Getting to answers about Biogen’s Aduhelm

, ,

Exploring RCT, RWE and alternative study designs as a path forward

The FDA’s recent approval of Aduhelm (aducanumab), the first new therapy targeting Alzheimer’s disease since 2003, has drawn both praise and criticism from the healthcare community. While previously approved drugs are intended to manage the symptoms of Alzheimer’s, Aduhelm is the first therapy intended to slow disease progression. However, despite the strong demand from the Alzheimer’s community, there has been intense speculation about the drug’s safety and effectiveness and what comes next for patients. In fact, Google Trends shows a 335% increase in searches for “Alzheimers drugs” compared to the most popular period in the past five years.

The controversy surrounding the FDA’s decision to grant Aduhelm accelerated approval has been well covered in the media. In this article, we will discuss options for Biogen’s path forward.

Understanding the rush to approval

FDA granted Biogen an accelerated approval for Aduhelm on June 7, after having noted that the clinical trials did not demonstrate sufficient evidence to merit full approval. The agency’s accelerated approval program was established in 1992 to make potentially lifesaving therapeutics available to patients in a timely manner by focusing on surrogate endpoints (typically laboratory or pathology markers such as cholesterol) instead of clinical outcomes (such as heart attack, stroke, or death) which take longer to assess. The accelerated approval pathway is available to drugs intended for serious diseases where there is an unmet need and an expectation of clinical benefit. Because Alzheimer’s is the sixth-leading cause of death in the US, with six million Americans currently living with the disease, and there were no approved drugs to treat its underlying pathophysiology, FDA’s decision to grant Biogen access to the accelerated approval track was certainly strongly influenced by unmet medical need.

The Aduhelm clinical trial data clearly demonstrated a reduction of amyloid plaques, which FDA believed was likely to predict a clinical benefit for Alzheimer’s patients. However, the relationship between amyloid plaques and Alzheimer’s disease is still hotly debated. As the condition progresses, amyloid plaques accumulate, and according to the amyloid hypothesis, the buildup of amyloid plaques in the brain is the primary driver of Alzheimer’s pathogenesis. If this hypothesis is true, amyloid plaques would be a valid surrogate for Alzheimer’s progression. However, the Aduhelm trials, consistent with many other studies, found no clear association between amyloid plaque reduction and disease progression. Further investigation of the relationship between amyloid plaques and Alzheimer’s progression is needed.

The postmarketing requirement

As with all drugs granted accelerated approval, Aduhelm’s full approval is contingent on a postmarketing requirement (PMR) to complete a confirmatory trial. FDA specified that Biogen must complete a randomized controlled trial (RCT), “of sufficient duration to observe changes on an acceptable endpoint in the patient population.” The agency has given Biogen nine years to design and complete the PMR.

FDA stipulated that Biogen must select an “acceptable endpoint” but did not dictate what that endpoint should be. Alzheimer’s remains a clinically complex disease that lacks appropriate quantitative response assessments (e.g., labs, images, biomarkers), making any successful trial moving forward incredibly difficult. As discussed, amyloid plaque reduction is not an established surrogate for Alzheimer’s progression, and therefore is not an appropriate endpoint to support full regulatory approval. Assessments of cognitive function are somewhat subjective, and can be susceptible to bias if the patient, or caregiver, is unblinded. Even in typical care settings, patients can demonstrate variability in assessment scores from one assessment to the next. Consequently, demonstrating a clinically meaningful therapeutic effect in cognitive function will be challenging. Given the complexities of assessing Alzheimer’s progression, improved patient survival would be the most definitive marker of delayed disease progression, though notably, one that takes more time to demonstrate. Ideally, the PMR study could be designed to demonstrates a clinically meaningful benefit in cognitive function, or an improvement in patient survival.

Deciphering study designs

As Biogen plans for its postmarketing requirement, it has some options far as study design. The company can go with a classic RCT, use an alternative study design, perform a real-world evidence (RWE) study or conduct an RCT/RWE hybrid design.

