Gene Therapies Need to Have Early Focus on Surrogate Endpoints, Manufacturing
Traditional FDA approval is based on large clinical trial populations and Phase III data, but that approach may not work for a variety of newer targeted drugs, including gene therapies. Jeremy Schafer discusses how an accelerated pathway using surrogate measures, while beneficial for speeding the approval process, may pose challenges for manufacturers and payers alike.
Read the full article below.
Gene Therapies Need to Have Early Focus On Surrogate Endpoints, Manufacturing
AIS Health – Market Access Spotlight
by Angela Maas
The FDA has multiple pathways for companies to get a review of promising drugs for conditions sorely in need of a treatment earlier in their development process than is typical, based on a surrogate endpoint. Following the FDA’s approval of the first three gene therapies last year, the agency recently has indicated that it will issue guidance on these products, including their potential approval based on surrogate measures. Traditional FDA approval is based on large clinical trial populations and Phase III data, but that approach doesn’t necessarily work for many newer targeted drugs, including gene therapies. But this accelerated pathway using surrogate measures holds challenges for manufacturers and payers alike that need to be addressed early in the development process.
The FDA has multiple programs that provide accelerated drug development and review: fast track designation, accelerated approval, breakthrough therapy designation and regenerative medicine advanced therapy (RMAT) designation, which was established in December 2016. The agency in November 2017 released draft guidance on applying the RMAT designation, which some gene therapies may qualify for. Speaking at the May 22 annual board meeting of the Alliance for Regenerative Medicine, FDA Commissioner Scott Gottlieb, M.D., revealed that his agency “soon” would be issuing a framework similar to the RMAT guidance “that will explain how we intend to address manufacturing issues and the development pathway for gene therapy products.”
During his speech, Gottlieb noted that the FDA “has more than 500 active investigational new drug applications involving gene therapy products. We’ve received more than one hundred such applications last year alone. This shows the intensity of scientific work going on in this field.” He also cited an MIT publication that projects the FDA will have approved about 40 gene therapies by 2022 from a pipeline that, as of 2017, had about 930 products in it. Almost half of the approvals, according to MIT, would be in oncology.
The guidance documents the FDA plans to issue will address “product-related issues,” as well as “provide recommendations for clinical development in certain areas,” said Gottlieb. Specifically, they will “lay out potential accelerated approval endpoints for certain gene therapy products,” starting with hemophilia, “where factor production may be sufficient in some cases as a surrogate measure of benefit where a gene therapy product can potentially normalize factor production. In these settings, the demonstration of a reduction in bleeding rates could be confirmed post approval, as we continue to study a product’s long-term safety and durability. The other guidance documents that we’ll release will deal with specific manufacturing and clinical issues related to gene therapy products.”
Traditionally, clinical trials “are generally designed to find a result with the smallest population and shortest duration possible,” explains Jeremy Schafer, Pharm.D., senior vice president and director of payer access solutions at Precision for Value, a consulting and marketing services firm for life sciences companies. “This is understandable from a cost of R&D perspective and will still impact gene therapy development. Gene therapy trials are precision medicine trials. Gene therapy products are highly efficacious with an impressive response rate and can be curative.”
Approvals Are Based on Less Data
“You’re typically seeing gene and cellular therapies go after highly targeted patient populations,” which means their “clinical trials are much smaller,” points out Peter Gilmore, a principal in the strategy practice at KPMG. Many of these approvals also are “accelerated approvals based on much smaller data sets. The concept of Phase III as we’ve known it has gone away” to a large extent as these products are “launching with Phase II data,” he says.
According to Schafer, “some gene therapies received regulatory approval based on a single trial. This is part of the current regulatory efforts to bring forward therapies for serious, life-threatening conditions without viable treatment options.”
“While some critics describe the use [of surrogate endpoints] as ‘lowering’ the approval standard, that characterization fails to recognize the rigor of choosing and evaluating surrogate endpoints that are reasonably likely to yield clinical benefit, particularly in patients with unmet need,” maintains Susan Winckler, president of Leavitt Partners Consulting.
One of the benefits of the RMAT designation, says Arshi Gupta, a director within the health care life sciences strategy practice at KPMG, is that “clients could really get feedback from the FDA quickly,” which can help define the surrogate endpoints a company chooses. She mentions a client that got an RMAT designation and met with the FDA essentially monthly.
Schafer asserts that “surrogate endpoints tend to show results faster than clinical outcomes and enable a clinical trial to show results earlier. If the relationship of the surrogate endpoint and the clinical outcome is strong, as shown by many clinical trials across different drug classes, it is reasonable to assume that improving the surrogate endpoint will lead to improved clinical outcomes.”
However, says Schafer, selecting appropriate surrogate endpoints can be challenging because they “do not always predict clinical outcomes. For example, alteration or removal of beta amyloid plaques with investigational drugs for Alzheimer’s was theorized to improve the disease, but numerous agents have failed to demonstrate clinical benefit despite surrogate endpoint improvement. CETP inhibitors raised HDL cholesterol, a surrogate (it was thought) that would predict decreased heart attacks. However, the CETP agents all failed. Manufacturers can overcome this hurdle by identifying surrogate endpoints that reliably predict improved outcomes with an emphasis on such surrogates being proven in prior work or other drug classes. Manufacturers should also ensure that they have a plan for measuring the clinical outcomes as well in the trial.”
Surrogate endpoints should be “meaningful among health care providers,” Gilmore tells AIS Health. If they aren’t, companies are “introducing an element of commercial risk.” For payers, “oftentimes the endpoints they want to see are not fully aligned with what regulators and health care providers” want to see, which can create reimbursement risk and potentially restrict access to the therapies.
