Predictors of Placebo Response in Pooled Lamotrigine Neuropathic Pain Clinical Trials

OBJECTIVESOne limitation of neuropathic pain clinical trials is the often large and variable extent of response in the placebo group, possibly obscuring true medication effects. We pooled data from 252 individuals in the placebo arms of 3 clinical trials of lamotrigine in patients with neuropathic pain to examine the relationship of baseline patient and study site characteristics with 12-week change in the Pain Intensity Numerical Rating Scale score (ΔPI-NRS). The 574 patients in the pooled lamotrigine treatment arms were used as a replication dataset. MATERIALS AND METHODSWe performed univariable and multivariable regression analysis of predictors of ΔPI-NRS. Clinical factors examined were baseline pain intensity score (mean daily PI-NRS over the week prior to randomization), age, sex, diagnosis, prior and concurrent gabapentin use, prior and concurrent tricyclic antidepressant use, pain duration, variability of daily pain scores during the baseline week, and slope of daily pain scores over the baseline week. Site factors evaluated were study site, US geographic region, recruitment rate, and recruitment period. RESULTSBaseline PI-NRS and site recruitment rate were independent predictors of the 12-week ΔPI-NRS in the last observation carried forward, observed case, and repeated measures analyses. Patients with higher baseline PI-NRS scores had a significantly greater 12-week reduction in pain intensity than patients with lower baseline scores. Patients within sites with a faster recruitment rate also had a significantly greater reduction of pain intensity than those in sites with slower recruitment. DISCUSSIONThese results suggest that both patient and study site characteristics can influence the response in the placebo arms of neuropathic pain studies.