Objective:

The objective of this study was to investigate long-term cognitive trajectories in patients with LGI1-AE using the Montreal Cognitive Assessment (MoCA) and its subcomponents, with a specific focus on the persistence of executive deficits and the influence of age at onset. 

Background:

Leucine-rich glioma-inactivated 1 (LGI1) antibody-associated autoimmune encephalitis (AE) presents with cognitive and behavioral disturbances. Although most patients improve with immunotherapy, long-term domain-specific cognitive outcomes have not been thoroughly studied. We examined long-term cognitive trajectories in LGI1-AE, with a focus on executive dysfunction.

Design/Methods:

We conducted a retrospective study of 18 LGI1-AE patients followed at a single tertiary center (2015–2025) for a median of 44 months [range 6–82]. Cognitive function was assessed using Montreal Cognitive Assessment (MoCA) and its subscales, including a broad Executive Index Score (EIS), a narrow executive composite, and delayed recall. We studied the course of cognitive function using repeated evaluations on follow up visits. Longitudinal changes were analyzed using the Wilcoxon signed-rank test, and predictors of outcomes were evaluated by regression analysis.

Results:

Global cognition improved significantly from first to last visit (median MoCA 20 to 24, p = 0.001). Executive function and delayed recall also showed gains (p = 0.001 and p = 0.024, respectively). Younger patients (≤65 y) outperformed older patients at presentation and follow-up, despite similar immunotherapy timing (MoCA: 26 vs. 19, p = 0.016 and EIS: 12.5 vs. 9.5, p = 0.009). Improvement magnitude was similar across ages. Regression showed initial MoCA predicted long-term global cognition (p = 0.001), while both initial EIS (p = 0.036) and age at onset (p = 0.007) independently predicted executive outcomes.

Conclusions:

Cognitive outcomes in LGI1 autoimmune encephalitis improve after immunotherapy, yet executive dysfunction frequently persists as a long-term deficit. Older age independently predicts poorer outcomes, suggesting that age-related vulnerability may limit recovery.