5α-Reductase-2 Deficiency: Clinical Findings, Endocrine Pitfalls, and Genetic Features in a Large Italian Cohort.

Clinical records (n = 24) with an established diagnosis of 5α-reductase-2 deficiency were reviewed. A previous misdiagnosis was present in about 70% (period from first observation to definitive diagnosis: 9.1 ± 10.8 years), and in 8 children gonadal removal was performed before certain diagnosis. Initial sex assignment was female in 16/24 (67%) and male in 8/24 (33%) cases. After diagnosis, sex re-assignment was performed in 5 babies (4 girls to male sex; 1 boy to female sex). Baseline testosterone/DHT ratio was diagnostic in 6/12 subjects (first months of life n = 4; puberty n = 2), while post-hCG testosterone/DHT ratio was diagnostic in all tested individuals (choosing both the cut-off value 15 or 10). Eighteen different mutations in the steroid-5α-reductase-2 (SRD5A2) gene were identified, 5 of which have never been reported. In conclusion, a time lag exists before the diagnosis of 5α-reductase-2 deficiency is established; sex assignment and gonadal removal may be performed before certain diagnosis. Sex re-assignment is usually female to male, but the contrary may occur. A large variability in clinical phenotypes and genetic mutations was present in this cohort. Accurate endocrine evaluation is recommended in babies possibly affected by 5α-reductase-2 deficiency, since the use of appropriate cut-off values of testosterone/DHT ratio after hCG stimulation may permit to select individuals for SRD5A2 gene analysis. A genotype-phenotype correlation was not found in this study.