Efficacy of biologically-directed daylight therapy on sleep and circadian rhythm in Parkinson's disease: a randomised, double-blind, parallel-group, active-controlled, phase 2 clinical trial.

Background: New non-pharmacological treatments for improving non-motor symptoms in Parkinson's disease (PD) are urgently needed. Previous light therapies for modifying sleep behaviour lacked standardised protocols and were not personalised for an individual patient chronotype. We aimed to assess the efficacy of a biologically-directed light therapy in PD that targets retinal inputs to the circadian system on sleep, as well as other non-motor and motor functions. Methods: In this randomised, double-blind, parallel-group, active-controlled trial at the Queensland University of Technology, Australia, participants with mild to moderate PD were computer randomised (1:1) to receive one of two light therapies that had the same photometric luminance and visual appearance to allow blinding of investigators and participants to the intervention. One of these biologically-directed lights matched natural daylight (Day Mel), which is known to stimulate melanopsin cells. The light therapy of the other treatment arm of the study, specifically supplemented the stimulation of retinal melanopsin cells (Enhanced Mel), targeting deficits to the circadian system. Both lights were administered 30 min per day over 4-weeks and personalised to an individual patient's chronotype, while monitoring environmental light exposure with actigraphy. Co-primary endpoints were a change from baseline in mean sleep macrostructure (polysomnography, PSG) and an endocrine biomarker of circadian phase (dim light melatonin secretion onset, DLMO) at weeks 4 and 6. Participants data were analysed using an intention to treat principle. All endpoints were evaluated by applying a mixed model analysis. The trial is registered with the Australian New Zealand Clinical Trials Registry, ACTRN12621000077864. Findings: Between February 4, 2021 and August 8, 2022, 144 participants with PD were consecutively screened, 60 enrolled and randomly assigned to a light intervention. There was no significant difference in co-primary outcomes between randomised groups overall or at any individual timepoint during follow-up. The mean (95% CI) for PSG, N3% was 24.15 (19.82-28.48) for Day Mel (n = 23) and 19.34 (15.20-23.47) for the Enhanced Mel group (n = 25) in week 4 (p = 0.12); and 21.13 (16.99-25.28) for Day Mel (n = 26) and 18.48 (14.34-22.62) for the Enhanced Mel group (n = 25) in week 6, (p = 0.37). The mean (95% CI) DLMO (decimal time) was 19.82 (19.20-20.44) for Day Mel (n = 22) and 19.44 (18.85-20.04) for the Enhanced Mel group (n = 24) in week 4 (p = 0.38); and 19.90 (19.27-20.53) for Day Mel (n = 23) and 19.04 (18.44-19.64) for the Enhanced Mel group (n = 25) in week 6 (p = 0.05). However, both the controlled daylight (Day Mel) and the enhanced melanopsin (Enhanced Mel) interventions demonstrated significant improvement in primary PSG sleep macrostructure. The restorative deep sleep phase (PSG, N3) significantly improved at week 6 in both groups [model-based mean difference to baseline (95% CI): -3.87 (-6.91 to -0.83), p = 0.04]. There was a phase-advance in DLMO in both groups which did not reach statistical significance between groups at any time-point. There were no safety concerns or severe adverse events related to the intervention. Interpretation: Both the controlled daylight and melanopsin booster light showed efficacy in improving measures of restorative deep sleep in people with mild to moderate PD. That there was no significant difference between the two intervention groups may be due to the early disease stage. The findings suggest that controlled indoor daylight that is personalised to the individuals' chronotype could be effective for improving sleep in early to moderate PD, and further studies evaluating controlled daylight interventions are now required utilising this standardised approach, including in advanced PD. Funding: The Michael J Fox Foundation for Parkinson's Research, Shake IT Up Australia, National Health and Medical Research Council, and Australian Research Council.