Integrated Clinical and In Silico Analysis of Hypomorphic DCLRE1C Variants: Predicting Malignancy and Autoimmunity Risk

Objective: Hypomorphic variants of DNA cross-link repair 1C (DCLRE1C) cause a spectrum of combined immunodeficiency phenotypes that range from mild autoimmunity to life-threatening malignancy. We aimed to integrate detailed clinical phenotyping with in silico pathogenicity metrics to improve the prediction of adverse outcomes in this rare disorder. Materials and Methods: We compiled a descriptive cohort of 41 patients carrying one of 12 published hypomorphic DCLRE1C variants. Inferential analyses focused on the molecularly homogeneous subgroup of 25 individuals harboring three recurrent missense changes within the metallo-β-lactamase/β-CASP domains—c.194C>T [p.Thr65Ile], c.500C>T [p.Thr167Met] and c.632G>T [p.Gly211Val]. Comprehensive clinical, immunological, and laboratory data were extracted. Pathogenicity was assessed with eight in silico algorithms, ΔΔG stability modelling, and structural mapping. Logistic regression identified variables associated with malignancy and autoimmunity, and composite risk scores were generated. Results: Median age at last follow-up was 9.4 years (range 0.5–27). Recurrent sinopulmonary infection (80%), chronic diarrhea (48%), and candidiasis (44%) were common across variants, while otitis media and growth retardation were confined to c-terminal changes. All three analyzed variants exceeded pathogenicity thresholds (mean Combined Annotation-Dependent Depletion [CADD] 26.7 ± 2.1) and reduced predicted protein stability (mean ΔΔG +1.8 kcal/mol). A revised malignancy risk score (lymphadenopathy, immunoglobulin G [IgG] <400 mg/dL, diarrhea, ΔΔG ≥ 1 kcal/mol, BayesDel >0.5, age > 60 months) > 4 achieved an area under the receiver operating characteristic curve (AUC) of 0.89 (95% confidence interval [CI] 0.57–1.00), with 75% sensitivity and 100% specificity. The analogous autoimmunity score (skin lesions, pulmonary infections, IgG low, ΔΔG ≥ 1, BayesDel > 0.5, CD19 low, age > 60 months) > 8 showed moderate discrimination (AUC 0.75; 95% CI, 0.54–0.97), with 83.3% sensitivity and 73.7% specificity. Conclusion: Integration of domain-restricted in silico metrics with core clinical parameters modestly improves malignancy risk prediction in hypomorphic DCLRE1C deficiency. Prospective, multi-center validation and functional assays are required before clinical implementation. Keywords: Artemis leaky, severe, combined, immunodeficiency combined, immune, deficiency DCLRE1C