GlaxoSmithKline and XenoPort have unveiled disappointing mid-stage data on gabapentin enacarbil, which has failed to show any improvement over placebo as a potential treatment for migraine.

In the 30-week, Phase IIb study, 526 patients received 1200, 1800, 2400 or 3000 mg/day of the drug, also known as GSK1838262, or placebo, administered twice daily. During the last four weeks of treatment, it did not demonstrate a statistically significant improvement over placebo in the number of migraine-free days, the study’s primary endpoint.

Also, eight patients who received GSK1838262 experienced serious adverse events including two deaths. However, one was due to bronchopneumonia “and was assessed by the investigator as not related to the study drug”, the companies said, while the second death resulted from an accidental overdose involving medications other than the GSK/XenoPort drug.

The companies said the failure of the study may be a consequence of the unexpectedly high placebo response rate observed. However it now seems likely that GSK will not proceed with a Phase III programme for the compound in migraine, and XenoPort’s shares fell some 25%.

Migraine is just one of three indications that gabapentin enacarbil has been studied for. The most advanced is for the treatment of moderate to severe primary restless leg syndrome, and in May, the US Food and Drug Administration requested further preclinical data on the drug in RLS where it is known as Horizant.

The drug, which is a modified version of Pfizer’s epilepsy treatment Neurontin, is also in trials for painful diabetic neuropathy.