Levonorgestrel‐releasing intrauterine system (Mirena®) as a therapy for endometrial hyperplasia and carcinoma

Adenocarcinoma of the endometrium usually occurs in postmenopausal women. The diagnosis is habitually made during early stages of the disease. Hysterectomy, bilateral salpingo-oophorectomy and staging of the abdomen are the standard therapies. However, these surgical therapies are not feasible in all women because some of them do not present clinical conditions to undergo a surgical procedure. There are several case reports in which women with endometrial carcinoma or hyperplasia have been successfully treated with progestin alone as the primary therapy (1–5). Most cases were treated with medroxyprogesterone acetate (MPA) (1–5), and there were some reports in which a complete clinical remission of the disease was obtained and some women became pregnant after the therapy (1–5). Although MPA was used in exceptional cases, some authors are against this kind of therapy because of the associated risks, such as metastatic disease (6). The report aims to present two women with endometrial carcinoma and one with endometrial hyperplasia treated with the 20-µg/day levonorgestrel-releasing intrauterine system (LNG-IUS, Mirena®, Leiras Oy, Turku, Finland). A 31-year-old woman with morbid obesity [weight 186 kg, height 162 cm, and a 70.9 body mass index (BMI) (kg/m2)] was referred to our service after being submitted for a curettage due to menorrhagia. A pathologic report showed a grade 1 endometrioid carcinoma. During a gynecologic examination, she presented with a hysterometry of 14 cm, and apparently no compromising of the adnexa and parametrium. An ultrasonography showed a uterine volume of 232.4 cm3 and an endometrial thickness of 21 mm with suspicion of an invasion of the myometrium. As she had unfavorable clinical conditions for a hysterectomy due to obesity, hypertension and diabetes, the risk of performing a hysterectomy was considered high and the patient was referred to the Surgery Division to undergo a gastroplasty and had the hysterectomy scheduled after the weight reduction. At the same time a LNG-IUS was inserted uneventfully. Three months after the insertion, an endometrial biopsy was performed with an aspirating cannula and showed an endometrium with metaplasia and chronic endometritis. An ultrasonography revealed a uterine volume of 112.5 cm3 and an endometrial thickness of 12 mm. After 100 days of having the device, it was expelled spontaneously and a new device was inserted. Six months after the initial visit, a hysteroscopy with an endometrial biopsy was performed. The hysterometry was 10 cm and the pathologic report showed the endometrium with a pseudodecidual reaction. One month later, an ultrasonography showed a uterine volume of 94.8 cm3 and an endometrium of 7.0 mm. Seven months after the gastroplasty and the use of the LNG-IUS, the woman's weight was 135 kg with a BMI (kg/m2) of 51.4. She was then submitted for a radical abdominal hysterectomy (Wertheim-Meigs), and the pathologic report showed a G1 well-differentiated endometrioid carcinoma with squamous differentiation restricted to the endometrial mucosa (Stage 1a) and decidualization of the endometrial stroma compatible with the use of exogenous progestin. A 62-year-old woman weighing 130 kg, height 156 cm, and a 53.4 body mass index (BMI) (kg/m2) was referred to our service after undergoing a curettage due to postmenopausal bleeding. The pathologic reports showed a well-differentiated endometrial carcinoma (G1). During an examination, the hysterometry was 9.0 cm and apparently without a compromising of the adnexa and parametrium. An ultrasonography showed a uterine volume of 205 cm3 and an endometrial thickness of 20 mm. Due to morbid obesity, the patient was referred to the Surgery Division to undergo a gastroplasty and a hysterectomy was scheduled after the weight reduction. At the same time a LNG-IUS was inserted uneventfully. One month after the initial visit, an ultrasonography showed a uterine volume of 198 cm3 and an endometrial thickness of 28 mm with a suspicion of myometrial invasion; the device was partially expelled spontaneously and a new one was inserted. Two months after the second insertion of the LNG-IUS, an ultrasonography showed a uterine volume of 188.1 cm3 and an endometrial thickness of 19 mm. Even though the patient had not undergone a gastroplasty or lost weight, she decided to take the risk of a radical abdominal hysterectomy. It was performed 6 months after the first insertion of the LNG-IUS. The pathologic report showed G1 mixed adenocarcinoma with endometrioid and mucinous well-differentiated areas with invasion of half of the myometrium (Stage 1b). A 68-year-old woman weighing 81 kg, height 148 cm, and a 36.97 body mass index (BMI) (kg/m2) consulted due to postmenopausal metrorrhagia. She had been treated unsuccessfully for 6 months with 200 mg of MPA daily elsewhere and the bleeding was not controlled. A