Smith, Steven R., O'Neil, Patrick M., Astrup, Arne, Finer, Nicholas, Sanchez-Kam, Matilde, Fraher, Kyle, Fain, Randi, Shanahan, William R.
funding text
See the online ICMJE Conflict of Interest Forms for this article. Editorial support was provided by Ericka Peterson, PhD, Tressa Chung, RPh, CMPP, and Stacey Reeber, PhD, of Medicus International New York, with funding from Eisai, Inc.
abstract
ObjectiveTo identify an early treatment milestone that optimizes sensitivity and specificity for predicting 5% weight loss at Week (W) 52 in patients with and without type 2 diabetes on lorcaserin or placebo. MethodsPost hoc area under the curve for receiver operating characteristic analyses of data from three phase 3 trials comparing lifestyle modification+placebo with lifestyle modification+lorcaserin. A total of 6897 patients (18-65 years; BMI, 30-45 or 27-29.9 kg/m(2) with 1 comorbidity) were randomized to placebo or lorcaserin 10 mg bid. Changes (baseline to W52) in cardiometabolic parameters were assessed. ResultsResponse (5% weight loss from baseline) at W12 was a strong predictor of W52 response. Lorcaserin patients with a W12 response achieved mean W52 weight losses of 10.6 kg (without diabetes) and 9.3 kg (with diabetes). Proportions achieving 5% and 10% weight loss at W52 were 85.5% and 49.8% (without diabetes), and 70.5% and 35.9% (with diabetes). Lorcaserin patients who did not achieve a W12 response lost 3.2 kg (without diabetes) and 2.8 kg (with diabetes) at W52. Responders had greater improvements in cardiometabolic risk factors than the modified intent-to-treat (MITT) population, consistent with greater weight loss. Conclusions5% weight loss by W12 predicts robust response to lorcaserin at 1 year.