A designer drug for amyloid polyneuropathy.

Neurology 2012;79:730–731 There are frustratingly few specific treatments available for the 100-odd etiologies of polyneuropathy. Some notable exceptions aside (e.g., dapsone and rifampin for leprosy, IV immunoglobulin for chronic inflammatory demyelinating polyradiculoneuropathy), symptomatic treatments, such as ankle-foot orthoses or drugs for neuropathic pain, form the bulk of the therapeutic armamentarium for polyneuropathy. Specific treatments are even more elusive in the genetic polyneuropathies. Recent hopes from animal models that vitamin C might be effective in type 1 Charcot-Marie-Tooth disease have been unrealized.1 In this context, the report about tafamidis in familial amyloid polyneuropathy (FAP) in this issue of Neurology is of particular interest.2 The amyloid polyneuropathies can be acquired or inherited, and they are classified according to the protein responsible for amyloid fibril formation. Proteins implicated in acquired amyloid neuropathy include immunoglobulin light chains, serum protein A, and -2 microglobulin, whereas in the familial amyloid polyneuropathies (usually referred to as FAP) amyloid formation is a result of misfolding of mutant proteins, including transthyretin (TTR), apolipoprotein A-1, or gelsolin.3 The most common variety of FAP is due to dominantly inherited TTR gene mutations, the most prevalent of which results in substitution of methionine for valine at position 30 (p.M30V) in the 127-amino acid TTR polypeptide monomer. TTR amyloidosis is particularly prevalent in Sweden, Japan, and the Oporto region of Portugal, although cases have been described worldwide. In endemic regions, the disease usually presents in the 30s as a distal painful sensory neuropathy followed by progressive autonomic and then motor dysfunction. Symptomatic cardiac involvement usually becomes evident within 7 to 8 years, and most patients succumb to heart failure or cardiac arrhythmias within 10 years of onset. As the mutated TTR is primarily generated in the liver, liver transplantation has been increasingly used as a treatment for TTRFAP. Randomized trials have not been performed, but there is a general consensus that liver transplantation does have an important and lasting benefit in this disease.4 However, cost, perioperative mortality, and the need for long-term immunosuppression are important drawbacks to transplantation. TTR is a homotetrameric serum protein with binding sites for thyroxine and retinol A. Mutated TTR is unstable and dissociates into mutated TTR monomers, which readily misfold, aggregate, and deposit in tissues as -pleated sheets of amyloid. An appreciation of this mechanism of amyloid formation led to the idea that if one could stabilize the mutated TTR tetramer, and inhibit its dissociation, this would reduce the availability of the amyloidogenic monomer, and thereby decrease amyloid formation and its deleterious consequences on peripheral nerve and heart.5 Tafamidis is a small molecule that occupies TTR’s thyroxine binding sites and stabilizes both normal and mutated TTR tetramers, thus shifting the monomer-tetramer equilibrium away from the amyloidogenic monomers. But does this attractive biochemical hypothesis lead to improvement in human FAP? In the trial reported in this issue of Neurology, Coelho et al.2 randomized 128 patients with V30M TTR-FAP to treatment with either tafamidis 20 mg daily or placebo for 18 months. Two primary outcome measures were used: the Neuropathy Impairment Score (a quantified neurologic examination) and a 35-item, patient-reported quality of life questionnaire. Using an intention-to-treat analysis, there was no significant change in either primary outcome measure in tafamidis-treated patients. The Neuropathy Impairment Score did improve, but the improvement fell slightly short of statistical significance. However, interpretation of these results is confounded by the unexpectedly high dropout rate (21%) in both the tafamidis and placebo groups due to liver transplantation (dropout patients were considered to be tafamidis nonresponders). If the transplant patients are removed from the analysis and only