Abstracts

Rationale: Next Generation Sequencing (NGS) panel testing only identifies a subset of causal variants, mainly in coding regions and limited intronic regions, in individuals with SCN1A-associated conditions. This leaves a gap where the genetic basis for disease is unknown, potentially depriving patients of the ability to access targeted treatment and use genetic data in family planning and risk assessment. The poison exon 20N region in SCN1A is a promising target to expand diagnoses of SCN1A-related conditions [AJHG 103.6 (2018): 1022-1029]. Our laboratory started reporting variants in the SCN1A poison exon 20N region (defined as NM_001165963.1: c.4002+1958_4002+2650) in 2021, and reviewed internal data to determine the impact of including this new region on NGS panels.

Methods: We conducted a retrospective analysis on 58,198 individuals who underwent diagnostic multigene panel testing from August 8, 2021-May 27, 2025 that included the SCN1A gene. Variants were classified using a semi-quantitative, validated framework based on the ACMG/AMP guidelines. Descriptive statistics were used to assess group characteristics, including ICD-10 codes, demographics, and clinical evidence application.

Results: There were 3504 individuals with at least one reportable variant in SCN1A, totaling 985 unique variants and classified as Variant of Uncertain Significance (VUS), Likely Pathogenic (LP), or Pathogenic (P), with diagnostic results (LP/P variant) in 1521 individuals.

Of the 684 (19.5%) individuals with a reportable variant in poison exon 20N, 11 (1.6%) had a second LP/P variant in SCN1A, 67 (9.7%) had a diagnostic result in a different gene, and 225 (32.9%) had no reported findings outside poison exon 20N. Individuals with reportable variants in the poison exon region were 57% male, had a mean age of 9 years and a median age of 5 years at time of testing. ICD-10 codes were provided in 38% of orders; the most common were F84.0 (Autistic disorder, 20%), R56.9 (Unspecified convulsions, 15%), R62 (Lack of expected normal physiological development in childhood and adults, 13%), and G40.909 (Epilepsy, unspecified, not intractable, without status epilepticus, 13%).

Variants in the poison exon accounted for 183/985 (18.6%) of the reportable variants in SCN1A. Of these, 167 (91.3%) were in autosomal dominant pathogenic range and 43 (23.5%) have clinical case evidence related to developmental and epileptic encephalopathy (DEE). Two variants in the poison exon, c.4002+2361C >G and c.4002+2451G >C reached LP/P based on multiple de novo occurrences and RNA splicing evidence. The remainder were classified as VUS.

Conclusions: Poison exon 20N is an important special target to include in NGS panels for individuals with relevant phenotypes. Additional experimental studies and segregation analysis may complement existing clinical evidence to reclassify the majority of poison exon 20N variants that remain VUS. With more than 1 in 6 reportable SCN1A variants located in this region, it is a target for future diagnoses, crucially in the 33% of individuals with no other findings.

Funding: Authors are employees of Labcorp.