Genotype–Phenotype Correlation and Mutational Burden in Colombian Patients with Congenital Adrenal Hyperplasia

Abstract Background Congenital adrenal hyperplasia (CAH) due to 21-hydroxylase deficiency (21OHD) is characterized by a broad clinical spectrum, ranging from salt-wasting to nonclassical forms. Genotype–phenotype correlations based on predicted residual enzymatic activity have been widely studied, but data from Latin American populations remain scarce. Additionally, the influence of mutational burden on phenotype prediction has not been fully explored. Objective To evaluate the genotype–phenotype correlation and the impact of mutational burden on predictive accuracy in a Colombian cohort of patients with CAH. Methods We conducted a cross-sectional study of patients with confirmed CAH enrolled in a specialized rare disease program. Genotypic classification was based on predicted residual enzymatic activity (Null, A, B, C), and clinical phenotype was categorized as salt-wasting (SW), simple virilizing (SV), or nonclassical (NC). Genotype–phenotype concordance was defined as exact category agreement. Mutational burden was defined as the total number of pathogenic variants, dichotomized as low (≤2 mutations) or high (>2). Penalized logistic regression (Firth method) was used to evaluate associations between mutational burden, sex, and concordance. Results Among 48 patients with available genetic data, genotype–phenotype concordance was highest in severe genotypes: 100% in Null and 85.7% in Group A. In contrast, concordance declined in Group B (33.3%) and Group C (44.4%). Individuals with high mutational burden had significantly lower odds of concordance (OR = 0.18; 95% CI: 0.03–0.94). No significant interaction between sex and mutational burden was observed. More than one-third of Group C patients exhibited more severe phenotypes than predicted. Conclusions Our findings support established genotype–phenotype correlations in CAH, particularly for severe genotypes. However, increased mutational burden was associated with reduced predictive accuracy, suggesting the need to consider total mutation load in clinical assessment and genetic counseling.