Year: 2022
Type: article
Abstract: Alzheimer’s disease represents a significant, and growing, public health problem (Bennett, 2021). Although the U.S. Congress has authorized billions of dollars for Alzheimer’s disease research, this spending has not yet led to a major breakthrough in the treatment of Alzheimer’s disease (Kaeberlein, 2019). In this context, a drug developed by Biogen for the treatment of Alzheimer’s disease, aducanumab (Aduhelm), was approved under the U.S. Food and Drug Administration’s (FDA’s) accelerated approval pathway in June 2021 over the strong objections of its own advisory committee (George, 2021). Three members of the committee resigned in protest. Dr. Aaron Kesseheim, one of the members, called the approval of Aduhelm “probably the worst drug approval decision in recent U.S. history.” Clinicians, bioethicists, and other medical experts have raised concerns about the drug’s safety and effectiveness, the quality of clinical trial data on which the approval was based, and the integrity of the FDA process, including accusations that it was inappropriately influenced by Biogen and patient groups who received funding from the company (Liu & Howard, 2021). In the wake of the FDA’s decision, private health insurance companies and Medicare may act as checks on the FDA’s decision by refusing to cover the drug. One month after the FDA’s approval, several private health insurers announced they would not cover the drug, which has helped to limit the sales of Aduhelm (Saltzman, 2021). Medicare is not required to cover a drug just because the FDA approves it, but it usually does. In this case, the Centers for Medicare and Medicaid Services (CMS) has launched a National Coverage Determination (NCD) process, which is expected to conclude in April 2022. The results will help the agency determine whether Medicare will cover Aduhelm (Gusmano, 2021). The NCD process could lead to a decision to cover the drug nationally; to limit coverage to a subset of the Medicare population; to leave the decision in the hands of Medicare Administrative Contractors; or to refuse to cover the drug at all. Until the NCD findings are ready, Medicare coverage will continue to be based on the decisions of Medicare Administrative Contractors in each of 12 jurisdictions. This has already led to an increase in costs for Medicare beneficiaries in the form of increased Part B premiums (O’Brien, 2021). In addition, Medicaid programs are required to cover outpatient drugs approved by the FDA (Gusmano, 2021). States may limit the use of the drug by developing stringent medical-necessity criteria, but this could still lead to substantial increases in Medicaid spending (Salo, 2021). The Aduhelm decision has been the target of significant criticism, but the approval of new drugs on the basis of limited evidence from clinical trials and the increased reliance on surrogate measures, rather than clinical endpoints, are not new. For most of the twentieth century, federal policy was increasingly focused on the need to keep unsafe drugs away from patients (Maschke & Gusmano, 2018). By the 1980s, this concern lost some of its salience. In its place emerged a “patient tragedy” narrative that blamed government regulation for slowing medical progress (Grob, 2011; Wills et al., 2021). These narratives aligned with free-market ideologies, which were gaining influence in the United States. Advocates claimed that patients were harmed by the FDA’s approval process because it took too long and relied on evidentiary standards for safety and effectiveness that were too high. An article published in JAMA in 1974 argued that because of the FDA’s complex approval process, there were fewer new drugs available in the United States than in other countries (Wardell, 1974). In 1987, a paper presented at an Institute of Medicine conference claimed that the slow speed of the FDA approval process inhibited medical innovation because smaller, innovative firms did not have the resources to sustain themselves while waiting for market approval (Roberts, 1987). The Cato Institute, a libertarian think tank, published dozens of reports arguing that the FDA’s approval process for drugs and devices harmed patients and pushed jobs overseas. In a 1995 article, the president of the Cato Institute, a company that produced catheters used to clear arteries, said “by the time we’re approved in the U.S., that product will have been available in Europe on the free market for three to four years,” and another medical device–company executive claimed, “we’re having to move out of the United States. We just can’t take it anymore” (Higgs, 1995, p. 15). Many of these same advocates, with the backing of the pharmaceutical and medical-device industry and some collaborators in academic medical centers, argued that the FDA’s approval process and evidentiary standards harmed medical-product innovation and the country’s reputation as a global leader