Rationale: Obstructive sleep apnea (OSA) is associated with cognitive impairment. The effects of continuous positive airway pressure (CPAP) on neuroimaging biomarkers and cognitive performance among middle-aged patients with OSA and normal cognition remain unclear. Objectives: To investigate the effects of CPAP therapy over 12 months on neuroimaging biomarkers and cognitive performance. Methods: In this multicenter, randomized clinical trial, we randomly assigned 148 participants with normal cognition and an apnea-hypopnea index >15/h into two groups: patients receiving CPAP with best supportive care (BSC); and patients receiving BSC alone. The primary endpoint was Montreal Cognitive Assessment (MoCA) score at 6 months after enrollment. The secondary endpoints were intranetwork functional connectivity (FC) of default mode network (DMN) and cortical thickness assessed by functional and structural magnetic resonance imaging, other neuroimaging biomarkers, and neurobehavioral tests. Measurements and Main Results: Between 2017 and 2021, 148 patients were recruited from five hospitals. Linear mixed models showed that there was no significant difference in MoCA scores at 6 months between the CPAP and BSC groups (difference,-0.04; 95% confidence interval [CI],-0.72 to 0.65; P = 0.91). However, there were significant differences in the FC of DMN (difference,-13.73; 95% CI,-23.40 to-4.06; P= 0.01) and cortical thickness (difference,-0.06 mm; 95% CI,-0.10 to-0.01 mm; P= 0.02) between CPAP and BSC groups at 6 months after treatment. No serious adverse events occurred. Conclusions: CPAP improved cortical thickness and FC of DMN, suggesting that patients with OSA may recover from brain atrophic processes after CPAP treatment. However, no improvement in MoCA was found.