Endocrine Abstracts

Background: Three 5α-reductase (5α-R1-3) isoenzymes are described. In humans, 5α-R2 deficiency (5α-R2D) causes pseudohermaphroditism and 5α-R3 neurological defects. The role of 5α-R1 is not clear. Urine steroid profiling (USP) by GC–MS can compare 5α-reduction activity based on excretion of 5α- and 5β-reduced metabolite ratios. Finasteride predominantly inhibits 5α-R2. USP in these patients shows concordance with 5α-R2D. Dutasteride is a dual inhibitor of 5α-R1 and 5α-R2. We compared USP data on patients using finasteride and dutasteride against 5α-R2D to uncover the role of 5α-R1 in generating urinary 5α-reduced metabolites.

Methods: We completed intercomparison of 5α/5β-reduced urinary steroid ratios by USP in genetically confirmed 5α-R2D (n=28), finasteride (n=6) and dutasteride (n=2) treatment with controls (n=36). The ratios studied were: androgens, androsterone (A)/aetiocholanolone and 11β-OH androsterone (11OHA)/11β-OH aetiocholanolone; and corticosteroids, allo-tetrahydrocorticosterone/tetrahydrocorticosterone and allo-tetrahydrocortisol/tetrahydrocortisol.

Results: Study groups showed significant decrease of all ratios relative to controls. Values (ranges) for the four ratios in the sequence listed above were: 5α-R2D, 0.10–10.0; 0.17–20.7; 0.10–1.67; 0.00–0.10, finasteride, 0.07–0.28; 0.20–2.31; 0.07–3.42; 0.00–0.09, dutasteride, 0.02–0.03; 0.07–0.29; 0.55–0.61; 0.00–0.01. Thus there was no difference for corticosteroids between the groups, but the dutasteride patients showed significantly lower androgen metabolite ratios compared to 5α-R2D.

Conclusions: These findings support the expectation that 5α-reduction of corticosteroids is almost exclusively dependent on 5α-R2 activity, this explains why these ratios are the most diagnostic for 5α-R2D. Lower excretion of A and 11OHA in the urine with dutasteride use indicates 5α-R1 has a significant role in androgen catabolism. This may be true for 5α-reduction of testosterone and dihydrotestosterone and explain the lower diagnostic sensitivity of their ratio in serum in 5α-R2D. There is evidence of derangement of androgen metabolism in animal 5α-R1 knock-out studies, thus potential for currently unidentified pathology if 5α-R1 deficiency exists in humans.