curettage was performed and the findings were: a hysterometry of 9.0 cm and a pathologic report showing complex atypical endometrial hyperplasia. An ultrasonography showed a uterine volume of 224.7 cm3 and an endometrial thickness of 23 mm. The patient presented with serious pulmonary and cardiovascular disease that contraindicated surgery. After discussing it with the patient, she agreed upon an LNG-IUS being inserted. Three months after the insertion of the device, an ultrasonography showed a uterine volume of 114.3 cm3 and an endometrium thickness of 4 mm. Six months later the hysterometry was 7.0 cm and an endometrial biopsy reported an endometrial mucosa with pseudodecidualization without hyperplasia. The women are still in control. The three women accepted the insertion of the LNG-IUS after an explanation about the medical conditions, treatment options, risks and benefits, and after signing an informed consent. We choose the LNG-IUS for the medical treatment of these three cases of endometrial cancer and hyperplasia because previous reports showed that progestin could be an alternative to surgery in a well-differentiated endometrial carcinoma and hyperplasia (1–5, 7). The LNG-IUS is able to provoke an endometrial glandular atrophy and a decidualization of the stroma (8) and it has been a successful treatment option for women with menorrhagia (9, 10). We believe that the LNG-IUS could present some advantages over other progestins due to the higher level of levonorgestrel in the uterus and especially in the endometrium than the level observed in the plasma (11–13). Previous reports showed that after the insertion of a LNG-IUS, the concentrations of levonorgestrel in myometria ranged from 1 to 5 ng/g of wet weight of tissue and in endometria were many folds higher than the level observed in women who received levonorgestrel orally for 7 days (11). Moreover, two studies showed that endometrial thickness was reduced in 50% or more cases after 3 months of LNG-IUS use (14, 15). Our results in treating the woman with endometrial hyperplasia was in accordance to a previous report (16) and to the findings of Randal & Kurman (1) who observed regression in 16 out of 17 patients with endometrial hyperplasia when they were treated with progestin. We believe that for women with atypical hyperplasia the LNG-IUS could be an alternative, because it avoids the daily intake of medication, probably increasing the adherence to the treatment, and it could be acceptable for women who do not present clinical conditions for surgery. Regarding the two women with endometrial carcinoma, the results are dubious. One woman used the device for a short time and consequently the result was disappointing in controlling the endometrial carcinoma. Probably, this woman had the LNG-IUS partially expelled during the first month of use and consequently she used the device for a shorter time, thus having the surgery performed sooner than the case no. 1. The specimen showed that the pathology was unaltered. Another explanation about the different performance of the LNG-IUS between both women with endometrial carcinoma was that the histopathology of the cancer was different. It is in agreement to a previous study with 12 women with endometrial carcinoma treated with progestins in which nine of them had the regression of their lesions, but they needed a median length of 9 months of treatment for the regression of the carcinoma (1). In another study, the women needed a median of 29 weeks for the regression of their endometrial lesions (3). For the other woman with morbid obesity and endometrial carcinoma the LNG-IUS was a temporary alternative to treat the disease until she gained clinical condition to undergo the definitive surgical treatment. It is not our intention to recommend the use of the LNG-IUS as a primary therapy for endometrial carcinoma. However, it could be a transitory medical therapy for endometrial hyperplasia and/or carcinoma for women with contraindications for surgery, or very old women for whom intrauterine levonorgestrel could be a better alternative than systemic progestins. The development of an endometrial cancer takes years, in most instances with a continuum of events from hyperplasia to the disease. Even in the cases in which the use of any kind of progestin indicates a regression of the disease, it is important to remember that the cells preserve their capacity to generate invasive cells and consequently, women who will not undergo the surgery must be protected by progestin during a long-time. In conclusion, women with endometrial carcinoma or hyperplasia and with contraindications to have the surgery performed could temporarily benefit from the use of the LNG-IUS. A large number of cases are necessary in order to show the LNG-IUS' real role in the medical treatment of those endometrial pathologies. Address for corresondence: Luis Bahamondes Caixa Postal 6181 13084-971 Campinas, SP, Brazil e-mail: bahamond@caism.unicamp.br