Randomized control trials

Randomized, placebo-controlled trials remain the gold standard in study design, because randomization makes them less prone to sources of bias and placebos offer a comparison to accurately assess treatment effect, but they are expensive and time-consuming. If an investigational product has a high expectation of a clinical benefit, there may be ethical implications to denying patients treatment by randomizing to placebo or standard of care. Furthermore, clinical trial populations tend to be younger, healthier, less diverse and have better access to healthcare compared to real-world patients. Consequently, findings from clinical trials are not always generalizable to a real-world patient population. There are also practical implications of conducting a clinical trial for a drug that is already available to patients. Patients interested in participating in a trial, and thus interested in taking Aduhelm, will not want to be randomized to placebo. Therefore, as is often the case with drugs granted accelerated approval, enrolling patients in a trial is likely to be difficult.

Real-world evidence

RWE studies are increasingly being used to provide evidence of safety and effectiveness for drug applications submitted to FDA. RWE studies are typically less expensive, can be conducted in a shorter timeframe and may be able to identify larger, more diverse patient cohorts compared to clinical trials. Because RWE studies are conducted in typical care settings, patient populations are often more representative of the indicated target population, and study results may be more generalizable compared to clinical trials.

Should a classic randomized control study prove infeasible, or unethical, there are opportunities to fulfill the PMR with RWE. A carefully designed RCT conducted in real-world clinical practice (i.e., a pragmatic trial) could fulfill the PMR as it is currently written, and would be both more representative of typical care and less susceptible to bias compared to a fully retrospective RWE study. This type of design could also incorporate ambidirectional controls, a combination of historical and prospective control patients.

Biogen could also conduct an observational study as supportive evidence in addition to a clinical trial. For example, if Aduhelm is effective and a sufficient percentage of the patient population is treated, we would expect patient survival to proportionately improve. An RWE study demonstrating improved patient survival after Aduhelm’s approval would be relatively simple to conduct and could provide strong supportive evidence of a clinical benefit. Additionally, it could provide earlier insights, which could be particularly valuable if it takes a long time to recruit patients via RCT.

Alternative designs

Recognizing that traditional medical product development and approval pathways are not always practical, efficient or even feasible in certain contexts, FDA has embraced multiple initiatives focused on modernizing medical product development and approval. In addition to expedited development, review and approval pathways, FDA has also promoted the use of alternative study designs when classical clinical trials are not a viable option. Adaptive study designs allow for prospectively planned modifications based on accumulating data. Decentralized trials are executed via telemedicine, mobile healthcare providers or local health services. During the Covid-19 pandemic, many trials were forced to become partially or fully decentralized. The primary advantage of decentralized trials is improved convenience and accessibility for patients and providers. Consequently, decentralized trials are likely to demonstrate superior recruitment and retention. A partially decentralized design would have obvious benefits for Alzheimer’s patients, who would not need to frequently travel to academic centers for assessment, and infusions could be given at home, or in local infusion centers. A decentralized design may allow Biogen to enroll a large representative sample of patients very quickly and may be less expensive than a traditional design.

Critical quest

With 6 million patients living with Alzheimer’s disease and an annual cost of more than $355 billion, the demand for a better understanding of the disease and effective treatments has never been greater. Yet, for Aduhelm and other Alzheimer’s therapies in the drug pipeline, answers about efficacy may not come quickly or easily. Well-designed trials, advanced research methodology and collaboration among researchers, providers and patients will be necessary to deepen our collective understanding of this deadly disease.

About the authors

Scott Swain, PhD, MPH, is Director of Regulatory Sciences and Real-World Evidence, Cardinal Health.

Bruce Feinberg, DO, is Vice President of Clinical Affairs and Chief Medical Officer, Cardinal Health Specialty Solutions.

Ajeet Gajra, MD MBBS, FACP, is Vice President and Chief Medical Officer, Cardinal Health Specialty Solutions.