Payers Question Durability of Response
As these gene therapies launch with expedited development and review, the manufacturer will continue to gather data, but for “the payer realm,” this reality can be “problematic,” says Gilmore.
“The biggest question payers have about gene therapy is the durability of response,” says Schafer, citing a survey his firm did in which 72% of payers cited this as “their top information need.…Manufacturers of gene therapies may still develop clinical trials based on small population and short length but should make every effort to track outcomes in a long-term extension.”
Those data are critical to addressing concerns, particularly among payers. According to Winckler, “Where a surrogate endpoint is considered a ‘lab value’ but not indicative of ‘clinical value,’ real-world evidence increases in importance — did the improved ‘lab value’ yield ‘clinical value’ to the patient? Monitoring and tracking the real-world data in a system that yields real-world evidence will be helpful in overcoming some of the skepticism.”
Gene therapies also pose an issue because of their high costs, which are for a one-time treatment (MAS 2/18, p. 1). Novartis Pharmaceuticals Corp.’s chimeric antigen receptor T cell (CAR-T) oncology therapy Kymriah (tisagenlecleucel) launched with a $475,000 price tag. Oncologic Yescarta (axicabtagene ciloleucel), a CAR-T drug from Kite Pharma, Inc. — which was purchased by Gilead Sciences, Inc. shortly before the therapy’s approval — costs $373,000. And Spark Therapeutics, Inc.’s Luxturna (voretigene neparvovec-rzyl), which treats a rare inherited retinal disease, costs $850,000 — $425,000 per eye.
The cost of the treatments is “a major issue,” contends Gilmore. However, administering the drugs, particularly the CAR-T therapies, is “a very complex process” that is “quite costly for everybody involved.” The manufacturing process is “so customized,” and there “are a lot of parties involved, and each bears a lot of the costs and the risks.”
The therapies’ one-and-done nature means that “payers are asked to pay the entire price on the assumption that the drug will work,” Schafer says. “If the drug fails or does not improve the patient’s condition, the payer will have made the full investment for a lifetime of therapy without a reasonable return of improved health. This contrasts with chronic therapy drugs where payers make payments for therapy, often on a monthly basis, and can stop paying if the drug isn’t working.”
This unique treatment and pricing paradigm has prompted payers to develop innovative reimbursement models for these therapies. Schafer tells AIS Health that his research revealed that “multiple payers already have outcomes-based arrangements in place for current gene therapy and CAR-T agents. Payers also showed significant operational readiness and a willingness to implement outcomes arrangements or require a gene therapy to go through a designated channel partner (specialty pharmacy) in order to reduce costs. Payers seemed less willing and able to make annuity payments, but there was still some interest.”
“Gene therapies are an area where the distance between FDA approval and coverage by health insurance may continue to increase,” Winckler tells AIS Health. “These are two different decisions — whether a medical product is safe and effective for marketing in the U.S. versus whether a health insurer/payer will invest their resources in that medical product. Payers will be compelled to continue to evaluate their approach to coverage, financing and benefit design.”
Manufacturing Presents Challenges
With gene therapies, “the handling of samples, especially for autologous therapies,…and the manufacturing of cells…is very different than it is for traditional drugs,” says Gupta. She tells AIS Health that she has seen “many programs shut down” due to manufacturing issues. “Manufacturing and price need to get sorted out.”
“The complexity of the product manufacturing and quality challenges…far exceed those of more traditional drugs or devices,” agrees Winckler.
In his speech, Gottlieb contended that the “current process for producing gene therapy vectors, both lentiviruses and adeno-associated viruses, is relatively inefficient. This has led to a situation where there’s insufficient manufacturing capacity and very high cost. It can cost up to a quarter of a million dollars or more to make [an] investigational product to treat one patient in a clinical trial. These challenges delay — and in some cases prevent — the more widespread development of potentially life-saving medical products. What’s needed are more efficient and standardized production processes for developing gene therapy vectors.”
Because these products “will be more manufacturing-based,” Gupta recommends that manufacturers of gene therapies do the following:
- “Clients should think of producing a quality manufacturing plan” that “should talk about capital investment.”
- Companies should be “thinking about scalability…not just in clinical trials but also in the commercial setting.”
- Firms need to think about “how to come up with viable cogs” to support the development, production and administration of a product. For example, manufacturers need their workforce to understand process automation, and centers administering a therapy need to be “trained in how to use the product.”
“Think of the manufacturing process not only for clinical trials but also for commercial” use, says Gupta.
So with all these dynamics, what should the FDA guidance on gene therapies address?
“Manufacturing parameters, safety measures and the pathway toward clinical development (including potential accelerated approval endpoints for certain gene therapy products),” says Winckler.
“A plan for assessing long-term outcomes in safety and durability of response is key,” contends Schafer. “It is important to note that gene therapies fundamentally change the genetics of patients. Unlike conventional drugs, we cannot assume that the effect of a gene therapy will ‘wear off’ due to drug elimination, half-life, etc. Thus, there should be a key provision that mandates long-term study and reporting. It would not be realistic to require long-term data be done before approval, but it should be an ongoing condition of approval that the FDA enforces. Per current FDA regulations, for gene therapies that satisfy pre-defined criteria, i.e., CAR T-cell products, a long-term follow-up is required up to 15 years to observe for delayed adverse events.”
Contact Gilmore and Gupta through Bill Borden at firstname.lastname@example.org, Schafer through Tess Rollano at email@example.com and Winckler through Jordana Choucair at firstname.lastname@example.org.