Title : Seizure trajectory subgroups in Christianson syndrome
Abstract:
Rationale: Christianson syndrome (CS), a rare X-linked neurodevelopmental disorder caused by mutations in NHE6, is frequently associated with epilepsy (>85%). However, longitudinal seizure trajectories and related clinical interventions are not well characterized and may have meaningful implications for patient quality of life and treatment approaches.
Methods: Data were collected as part of the International Christianson Syndrome and NHE6 (SLC9A6) Gene Network Study. Longitudinal seizure-frequency scores from baseline and annual follow-up visits were clustered using an unsupervised approach to identify seizure trajectory groups. Phenotypic data, including age of seizure onset, seizure types, antiseizure medication (ASM) use, status epilepticus at intake, and genetic variant class were analyzed. Group differences were determined using Mann-Whitney U and Fisher exact tests.
Results: Seventy-one male individuals (mean age at baseline = 10.7 ± 8.0 years; Range 1-32 years) were assessed across time points, with a mean of 3.3 follow-up time points per participant. Cluster analyses revealed multiple distinct trajectory groups; here, the least severe (cluster 1: ‘remitting’; n=14) was compared to the most severe (cluster 5: ‘persistent severe’; n=9).
The ‘persistent severe’ trajectory group showed earlier mean seizure onset (25.6 vs 18 months, U= 90.0, p = 0.089;). Generalized motor seizure types (myoclonic, tonic-clonic, or atonic) were more frequent in the ‘persistent severe’ trajectory group (55.6% vs 7.1%; OR= 16.3 (1.4-188), p=0.019); this group also experienced greater treatment burden, including higher lifetime ASM exposure (7.00 ± 5.63 medications compared to 2.43 ± 1.99 ; U= 17.0, p=0.004) and more ASMs at intake (3.78 ± 1.39 medications vs 1.85 ± 0.9; U= 12.0, p=0.002). Status epilepticus at intake was more common in the ‘persistent severe’ trajectory group (88.9% vs 38.5%; OR= 12.8 (1.2-137), p=0.031). Genetically, the ‘remitting’ trajectory group was enriched for predicted loss-of-function variants (nonsense, frameshift, or canonical splice; 85.7% vs 11.1%; OR= 48.0 (3.8-607), p<0.001), whereas the ‘persistent severe’ trajectory group was enriched for missense, in-frame deletion, and noncanonical splice variants (66.7% vs 7.1%; OR= 26.0 (2.1- 319), p=0.005).
Conclusions: In Christianson syndrome, trajectory-defined ‘remitting’ and ‘persistent severe’ seizure phenotypes differed in age at onset, seizure semiology, number of AEDs, SE history, and underlying variant class. The enrichment of missense variants in the severe cluster raises the possibility that mechanisms beyond loss-of-function, perhaps gain-of-function mechanisms, may contribute to more refractory epilepsy in a subset of patients. While more work is needed to fully elucidate the mechanisms underlying genotypes and epilepsy-related phenotypes in CS, these findings support further investigation of seizure-trajectory phenotyping.
