The purpose of this study was to determine the association between the level of salivary dehydroepiandrosterone sulfate (DHEA-S) and the magnitude of adaptation to exercise training in insulin sensitivity for aged females. A group of 16 females, aged 80-93 years old, was divided into 2 groups according to their baseline DHEA-S levels: Lower Halves (N = 8) and Upper Halves (N = 8), and participated in a 4-month exercise intervention trial. Insulin response with an oral glucose tolerance test (OGTT), cholesterol, blood pressure (BP), motor performance, and DHEA-S were determined at baseline and 4 months after the training program. Glucose tolerance and body mass index (BMI) remained unchanged with training for both groups. Insulin, fasted cholesterol, diastolic blood pressure, reaction time, and locomotive function were significantly lowered by training only in the Upper Halves group. Changes in the area under curve of insulin (IAUC) were negatively correlated with the baseline DHEA-S level (R= - 0.60, P < 0.05). The current study provides the first evidence that oldest-old subjects with low DHEA-S level appear to be poor responders to exercise-training adaptations.