Background: Exposure to bright white light (BWL) has been shown to improve outcomes including fatigue, depression, and sleep disturbances among the general older population but the benefit has not been demonstrated as yet in prostate cancer. Older men with prostate cancer on androgen deprivation therapy (ADT) are at risk for obese frailty (components: fatigue, weakness, poor mood, slowness and obesity). In this pilot study, we aimed to determine if BWL, compared to dim white wight (DWL), reduces obese frailty in older prostate cancer patients starting anti-androgens. Methods: Men age ≥ 65 with prostate cancer initiating ADT were randomly assigned to either the BWL (exposure to full spectrum light, 500-1500 lux) or DWL cohort (exposure to low dose white light (<50 lux)) and were blinded to assignment. Men received daily 30-minute morning light session from special glasses (Luminette) for 3 months. Participants were assessed at the beginning of treatment and 3 months later with the Short Physical Performance Battery (SPPB) - including timed 4m walk, tandem balance, and timed chair stands (range: 0 to 12) - energy levels (self-reported 5-point Likert scale), waist circumference (measured via tape (mm)), and muscle strength (handgrip: assessed via dynamometer (kg)). Pre-post outcomes (means differences) were evaluated using paired t-tests with a p-value of 0.05 considered significant. Results: 18 patients (9 per cohort) were recruited. Patients in the BWL arm showed a statistically significant improvement in muscle strength (BWL p=0.012 versus in DWL p=0.22). Compared with DWL, BWL arm showed no decline in energy levels, (BWL p=0.28 vs. DWL p=0.035) nor gain in waist circumference (BWL p=0.51, DWL p=0.046). There were no statistically significant differences in either arm on the SPPB (BWL p=0.44, and in DWL p=0.09). Conclusions: BWL may offset negative effects of ADT for older men with prostate cancer, through improved muscle strength, maintained energy levels, and no waist circumference gain. While a small pilot study, the intervention warrants further research in a larger sample.