Background Substantial heterogeneity exists in the cardiorespiratory fitness (CRF) change in response to exercise training, and its long-term prognostic implication is not well understood. We evaluated the association between the short-term supervised training-related changes in CRF and CRF levels 10 years later. Methods and Results STRRIDE (Studies of a Targeted Risk Reduction Intervention Through Defined Exercise) trial participants who were originally randomized to exercise training for 8 months and participated in the 10-year follow-up visit were included. CRF levels were measured at baseline, after training (8 months), and at 10-year follow-up as peak oxygen uptake (vo, mL/kg per min) using the maximal treadmill test. Participants were stratified into low, moderate, and high CRF response groups according to the training regimen-specific tertiles of CRF change. The study included 80 participants (age: 52 years; 35% female). At 10-year follow-up, the high-response CRF group had the least decline in CRF compared with the moderate- and low-response CRF groups (-0.35 versus -2.20 and -4.25 mL/kg per minute, respectively; =0.02). This result was largely related to the differential age-related changes in peak oxygen pulse across the 3 groups (0.58, -0.23, and -0.86 mL/beat, respectively; =0.03) with no difference in the peak heart rate change. In adjusted linear regression analysis, high response was significantly associated with greater CRF at follow-up independent of other baseline characteristics (high versus low [reference] CRF response: standard β=0.25; =0.004). Conclusions Greater CRF improvement in response to short-term training is associated with higher CRF levels 10 years later. Lack of CRF improvements in response to short-term training may identify individuals at risk for exaggerated CRF decline with aging.