Review question
Cochrane authors investigated the efficacy and safety of oral and intrauterine progestogens for the precancerous condition of thickening in the lining of the womb (endometrium) called atypical endometrial hyperplasia.
Background
Oral progestins have been used as conservative (non-surgical) treatment in young women with atypical endometrial hyperplasia. However, oral progestins are associated with side effects that may limit their overall efficacy. The levonorgestrel-releasing intrauterine system (LNG-IUS) is thought to reduce the risk of endometrial hyperplasia because it slows thickening in the lining of the womb, and studies have reported that it may be an appropriate conservative treatment for atypical hyperplasia. So it is important to determine the efficacy and safety of the LNG-IUS in treating atypical endometrial hyperplasia.
Study characteristics
This review of the evidence found one randomised controlled trial (RCT) of the LNG-IUS versus oral progestin therapy in women with any type of endometrial hyperplasia. It was conducted in Norway and included 153 women, but only 19 of the women had a confirmed histological diagnosis of atypical endometrial hyperplasia; the other women had other types of endometrial hyperplasia. The evidence is current to July 2018.
Key results
The included RCT compared LNG-IUS versus oral continuous or cyclic medroxyprogesterone (MPA) for treating endometrial hyperplasia. After six months of treatment, there was insufficient evidence to determine whether there was a difference in regression rates between the LNG-IUS group and the MPA group. The rate of regression was 100% in the LNG-IUS group (n = 6/6) and 77% in the MPA group (n = 10/13). Among the total study population (N = 153), over the six months' treatment the main adverse effects were nausea and vaginal bleeding. There was no evidence of a difference between the groups in rates of nausea, but vaginal bleeding was more common in the LNG-IUS group.
Quality of the evidence
The quality of the evidence was low or very low for all outcomes. The included study was at low risk of bias, but the quality of the evidence was very seriously limited by imprecision and indirectness.
We did not find any RCTS of women with atypical endometrial hyperplasia, and our findings derive from a subgroup of 19 women in a larger RCT. All six women who used the LNG-IUS system achieved regression of atypical hyperplasia, but there was insufficient evidence to draw any conclusions regarding the relative efficacy of LNG-IUS versus oral progesterone (MPA) in this group of women. When assessed in a population of women with any type of endometrial hyperplasia, there was no clear evidence of a difference between LNG-IUS and oral progesterone (MPA) in risk of nausea, but vaginal bleeding was more likely to occur in women using the LNG-IUS. Larger studies are necessary to assess the efficacy and safety of oral and intrauterine progestogens in treating atypical endometrial hyperplasia.
Endometrial carcinoma is the most common gynaecologic malignancy in the world and develops through preliminary stages of endometrial hyperplasia. Atypical endometrial hyperplasia suggests a significant pre-malignant state with frank progression to endometrial carcinoma, and tends to occur at a young age. Oral progestins have been used as conservative treatment in young women with atypical endometrial hyperplasia, but they are associated with poor tolerability and side effects that may limit their overall efficacy. So it has become increasingly important and necessary to find a safe and effective fertility-sparing treatment with better tolerability and fewer side effects than the options currently available. The levonorgestrel-releasing intrauterine system (LNG-IUS) has been used to provide endometrial protection in women with breast cancer who are on adjuvant tamoxifen. The antiproliferative function of levonorgestrel is thought to reduce the risk of endometrial hyperplasia.
To determine the efficacy and safety of oral and intrauterine progestogens in treating atypical endometrial hyperplasia.
In July 2018 we searched CENTRAL; MEDLINE; Embase; CINAHL, PsycINFO and the China National Knowledge Infrastructure for relevant trials. Cochrane Gynaecology and Fertility (CGF) Specialised Register and Embase were searched in November 2018. We attempted to identify trials from references in published studies. We also searched for ongoing trials in five major clinical trials registries.
Randomised controlled trials (RCTs) of oral and intrauterine progestogens (LNG-IUS) versus each other or placebo in women with a confirmed histological diagnosis of simple or complex endometrial hyperplasia with atypia.
Two review authors assessed trial eligibility and risk of bias and extracted the data. The primary outcomes of the review were rate of regression and adverse effects. Secondary outcomes included rate of recurrence and proportion of women undergoing hysterectomy. We have used GRADE methodology to judge the quality of the evidence.
We included one RCT (153 women) comparing the LNG-IUS administering 20 micrograms (μu) levonorgestrel per day versus 10 milligrams of continuous or cyclical oral medroxyprogesterone (MPA) for treating any type of endometrial hyperplasia. Only 19 women in this study were histologically confirmed with atypical complex hyperplasia before treatment. The evidence was of low or very low quality. The included study was at low risk of bias, but the quality of the evidence was very seriously limited by imprecision and indirectness. We did not find any RCTS comparing the LNG-IUS or oral progestogens versus placebo in women with atypical endometrial hyperplasia.
Among the 19 women with atypical complex hyperplasia, after six months of treatment there was insufficient evidence to determine whether there was a difference in regression rates between the LNG-IUS group and the progesterone group (odds ratio (OR) 2.76, 95% confidence interval (CI) 0.26 to 29.73; 1 RCT subgroup, 19 women, very low-quality evidence). The rate of regression was 100% in the LNG-IUS group (n = 6/6) and 77% in the progesterone group (n = 10/13).
Among the total study population (N = 153), over the six months' treatment the main adverse effects were nausea and vaginal bleeding. There was no evidence of a difference between the groups in rates of nausea (OR 0.58, 95% CI 0.28 to 1.18; 1 RCT, 153 women, very low-quality evidence). Vaginal bleeding was more common in the LNG-IUS group (OR 2.89, 95% CI 1.11 to 7.52; 1 RCT, 153 women, low-quality evidence). Except for nausea and vaginal bleeding, no other adverse effects were reported.