Does Speed Kill? Managing the Commercial Risks Associated with FDA Accelerated Approval

No one in the biopharmaceutical industry debates that fact that “Time is Money” or that reducing the time a drug spends in development will substantially increase its value. However, as regulatory authorities increase their use of Accelerated Approvals for new cell and gene therapies and other innovative treatments, developers must evaluate an array of potential strategic issues to determine whether or not “speed kills”... ?

Accelerating the development of a novel medicine can substantially increase the value of a development program and provide patients with access to life-saving treatments they might not otherwise have. In recent years, the US Food and Drug Administration has made efforts to increase the speed of development and regulatory approval, including a public commitment to increasing the use of Accelerated Approvals (AA) for novel cell and gene-based therapies. The Accelerated Approval Program allows the FDA to approve therapies for “serious” or “life-threatening” conditions based on surrogate endpoints or biomarkers that are "reasonably likely" to predict clinical outcomes. While the program has created earlier access to products, it has also created controversial and insurers and government payors have restricted access and limited the uptake of products granted AA. The forces were clearly visible during the approval and launch of ADUHELM (aducanumab) which was granted AA for the treatment of Alzheimer’s Disease (AD) by the FDA in 2021. Although it was the first new product approved for AD in nearly 20 years, there were numerous controversies around ADUHELM, and, after generating less than $13 million in total revenue, the product was withdrawn from the US market at the end of 2023.

The ADUHELM experience highlights the risks that developers must consider when designing pivotal clinical studies and developing launch strategies for products that might receive AA. ?While some recent analyses have suggested that payors’ coverage of drugs receiving AA is in alignment with their FDA-approved indications, this is somewhat misleading. A deeper evaluation shows that 4 out of 5 non-oncology drugs that receive AA are restricted for indicated patients and insurers are 4x more likely to place incremental restrictions on non-oncology drugs granted AA compared to oncology drugs. These challenges highlight the need for developers to fully evaluate the potential implications of pursuing accelerated approval when determining a product’s clinical development strategy. This is especially important as the FDA has publicly committed to increasing the use of AA for new cell and gene-based therapies. To mitigate potential commercial risks and to maximize the likelihood of patients being able to access these novel products, developers need to evaluate a variety of factors including the etiology of the disease and whether treatment is considered to be “life-saving” and the capacity of the surrogate endpoints or biomarkers used to predict clinical outcomes.

Despite questions and controversies, Accelerated Approvals are increasing

The Accelerated Approval Program was first created in 1992, during the HIV/AIDS crisis, to expedite the approval of certain drugs for serious or life-threatening conditions. The program provides developers with the opportunity to gain early approval of a product that:

1. Treats a ?“serious” (i.e. “associated with morbidity that has a substantial impact on day-to-day functioning.”) or “life-threatening” condition,
2. Offers a “meaningful advantage over available therap(ies)”, (i.e. no other treatments available or a large proportion of patients have an inadequate response to the available treatments), and
3. Demonstrates “an effect on an endpoint that is reasonably likely to predict clinical benefit”.

After approval, developers are required to complete studies that validate the impact of the product on clinical outcomes. ?Since its inception over 30 years ago, the FDA has granted AAs to 310 different submissions, but, over the past few years, use of the program has skyrocketed, with ~33% (n=102) of all AAs being granted over the last 5 years. Nearly ~85% of the AAs granted during this period have been for oncology products. ???

As the number of AAs has increased, so have the concerns and questions about the program, most notably about the mechanisms available to ensure developers complete on their post-approval commitments in a timely fashion and the appropriateness of approvals based on unqualified biomarkers.

Public health advocates have raised concerns about the AA program and the lack of follow-through by developers on their post-approval commitments on numerous occasions. These concerns were broadcast publicly through a 2022 study conducted by National Public Radio? (NPR) ?which found ~42% of the required post-approval studies had not been started a year after the product was approved and over 16% of the studies had not been started within 3 years of approval.

“Some companies have taken advantage of the Accelerated Approval Pathway, falling behind on providing confirmatory evidence, while FDA has shied away from using its authority to hold drug companies accountable for fulfilling their obligations," Sen. Ron Wyden (D-Ore.), Chair of the Senate Finance Committee,?wrote in a letter to Rob Califf, the Commissioner of the FDA, in 2022. ?

