Skip to main content

Conference Program

Subpage Hero

Loading

Clinical Exome Sequencing Towards Molecular Diagnosis: Insights and Outcomes from 822 Pediatric Cases

Laboratory Genetics and Genomics
  • Primary Categories:
    • Laboratory Genetics
  • Secondary Categories:
    • Laboratory Genetics
Introduction:
In pediatric populations, exome sequencing (ES) increases diagnostic yield, improves treatment options, reduces the time of diagnostic odyssey, and provides risk information for families.

 

Methods:
Results from 822 pediatric patients (0-18 years old) who underwent ES between 2021 to July 2024 at our clinical laboratory were reviewed. This retrospective study examined the clinical and genetic features of these patients and evaluated the clinical utility of the ES as well as the performance of artificial intelligence (AI) integration into the analysis.

 

Results:
Overall, 22% (179/822) of index cases received a definitive molecular diagnosis associated with their primary indication for testing. An additional 40% cases (326/822) had uncertain results with variant of unknown significance (VUS) findings associated with the clinical presentation, and 2% cases (16/822) had a secondary finding. Among patients with a definitive molecular diagnosis, 93% (166/179) had a single finding and 7% (13/179) had multiple potentially relevant genetic findings. The diagnostic yield was influenced by age, with the highest yield (28%) being in the age group 0-2 years. Within this cohort, 60% (496/822) were trio cases. There is a greater diagnostic certainty for the trio group when compared to the proband only group; the trios had a significantly higher negative result percentage (44% VS. 22%) and a lower uncertain result percentage (33% VS. 55%) than the proband only group.

The proband’s phenotype, typically described using Human Phenotype Ontology (HPO) terms, dictates the interpretation of ES genetic data and impacts the diagnostic yield of ES. Autism, intellectual/developmental delay, multiple congenital anomalies, connective tissue abnormalities, and seizures are the most frequent phenotype categories reported in this cohort. Patients with HPO terms ‘Patent ductus arteriosus’, ‘Microcephaly’, ‘Intellectual disability’, and ‘Absent speech’ had high diagnostic yields at 46% (22/48), 35% (28/81), 35% (26/75), and 33% (23/70) respectively.

AI-based tools within our analytical platform Emedgene, were used to jointly assess the genetic information and the proband’s HPO terms for variant prioritization during the initial case review. Among the 844 variants reported in this cohort, 47% (395/844) variants were initially picked by AI and tagged as ‘most_likely’ and 43% (366/844) variants were tagged as 'candidate’, and 9% (77/844) variants were picked manually after review.

 

Conclusion:
The overall diagnostic yield of 22% and the identification of multiple relevant genetic findings in 7% of the pediatric patients in this study underscores the clinical utility of exome sequencing. Relying solely on a more limited targeted sequencing panel may prevent complete molecular diagnosis in complex patients. We demonstrate that parental samples in trio cases provide the segregation evidence and are extremely helpful to interpret the VUSs, and the diagnostic yield is significantly associated with a patient's phenotype and their age. Among all the reportable variants, 90% picked for prioritization by AI tool suggests the use of AI in clinical laboratories can enhances exome sequencing analysis by reducing analysis time and increasing efficiency.  

Agenda

Sponsors