Abstract
Abstract
Introduction
The purpose of this review is to evaluate the risk factors for endometrial cancer in young women, as well as the fertility preservation options. A literature search using the key words fertility preservation, endometrial cancer, uterine cancer, and risk factors was conducted in Medline, EMBASE, and the Cochrane Database of Systematic Reviews.
Etiologic Factors
Based on etiologic factors; endometrial cancer can be divided into type I and type II. 8 Type I is estrogen related, associated with endometrial hyperplasia, and is mostly low-grade. Type II is not related to estrogen stimulation and usually high grade. 8 Patients with type I endometrial cancers tend to have risk factors including obesity, nulliparity, excess endogenous or exogenous estrogen, diabetes mellitus (DM), or hypertension (HTN). In this review, only type I endometrial cancer is discussed.
Hyperestrogenism
Early menarche and late menopause are considered risk factors for the development of endometrial cancer, suggesting a relationship with longer lifetime exposure to estrogens. 9 Hyperestrogenism may cause endometrial cancer by abnormal, unopposed endogenous mitogenic estrogenic stimulation of the endometrium. Cyclical progesterone is important because of its antiestrogenic activity on the endometrium. Accordingly, regularly menstruating women are protected during their reproductive years by the periodic shedding of endometrial tissue following withdrawal of ovarian estradiol and progesterone eliminating cells with neoplastic potential. 10
Studies have shown the association between exogenous estrogen and increased risk of endometrial carcinoma.11,12 A state of high estrogen is seen in obesity, chronic anovulation, estrogen-secreting tumors, and treatment with exogenous estrogen and selective estrogen modulators such as tamoxifen. 13 Women with >5 years of exposure to tamoxifen had a 4.06-fold greater odds of developing endometrial cancer than did nonusers (95% confidence interval [CI]: 1.74–9.47). 9
Obesity
Obese women tend to have high levels of endogenous estrogen because of the conversion of androgen to estrogen in peripheral adipose tissue. Overweight or obesity is an important risk factor present in almost 50% of women with endometrial cancer.14,15 It is estimated that body mass index (BMI)>25 kg/m2 doubles a woman's risk of endometrial cancer, and a BMI>30 kg/m2 triples the risk. 16 Overweight women usually have insulin resistance, ovarian androgen excess, anovulation, and chronic progesterone deficiency. 17
Obesity is considered a risk factor for endometrial cancer even when the circulating concentrations of estrogen are normal. 4 In a case–control study of 283 women with endometrial cancer and 566 age-matched controls, endometrial cancer risk was found to be proportional to the BMI. Obesity was related strongly to the risk of endometrial cancer both in young premenopausal and in older women, with risk estimates of 20.3 and 7.7, respectively. 18
Nulliparity
Nulliparity or low parity increases the risk of developing endometrial cancer, suggesting the role of long periods of estrogen stimulation. 19 In multiparous women, protection against endometrial cancer is caused by placental production of progesterone.19–22 In a case–control study of 405 endometrial cancer cases and 297 population controls, a major risk factor was the absence of a prior pregnancy (relative risk 2.8, 95% CI 1.7–4.6). 21
Chronic anovulation
In chronic anovulation, estrogen levels are higher than normal and the estrogen is unopposed by progesterone. In a cohort of 1270 patients, the researchers found that the relative risk of developing carcinoma of the endometrium after the diagnosis of chronic anovulation syndrome was 3.1 (95% CI 1.1–7.3). 23 Continuous estrogenic stimulation to the endometrium leads to endometrial hyperplasia. If untreated, it may transform into endometrial cancer. PCOS-related anovulation is a risk factor for endometrial cancer.24,25
Estrogen-secreting tumors
Some ovarian sex-cord stromal tumors produce estrogen and may present with signs of estrogen excess in the reproductive age, mainly abnormal uterine bleeding. The association between granulosa-cell tumors and both endometrial hyperplasia and adenocarcinoma is well documented.26,27 Endometrial biopsy will detect endometrial hyperplasia in 25%–50% of women with granulosa cell tumors, and carcinoma in 5%–10% of cases. 28 Thecoma and pure Sertoli cell tumors can also be associated with endometrial cancer.