To address these concerns, in December of 2022, Congress passed the Food and Drug Omnibus Reform Act (FDORA) which made several amendments to the Accelerated Approval Program. The amendments:

• Allow the FDA to require the sponsor to have its confirmatory clinical trial underway before the accelerated approval is awarded.
• Require that sponsors of products granted accelerated approval to submit progress reports on their?post-approval studies to the FDA every six months (until the mandated study is completed). The agency will also have to make the information from such progress reports publicly available on its website;?
• Give FDA the option of using expedited procedures to withdraw approval after the confirmatory trial fails to verify the product’s clinical benefit; and
• Require the agency to publish on its website “the rationale for why [a post-approval study] is not appropriate or necessary” whenever it decides not to require such a study upon granting accelerated approval; and
• Order the creation of an intra-agency Accelerated Approval Council, with members from many different FDA constituencies, to “ensure the consistent and appropriate use” of accelerated approval across the agency, including through direct engagement with product review teams that may include training and development of guidance. The Council is also expected to meet at least three times a year and to publish an annual report detailing its activities.

While the FDORA amendments have given the FDA broader authority to manage post-approval commitments, the questions about using unqualified biomarkers for approval remain unanswered. ?These questions have some merit since one out of every six accelerated approvals (17%) has ultimately failed to meet its confirmatory endpoints and has been withdrawn. That noted, the FDA has reinforced its commitment to use even unqualified biomarkers for approval ?When speaking at a workshop at the Reagan-Udall Foundation earlier this year, ?Dr. Peter Marks, the current head of the US FDA’s Center for Biologics Evaluation and Research division (CBER), who is also the chair of the FDA’s newly formed Accelerated Approval Council, noted that the use of unqualified biomarkers is common, “There are plenty of [unqualified] ?biomarkers that have been used over the years … As long as you can show that you have a biomarker that can be measured accurately and reproducibly, we don’t necessarily always need all of the additional work that gets to qualification.”

Ultimately, the FDA is planning to increase its use of AAs over the next few years as the FDA, most notably with novel cell and gene-based therapies. On numerous occasions, Dr. Marks has highlighted the FDA’s commitment to expand its use of the AA program. “If we don’t lean into accelerated approval, we’re going to leave a lot of patients behind,” Marks said during his remarks at the Reagan-Udall Foundation. “Accelerated approval is going to be the norm for a lot of gene therapies [moving forward].”

As Accelerated Approvals Increase, so have the Commercial Challenges

While the FDA is moving towards the increased use of biomarkers and accelerated approvals, commercial and government payors have been far less enthusiastic about this trend.?

The most notable commercial challenge for a product receiving accelerated approval was faced by ADUHELM (aducanumab), a product that was developed by Biogen and Eisai for the treatment of Alzheimer’s Disease (AD). ADUHELM received AA in 2021 based on its ability to reduce amyloid-β?plaque, a controversial biomarker that was connected to the pathophysiology of AD but had not been demonstrated to be linked to the progression of the disease. Upon granting AA, the FDA received strong pushback primarily due to questions about the product’s efficacy, as ADUHELM had failed to meet the pre-specified efficacy endpoints in its pivotal trials, and its safety, as it was associated with an increased risk of brain bleeds and swelling. ?The FDA defended their decision to approve ADUHELM with Commissioner Janet Woodcock stating at the time that the approval was “very solid” and the submission contained “a lot more supportive data” than other products that had received AA. The FDA’s support notwithstanding, in 2022 the Centers for Medicare and Medicaid Services (CMS) published its National Coverage Decision and declined to reimburse ADUHELM unless the product was provided within an FDA-approved randomized controlled trial. Other commercial insurers followed suit, effectively blocking the commercial use of ADUHELM. Without a pathway to reimbursement, ADUHELM was unable to gain commercial traction, and at the end of 2023, after generating a total of $13 million in cumulative revenue, Biogen withdrew the product from the market.

Analyzing Commercial Restrictions on Products with Accelerated Approvals

?So how can developers take advantage of opportunities to accelerate regulatory approval while minimizing the risks of commercial restrictions like those placed on ADUHELM? ?The answer can be found, in part, by analyzing insurers’ coverage policies for products that have previously received AAs.

