Background: Continuous administration of progesterone antagonists (PAs) results in endometrial suppression and amenorrhoea in several model systems. We compared the effects of intrauterine release of a highly specific PA, ZK230211, to those of a progestin using the levonorgestrel-releasing intrauterine system (LNG-IUS).
Methods: Forty-two women were randomly fitted with an IUS releasing either ZK230211 at a rate 1, 4 or 8 microg/24 h (ZK-IUS) or LNG (at 20 microg/24 h, LNG-IUS) at 4-8 weeks before hysterectomy. Bleeding patterns, endometrial morphology and content of ZK230211, and various immunohistochemistries (IHCs) were evaluated.
Results: Days of bleeding and spotting were unchanged by the use of ZK-IUSs but were increased by LNG-IUS (P < 0.01). ZK230211 was measurable in all endometrial specimens. Endometrium was partly suppressed in 9-30% of women following the use of ZK-IUSs, and in 67% after LNG-IUS. IHCs for Ki-67 and phosphorylated histone H3 were not suggestive of proliferative activity in any group. Compared to LNG, progesterone receptor (PR) was increased following ZK230211 in surface epithelium (all three doses P < 0.01-P < 0.05) and stroma at 4 microg/24 h (P < 0.05). Although low, androgen receptor staining was higher in endothelial epithelium following LNG than ZK230211 (P < 0.05). Insulin-like growth factor-binding protein-1 (IGFBP-1) was detectable only following LNG (P < 0.0001).
Conclusions: Short-term intrauterine release of ZK230211 did not change bleeding patterns or result in endometrial suppression. Expression of proliferation markers was low following the use of both IUSs. Absence of IGFBP-1 and increase in PR reflect the PA effects of ZK230211.