Background: Next-generation sequencing has reshaped pediatric neurogenetics, where neurological and neurodevelopmental disorders exhibit overlapping phenotypes and substantial genetic heterogeneity. Whole-exome sequencing is widely adopted as a first- or second-tier test, though real-world performance may differ from research-based estimates. Rationale and Objectives: Reported diagnostic yields often derive from selected or multicenter cohorts and may not reflect routine tertiary-care practice. This single-center exploratory analysis aimed to estimate real-world exome sequencing (ES) diagnostic yield, compare panel-restricted and agnostic exome-wide interpretation, evaluate the impact of family configuration, and quantify the incremental contribution of systematic reanalysis and read-depth–based copy number variants (CNVs) detection. Materials and Methods: We analyzed 817 pediatric patients who underwent ES at IRCCS Giannina Gaslini Institute (November 2021–October 2025). Testing was performed as singletons (43.9%), trios (51.3%), or duos (4.8%), and interpreted using either virtual phenotype-driven panels (37.6%) or an agnostic exome-wide approach (62.4%). Reanalysis was conducted in selected non-diagnostic cases. Read-depth–based CNV detection was implemented in 2024 in a defined subset with orthogonal validation. Results: Baseline diagnostic yield was 23.5% (192/817), significantly higher with agnostic than panel-based interpretation (29.6% vs 13.4%, p < 0.001). In the overall cohort, trio testing outperformed singletons (29.1% vs 15.0%). Reanalysis (10.0% of cases) achieved a 27.2% incremental yield among initially negative individuals (22/81), corresponding to a +2.7% absolute increase. In the agnostic-only strict-exome subset, incremental yield was 24.3% (+1.8%), mainly driven by periodic bioinformatic reanalysis (55.6%) and updated clinical reassessment (33.3%). CNV calling (n = 215) provided 12 additional diagnoses (5.6%) and secondary findings were identified in 6.4%. Discussion and Conclusions: Despite the inherent limitations of this single-center exploratory analysis, ES demonstrated superior performance with agnostic exome-wide interpretation and trio-based testing in a real-world tertiary pediatric cohort. Integration of CNV analysis and systematic reanalysis further increased diagnostic yield, supporting the role of ES as a dynamic and expandable platform for individualized, clinically integrated genomic diagnostics, particularly in neurological and neurodevelopmental disorders.
Exome Sequencing as a Dynamic and Integrated Diagnostic Platform in Pediatric Neurological and Neurodevelopmental Disorders: A Real-World Single-Center Analysis
AMADORI, ELISABETTA
2026-05-27
Abstract
Background: Next-generation sequencing has reshaped pediatric neurogenetics, where neurological and neurodevelopmental disorders exhibit overlapping phenotypes and substantial genetic heterogeneity. Whole-exome sequencing is widely adopted as a first- or second-tier test, though real-world performance may differ from research-based estimates. Rationale and Objectives: Reported diagnostic yields often derive from selected or multicenter cohorts and may not reflect routine tertiary-care practice. This single-center exploratory analysis aimed to estimate real-world exome sequencing (ES) diagnostic yield, compare panel-restricted and agnostic exome-wide interpretation, evaluate the impact of family configuration, and quantify the incremental contribution of systematic reanalysis and read-depth–based copy number variants (CNVs) detection. Materials and Methods: We analyzed 817 pediatric patients who underwent ES at IRCCS Giannina Gaslini Institute (November 2021–October 2025). Testing was performed as singletons (43.9%), trios (51.3%), or duos (4.8%), and interpreted using either virtual phenotype-driven panels (37.6%) or an agnostic exome-wide approach (62.4%). Reanalysis was conducted in selected non-diagnostic cases. Read-depth–based CNV detection was implemented in 2024 in a defined subset with orthogonal validation. Results: Baseline diagnostic yield was 23.5% (192/817), significantly higher with agnostic than panel-based interpretation (29.6% vs 13.4%, p < 0.001). In the overall cohort, trio testing outperformed singletons (29.1% vs 15.0%). Reanalysis (10.0% of cases) achieved a 27.2% incremental yield among initially negative individuals (22/81), corresponding to a +2.7% absolute increase. In the agnostic-only strict-exome subset, incremental yield was 24.3% (+1.8%), mainly driven by periodic bioinformatic reanalysis (55.6%) and updated clinical reassessment (33.3%). CNV calling (n = 215) provided 12 additional diagnoses (5.6%) and secondary findings were identified in 6.4%. Discussion and Conclusions: Despite the inherent limitations of this single-center exploratory analysis, ES demonstrated superior performance with agnostic exome-wide interpretation and trio-based testing in a real-world tertiary pediatric cohort. Integration of CNV analysis and systematic reanalysis further increased diagnostic yield, supporting the role of ES as a dynamic and expandable platform for individualized, clinically integrated genomic diagnostics, particularly in neurological and neurodevelopmental disorders.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
