The aging of humans is associated with body composition and function deterioration creating a burden on an individual level, but also on a societal one, resulting in an economic burden that is socially unsustainable. This study aimed to evaluate changes in body composition after a 6-month MCT (multicomponent training) and a 4-month detraining period, and to examine the possible influence of energy and macronutrient intake in these changes in frail and pre-frail older adults. A total of 43 participants from the training group (TRAIN) and 28 controls (CON) completed the study protocol. Body weight, body mass index (BMI), waist and hip circumferences, fat mass, fat free mass and fat mass percentage were recorded, with a bio-electrical impedance analyzer, at baseline, after 6 months and four months after finishing the MCT. A food frequency questionnaire was used to estimate energy intake. Mixed effect models did not show differences between groups. CON showed increases in hip circumference and waist (3.20 ± 1.41 and 3.06 ± 1.66 cm, respectively) during the first 6 months. TRAIN showed decreases in BMI (−0.29 ± 0.14), fat mass (−0.86 ± 0.38 kg), body fat percentage (−0.98 ± 0.36%) and increases in waist circumference (3.20 ± 1.41). After detraining, TRAIN group showed increases in fat mas (1.07 ± 0.30 kg), body fat percentage (1.43 ± 0.31%) and waist (3.92 ± 1.38 cm), and decreases in fat free mass (−0.90 ± 0.30 kg). CON group only showed an increase in body fat (1.32 ± 0.47%). Energy intake was negatively associated with hip circumference in the first six months and fat mass during detraining in CON. Energy intake showed positive associations with fat mass in TRAIN during detraining. Only carbohydrates were negatively related to detraining changes in fat free mass and BMI in CON. In conclusion, the MCT reduces adiposity of frail and pre-frail older people, leading to a maintenance of fat free mass. In addition, these interventions should not be stopped in this population in order to improve health sustainability.