Diabetes and hypertension
The association of DM and HTN and the development of endometrial cancer has been reported.29–31 In a case–control study including 709 women with histopathologically proven endometrial cancer and 3368 controls, the authors reported that overweight, obesity, and DM are associated with increased endometrial cancer. HTN increases the risk among obese women. 32
Overweight and obesity are commonly seen in both DM and HTN. The association between DM and the development of endometrial cancer could be explained by overweight-related increases in endogenous oestrogen levels. 33 The relative risk (RR) of developing endometrial cancer in women with DM is 3.8. 18 In diabetic women, hyperinsulinemia is also considered causative, leading to higher levels of insulin-like growth factor 1 (IGF-1), which acts as a direct stimulus for estrogen synthesis and production from the ovaries. 9
Familial and genetic syndrome
Endometrial cancer in a first-degree female relative increases the risk of endometrial cancer by nearly 3-fold, and almost 5% of endometrial cancer among women between the ages of 20 and 54 may be attributable to a family history of endometrial cancer. 34 Women with hereditary non-polyposis colorectal cancer (HNPCC) also known as Lynch syndrome, are at a 40%–60 % risk of developing endometrial cancer, and endometrial cancer may be the first manifestation of malignancy.35,36 The American Cancer Society recommends that annual screening by endometrial biopsy be initiated by age 35. 37
Ovulation induction
Ovulation induction leads to a hyperestrogenic state, raising concerns about its association with endometrial cancer. However, the literature on this subject is conflicting. In an Australian cohort consisting of 29,700 women, 20,656 women were exposed to fertility drugs and 9044 were not. There were 143 breast cancers, 13 ovarian cancers, and 12 cancers of the uterus among these women. Analysis of cancer incidence within 12 months of exposure to fertility drugs with in vitro fertilization (IVF) showed that incidence was significantly higher than expected for breast and uterine cancer (1.96 [1.22–3.15 CI] and 4.96 [1.24–19.8 CI]). Women who have been exposed to fertility drugs with IVF seem to have a slightly transient increase in the risk of having breast or uterine cancer diagnosed in the first year of treatment. 38
In a case–control study, the authors evaluated 128 women, ages 35–64, with a histologically confirmed diagnosis of endometrial carcinoma and a control sample of 255 women. Seven women with endometrial carcinoma (5.5%) and 10 healthy controls (3.9%) reported that they had used any fertility drug (crude odds ratio [OR] 1.4; 95% CI: 0.47–4.2). They found no evidence that the use of ovulation induction agents, including clomiphene citrate, was associated with a higher risk of endometrial carcinoma. 2 In a multicenter U.S. cohort study including 8431 women evaluated for infertility, 39 cases of endometrial cancer were detected among cohort members. The authors reported that clomiphene may increase uterine cancer risk (RR=1.79, 95% CI: 0.9, 3.4) and present evidence of a dose response (p=0.07) and latency effect (p=0.04). The risk substantially elevated among women who were both obese and nulligravid (RR=12.52, 95% CI: 1.5, 108.0). 39
Presentation and Diagnosis
Bleeding is the most common symptom of endometrial cancer. Accordingly, endometrial sampling should be performed in women with risk factors for endometrial cancer who experience abnormal uterine bleeding. Those patients usually have a history of oligomenorrhea and chronic anovulation.40–42 In general, endometrial cancers in these women are well differentiated with minimal or no myometrial invasion.
Office endometrial biopsy versus dilation and curettage (D & C)
Endometrial biopsy should be the initial diagnostic test to rule out endometrial cancer. However, it might miss a localized lesion. 43 Larson et al. evaluated 183 women with endometrial cancer who underwent an endometrial biopsy or a D & C before hysterectomy. The tumor grade at diagnosis matched the tumor grade determined after hysterectomy in 58% of the patients diagnosed with endometrial biopsy, compared with 77% of the patients diagnosed with a D & C. They recommended a diagnostic D & C. 44
A better technique is hysteroscopy-directed biopsy. It allows direct visualization of endometrial cavity and directed biopsy from a suspicious area. 45 The concern that hysteroscopy may lead to malignant cell dissemination into the peritoneal cavity appears to be unfounded. In a meta-analysis of five studies including 756 patients, the authors reported that hysteroscopy examination did not increase the risk of tumor cell dissemination into the abdominal cavity. 46 Hysteroscopy allows sampling of tissue that could be missed by endometrial biopsy or D & C.47–50
Pretreatment Evaluation
The treatment of endometrial cancer depends upon the patient's desire to conceive, and more importantly, on the type and extent of the endometrial cancer. Conservative treatment should be considered only in those with disease confined to the endometrium with a well-differentiated tumor; that is, stage 1 grade 1. In addition, as demonstrated by magnetic resonance imaging (MRI), the tumor should not invade the myometrium, and no evidence of suspicious pelvic or para-aortic nodes and synchronous ovarian tumor. Counseling is essential, as medical treatments are not definitive, and require compliance for follow-up. Proper evaluations of the disease include hematologic and radiologic studies to ascertain the diagnosis and exclude invasion, metastasis, and the coexistence of another primary tumor.
