Inhibin-A and inhibin-B in cyclic and pregnant mares, and mares with granulosa-theca cell tumors: Physiological and diagnostic implications

Abstract

Studies in mares have examined serum inhibin concentrations using immuno-assays unable to distinguish dimeric inhibin-A from inhibin-B isoforms. Inhibin-A and inhibin-B immuno-assays were used to investigate concentrations in cyclic mares, young and old (6 vs 19 years old, respectively) mares following hemi-ovariectomy, mares during pregnancy and in mares with confirmed granulosa cell tumors (GCTs). Mares with inter-ovulatory intervals of 26 days had ovulatory peaks of inhibin-A averaging 80 pg/mL with a mid-cycle nadir of 5 pg/mL. Inhibin-A and inhibin-B concentrations were highly correlated (r = + 0.79, P < 0.01) though peak and nadir concentrations of inhibin-B were not significantly different. However, the ratio of inhibin-A to inhibin-B (A/B) changed significantly through the cycle, highest at ovulation and <1 (more inhibin-B than –A) at mid-cycle. Two mares with grossly extended inter-ovulatory intervals demonstrated mid-cycle inhibin-A (and inhibin-B) excursions suggestive of follicular waves. Follicle-stimulating hormone was negatively correlated with inhibin-A and –B concentrations in all 6 mares. Hemi-ovariectomy in young mares resulted in a significant decrease in inhibin-A and inhibin-B concentrations one day later (P < 0.05) but older mares did not, suggesting a possible extra-ovarian source(s) of these hormones. Both inhibin isoforms dropped to very low levels during pregnancy (P < 0.0001), inhibin-A (P < 0.0001) more rapidly than –B (P < 0.05), so that inhibin-B became the predominant measured form throughout most of gestation (P < 0.05). Mares with confirmed GCTs had elevated inhibin-B concentrations more reliably than inhibin-A but neither inhibin-A or -B was correlated with anti-Müllerian hormone concentrations. Collectively, concentrations of inhibin-A and –B were aligned with physiological events in healthy mares, though more pronounced cyclic changes were seen with inhibin-A. Inhibin-B concentrations were significantly associated with GCTs (P < 0.01), inhibin-A concentrations were not. While both inhibin-A and –B concentrations track physiological events such as cyclic follicular activity, only inhibin-B concentrations effectively signal ovarian neoplasia in mares.