ESPE Abstracts

Background: HSD17B3 deficiency is diagnosed when a testosterone/androstenedione (T/A-dione) ratio after hCG stimulation is below 0.8, and this cut-off value is primarily based on hormonal data measured by conventional immunoassay (IA) in patients with feminized or ambiguous genitalia suggesting severely compromised T production. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) has been regarded as the gold standard for steroid measurement, and it is necessary to examine whether the cut-off value can be utilized for hormonal diagnosis with LC-MS/MS in a wide range of patients including phenotypic male patients.

Case presentation: We examined two 46,XY Japanese patients with undermasculinized genitalia including hypospadias (patient 1 and patient 2). Endocrine studies by IA demonstrated a well increased serum T value following hCG stimulation (2.91 ng/mL) and a high T/A-dione ratio (4.04) in patient 1 at 2 weeks of age and sufficiently elevated basal serum T value (2.60 ng/mL) in patient 2 at 1.5 months of age. Despite partial androgen insensitivity syndrome (PAIS)-like findings, whole exome sequencing identified biallelic pathogenic or likely pathogenic variants in HSD17B (c.188C>T:p.(Ala63Val) and c.194C>T:p.(Ser65Leu) in patient 1, and c.139A>G:p.(Met47Val) and c.672+1g>a in patient 2) (NM_000197.2). Functional analysis revealed that the missense variants resulted in reduced HSD17B3 activities (~ 43% for p.Met47Val, ~ 14% for p.Ala63Val, and ~ 0% for p.Ser65Leu). Thus, we investigated hCG-stimulated serum steroid metabolite profiles by LC-MS/MS in patient 1 at 7 months of age and in patient 2 at 11 months of age as well as those of five control males with idiopathic micropenis aged 1 – 8 years, and revealed (1) markedly high T/A-dione ratios (12.3 in patient 1 and 5.4 in patient 2), which were, however, obviously lower than those in the control boys (25.3 – 56.1) and sufficiently increased T values comparable to those of the control males, and (2) markedly low androstenediol/dehydroepiandrosterone (A-diol/DHEA) ratios and severely decreased A-diol values as compared to those of the control males.

Conclusion: The elevated T/A-dione ratios are considered to be the result of residual HSD17B3 function and the measurement by LC-MS/MS. It is therefore recommended that a cut-off value for the T/A-dione ratio be established according to the phenotypic sex, reflecting the residual function and the measurement method, and to examine the possible value of A-diol/DHEA ratio in the hormonal diagnosis of HSD17B3 deficiency in a wide range of patients including those with PAIS-like phenotype.