in the medical-product sector. The Aduhelm decision has been the target of significant criticism, but the approval of new drugs on the basis of limited evidence from clinical trials and the increased reliance on surrogate measures, rather than clinical endpoints, are not new. In response, Congress changed the regulatory approach by instructing the FDA to expedite the drug development and review process. Between 1983 and 2016, Congress adopted a series of measures designed to speed up the FDA approval process and loosen evidentiary standards (Darrow et al., 2020; Maschke & Gusmano, 2018). In 1983, Congress passed the Orphan Drug Act for development of rare-disease treatments. In 1988, it adopted the Fast Track Program, which gave the FDA the ability to approve new drugs and devices based on Phase 2 trials alone. In 1992, the Accelerated Approval Program gave the FDA the ability to issue approvals based on surrogate measures reasonably likely to predict clinical benefit (Food and Drug Administration, n.d.). This was accompanied by the requirement of post-approval studies to verify actual clinical benefits. A surrogate endpoint can take many forms. It may involve a laboratory measurement, a radiograph, or a biomarker that is thought to predict the clinical benefit (Kraus et al., 2019). In the case of Aduhelm, for example, the clinical trials and the FDA relied heavily on measures of amyloid-beta plaque. In contrast, an intermediate clinical endpoint is a measure of a therapeutic effect (Kraus et al., 2019). Using surrogate measures has the potential to speed up the time it takes a new drug to reach the market compared to what might have been required had the trial used a clinical endpoint. Unfortunately, surrogate measures are also more prone to bias than the measurement of clinical endpoints (Gyawali et al., 2019). Moreover, the FDA has less influence after a drug is approved to ensure completion of these studies and examples of withdrawal of approval are rare (Gyawali et al., 2019; Maschke & Gusmano, 2018). In 2012, the Breakthrough Therapy Program was adopted to speed up the approval of drugs that, either alone or in combination with other drugs, may be used to treat a serious or life-threatening disease or condition. The law authorized the FDA to approve such drugs if “preliminary clinical evidence indicates that the drug may demonstrate substantial improvement over existing therapies on one or more clinically significant endpoints, such as substantial treatment effects observed early in clinical development” (Public Law No: 112-144, 2012, p. 126). Even after the adoption of several policies designed to speed up the FDA approval process, some members of Congress argued that the FDA was a barrier to innovation. Senator Lamar Alexander (R-TN) argued that the “disparity between the pace of scientific discovery and [the] FDA’s scientific knowledge is threatening America’s position as a global leader in medical innovation” (U.S. Government Publishing Office, 2015). The next year (2016), Congress adopted the 21st Century Cures Act and continued the trend toward pushing for faster FDA approval. This law authorized the FDA to use evidence from a wider range of sources when making determinations about safety and effectiveness. Along with instructing the FDA to issue guidance addressing the use of complex adaptive and other novel trial designs for premarket approval, it instructed the agency to establish a program to evaluate the potential use of real-world evidence, from sources other than randomized clinical trials (RCTs), which are considered the gold standard of evidence for the approval of biotechnologies. These changes were accompanied by efforts to expand post-marketing surveillance of drugs to mitigate the risks associated with faster approval. The 2007 FDA Amendments Act directed the FDA to establish a system to conduct active post-marketing surveillance of approved drugs and devices using data from multiple sources, including Medicare, the Veterans Affairs’ health system, and private health insurance claims data. But even though drugs that receive accelerated approval must undergo confirmatory studies, these studies are not always completed on time and the FDA is often reluctant to withdraw approval (Rome et al., 2021). Proponents argue that only a small percentage of drugs approved under an accelerated pathway have failed to verify clinical benefits with subsequent information (Beaver et al., 2018). This approach, they argue, provides enormous benefit to patients with life-limiting illnesses by bringing treatments to market years earlier than would have been the case under the previous FDA approval process (Beaver et al., 2018). Critics, however, argue that these changes have led to an expansion of approvals for drugs of questionable value. A review of FDA decisions between 1987 and 2014 found that drugs approved through “expedited development or review