Clinical significance of mRNA nonstop decay in rare disease diagnosis and recommendations for its application in variant classification

Background : Nonsense-mediated decay (NMD) is routinely considered during clinical interpretation of predicted loss-of-function (LoF) variants, whereas mRNA nonstop decay (NSD)—a translation-coupled surveillance pathway that degrades transcripts lacking an in-frame termination codon—is rarely considered. Variants that abolish the canonical stop codon or remove the final stop-containing exon(s) can create nonstop transcripts, but current American College of Medical Genetics and Genomics/Association for Molecular Pathology (ACMG/AMP) variant classification guidelines and the Clinical Genome Resource (ClinGen) PVS1 recommendations do not explicitly address NSD, creating a potential mechanism-specific gap in variant classification. Methods : We developed a computational workflow to identify disease-associated genes and transcript regions in which variants are predicted to generate nonstop transcripts in any of the three reading frames. Disease genes were obtained from OMIM and transcript annotations were retrieved from Ensembl. NSD-potential genes were defined as those lacking an in-frame stop codon in the 3′ UTR prior to the annotated polyadenylation site for at least one reading frame. We then screened 2,439 clinical genome sequencing (GS) variant call format (VCF) files for variants overlapping NSD-susceptible regions and performed case-level review. Results : Among 4,954 OMIM disease-associated protein-coding genes, 333 (6.72%) had NSD potential in ≥1 reading frame, yielding 546 genomic NSD-susceptible loci spanning 34,744 bp (~0.0011% of the human genome). Screening of 2,439 GS cases identified 359 candidate NSD variants; six variants in six probands were consistent with the individuals’ phenotypes and the genes’ disease mechanism. In these cases, incorporation of NSD as a transcript-null mechanism would increase the weight of LoF evidence and support reclassification from variant of uncertain significance to likely pathogenic or pathogenic when combined with additional clinical and genetic evidence. Conclusions : NSD-predicted variants are not ultra-rare among disease genes and can be clinically relevant in rare disease diagnosis. We propose practical recommendations for recognizing NSD-susceptible variants and incorporating NSD into gene-/disease-specific LoF frameworks (e.g., PVS1) to improve consistency and diagnostic yield.