Histopathology
Women with poorly differentiated carcinoma, even if confined to the endometrium, are not candidates for fertility preservation, because the incidence of lymph node metastasis and recurrent is high.51,52 The presence of progesterone receptors, which is vital in the response to treatment, should be evaluated. The clinical response rate to progestin therapy was 72% for progesterone-positive tumors and 12% for progesterone-negative tumors. 53 Absence of progesterone receptors may lead to treatment failure.54,55
Coexistent Ovarian Cancer
Endometrial cancer is an estrogen-dependent tumor. Removal of the ovaries is typically recommended with hysterectomy to exclude occult ovarian metastases and to decrease estrogen production. However, preservation of the ovaries in young women with early endometrial cancer has been performed. Wright et al. studied a total of 3269 women, including 402 (12%) perimenopausal patients with early well-differentiated endometrial cancer who had ovarian preservation, and found that ovarian preservation had no effect on either cancer-specific hazard ratio (HR) (HR: 0.58; 95% CI: 0.14–2.44) or overall survival (HR: 0.68; 95% CI: 0.34–1.35) survival. 56
Ovarian and uterine cancers coexist in 5%–29% of premenopausal women.1,57 Risk factors for concomitant cancers include young age, obesity, premenopausal status, and nulliparity. 58
Cancer Antigen (CA-125)
Several studies found that serum CA-125 level of > 35 U/mL strongly predicted extrauterine disease.59–61 However, it is less useful in perimenopausal women. In addition, many benign conditions may elevate the level.
Radiologic Staging
Endometrial cancer is a surgically staged disease. Imaging studies evaluate coexisting ovarian carcinoma, myometrial invasion, and metastasis to lymph nodes.58,62 While ultrasound (US) or computed tomography (CT) scan is good at detecting adnexal pathology, magnetic resonance imaging (MRI) is more sensitive for detecting myometrial invasion and lymph nodes metastasis.63,64
A meta-analysis demonstrated that in the preoperative assessment of myometrial invasion, contrast-enhanced MRI of the pelvis performed significantly better than US, CT, and noncontrast enhanced MRI. 65
Treatment Options
The standard treatment of an early stage endometrial cancer is staging in the form of TH and BSO, and peritoneal washings, with or without lymph node dissection. This conventional treatment leads to an abrupt loss of fertility as well as surgical menopause. Primary radiation therapy has also been advocated as an option for nonsurgical candidates, but not an option for young women. 66 Recently, several case reports and case series showed a reasonable treatment response to hormonal therapy, giving a hope to young patients with endometrial cancer.
Progestational agents have been used in recurrent and advanced endometrial cancer since 1961. 67 Objective remissions may be expected in ∼ 35% of patients and subjective remission may be expected in ∼ 70%.68–70 Progestin treatment induces secretory differentiation, inhibits estrogen receptor function and endometrial cell mitosis, and promotes apoptosis. In addition, progestin has an antiangiogenic effect. 53 The effects of progestin on endometrial cancer cells become apparent 10–12 weeks after initiation of treatment.53,70
Intrauterine Progesterone
Intrauterine application of progestins in the form of a norgestrel-containing intrauterine device (norgestrel-IUD) has a similar effect on the endometrium as a systemic treatment. It is associated with minimal systemic side effects and a high dose of progesterone effect at the site of the cancer. Montz et al. studied 14 patients with International Federation of Gynecology and Obstetrics (FIGO) stage I, well-differentiated endometrial cancer treated with norgestrel IUD. Twelve patients were followed up to 36 months. Biopsies were negative in 7 of 11 patients at 6 months and in 6 of 8 patients at 12 months. There were no IUD-related complications, except for IUD expulsion in 3 women. 71
Dhar et al. reported 4 women with FIGO stage 1, well-differentiated endometrial cancer with positive progesterone receptors treated with norgestrel-IUD. One woman had complete histological regression of disease within 6 months. One of 3 women who did not respond to treatment subsequently had a vaginal hysterectomy, which showed endometrial cancer with superficial myometrial invasion. 72 In another case report, the authors reported a woman with stage 1 well-differentiated endometrial cancer who was treated with combined norgestrel-releasing intrauterine system and oral medroxyprogesterone acetate (MPA) (200 mg, twice a day). Endometrial biopsy at 5, 9, and 13 months revealed no hyperplasia or malignancy. 73
Repeated D & C
Treatment by repeated curettage of localized stage I well-differentiated endometrial cancer has recently been advocated. Kung et al. reported a 22-year-old nullipara with polycystic ovaries who was treated with repeated D & C. The patient underwent endometrial curettage at the beginning of treatment and at 3- and 6-month intervals. In addition, she was treated with megestrol acetate and tamoxifen for 6 months. One year after the last curettage, there was no evidence of disease. 74 It is unclear whether the resolution of the cancer resulted from repeated D & C or from hormonal treatment alone.
