ORF15-04 - Hyperandrogenism in the Absence of Metabolic Dysfunction Is Not Associated with MASLD

Background:
Hyperandrogenism has been associated with metabolic dysfunction–associated steatotic liver disease (MASLD), particularly in women with polycystic ovary syndrome (PCOS), obesity, and insulin resistance. Whether androgen excess independently contributes to MASLD in metabolically healthy women remains unclear. The aim of this study was to evaluate the association between hyperandrogenism and MASLD in nonobese, insulin‑sensitive women.

Methods:
MASLD was defined by a controlled attenuation parameter (CAP ≥274 dB/m). Clinically significant and advanced liver fibrosis were defined based on liver stiffness measurement (LSM) by transient elastography (FibroScan®) of ≥8.2 kPa and ≥9.7 kPa, respectively. Hyperandrogenism was defined as total testosterone ≥55 ng/dL, measured by highly accurate Isotope Dilution Liquid Chromatography-Tandem Mass Spectrometry (ID-LC-MS/MS). Metabolic, hepatic, and hormonal profiles were compared between women with and without hyperandrogenism. Data were obtained from NHANES 2021–2023, and sampling weights were applied to maintain the representativeness of the U.S. population.

Results:
A total of 717 women with BMI < 30 kg/m² and HOMA‑IR < 3 were included in the study. The mean age was 46 ± 1 years, and the mean BMI was 24.1 ± 0.2 kg/m²; none had diabetes. The overall prevalence of MASLD was 8.5%, and 3.5% had hyperandrogenism. Women with hyperandrogenism had metabolic profiles similar to controls, with no differences in BMI, HbA1c, blood pressure, triglycerides, or HDL‑C. Despite comparable metabolic risk profiles, MASLD prevalence was lower in women with hyperandrogenism (2.1% vs. 8.7%, p = 0.014). In contrast, a numerically higher prevalence of clinically significant fibrosis was observed (8.4% vs. 2.7%), although this difference was not statistically significant. Women with hyperandrogenism had higher androstenedione levels (152 ± 19 vs. 85 ± 3 ng/dL, p = 0.003), a trend toward higher estradiol, estrone, and estrone sulfate levels, and markedly lower progesterone levels (209 ± 24 vs. 27 ± 9 ng/dL, p < 0.001).

Conclusions:
In non‑obese, insulin‑sensitive women, hyperandrogenism was not associated with an increased prevalence of MASLD and was paradoxically associated with lower hepatic steatosis. These findings suggest that previously reported associations between hyperandrogenism and MASLD are likely mediated by insulin resistance, adipose tissue dysfunction, and obesity rather than by androgen excess alone. Relative estrogen predominance may exert hepatoprotective effects, while the discordance between steatosis and fibrosis raises the possibility of androgen‑mediated, nonsteatotic pathways of liver injury. Future mechanistic studies should focus on hepatic sex‑steroid signaling across hyperandrogenic phenotypes, independent of metabolic disease.