OBJECTIVE The fat mass and obesity-associated gene (FTO) variant has shown the strongest association with obesity. Recent studies suggest that dietary intake may modify the genetic effects of FTO. We tested the effect of FTO variant on long-term weight loss and change in body composition in a 2-year randomized intervention trial. RESEARCH DESIGN AND METHODS FTO SNP rs1558902 was genotyped in 742 overweight or obese adults who were randomly assigned to one of four diets differing in the percentages of energy derived from fat, protein and carbohydrate (20, 15, and 65%; 20, 25, and 55%; 40, 15, and 45%; and 40, 25, and 35%) in the Pounds Lost Trial for 2 years. Body composition and fat distribution were measured in 424 and 195 participants by Dual-energy X-ray absorptiometry (DXA) and computed tomography (CT), respectively. RESULTS We found significant modification effects for diet intervention varying in protein (low vs high), but not in fat, on 2-year changes in total fat, fat free mass (FFM), fat mass% (FM%), trunk fat%, total adipose tissue mass (TAT), visceral adipose tissue mass (VAT) and superficial adipose tissue mass (SAT) (P for interactions=0.045, 0.036, 0.033, 0.048, 0.001, 0.008 and 0.002, respectively). Carriers of the risk allele (A allele) had a greater loss of weight and regional fat in response to high protein diet, while an opposite genetic effect was observed on changes in TAT and SAT in response to low protein diet at 2 years. Significant gene by diet interventions (low vs high protein) were also observed at 6 months of intervention, when the maximum of weight-loss was achieved, for changes in FFM, TAT, VAT and SAT (P for interactions=0.007, 0.019, 0.036 and 0.041, respectively). Patterns of change in body composition and fat distribution by diet intervention were similar at 6 month and 2 year. CONCLUSIONS Our data suggest that a high-protein diet may be beneficial for weight loss in individuals with the risk allele of the FTO variant rs1558902.