“Genotype-first” approaches on a curious case of idiopathic progressive cognitive decline

Abstract

BackgroundIn developing countries, many cases with rare neurological diseases remain undiagnosed due to limited diagnostic experience. We encountered a case in China where two siblings both began to develop idiopathic progressive cognitive decline starting from age six, and were suspected to have an undiagnosed neurological disease.MethodsInitial clinical assessments included review of medical history, comprehensive physical examination, genetic testing for metabolic diseases, blood tests and brain imaging. We performed exome sequencing with Agilent SureSelect exon capture and Illumina HiSeq2000 platform, followed by variant annotation and selection of rare, shared mutations that fit a recessive model of inheritance. To assess functional impacts of candidate variants, we performed extensive biochemical tests in blood and urine, and examined their possible roles by protein structure modeling.ResultsExome sequencing identified NAGLU as the most likely candidate gene with compound heterozygous mutations (chr17:40695717C > T and chr17:40693129A > G in hg19 coordinate), which were documented to be pathogenic. Sanger sequencing confirmed the recessive patterns of inheritance, leading to a genetic diagnosis of Sanfilippo syndrome (mucopolysaccharidosis IIIB). Biochemical tests confirmed the complete loss of activity of alpha-N-acetylglucosaminidase (encoded by NAGLU) in blood, as well as significantly elevated dermatan sulfate and heparan sulfate in urine. Structure modeling revealed the mechanism on how the two variants affect protein structural stability.ConclusionsSuccessful diagnosis of a rare genetic disorder with an atypical phenotypic presentation confirmed that such “genotype-first” approaches can particularly succeed in areas of the world with insufficient medical genetics expertise and with cost-prohibitive in-depth phenotyping.

Clinical, biochemical, and molecular characterization of mucopolysaccharidosis type III in 34 Egyptian patients

“…Although the position of this variant did not induce nonsense‐mediated decay, they found that the mRNA was greatly reduced in those patients, possibly due to increased mRNA instability and degradation (Mangas et al, 2008). The c.1558C > T p.Arg520Trp and c.926A > G p.Tyr309Cys missense variants in NAGLU that were detected in P12 and P18 in our cohort respectively, had previously been identified several times with a severe phenotype (Beesley et al, 2005; Lee‐Chen et al, 2002; Shi et al, 2014; Tessitore et al, 2000; Truxal et al, 2016). Patient 16, who had the c.678 + 1G > A variant in our cohort, had undetectable enzyme activity.…”

Deep Phenotyping on Electronic Health Records Facilitates Genetic Diagnosis by Clinical Exomes

“…We previously identified compound-heterozygous mutations in N-acetylalpha-glucosaminidase (NAGLU [MIM: 609701]), leading to a genetic diagnosis of Sanfilippo syndrome (mucopolysaccharidosis IIIB). 41 Biochemical tests confirmed the complete loss of activity of alpha-N-acetylglucosaminidase (encoded by NAGLU) in both individuals. In the current study, we did not filter for shared variants between the siblings and instead analyzed each individual's exome separately.…”

“…Our analysis above focused on phenotype-driven prioritization of genes and demonstrated that genes with causal variants can be ranked much higher than other genes with the use of phenotype information extracted from EHRs. To further demonstrate the applicability of this method in real-world settings to facilitate the identification of disease-causing variants, we analyzed several previously published cases, 40,41 for which we performed a combined analysis of genotype data (VCF files) and clinical descriptions from the methods sections of the published manuscripts. We observed that the clinical descriptions in scientific manuscripts were professionally edited and could be of higher quality than typical EHR narratives, but extracting HPO terms from the public case reports poses challenges similar to those faced in EHR settings.…”

Fine-tuning Large Language Models for Rare Disease Concept Normalization

“…To illustrate how the proposed approach can be used in an end-to-end setting to facilitate genetic diagnosis of rare diseases, we illustrate an example of using a public, de-identified clinical note on a patient with idiopathic progressive cognitive decline and other phenotypic features, previously reported [29]( Figure 3 ). In a two-step approach (shown as in Figure 3A ), various concept extraction tools, including traditional named entity recognition tools, ChatGPT, or more recent GPT-based phenotype extraction tools [23] can extract specific mentions of phenotypes from clinical notes first.…”