Background: In a large sample of obese Sardinian children and controls we have confirmed that the rs9939609 variant of the FTO gene is strongly associated with BMI, with a frequency of the A allele of 55% in patients’ and of 43% in controls’ chromosomes. It has been recently suggested that this association may be age-dependent. Objective: Aim of this study is to evaluate whether this association is uniformly present in all ages. Methods: 1004 Sardinian obese children and adolescents were genotyped for rs9939609 by using TaqMan probes (472 boys and 532 girls, age 1-20 y). Association analyses were performed by chi-square test (2x1 table), after evaluation of the sample statistical power. Association analyses were then repeated separately in four age groups: ≤ 5 years, 6-10, 11-15 and ≥ 16. Results: In the whole dataset the A allele was significantly more frequent (54.6%) than the T variant (P = 3.3 x 10-5). Analysis in the different age groups showed that the A allele was associated to obesity only in the age groups 6-10 years (55.4%, P = 8.2 x 10-4) and 11-15 years (56.1%, P = 7.1 x 10-4). In the older subjects the frequency of the two alleles was almost the same (A allele frequency = 49.5%), while in children ≤ 5 years the T allele was even more frequent than the A allele (44.0%), similarly to a group of lean controls. Conclusions: Our preliminary results seem to confirm that the effect of the FTO rs9939609 A variant is age-dependent. It may be speculated that the effects of the FTO gene variants predominate in the age groups which escape control of food intake, either by the parents (≤5 years) or self-control (≥16 years). This is in line with the theory that FTO exerts its effect by an action on feeding behavior.