Shibahara et al. reported a successful pregnancy in a woman with early stage endometrial adenocarcinoma treated with repeated curettage only 3 weeks apart. A transvaginal US examination 2 months after delivery showed a smooth, linear endometrium, and subsequent endometrial biopsy revealed no malignancies. 75
Hysteroscopic Resection
Endometrial cancer occasionally is found on endometrial polyps, especially in postmenopausal women and in those treated with tamoxifen. Most of these polyps are high-grade carcinoma.76,77
Several authors have advocated hysteroscopic resection for endometrial cancer. Mazzon et al. reported a young woman with stage I endometrial adenocarcinoma successfully treated with hysteroscopic resection followed by hormonal treatment (megestrol acetate, 160 mg) daily. Repeated biopsy at 10 months of treatment was negative. 78 Sparac et al. reported successful hysteroscopic resection of well-differentiated stage I endometrial cancer in a Lynch syndrome patient followed by hormonal treatment for 3 months consisting of medroxyprogesterone (400 mg/day). 79 Zanetta et al. similarly reported successful treatment in a young patient with repeated hysteroscopic resection who conceived twice after conservative treatment. 80
Repeated D & C and hysteroscopy without progestin treatment are not the best treatments for women with endometrial cancer.
Systemic Progesterone
Although many different protocols are available for medical treatment of endometrial cancer, there are no clear guidelines on progestational agents and dosage, the duration of treatment, and long-term follow-up. The largest numbers of data are from women treated with MPA (Table 1).81–89 Of a total of 97 women treated with MPA, regression was seen in 60 women (60.6%) and relapse in 34 others (35.1%). Of 46 women who attempted to conceive, 36 women (78.3%) achieved pregnancy and 29 (63%) had a live birth.
Table 2 shows that regression occurred in 25 of 30 women (83.3%) treated with megestrol acetate.90–92 In this group, 10 pregnancies were found and resulted in 7 live births. Although the number of patients in this group was small, the results are promising. Treatment with a variety of progestins is demonstrated in Table 3.93–97 Similar to treatment with MPA or megestrol acetate, the results are encouraging. Side effects of progestins include phlebitis (5%–10% with high dose), 98 loss of libido, weight gain, and mood changes.
MPA, medroxyprogesterone acetate; IUD, intrauterine device; GnRH, gonadotropin-releasing hormone.
Conclusions: The Future and Safety of Hormonal Treatment
In women with early stage of well-differentiated endometrial cancer, conservative treatment with progestins should be considered. However, pretreatment evaluation should be performed thoroughly. Coexisting ovarian cancer, which is high among this group of young women, should be ruled out. It is important to note that this is a temporary treatment that has to be followed by a definitive treatment after completion of the patient's family.
Because of the paucity of information, there is no standard recommendation for conservative treatment of endometrial cancer. However, the endometrium should be reevaluated 12 weeks after hormonal treatment. The patient is encouraged to conceive if the endometrium has reversed to normal. Indeed, pregnancy is considered a part of the treatment protocol. In the presence of persistent hyperplasia or endometrial cancer, the treatment is continued with 3–4 months periodic evaluation of the endometrium. Following a pregnancy, endometrial sampling should be performed at 6 weeks postpartum. Once childbearing is completed, the standard treatment should be performed.
Proper counseling and close follow-up are mandatory. Because of the rarity of the disease in young women, a multicenter trial or a registry is needed. Another issue is to identify which group of patients will respond to progestin therapy, perhaps by molecular and gene analysis.
Footnotes
Disclosure Statement
Dr. Tulandi is an advisor for Genzyme in adhesion-related research and one of the clinical investigators for HALT Medical.
