Randomized sequential trial of valproic acid in amyotrophic lateral sclerosis.

OBJECTIVE: To determine whether valproic acid (VPA), a histone deacetylase inhibitor that showed antioxidative and antiapoptotic properties and reduced glutamate toxicity in preclinical studies, is safe and effective in amyotrophic lateral sclerosis (ALS) using a sequential trial design. METHODS: Between April 2005 and January 2007, 163 ALS patients received VPA 1,500mg or placebo daily. Primary end point was survival. Secondary outcome measure was decline of functional status measured by the revised ALS Functional Rating Scale. Analysis was by intention to treat and according to a sequential trial design. This trial was registered with ClinicalTrials.gov (number NCT00136110). RESULTS: VPA did not affect survival (cumulative survival probability of 0.72 in the VPA group [standard error (SE), 0.06] vs 0.88 in the placebo group [SE, 0.04] at 12 months, and 0.59 in the VPA group [SE, 0.07] vs 0.68 in the placebo group [SE, 0.08] at 16 months) or the rate of decline of functional status. VPA intake did not cause serious adverse reactions. INTERPRETATION: Our finding that VPA, at a dose used in epilepsy, does not show a beneficial effect on survival or disease progression in patients with ALS has implications for future trials with histone deacetylase inhibitors in ALS and other neurodegenerative diseases. The use of a sequential trial design allowed inclusion of only half the number of patients required for a classic trial design and prevented patients from unnecessarily continuing potentially harmful study medication.