Customize your JAMA Network experience by selecting one or more topics from the list below.
The recent US Food and Drug Administration (FDA) approval of aducanumab for the treatment of Alzheimer disease triggered both praise and criticism from various advocacy groups, but a more important decision regarding the drug—whether to pay for it—will now need to be undertaken by the federal government. This next decision offers a window into both the challenging politics of value-based health care and the difficult administration of pharmaceutical policy in the US.
Alzheimer disease is a devastating illness, with significant unmet clinical need, and effective therapies for the prevention or treatment of the disease would be welcome. There is significant controversy around the assessment of whether aducanumab (a monoclonal antibody that targets amyloid-β) met the FDA’s legislative standard of “safe and effective”1 required for approval. The 2 primary clinical trials (EMERGE and ENGAGE) were stopped for futility according to a prespecified pooled interim analysis, but the final analysis of the individual trials found a statistically significant yet modest slowing of decline in cognition compared with placebo at 78 weeks in the high-dose treatment group of one trial (in the FDA analysis, a decline of 1.35 in the treatment group vs 1.74 in the placebo group on the 18-point Clinical Dementia Rating Scale Sum of Boxes score2). Differences in rates of cognitive decline in the placebo groups in the 2 trials, among other factors, could have accounted for the observed benefit in the trial with positive results.3
An FDA advisory committee recommended, nearly unanimously, against approval because of inadequate evidence of the effectiveness of the drug, and even FDA was left unsure, suggesting that the clinical trial results “left residual uncertainties regarding clinical benefit.”4 Ultimately, FDA reasoned that the drug does reduce a surrogate end point, amyloid plaques in the brain, and that the significant unmet clinical need was a factor in the decision to provide accelerated approval. Moreover, a bayesian approach, given the many negative-outcome clinical trials of other drugs that target amyloid-β, would urge regulators to demand a much larger effect size before approval.
However, the FDA approval does not dictate subsequent payment decisions. As with most countries, approval and reimbursement decisions in the US follow separate processes. Because most patients with Alzheimer disease will be covered under Medicare, the coverage decision is now before Centers for Medicare & Medicaid Services (CMS) administrators. In contrast to the FDA’s “safe and effective” standard, the statutory requirement for Medicare payment is that the product must be “reasonable and necessary” for the treatment of beneficiaries.5 Although the move is highly unusual, CMS has declined to provide coverage for some FDA-approved therapies and diagnostics (including nesiritide for heart failure6 and, of particular relevance, PET amyloid-β imaging).
Beyond the issue of medication costs, a decision to cover aducanumab would prompt a significant change in the treatment of Alzheimer disease nationwide. Because aducanumab is approved based on a reduction of amyloid-β in the brain and the clinical trials required patients to be amyloid positive, screening patients to identify appropriate candidates for treatment might require amyloid PET scans to confirm the presence of amyloid in the brain. Amyloid PET scans are currently covered as a benefit under Medicare only when performed as part of a CMS-approved clinical trial. Furthermore, patients will need ongoing surveillance MRI scans for evaluation of adverse effects, including brain swelling, microhemorrhages, or both, which occurred in up to 40% of patients in the trials.2
One kind of reasonableness assessment would inquire whether the therapeutic value of a drug is at least equal to its cost. Biogen announced (only after the product was approved7) that aducanumab will come to market at an annual price of $56 000. Yet CMS, unlike other national payers, does not require a formal economic analysis of price in its coverage process, nor are CMS coverage decisions constrained by an annual budget (although states do have an annual budget for individuals who are dual eligible for Medicaid coverage and will experience the consequences of the CMS decision). How, then, should CMS assess whether this product is reasonable and necessary?
One approach might be to quantify the benefits that accrue to patients with Alzheimer disease. As patient advocate groups have articulated, the value of an effective therapy to patients with this disease would be significant. Treatment could help reduce the loss of function of patients with Alzheimer disease, reduce the need for supportive care, and markedly improve quality of life. But to date, clinical trials have not shown conclusively that aducanumab provides any of these benefits. Providing hope to patients with Alzheimer disease is another potential benefit. Although conceptually there is an infinite value assigned to such a benefit, reimbursement decisions need to be based on expectations of demonstrated clinical benefit. Furthermore, there could be significant negative consequences for patients and families from creating false hope around a drug.
Another approach is to consider the social benefit from the perspective of the general public. If approved by CMS, aducanumab would be covered as an outpatient physician-administered therapy under Medicare Part B, 72% of which is financed by general tax revenues.8 Approximately 6 million US adults have Alzheimer disease.9 Even at modest levels of uptake, aducanumab would substantially increase Part B drug spending. Assuming 10% of the nearly 6 million US patients with Alzheimer disease receive the drug, overall drug spending could increase by 88%, from an estimated $37 billion per year to $69 billion per year, and at 30% uptake, drug spending could increase by 262%, to an estimated $134 billion per year. Funding this one product could potentially cost every person in the US (current population, 332 406 00010) an estimated $70 annually at 10% drug uptake among patients with Alzheimer disease and $211 annually at 30% uptake. At only 10% drug uptake, Biogen could collect an estimated $32.5 billion in annual revenue, the most ever for a single drug.
