Redefining the Diagnostic Intermediate in Trio Genome Sequencing for Neurodevelopmental Disorders

Purpose. To develop a proband-level five-tier outcome classification for trio genome sequencing (GS) in neurodevelopmental disorders (NDD), and, in a small subgroup, to explore what is associated with a full versus partial phenotype match to a candidate gene’s disease spectrum. Methods. A rule-based five-tier framework integrating ACMG/AMP classification, ClinGen gene-disease validity, phenotype concordance, and orthogonal validation was applied to 40 probands with unresolved NDD (36 after exclusions); two reviewers assigned tiers by consensus. Phenotype matching used clinician-documented Human Phenotype Ontology terms, characterised against documentation-derived metrics. Results. Of 36 probands, 7 (19.4%; 95% CI 8.2–36.0%) received a confirmed diagnosis (Tier 1) and 5 (13.9%) a high-priority, unvalidated candidate (Tier 3); none were Tier 2 (Provisional). The remainder had a low-priority candidate (22, 61.1%) or no candidate (2, 5.6%). Among 12 probands with strong disease-association evidence, a full phenotype match was associated with richer documentation than a partial match, not a higher similarity score (p = 0.005); this preliminary finding warrants replication. All confirmed diagnoses involved variant types detectable by exome sequencing (ES), not genome-specific findings like deep intronic variants. Conclusion. The five-tier framework resolves the diagnostic intermediate collapsed by binary solved/unsolved reporting. Comprehensive, disease-specific phenotyping is the practical lever for framework performance.