Excess upper-body (android) fat is considered an health hazard. Exercise training is known to have the potential to modify body composition and to induce a preferential loss of abdominal fat. We studied and compared the composition of whole body and major body regions using dual-energy X-ray absorptiometry (DEXA) in 21 exercising (3-4 hours of intense physical activity/day) and 21 sedentary eumenorrhoic women of similar ages, body mass index (BMI), waist-to-hip ratio (WHR) and age of menarche. In a small number of women in each group (6 out of 21), the ACTH and cortisol response to CRH test and the 24-h urinary cortisol excretion was evaluated. Exercising women had 10% higher total and leg lean mass (p<0.05), and 38% lower total fat mass (p<0.01) than sedentary women. Furthermore, the proportion of android fat was 22% lower in exercising than sedentary women (p<0.01), while the proportion of lower-body (gynoid fat) was unchanged. BMI and WHR were not different between the two groups, while the android/gynoid fat ratios were 16% lower in exercising than in sedentary women (p<0.01). In the exercising women, ACTH and cortisol plasma levels, as well as the 24-h urinary cortisol excretion, were significantly (p<0.01) higher than in the sedentary women studied. In these subjects, a direct relationship between the peak delta percentage increases of ACTH and cortisol after the CRH test and the proportion of android fat was found (r=0.60, p<0.05 and r=0.69, p<0.02, respectively). These results demonstrate that in women who practise intense exercise there are significant differences in body fat distribution in comparison to sedentary women, with a marked less amount of android fat, and suggest that this difference may be related to a reduced response of the pituitary-adrenal axis to CRH.