Rare Predicted Pathogenic Variants in GWAS-Implicated Genes Reveal a Distinct Genetic Architecture in Familial Multiple Sclerosis
Multiple sclerosis (MS) is a chronic immune-mediated neurodegenerative disease of the central nervous system in which both environmental exposures and genetic predisposition contribute to risk. Large genome-wide association studies (GWAS) have robustly identified hundreds of common MS-associated variants, but these typically confer small effect sizes and collectively account for only a portion of the estimated heritable component. Against this backdrop, Turk and colleagues interrogate a complementary hypothesis: that rare variants—particularly those with higher predicted functional impact—may disproportionately contribute to familial multiple sclerosis (FMS), where clustering within families suggests stronger genetic loading than in sporadic MS (SMS).
Study objective and conceptual rationale
The authors explicitly test whether rare, predicted pathogenic (RPP) variants in genes previously implicated by MS GWAS are enriched in FMS relative to SMS and population controls. This design leverages GWAS not as an endpoint but as a gene-prioritization strategy: if GWAS reliably points to biological pathways relevant to MS, then rare deleterious alleles within the same genes might represent higher-impact contributors in pedigrees. In essence, the study evaluates whether the “common disease–common variant” signals from GWAS can be complemented by a “common disease–rare variant” burden in a clinically and epidemiologically meaningful subgroup—familial cases.
Cohorts, sequencing strategy, and gene panel construction
Whole-exome sequencing (WES) was performed in 87 individuals with FMS and 89 with SMS, and results were compared against 3,866 controls. Sequencing reads were aligned to hg38 and processed following Genome Analysis Toolkit (GATK) best practices, with quality filters including read depth ≥10 and genotype quality ≥20. For the target gene set, the authors derived a 111-gene panel from the International Multiple Sclerosis Genetics Consortium (IMSGC) GWAS loci list, including 99 genes outside the extended major histocompatibility complex (MHC) and 12 MHC genes, focusing on autosomal protein-coding genes linked to GWAS-significant regions.
Defining “rare, predicted pathogenic”: stringent variant filtration
A key methodological feature is the conservative operational definition of RPP variants. Synonymous variants were excluded, and only variants with allele frequency (AF) < 0.01 in gnomAD exomes (gnomADe) were considered rare; variants without a recorded AF were also treated as rare. Pathogenicity prediction combined consequence-based and algorithmic evidence: loss-of-function classes called by VEP (e.g., frameshift, splice donor/acceptor, stop gained/lost, start lost) were included, while missense variants required concordant deleterious predictions across five tools (SIFT, PolyPhen2 HDIV, PolyPhen2 HVAR, LRT, MutationTaster) and a Combined Annotation Dependent Depletion (CADD) score ≥20. This multi-layer filter increases specificity for functionally disruptive alleles, at the cost of potentially excluding true-risk variants with subtler annotations.
Statistical framework: gene-level aggregation and burden modeling
RPP variants passing filters were aggregated by gene and tested via generalized linear modeling (GLM) implemented in R (snpStats and CMGgenomics). The analysis was run in three comparisons: (i) all MS cases vs controls, (ii) SMS vs controls, and (iii) FMS vs controls. Multiple-testing control was applied using false discovery rate (FDR) correction with α=0.05, and candidate variants were manually reviewed in IGV for read-level support. This gene-burden paradigm is well-suited to rare-variant inference because it trades single-variant power (often minimal for rare alleles) for a cumulative signal across multiple rare events within the same functional unit.
Core findings: a striking enrichment in familial MS but not sporadic MS
The principal result is a pronounced overrepresentation of RPP variants in the FMS cohort compared with controls (FDR-adjusted p = 5.27 × 10⁻⁷⁴), while the SMS cohort showed no enrichment (p = 1.00). The combined MS cohort was also significant (p = 0.00337), but the magnitude and direction strongly indicate that the familial subgroup drives this signal. At the gene level, six genes significantly contributed to the FMS burden—ALPK2, ANKRD55, INTS8, IQCB1, JADE2, and MALT1—whereas SMS showed a notable signal primarily for LIMK2. Importantly, many of the enriched variants (10 of 16 across analyses) lacked an allele frequency record in gnomADe, underscoring how a non-trivial fraction of candidate risk alleles may be absent or extremely sparse in reference population datasets.
Biological interpretation, limitations, and why this matters
The discussion positions the implicated genes within immune regulation, neuroinflammatory signaling, and neurodevelopmental or neuronal processes—domains consistent with MS biology. Particularly compelling is MALT1, a key regulator of NF-κB signaling and lymphocyte activation, with experimental evidence (in murine models) that MALT1 protease inhibition can attenuate demyelination and inflammatory infiltration; this aligns mechanistically with the concept that dysregulated immune activation can sustain CNS injury. At the same time, the authors appropriately qualify inference: only one affected individual per family was included (precluding segregation analysis), cohort size may limit detection of smaller rare-variant effects, linkage disequilibrium with nearby common variants could partially account for observed signals, and no functional validation was performed for specific alleles. Even with these constraints, the study makes a clean, clinically relevant point: familial MS may have a different genetic architecture than sporadic MS, in which rare, high-impact variants within GWAS-implicated genes contribute materially to risk. This supports a pragmatic roadmap for future work—larger FMS cohorts with family-based designs, functional assays for top variants, and integrated modeling that jointly considers common polygenic risk and rare-variant burden to sharpen patient stratification and mechanistic discovery.
Disclaimer: This blog post is based on the provided research article and is intended for informational purposes only. It is not intended to provide medical advice. Please consult with a healthcare professional for any health concerns.
References:
Turk, A., Maver, A., Juvan, P., Drulović, J., Mesaroš, Š., Novaković, I., ... & Peterlin, B. (2025). Increased burden of rare variants in GWAS associated genes in familial multiple sclerosis. Scientific reports, 15(1), 21200.
