FDA Panel Gives Qualified Nod to Drug for FAP

Pauline Anderson

May 25, 2012

May 25, 2012 — In a 13 to 4 vote, the Food and Drug Administration's (FDA's) Peripheral and Central Nervous System Drugs Advisory Committee decided that findings of a single study assessing the drug tafamidis in patients with familial amyloid polyneuropathy (FAP) don't meet criteria for efficacy on a clinical endpoint.

However, the committee did decide — in a complete vote reversal — that the study had substantial evidence of effectiveness for a biomarker or surrogate marker endpoint, in this case, tests of muscle strength and small fiber function. FAP is a rare genetic disorder causing build-up of abnormal amyloid proteins.

The FDA will now consider the committee's recommendation when making its decision on the New Drug Application (NDA) for tafamidis meglumine capsules (proposed trade name, Vyndaqel) submitted by FoldRx Pharmaceuticals Inc, a subsidiary of Pfizer Inc.

In advance of the meeting, the FDA's own drug reviewers had recommended rejecting the drug.

Approval Process

Although in the past, drug approval depended on positive results from at least 2 trials, a drug can now be considered for approval with a single adequate and well-controlled clinical investigation, along with confirmatory evidence. However, the findings from that single study are expected to be particularly robust. The agency can also approve a drug on the basis of its effect on a surrogate marker that is reasonably likely to, but not necessarily known to, predict the desired clinical effect.

The new drug is a treatment for FAP, a condition caused by a mutation in the gene that encodes for transthyretin or transporter of thyroxine and retinal (TTR). The drug is a tetramer stabilizer and helps block amyloidogenesis.

FAP is 1 of 2 major phenotypes of familial amyloidoses; the other is familial amyloid cardiomyopathy (FAC). Both are autosomal dominant diseases. FAP is a progressive disease that often manifests in early middle age and results in increasing deposition of amyloid, primarily in the peripheral nerves. It can be accompanied by significant weight loss and can affect the heart, lungs, and kidney. Life expectancy from the first symptom is 10 to 15 years.

An estimated 2500 patients in the United States have FAP; the actual numbers are hard to pinpoint because the condition is often misdiagnosed. The worldwide prevalence of the disease is between 5000 and 10,000, with large clusters in Portugal, Sweden, and Japan. Tafamidis is approved in Europe for this indication under an exceptional circumstances category.

The only available treatment for FAP in the United States is liver transplantation, which removes the abnormal TTR from the circulation. However, such transplantations are costly, are subject to availability, have a significant risk for death, and require lifetime use of immunosuppressants. In addition, not all patients benefit from a transplant and some continue to deteriorate.

Common Mutation

The study under review by the committee, study 005, was an 18-month multicenter, randomized, double-blind, placebo-controlled trial that included 128 patients, all with the same common mutation, V30M. Patients were assigned to daily placebo or 20 mg of tafamidis.

The co-primary outcomes for this study were the Neuropathy Impairment Score–Lower Limb (NIS-LL) (which included assessments of muscle weakness in the ankle, knee, and hip and reflexes and sensations) and the Norfolk Quality of Life Questionnaire (TQOL) domains, which include physical functioning/large fiber; activities of daily living; symptoms, small fiber, and autonomic measures.

Neither of the tafamidis-placebo comparisons for the co-primary outcomes achieved a P value of .05; the P values for the NIS-LL comparison and TQOL were 0.07 and 0.12, respectively.

Secondary outcomes for this study included nerve conduction studies, quantitative sensory testing, and modified body mass index.

Study 005 was followed by a 12-month, open-label extension trial (study 006) that enrolled 71 patients who had completed study 005. The company also carried out several other open-label uncontrolled studies of the drug in patients with other mutations. Committee members heard that the pattern of progression is similar across all mutations.

