Natural History and Prognostic Factors in Pediatric Alexander Disease: A Cohort Study

This study aims to clarify the natural course of Alexander disease and explore genotype‒phenotype correlations and prognostic factors.

This single-center, bidirectional cohort study included patients genetically confirmed with Alexander disease, aged between 0 and 18 years. Survival curve analysis was conducted to evaluate the acquisition and loss of motor/cognitive skills to clarify the natural course of the disease. Statistical methods such as survival curve analysis and Cox regression analysis were used to analyze genotype‒phenotype correlations and prognostic factors.

A total of 81 patients were included. A total of 27 types of gene variants were found among all the children, with 40.7% (11/27) in the 1A domain. At the last follow-up, 12.3% (10/81) of the patients had died. Survival curve analysis for the ability of "walking without support," "sitting down without support," "holding head upright," "understanding and following simple instructions," and "saying simple words" were lost at an average age of 15.2 ± 1.2 years, 17.3 ± 1.4 years, 17.2 ± 1.3 years, 18.8 ± 14.6 years, and 18.2 ± 1.3 years, respectively. Prognostic factor analysis via Cox single-factor and multifactor regression analysis found that patients with variants of R239 had a greater incidence of poor outcome than other variants did (hazard ratio: 2.597 [95% confidence interval: 1.052, 6.409], P = 0.038).

The overall prognosis of Alexander disease is poor, with an average age at death of 18.6 ± 1.4 years and a high incidence of epilepsy (81.5%). Variants of R239 are potential prognostic factors for poor outcome.