Tufts Medical Center has compiled a unique dataset as part of its Specialty Drug Evidence and Coverage (SPEC) Database, which was designed to enhance the transparency of commercial health plan specialty drug coverage. SPEC contains data on publicly available medical and pharmacy coverage policies issued by 18 of the largest commercial health plans accounting for ~170 million lives, or ~70% of the market for specialty products, including those that have received accelerated approval.

An analysis of the SPEC data published on the Tufts Website in March 2024 titled “Despite Criticisms of Accelerated Approval Pathway— Commercial Payers Defer to FDA” showed that 49% of the coverage policies for drugs receiving AA between 2016 -2021 were equivalent to the drug’s labeled indications (i.e. drugs were reimbursed for the set of patients included in the product’s initial label). Twenty-eight percent (28%) of insurers’ coverage policies for AA drugs included restrictions beyond the FDA label, and 23% of coverage policies included patients that were outside of the labeled indications (Figure 1)

.

The authors’ primary conclusion states “that, despite criticisms of the AA program, health plans frequently permit their enrollees’ access to AA drugs through coverage policies.” This statement is true, but it underemphasizes the stark differences between products that received AA for oncology indications vs. non-oncology indications. ?

Digging deeper, the SPEC database shows that non-oncology products receiving AA were ~4x more likely to have additional access restrictions (82% of coverage policies restricted the use of the product to a subsegment of the indicated population) than oncology products (only 21% of policies had similar restrictions) (Figure 2). ?The difference between the two segments is even more noticeable when considering that more than one in four oncology drugs receiving AA were reimbursed for patients for which the product was not even indicated

This data confirms the different standards that insurers apply when evaluating oncology products receiving AA vs. non-oncology products. It also highlights the different level of commercial risks and challenges that companies developing non-cancer treatments eligible for AA will need to address.

This difference between oncology and non-oncology products is even more important when considering the FDA’s commitment to increasing the use of the program, specifically to bring more cell and gene therapies to the market. According to the Alliance for Regenerative Medicine, ~75% of the ongoing ?Phase II and Phase III studies for cell and gene therapies are for non-cancer-related indications.

Key Considerations When Evaluating the Risks and Benefits of Accelerated Approval?

Regardless of the technology or indication, there are a variety of factors that developers need to evaluate when trying to determine the commercial risks and benefits of seeking AA. Leveraging some of the insights gained from the SPEC database and considering the differences between oncology and non-oncology indications, two key factors come into focus: ?(1) the etiology and urgency of treating the disease and (2) the capacity of surrogate endpoints or biomarkers used to predict the desired clinical outcome.

1.??????? “Serious” vs. “Life-Threatening” Diseases and the perceived urgency of treatment

Although accelerated approval can be granted to products that treat either “serious” or “life-threatening” diseases, analysis suggests that payors are significantly less likely to restrict access to drugs granted AA that treat “life-threatening” conditions and more likely to restrict those that treat only “serious” diseases. No one would debate that cancer is “life-threatening”. The consequences of not aggressively treating most forms of cancers is life or death with survival times measured in months. Unless diagnosis occurs very early in the disease progression, both physicians and patients have a tremendous sense of urgency in gaining access to potential new cancer therapies which, in turn, puts significant pressure on payors to provide access to these new, potentially “life-saving” therapies.

The story for most non-oncology products receiving AA is different. The majority of non-oncology products receiving AA have been approved to treat congenital diseases associated with progressive loss of function (e.g. Duchenne Muscular Dystrophy (DMD), Sickle Cell Disease, etc). While these diseases cause significant morbidity and ultimately lead to early mortality, their progression can take years. The average life expectancy for a patient with DMD is 29.9 years. Despite patient and physician advocacy efforts, insurers do not appear to have the same sense of urgency when evaluating treatments for these congenital, progressive diseases. After the approval of ELEVYDIS for boys ages 4-5 with DMD last year, numerous articles were published that highlighted families’ struggles with insurers denying reimbursement for the product, in some cases stating that the treatment was not a ?“medical necessity” or ?“questioning the urgency” of the treatment. While some questions about reimbursing ELEVYDIS may be linked to the inconclusive efficacy of the drug (it was disclosed that ELEVYDIS missed an efficacy endpoint in one of its pivotal clinical studies) or its high price ($3.2 million for a dose of the single-use, gene-editing therapy), the urgency of treatment had a clear impact. ??