programs” were frequently not first in class and “potentially less innovative” (Kesselheim et al., 2015). Furthermore, the price of these drugs is often high, which has led to increased spending and put additional pressure on Medicare and Medicaid budgets to pay for drugs with no established clinical benefits (Rome et al., 2021). In the case of Aduhelm, the FDA followed the procedures that have developed since the 1990s and based its approval on changes in a surrogate measure, rather than evidence of changes in clinical outcomes. Under Aduhelm’s accelerated approval, Biogen has 9 years to show evidence that the drug actually reduces cognitive decline. For the accelerated approval decision, the FDA focused on changes in a surrogate measure, amyloid-beta plaque, which are deposits of the protein fragment beta amyloid that build up in the spaces between nerve cells. The use of surrogate measures is common in Phase 2 clinical trials, and surrogate measures may be used as a basis for FDA approval if there is evidence that they will reasonably predict subsequent clinical outcomes. Beyond this, however, there were decisions specific to Aduhelm. For example, the FDA also allowed Biogen to skip Phase 2 clinical testing, which is designed to confirm results from the Phase 1 trials. Combining Phase 2 and 3 trials has become a more accepted practice in recent years (Korn et al., 2012) and, in the case of coronavirus disease 2019 vaccination clinical trials, manufacturers were allowed to combine Phase 1 and 2 trials (Lurie et al., 2020). Skipping Phase 2 trials entirely is not a common practice. As a result of skipping Phase 2 trials, there was insufficient evidence about how to dose the drug to achieve a balance between the desired clinical outcomes and safety when the drug went into Phase 3 trials. There were mixed results in the Phase 3 trials regarding reductions in cognitive decline. Along with the questions about clinical benefits, there are some clear indications of serious side effects and questions about whether the administration of the treatment, which requires that patients undergo regular magnetic resonance imaging, will be feasible for many of the people who might use the drug. Perhaps most disturbing are the reports that prior to final approval the FDA “avoided conventional scientific scrutiny” (Liu & Howard, 2021) and that the FDA and Biogen had what watchdog organizations describe as “an inappropriately close collaboration” (Public Citizen, 2021, p. 1). These claims are now being investigated by the Office of the Inspector General (Sullivan, 2021). Policy decisions are often justified in terms of their effects on the economy (Berman, 2014; Lindblom, 1977). Calls for lowering evidentiary standards when allowing new health technologies on the market or when using public dollars to purchase these technologies may have greater resonance when proponents emphasize the contributions of technology innovation to economic growth or, conversely, when they suggest that more rigorous standards will drive jobs overseas (Gottlieb, 2011). Debates about epistemology often mask underlying political disputes (Heather, 2015; Rodwin, 2011). As the coronavirus disease 2019 pandemic has illustrated, scientific disputes are often viewed through a partisan or ideological lens (Gusmano et al., 2020). Commenting on the FDA’s 2011 decision to withdraw Avastin’s authorization for the treatment of advanced breast cancer, Daniel Carpenter argued that “the scientific debate over Avastin has been imbued with the ideological overtones of a debate over rationing and the role of regulation in American society” (Silverman, 2011, p. 1). Free-market advocates who want few regulatory controls over the market entry of medical technologies push for relaxing or eliminating regulations because it is consistent with an ideology that privileges freedom from state intervention above other values. While not all patients and advocacy organizations agree with those in favor of greater risk taking when the evidence base regarding safety and effectiveness is incomplete or in dispute, in media accounts the dominant voices are often the “urgency narratives” of patients pushing for access or large advocacy organizations with corporate connections that have a stake in continuing to push for faster adoption of medical technologies (Gusmano, 2013). In the case of Aduhelm, for example, the Alzheimer’s Association was an outspoken proponent of Aduhelm’s approval (Higgins-Dunn, 2021). The efforts of the Alzheimer’s Association in promoting this drug are consistent with their recent efforts to promote biomedical research. The Association is the largest noncorporate sponsor of Alzheimer’s disease research (Caspi, 2019; Cummings et al., 2018). The political pressure that has resulted in faster FDA approval of many drugs and devices is not a guarantee that public and private insurers will pay for the technology. Insurers may want to