Preprint from University of Exeter and King’s College London:
Two-arm parallel 24-month randomised controlled trial, with Vitamin D supplementation compared with a placebo. This was a remote trial, completed from home involving 620 adults 50 years or older with mild to moderate vitamin D deficiency and early cognitive impairment. The primary outcome was executive function measured through Trail making B and other secondary measures of cognition, function and wellbeing. Vitamin D supplementation conferred no significant benefit to executive function compared to placebo at follow-up on the primary outcome (between-group difference: 5770, 95% CI: -2189 to 13730) or cognition, function, or wellbeing.
This MR study found a causal protective factor of vitamin D for AD: The role of 25-OH vitamin D in Alzheimer’s disease through Mendelian randomization and MRI 2024
But according to this association study, the optimal level might be low: Association of Vitamin D Levels with Risk of Cognitive Impairment and Dementia: A Systematic Review and Meta-Analysis of Prospective Studies 2024: “The approximate 77.5–100 nmol/L 25-hydroxyvitamin D [25(OH)D] was optimal for reducing dementia risk. And the AD risk seemed to be decreased when the 25(OH)D level >40.1 nmol/L.” (check the units, 75 nmol/L = 30 ng/mL)
Based on all the above, I think that, for neuroprotection, vitamin D isn’t a priority, unless severely deficient.