The quick progression and “life-threatening” nature of cancer also reduces the amount of time required for a new product to demonstrate a significant clinical benefit when compared to the standard of care. As an illustration, in March 2022, Immunogenics submitted an NDA for ELAHERE (mirvetuximab soravtansine-gynx) for use in adult patients with FRα positive, platinum-resistant, epithelial ovarian, fallopian tube, or peritoneal cancer, who have?received?one to three prior systemic treatment regimens. The product was granted AA in November 2022. The required post-approval outcomes study, a continuation of the study used to gain AA in ELAHERE demonstrated an increase in progression-free survival of 5.6 months compared to 4.0 months with chemotherapy regimens was completed only 5 months later in March 2023. ELAHERE was granted full approval from the FDA in March 2024, two (2) years after its initial NDA submission.

On the other hand, demonstrating a statistically significant impact on morbidity or mortality for ?“serious” non-oncology diseases can often require a long follow-up period. Bluebird bio’s SKYSONA (elivaldogene autotemcel) was granted AA in 2022 for children aged 4-17 with Cerebral Adrenoleukodystrophy, a rare congenital disorder. The required post-marketing commitment is an outcomes study measuring the time to progress to a “Major Functional Disability”. The FDA is expecting an initial interim report from the study to be submitted in 2027 and the final study report is not expected to be submitted until 2032, ten (10) years after its approval. While some insurers have decided to provide access to SKYSONA without a demonstrated long-term outcomes benefit, others are skeptical. Regardless, it will be a long time before any new clinical data on the product will be available.

2.??????? The strength and predictive capability of the proposed biomarkers

The FDA requires that biomarkers used as the basis for AA be “reasonably likely to predict effect,” and has defined a rubric by which this standard can judged. The rubric is based on the following criteria:

The above noted, the determination of what is “reasonably likely” is highly subjective and it has been applied differently by the FDA across different disease states. Furthermore, the FDA does not have a framework for determining the minimum change in the surrogate endpoint that would be considered “reasonably likely” to translate into a meaningful clinical benefit. Ultimately, payors are not held to the same standards as the FDA and they are much more likely to restrict products that are approved based on a biomarker that has questions about its ability to predict outcomes. The FDA’s subjectivity and inconsistency in evaluating surrogate endpoints creates additional complexity and risk for developers trying to define a clear economic value proposition for their product.

The SPEC data provides a clear illustration of this point. Given the amount of cancer research that has been conducted, there are reasonably well-established links between biomarkers, surrogate endpoints (e.g. response rate, tumor shrinkage, etc.), and long-term outcomes (e.g. survival). This is not the case with non-cancerous conditions, where the links between biomarkers (e.g. reduction of amyloid-β?plaque, increase in levels of dystrophin) and outcomes (e.g. improved cognitive function patients, delayed loss of motor function) may not be as conclusive. As the uncertainty around the predictive nature of a biomarker increases, so does the likelihood that payors will question the value proposition of a product and restrict its access.

It should be noted that even though biomarkers for oncology are better established than those for non-oncology products, there are still questions about their ability to predict clinical outcomes. ?A study by Liu, et al, presented at the American Association for Cancer Research (AACR) Annual Meeting in April 2024 investigated whether cancer drugs granted AA ultimately demonstrated clinical benefit. The study found that only 43% of the cancer drugs granted AA had demonstrated an overall survival or quality of life benefit in confirmatory trials within 5 years of the approval. ?Regardless of the therapeutic area, developers need to understand the commercial risks and benefits associated with different biomarker(s) and surrogate endpoints that might be used to gain accelerated approval.

Increasing Acceleration, Limiting Commercial Risk

The factors and risks mentioned above decrease a developer's leverage in establishing an optimal price for their product and constructing a clear value proposition for payors. ?However, regardless of the pricing strategy or the indication being pursued, developers planning to use the AA pathway should involve payors early in the development process to gain a detailed understanding of what they will expect and/or require for reimbursement. Commercial risks need to be identified early and mitigation plans developed to avoid commercial failures like ADUHELM. Biomarkers with questionable links to outcomes should be fully evaluated and developers should publish as much evidence as possible to demonstrate the predictive nature of the surrogate endpoint selected. If the targeted disease has a "slower" progression, developers will need to work with advocacy groups and develop programs to increase awareness and reinforce the need to aggressively treat and manage the disease. Identifying and addressing these challenges will help developers realize the benefits of accelerated development and approval without increasing the risk of a commercial failure.

?