(A-135) Impact of Immune-Modulating Treatment on Progression from Mild Cognitive Impairment to Alzheimer’s Disease: A US Administrative Healthcare Claims Cohort Study

Background: Mild cognitive impairment (MCI) is the prodromal phase of Alzheimer’s Disease (AD) and related dementias (ADRD). Preclinical studies demonstrate that immune-modulating drug use reduces the risk of AD, while clinical results remain inconclusive.

Objectives: To assess the impact of immune-modulating drug exposure on AD progression among older adults with MCI in the US.

Methods: Using Optum’s de-identified Clinformatics® Data Mart (CDM) Database, we identified patients aged ≥ 50 years with incident MCI 2007-2024. For each initiator of immune-modulating drug use (medical or pharmacy claim with WHO ATC L04: Immunosuppressants or L01F: Monoclonal antibodies; 365-day washout), up to 5 patients without treatment were exact matched on calendar date (±30 days) and by age, sex, and time since MCI diagnosis. Patients with ADRD diagnosis or treatment prior to matched index and those without continuous enrollment during the 365-day baseline period prior to the matched index were excluded. Patients were followed from matched index until the outcome event of AD (≥1 medical claim with relevant ICD-9/10 codes), death, disenrollment, a maximum follow-up time of 730 days, or treatment initiation (referent patients only). Confounding for measured baseline covariates was controlled via inverse probability of treatment weighting (IPTW). Hazard ratios (HR) with 95% confidence intervals (CIs) were derived via Cox proportional models using Aetion® Substantiate (2024).

Results: A total of 5,471 exposed and 21,496 referent patients were included. Exposed and referent patients were of similar age [median (IQR) 76 (69, 82) for both groups]. Exposed patients had a higher prevalence of rheumatologic, autoimmune, sleep disorders, and liver disease than referent. However, all covariates were balanced after IPTW (absolute standardized difference ≤0.1). Fewer exposed patients than referent patients developed AD [382 (6.9%) vs 1,740 (8.9%)] over a median (IQR) follow-up of 531 (216, 1,030) and 577 (236, 1,098) days. Crude unadjusted AD hazard was lower among exposed compared to referent patients [HR, 0.92 (95% CI: 0.82-1.03)]. However, the effect was attenuated after IPTW [adjusted HR, 0.99 (95% CI: 0.88-1.12)].

Conclusions: While unadjusted results suggested a reduction of AD among MCI patients with immune-modulating drug use, there was no evidence to support a protective effect after IPTW. Additional causal relationships may be masked due to the broad classification of immune-modulators or unaccounted for dose-dependent effects.