costs $30 000, insurers complain but still cover the drug. The next company to introduce a drug, seeing that the first was able to charge $30 000, sets its price at $40 000, and the cycle continues.

Although the introduction of sipuleucel-T will have only a small effect on total health care costs, the cumulative effect of the continuous introduction of expensive treatments is that Medicare spending is increasing at an unsustainable rate. Any credible approach to reducing the rate of growth in spending must curb some combination of the introduction, use, and prices of new technologies.

The Patient Protection and Affordable Care Act (PPACA) seeks to reduce costs by enrolling Medicare beneficiaries in accountable care organizations (ACOs). Although cancer patients would benefit from care coordination,⁴ ACOs do not seem particularly well suited for restraining spending on medical technologies. The final rule for ACOs⁵ states, “Nothing in the Shared Savings Program shall be construed to affect the payment, coverage, program integrity, and other requirements that apply to providers and suppliers under [fee-for-service] Medicare,” and that ACOs will not achieve savings by “withholding any needed care that helps beneficiaries.”

President Obama recently proposed that the Independent Payment Advisory Board, a presidentially appointed committee authorized under the PPACA, be given more power to “reduce unnecessary spending.”⁶ “Unnecessary” is a vague term, but, according to common usage in health care settings, it does not apply to costly but effective treatments. In any event, the act explicitly restricts the ability of the board to limit access to new technology, stating that the board's proposals “shall not include any recommendation to ration health care.”

Congress could go one step farther and grant CMS the authority to set coverage policies and reimbursement rates based on drugs' cost-effectiveness, as is the practice in many European countries.⁷ Advocates hope this approach will rationalize medical care. However, organized interest groups may make it difficult to reach “rational” decisions. It is telling that of the 657 public comments received (as of May 7, 2011) by CMS regarding sipuleucel-T, only 11 opposed coverage.⁸

Turning Medicare into a premium-support plan, as proposed in Rep Paul Ryan's “Path to Prosperity,”⁹ is another option. Under the plan, the government would cover a portion of the cost of competing private health plans. The government's contribution would increase at the same rate as the consumer price index, which in the past has increased more slowly than health care spending.

Ryan's plan was widely criticized for shifting the burden of future medical costs from the government to beneficiaries, but these projections assume that health care spending continues to grow at historic rates. By providing incentives to beneficiaries to select less expensive health plans, a premium support reform could reduce costs by completely changing the dynamics of medical innovation.

If beneficiaries value unrestricted access to new technologies, they would migrate to plans that maintain current Medicare coverage policies. But perhaps they would be willing to give up access to expensive treatments in exchange for lower premiums. Health plans with restrictive benefit packages would gain market share. Pharmaceutical and device manufacturers would respond by setting lower prices or by refocusing their efforts on developing products that provide similar benefits to existing treatments but at a lower cost. The rate of increase in health care costs would decline, and coverage policies would better reflect beneficiaries' willingness to pay for medical care.

There are many details to be worked out. Health plans would have to offer a set of standard benefits, defined by the Office of Personnel Management.¹⁰ If the standard benefit package is overly rigid, plans will have no room to compete according to how they cover and authorize use of costly medical technologies. If the benefit package is too flexible, plans may adjust their coverage policies to discourage enrollment by high-cost beneficiaries. For this reason, it is critical that payments to health plans be properly “risk adjusted.”

There is little historical evidence with premium-support plans on which to base spending projections. It is tempting to draw lessons from the handful of large employers, such as the federal government, that use similar schemes in their employee benefits plan. However, Medicare is such a dominant payer that its adoption of a premium-support system would be transformational in a way that employers' use of premium support is not.

From today's vantage point, it is difficult to imagine a health plan refusing to cover an FDA-approved chemotherapeutic for cancer. It is equally difficult to imagine how the United States can reduce the rate of growth in health care costs to a sustainable level if Medicare and other insurers maintain carte blanche coverage policies for chemotherapeutics and other medical technologies.