Evaluation of composite responder outcomes of pain intensity and physical function in neuropathic pain clinical trials: an ACTTION individual patient data analysis

Integrating information on physical function and pain intensity into a composite measure may provide a useful method for assessing treatment efficacy in clinical trials of chronic pain. Accordingly, we evaluated composite outcomes in trials of duloxetine, gabapentin, and pregabalin. Data on 2287 patients in 9 trials for painful diabetic peripheral neuropathy (DPN) and 1513 patients in 6 trials for postherpetic neuralgia (PHN) were analyzed. All trials assessed pain intensity on a 0 to 10 numeric rating scale and physical function with the 10-item subscale of the Short Form-36, ranging 0 to 100 with higher scores indicating better function. Correlation between change in pain intensity from baseline to posttreatment and change in physical function was small in DPN (ρ = -0.22; P < 0.001) and nonsignificant in PHN (ρ = -0.05; P = 0.08). Assay sensitivities of 10 composite outcomes were examined in a random subsample of patients enrolled in pregabalin trials for DPN and PHN. Of these, a responder outcome of ≥50% improvement in pain intensity, or a ≥20% improvement in pain intensity and ≥30% improvement in physical function was not only significantly associated with pregabalin vs placebo in the development cohorts for both pain conditions but also in the validation cohorts. Furthermore, this composite outcome was cross-validated in trials of gabapentin for PHN and duloxetine for DPN, and had slightly lower number needed to treat than a standard responder outcome of ≥50% reduction in pain intensity. In summary, this study identified a composite outcome of pain intensity and physical function that may improve the assay sensitivity of future neuropathic pain trials.