Multiple Sclerosis Research: XenoPort to drop Arbaclofen for MS spasticity

Another one bites the dust. Pharma is an expensive and risky business. #MSBlog #MSResearch

Vrinda Manocha. XenoPort to drop multiple sclerosis drug, shares fall. Reuters May 20, 2013 12:32pm EDT.

XenoPort will stop development of Arbaclofen placarbil an experimental MS treatment it hoped to launch in 2015 after a late-stage trial failed to show significant improvement over a placebo. Arbaclofen placarbil, which was intended to treat spasticity, stiffness and involuntary multiple spasms, was tested on 228 MSer in the United States. Dosages of 30 mg and 45 mg were administered twice a day. XenoPort said seven patients experienced adverse events, none of which were related to the treatment. XenoPort said on Monday that arbaclofen placarbil did not show statistical significance compared to the placebo on two clinical scales - severity of symptoms and response to treatment, and spasticity.

The drug, arbaclofen placarbil, was XenoPort's only product in late-stage trial, according to information on the company's website. Xenoport spent about $13.2 million in 2012 on the development of the drug, according to a March regulatory filing. A XenoPort executive said more resources could be allocated to the company's experimental drug XP23829 to treat relapsing-remitting multiple sclerosis. XP23829, currently being tested in an early-stage trial, is a derivative of Biogen Idec's multiple sclerosis drug dimethyl fumarate (Tecfidera).

For full press release

"This study shows that even in late stage development (phase 3), with a drug that has more than a fighting chance, things can go wrong and you get a negative result. The www.clinicaltrials.gov trial entry (NCT01359566) suggest they did not use an enrichment design. Why? GW Pharma have shown so elegantly in their Sativex trials that this is the only way to do spasticity trials. In addition, the study used a co-primary outcome; i.e. (1) change from Baseline in maximum Ashworth scale score (6 hour post-dose time point) and (2) a Patient Global Impression of Change (PGIC) score. We now know that the Ashworth is not a good outcome measure; it is not responsive to change and unreliable. At first glance 228 MSers seems to be massively underpowered for these outcomes; there appears to have been 4 arms (placebo vs. 15mg vs. 30mg vs. 45mg) or 57 MSers per arm. For the Ashworth scale you need several hundred per group. I would be very keen to do a post-mortem of this study to see if we can learn something useful from this failure. I sincerely hope that this drug did not fail because of poor trial design."

"Is this another story of a baby being thrown out with the bathwater?"

"You may be interested to know that Acorda's trials of fampridine, also used an enrichment design. This means that only responders are entered into the double-blind placebo-controlled study. This design may seem counterintuitive, but it is what the Regulators now expect for symptomatic therapies."

CoI: multiple