Authors :
Presenting Author: Ian McSalley, MS – Childrens Hospital of Philadelphia
Anna Prentice, MS, LCGC – CHOP
Jan Magielski, BA – CHOP
Shiva Ganesan, MS – Childrens Hospital of Philadelphia
Jillian McKee, MD, PhD – Children's Hospital of Philadelphia & University of Pennsylvania
Aakash Rathod, MS – Ambit Inc
Chen Chen, MA – Ambit Inc
Dan Kim, PhD – Ambit Inc
Charlotte Atlee, MPH – Ambit Inc
Rob Sederman, MBA – Ambit Inc
Ingo Helbig, MD – Children's Hospital of Philadelphia
Rationale:
Within the epilepsies, channelopathies represent a set of distinct monogenic disorders with similar mechanisms, but unique clinical presentations both across and within specific etiologies. Traditional investigations have relied on prospective clinical cohort recruitment; an approach for which data collection is constrained by time and offers limited scalability. With ever-growing conformity and scope in the curation of administrative healthcare claims data, cohort delineation and recapitulation of clinical histories can be performed comprehensively to emulate previous findings and can readily increase cohort size to an unprecedented scale.
Methods:
Through retrospective analysis of administrative healthcare claims data, we identify cohorts of patients with genetic variants in SCN2A (n=1760) and SCN8A (n=1191) classified as pathogenic (P), likely pathogenic (LP), or variants of uncertain significance (VUS)(gnomAD allele frequency < 3.75x10-6). With associated ICD10 codes, we performed independence testing of the frequency of phenotypic features among patients with specific recurrent variants against their respective cohorts. Additionally, applying functional variant characterizations of ‘gain-of-function’ (GOF) or ‘loss-of-function’ (LOF), we assess the fidelity of age of seizure onset as a clinical delineator of variant consequence.
Results:
Among recurrent variants in SCN2A, p.R853Q (n=10) shows a marked increase in epileptic spasm (p< 0.0001) and developmental regression (p=0.0003), p.R1319Q (n=17) is associated with increased occurrence of focal-onset seizures (p=0.01) with decreased neurodevelopmental abnormality (p=0.05), and p.V261M (n=7) demonstrates a higher frequency of neonatal seizures (p=0.0004). In SCN8A, p.N1877S (n=37) is associated with a higher incidence of status epilepticus (p=0.0002) and seizure (p< 0.0001), p.R1872W/Q/L (n=14) shows an increased occurrence of generalized-onset motor seizures (p=0.005) and developmental regression (p=0.004), and p.R850Q (n=6) had higher frequency of motor seizures (p=0.04) and neurodevelopmental delay (p=0.03). In assessing the viability of age of seizure onset as a clinical predictor of functional consequence, among functionally classified (i.e. GOF/LOF) variants, we report minimum misclassification rates (8%) at six months (LOF >