Abstract
Newborn screening (NBS) for X-linked adrenoleukodystrophy (ALD) enables early identification of boys at risk for adrenal insufficiency (AI) and cerebral ALD (CALD). However, NBS frequently identifies ABCD1 variants of uncertain significance (VUS), which are associated with only borderline-elevated C26:0-lysophosphatidylcholine (LPC(26:0)) levels. Traditional American College of Medical Genetics and Genomics (ACMG) pathogenicity classification does not account for age-dependent penetrance or the broader phenotypic spectrum, complicating risk assessment and clinical management. Through the Grey Zone Project, we developed a risk-stratification framework using a receiver operating characteristic (ROC)-based approach prioritizing 95% sensitivity. This framework incorporates biochemical and longitudinal clinical data from 1627 control subjects and 196 confirmed ALD patients. Three pediatric risk categories were defined: "no ALD" (<110 nmol/L LPC(26:0)), "lower-risk AI/CALD" (110-177 nmol/L), and "at-risk AI/CALD" (>177 nmol/L). When applied to 108 samples carrying 51 unique ABCD1 VUSs, 26 variants were reclassified as "no ALD," 15 as "lower-risk AI/CALD," and 10 as "at-risk AI/CALD." The framework reclassifies ABCD1 variants based on biochemical risk profiles, reducing false-positive referrals, avoiding unnecessary MRI surveillance, and alleviating parental anxiety by identifying children who are unlikely to develop childhood-onset disease. Integrating biochemical thresholds with genetic and longitudinal clinical data improves the specificity of NBS without compromising its sensitivity. Providing systematic feedback on false-positive cases to screening laboratories will further refine cut-offs. This framework provides a scalable, evidence-based model for interpreting variants and enabling personalized follow-up in ALD and other disorders with a variable age of onset.