To evaluate the efficacy and safety of Intranasal insulin therapy in improving cognitive, functional, and behavioral outcomes among individuals with mild cognitive impairment (MCI) and Alzheimer’s disease (AD).

Newer evidence suggests that disrupted brain insulin signaling contributes to cognitive decline in AD. Intranasal insulin provides a safer means for delivering insulin directly to the CNS, potentially improving cognition without causing systemic metabolic effects. However, finding RCTs remains inconsistent, necessitating quantitative synthesis.

A systematic review and meta-analysis was performed in accordance with PRISMA guidelines. A comprehensive search was conducted across major databases from inception to May 2025. Eligible studies were RCTs comparing intranasal insulin (regular insulin, detemir, glulisine) to placebo in participants with MCI or AD. Primary outcomes included change in ADAS-cog and memory composite scores, while secondary outcomes assessed activities of daily living (ADL-MCI, ADCS-ADL), CDR-SOB, and GDS-15. Pooled mean differences (MDs) were calculated using random-effect models, and heterogeneity was assessed using the I² statistic. Adverse effects were summarized descriptively.

Six RCTs (N=611 participants) administering intranasal regular insulin over 3 weeks to 12 months were included. Pooled analysis showed a significant improvement in ADAS-cog scores at 2 months (MD =–0.45,[ –0.48 to –0.42])AND AT 6 months (MD = –0.38[–0.44 to –0.33] I²=12%). Memory composite scores also improved significantly in short-term follow-up, particularly among APOE4 carriers. In functional outcomes (ADL-MCI, ADCS-ADL, CDR-SOB), no consistent benefits were seen with insulin therapy. Behavioral improvement was modest, with reduced GDS-15 scores at 6 months. Adverse events ranged from mild to moderate, with no significant differences between the two groups.

Intranasal insulin demonstrates a modest but significant short-term benefit in individuals with MCI and early AD, without increasing AEs. While promising, the variability in formulations, sample sizes, and limited long-term data warrants the need for larger RCTs.