After extensive debate, the committee voted against the 005 study having sufficient evidence for a clinical endpoint. Some committee members felt the trial was underpowered, whereas others raised the issue of baseline differences: There was some suggestion that patients randomly assigned to tafamidis were less impaired than those assigned to placebo.

Another issue was that the efficacy seemed to be derived from a single site in Portugal, where 58% of the patients were enrolled. In his briefing materials, Russell Katz, MD, director, Division of Neurology Products (DNP), FDA, noted that 76% of the total number of responders in the tafamidis group were patients in Portugal, with the remaining 24% coming from the other centers combined.

Ronald Farkas, MD, PhD, clinical team leader, DNP, FDA, also raised the issue of multiplicity — if a large enough number of endpoints are used, there's a better chance of concluding that a drug is effective when it may not be.

Flawed Study

"This study is not just flawed in one regard, it's flawed in virtually every single way," commented Anne L. Oaklander, MD, PhD, associate professor of neurology, HMS, assistant in pathology (neuropathology), Massachusetts General Hospital, Boston. "We really have to draw the line between wishful thinking and looking at the data."

Also at issue for many committee members was the large drop-out rate — 20% of the patients dropped out because they were selected for a liver transplantation — and the subsequent classification of these drop-outs as nonresponders. There was some discussion that the loss of these patients might have decreased the power to detect any significant between-treatment differences.

Some committee members questioned whether the study results applied to affected patients in the United States. According to the sponsor, about 40% of US patients with FAP have the V30M mutation.

The drug was generally well tolerated, with mostly mild adverse effects and a low discontinuation rate due to adverse events. According to the company, 162 patients have to date been exposed to tafamidis in phase 1 studies.

Although many other committee members agreed there are no significant safety concerns regarding this drug, Dr. Oaklander said she doesn't feel there are enough data to judge safety because of the few patients studied and the short duration they were studied for in relation to the long course of the disease.

Biomarker Evidence

However, the views of the committee were reversed when they voted on whether the study provided sufficient evidence of efficacy for a biomarker endpoint. The vote here was 13 to 4 in favor.

Each member was asked which surrogate marker was most important. Several members said they were impressed with the outcomes for small fiber function, muscle strength, or TTR stabilization. More than 97% of patients receiving the drug exhibited TTR stabilization vs 0% receiving placebo.

For Clifton L. Gooch, MD, professor and chair, Department of Neurology, University of South Florida College of Medicine, and director, USF Neuroscience Collaborative, Tampa, small fiber function is an important endpoint for a patient with neuropathy. It's a part of the nervous system that would be affected early in the disease, he said.

"Small fiber dysfunction is correlated with significant pathology, so it has a direct clinical relevance." He also noted that test results for muscle strength are important since they indicate mobility and the risk of injury.

However, Dr. Oaklander pointed out that the measures used to evaluate small fiber function, including cooling threshold and heat pain threshold for lower limbs, and heart rate response to deep breathing, are not ones identified as optimal by the American Academy of Neurology or the European Federation of Neurological Societies.

"So I would suggest consideration in future studies of the addition of skin biopsy and of other small fiber measures."

TTR Stabilization

For his part, David C. Preston, MD, vice chairman of neurology, professor of neurology, and program director, neurology residency, University Hospitals, Case Medical Center, Cleveland, was not impressed with the results of the muscle strength or small fiber testing because of "problems of multiple testing," but he was impressed by the robustness of the TTR stabilization outcomes.

"This drug was so convincing as far as stabilizing that tetramer," he said. "It makes so much sense that I think a longer trial will show clinical efficacy."

Although he was also "completely convinced" there is an effect on the TTR stabilization, Jeremy Shefner, MD, PhD, professor and chair of neurology, State University of New York Upstate, Medical University, Syracuse, said no strong argument could be made at this point that this has an effect on clinical outcome.

"All of the clinical endpoints, either in aggregate or broken down, are not particularly robust; they didn't reproduce when the study was looked at without that single site."


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: