Early pain reduction can predict treatment response: results of integrated efficacy analyses of a once-daily gastroretentive formulation of gabapentin in patients with postherpetic neuralgia.

Jensen, Mark P; Hsu, Ping H; Vanhove, Geertrui F

OBJECTIVE The objectives of this study were to identify and determine the validity of early decision criteria following once-daily gastroretentive gabapentin (G-GR) treatment in patients with postherpetic neuralgia (PHN). DESIGN In two placebo-controlled studies, 279 patients were randomized to 1,800 mg G-GR and 270 to placebo with the evening meal; patients underwent a 2-week dose titration, followed by 8 weeks of stable dosing, and 1 week of dose tapering. Patients. Adults with PHN for ≥6 months and an average baseline Numerical Pain Rating Scale (NPRS) score of ≥4 were included in the study. OUTCOME MEASURES Percent change from baseline to week 10 in NPRS scores and the percentage of responders (defined as ≥30% reduction in NPRS scores from baseline to week 10) were determined. METHODS Patients randomized to G-GR were categorized at each week based on their percent pain reduction up to that week, and for each category, the percentage of week 10 responders was computed. For several early-improvement criteria, the percentage of week 10 responders, odds ratios for achieving week 10 treatment response, sensitivity, and specificity were calculated. RESULTS There was a significant positive association between early pain reduction and week 10 treatment response. Pain reduction of <10% at week 5 of G-GR treatment was the best early predictor of lack of endpoint response, with only 8% of these patients moving on to become week 10 treatment responders. CONCLUSIONS Early response was a reliable predictor of final response. This approach holds promise for aiding clinicians in decision making regarding the need for alternative or supplemental treatment during G-GR therapy for PHN.