OBJECTIVE: This study was designed to determine if serum leptin concentrations (adjusted for fat mass) after weight loss on a low-calorie diet predict subsequent weight gain. RESEARCH METHODS AND PROCEDURES: Body composition and serum leptin concentrations were determined on 14 moderately obese, postmenopausal, nondiabetic women with a familial predisposition to obesity. Assessments were obtained under tightly controlled metabolic ward conditions of macronutrient intake and weight maintenance both before (obese state) and after a mean weight loss of 12.0 kg to normal body weight (postobese state). Four years later, without intervention, body weight and body composition were reassessed. RESULTS: Weight loss resulted in significant decreases in fat mass (29.7 +/- 5.4 vs. 20.3 +/- 4.7; kg), body mass index (27.7 +/ 1.6 vs. 23.0 +/- 1.5; kg/m2), percent body fat (40.7 +/- 4.3 vs. 33.1 +/- 5.0), and serum leptin concentrations (31.8 +/- 16.0 vs. 11.5 +/- 5.4; ng/mL). Serum leptin concentrations were positively correlated (p<0.05) with fat mass in both the obese and postobese states (r=0.67 and r=0.56, respectively). However, residual serum leptin concentrations (adjusted for fat mass) in the obese and postobese states were not related to changes in body weight (p = 0.61 and 0.52), fat mass (p = 0.72 and 0.42), body mass index (p = 0.59 and 0.33), or percent body fat (p=0.84 and 0.46) over the follow-up period. DISCUSSION: These finding do not support the hypothesis that relatively low concentrations of leptin predict weight regain after weight loss. However, because the number of subjects in this study was limited, further studies are warranted.