see different outcome data than what the FDA reviewed for granting market approval, and their evidentiary bar for showing safety and effectiveness may be higher than that needed for FDA approval (Gusmano & Maschke 2021; Maschke & Gusmano, 2018). The political pressure that has resulted in faster FDA approval of many drugs and devices is not a guarantee that public and private insurers will pay for the technology. The Centers for Medicare and Medicaid Services is prohibited from taking cost into consideration when making coverage decisions, but questions about the value of medical interventions in relation to their cost may now take on greater significance, because the FDA approves more drugs and biologics on the basis of evidence from early-phase RCTs with a few patients or from alternative sources of data that may have been collected with less rigor than RCTs. Private health insurers have often taken over as gatekeepers in determining who will get drugs approved by the FDA under its expedited pathways (Thomas, 2017). Answering the question of whether public and private insurers should pay for medical technologies that may harm patients, have an incomplete evidence base, don’t work, or don’t work any better than less expensive technologies requires deciding whether the evidence base justifies payer coverage. Importantly, some payers are beginning to use novel coverage approaches. One disadvantage of this approach is that public and private insurers are subsidizing the research to obtain evidence of effectiveness. An advantage is that insurers are paying for a narrowly defined population of patients to get access to an intervention for a specific period of time, but not paying treatment for all patients diagnosed with the relevant condition. If outcomes data of interest from a coverage study do not meet the requisite evidentiary standards, the insurer is not required to expand coverage to all relevant patients in the clinical setting (Weintraub, 2017). Whether public and private insurers will pay for medical technologies the FDA approves on the basis of real-world evidence, rather than data from RCTs, or with limited data from RCTs remains to be seen. While the FDA is more willing to accept less rigorous data for the purposes of approval, many health insurers, including CMS, remain skeptical. Unless payers have reason to believe that real-world evidence is accurate and representative, they will be reluctant to cover therapies that are expensive and may be ineffective, if not harmful (Szabo, 2017). While the reluctance of public and private payers to pay for new drugs and devices for which there is limited evidence may protect against instances in which the FDA approval process moved too fast, there is a concern that the reliance on payers to call for additional evidence makes it easier for the FDA to issue approvals on the basis of limited evidence, and allows pharmaceutical companies and device manufacturers to get away with failing to conduct adequate clinical trials. It may be prudent for CMS to engage in additional investigation before agreeing to pay when it has questions about the evidentiary standards used by the FDA, but it is important to acknowledge that taxpayers are now subsidizing industry-led research and paying for the FDA’s questionable approval decisions. At best, relying on payers is a patchwork solution to this issue. A longer-term solution must involve strengthening the FDA’s evidentiary standards for product approvals and drug developers conducting rigorous clinical trials that focus on clinical outcomes and accept negative findings. Choices about the approval of new drugs and medical technologies, like Aduhelm, are not purely scientific matters and never will be. They involve debates about the goals of society and how we define what is fair. These decisions also have major consequences for the economy, the size and shape of government, the distribution of power, and the allocation of resources. It is not possible to make decisions of this sort by getting rid of, or getting around, politics. Choices of this sort are what politics is about. Clarifying the nature of these choices and raising questions about the evidence and arguments used to justify competing options will help policymakers muddle through the difficult choices. The Aduhelm case offers a window into how values and interests shape the way evidence is defined, gathered, and evaluated, as well as how the fluidity of those values and interests plays out in relation to the type of technology at issue, the setting in which it is evaluated, and the complex issues associated with its value and cost. None declared. None declared.
Source: Public Policy & Aging Report
DOI: 10.1093/ppar/prac005
Author: Michael K Gusmano
Institutions: Hastings Center, Lehigh University
Language: English
Subfield: Economics and Econometrics
Domain: Social Sciences
Open Access status: bronze
OpenAlex ID: W4214879025