Determining that aducanumab coverage is unreasonable would not mean the drug is unavailable. Such a finding would instead prompt a transparent and direct discussion among patients, their families, and physicians concerning the value of the drug. Patients with higher incomes might decide to pay directly for the therapy at this price despite the uncertainty regarding efficacy. To achieve sales targets without reimbursement, Biogen might be forced to implement a tiered pricing structure that might better match the value of the therapy to patients with the price of the drug. However, a major concern with this approach is that low-income populations might be excluded from treatment altogether, absent a major charity program from Biogen.
There are a number of intermediate pathways in which CMS could offer some coverage for aducanumab without exposing the nation to unreasonable and unnecessary expenses. First, there is a question of indication for use. While the initial FDA label was for all patients with Alzheimer disease, FDA has since revised the indication to limit the label to the populations studied in the clinical trials (mild cognitive impairment or mild dementia). On further review of the trial data, CMS could further refine the criteria establishing who may be eligible for this therapy.
Another path could have Congress authorize CMS to pursue creative arrangements with Biogen over the price of the drug. Beyond direct price negotiations, CMS could pursue value-based payment approaches to move reimbursement policy forward through the Center for Medicare and Medicaid Innovation. One value-based concept is a model that relies on patient outcomes, whereby CMS would pay only for patients who demonstrated benefit from the drug. Here, the clinical trials suggest a significant challenge because even patients in the high-dose treatment group had continued decline in cognitive status, and the difference in the rate of decline compared with placebo, even in the trial with positive results, was modest. Payment could be based on changes in amyloid plaque burden, but to date there has been no clinical correlation with these findings in the literature, to our knowledge. In fact, several prior antibody studies have shown substantial reduction in amyloid plaque with no association with improvement in cognitive outcomes. Reimbursement for plaque reduction could result in the resumption of development of many of these failed products.
Another value-based model could involve a conditional payment program whereby patients begin therapy for a fixed period, but then treatment would be discontinued if there were no objective signs of clinical benefit, or if patients met specific criteria for disease progression (for instance, progression from mild to moderate disease). Value-based payment approaches would be novel and would be challenging to implement, but might be justified, considering the daunting clinical and economic aspects of this product.
Currently, the FDA’s accelerated approval of aducanumab still allows the agency to reverse course. The accelerated approval includes an FDA mandate for a phase 4 study to verify the clinical benefit of aducanumab, and the agency could rescind approval if clinical benefits were lacking. However, given the challenges of interpreting the phase 3 trials, the successful design and analysis of a phase 4 program is difficult to envision. Such a clinical trial will also be challenging to implement and complete. Withdrawal of approval, although possible, would also be as challenging as the initial accelerated approval decision, if not more so.
Considering this additional FDA data requirement, CMS could pursue a pathway of coverage with evidence development whereby patients could receive the therapy only through a clinical trial (potentially the study required by the FDA). If Congress would not authorize direct price negotiations with Biogen for general consideration, it might consider authorizing CMS to allow direct price negotiation limited to the research setting.
Aducanumab illustrates the benefits of separating drug approval from drug coverage and the importance of preserving the independence of both the FDA and CMS. The responsibilities of CMS to Medicare beneficiaries and to taxpayers sometimes require limiting or declining what the FDA approves, and the agency owes the public both the creativity and the vigilance required in making difficult coverage decisions.
Corresponding Author: Kevin A. Schulman, MD, Clinical Excellence Research Center, Department of Medicine, 366 Galvez St, John A. and Cynthia Fry Gunn Building, Room 326, Stanford, CA 94305 (firstname.lastname@example.org).
Published Online: July 19, 2021. doi:10.1001/jama.2021.11768
Conflict of Interest Disclosures: Dr Schulman reported receiving grants from NBA Inc, serving as advisor or board member and shareholder for Grid Therapeutics, Faculty Connection, Prealize, and Reserve Therapeutics, investor in Altitude Ventures and Excelerate Health Ventures, receiving consulting fees from Novartis, Cytokinetics, HealthQuest, Business Roundtable, Motley Rice, Frazier Healthcare Partners, and ISMIE, president of the Business School Alliance for Health Management, senior associate editor for Health Services Research at AHA, on the advisory board for Civica RX, and on the board of directors for Catalysis. Dr Schulman also has a patent pending (US 2019/0172566 for mobile patient-centric electronic health record data). Dr Greicius reported being cofounder and shareholder of SBGneuro, which analyzes imaging data for clinical trials. SBGneuro had a contract with Bioclinica to analyze Biogen’s phase 1B aducanumab functional MRI data (contract ended October 2018). SBGneuro was not involved with the EMERGE or ENGAGE aducanumab trials. Dr Richman reported being of counsel at Fairmark Partners.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Schulman KA, Greicius MD, Richman B. Will CMS Find Aducanumab Reasonable and Necessary for Alzheimer Disease After FDA Approval? JAMA. Published online July 19, 2021. doi:10.1001/jama.2021.11768
Create a personal account or sign in to: