Introduction

Endometrial cancer (EC) remains the leading female genital tract malignancy in industrialized countries [1]. A total of 47,130 estimated new cases and 8,010 deaths from EC are projected to occur in 2012 in the U.S. population [2]. Incidence rates of EC are increasing in many Asia countries including Taiwan [1, 3, 4•], Korea, and Japan [1, 4•, 5]. EC is a disease of postmenopausal women, 4-5 % occurred in women aged 40 years or younger according to population-based studies from the U.S. and western countries [2, 68]. In the recent Cancer Registry Report in Taiwan, 10.3 % of all EC were diagnosed in women younger than 40 years [3]. A trend of increased fractions of cases occurring in young women is observed in Taiwan [3, 4•].

Clinical Features of EC in Young Women

Standard treatment for EC comprises hysterectomy, bilateral salpingo-oophorectomy, and pelvic and para-aortic lymphadenectomy according to risk evaluation and tailored adjuvant therapy [79]. Traditionally, EC of younger women is mostly lower-stage, well-differentiated, endometrioid, and has a better prognosis than older EC patients but some studies have suggested that age was not independently prognostic adjusted for type and grade [7, 1012]. In contrast, Duska et al. found that women younger than age 40 years who were not obese were at higher risk of both advanced disease and high-risk histology [13]. A recent Surveillance, Epidemiology and End Results (SEER)-based analysis indicated that stage IA (by American Joint Committee on Cancer codes) EC patients aged ≤40 years had 5-year survival rates of 98 % with or without ovarian preservation [14]. It is therefore advocated that ovarian preservation at the time of definitive surgery can be performed in selected cases [15]. Fertility-sparing options should be considered for young EC patients with knowledge-based approaches [16•].

Preserve Ovary at Definitive Surgery for Young EC Patients

There are several issues that both physicians and patients have to consider before practice conservations for ovary alone or both uterus and ovary for young EC patients. Incidence rates of synchronous ovarian malignancies in young EC patients (younger than aged 40 or 45 years) are substantial, varied from 9.4-25 % [17, 18•, 19]. Synchronous or metachronous ovarian malignancies also were not infrequently found upon cancer recurrence in the uterus for young women undergoing fertility-sparing treatment (FST) or subsequently after definitive hysterectomy with ovary conserved [2023, 24•].

Criteria of Selecting Candidates for Fertility Preservation

High remission rates with subsequent pregnancies are seen in clinical stage IA without myometrial invasion, grade 1 EC of young women using hormone therapy (HT; usually progestins) as FST [2023, 24•, 25•, 2635, 36•, 37]. Criteria of selecting candidates for FST varied. Most of the series requires (1) a strong desire to reserve fertility potential; (2) age younger than 40 years; (2) nulliparous (or more children required for family); (3) endometrioid carcinoma; (4) grade 1 differentiation (some series allowed grade 2); (5) presence of progesterone receptor (PgR; some do not perform PgR or estrogen receptor [ER] analysis [19, 21, 27]); (6) normal serum levels of CA125 (<35 U/mL); (7) clinical stage IA [38] with absence of myometrial, cervical invasion, or extrauterine spread by some imaging criteria (vaginal ultrasound or computed tomography/magnetic resonance imaging [MRI]); and (8) should be reliable for follow-up.

Sometimes, patients of moderate, differentiated EC were treated with FST, where data were scanty but seemed unwarranted [39, 40]. Rose et al. reported a 39-year-old patient with grade 2 endometrial carcinoma with superficial myometrial invasion, whose lesions had been hysteroscopically resected with margins free and treated conservatively with megestrol acetate (MA). Pelvic recurrence occurred 2 years later despite a document of complete remission at 6 months of FST [39]. Brown et al. reported another 18-year-old grade 2 EC patients treated with levonorgestrel-releasing intrauterine device (IUD) and was without disease at 13 months [40].

Although method of assessment for myometrial invasion varied, the consensus of previous conservation treatments for EC was usually restricted to presumed stage IA disease without myometrial invasion. However, the accuracy of using imaging studies to evaluate the myometrial invasion remains equivocal. A meta-analysis showed that contrast-enhanced MRI was the most accurate modality for the pretreatment assessment of myometrial involvement, but detecting cervical extension was suboptimal [41]. Results of a recent study reported that fused T2-weighted and high-b-value diffusion-weighted images at 3 Tesla provided 90 % accuracy for preoperative evaluation of myometrial involvement and 98 % accuracy for deep myometrial invasion [42]. Some used diagnostic laparoscopy to confirm negative ovarian neoplasms and peritoneal cytology [36•]; however, the benefits of invasive assessment were unproven. A deliberate family history and potential risk of progression should be conducted and genetic testing to rule out Lynch syndrome or Cowden syndrome may be indicated in cases with strong family history [16•].

Fertility-Preserving Treatment and Posttherapy Surveillance

In the literature, HT agents, dose/scheduling, and route of administration were highly variable (Table 1) [22, 25•, 2735, 36•]. Most early series used progestins, either medroxyprogesterone (MPA) or MA [27, 28]. In our hospital, we achieved a first success case using MA 160 mg/day (D) plus tamoxifen 20 mg/D [43]. However, from 1997, the protocol of our hospital was amended that tamoxifen was optional [33] but always required a positive PgR. Studies did not require a positive PgR for enrollment into FST tended to find lower response rates (Table 1).

Table 1 Hormone therapy for fertility-sparing treatment in young women with early-stage endometrial cancer

A progesterone-containing IUD releasing 65 μg of levonorgestrel daily was used to treat inoperable, early EC (age 60 ± 16 years) with 75 % (8/12 patients) complete response [44]. However, the results of FST studies in young women showed that success of two studies was achieved by concurrently using gonadotrophin-releasing hormone analogs (GnRHa) [34] or MPA [35] in addition to IUD releasing 20 μg of levonorgestrel, whereas complete remission using levonorgestrel IUD was only documented in eight patients of another study [36•].

Selective ER modifiers, such as tamoxifen and arzoxifene or GnRHa, thought had definite activity as a single agent, but they were used only in advanced and recurrent EC [4547]. Aromatase inhibitors have shown their potential to treat EC as single agent for advanced or recurrent EC or endometrial stromal sarcoma [7, 8, 48•], but only one report of two cases using anastrazole in combination with MA 160 mg/D has been reported in young women as FST [36•].

Live Birth Rates and Recurrences After FST

According to a recent systemic review of 32 studies (n = 408) of EC women with FST, 76.2 % regressed and the live birth rate was 28 % [24•]. Because these women are frequent sub/infertility patients, assisted reproductive technology (ART) could be introduced for patients with a history of ovulation and fertility problems. Including patients with atypical complex hyperplasia and EC with FST, those who received ART had a 39.4 % live birth rate compared with 14.9 % of those who tried spontaneous conception [24•]. Concerns of ART ovulation induction agents might be associated with increasing the risk of EC recurrence; however, according to a large multicenter study [22], the relapse rate did not worsen from the hyperestrogenic state by ART.

Of the 29 studies, follow-up time was reported to range from 11 to 76.5 months, the pooled relapse rate was 40.6 % (95 % confidence interval (CI) 33.1-49.8 %) and only two cases with documented death [24•]. Many recurrences after FST were of short follow-up after salvage, therefore, mortality rate could be underreported. Such as the case of Kaku’s series [32], the patient who relapsed and underwent hysterectomy and has a second relapse involving left obturator node and vagina was free of disease at the time of publication [32]. This patient again had liver and brain metastases at 46 months, and partial resection of the liver was performed at 48 months; she died of disease at 69 months after the initial MPA therapy (personal communication, Profs T. Kaku and H. Yoshikawa) (Table 1).

Conservative treatment could be offered only to strictly selected patients who desire to preserve fertility and for whom close follow-up is possible during HT and continued surveillance after complete remission. These patients should be followed for as long as possible.

Maintenance and Consolidation Hysterectomy

For women who do not want pregnancy at the time being, a maintenance oral contraceptive agent or depo-Provera (MPA 150 mg intramuscular injection every 12 months) [22, 23, 33, 49] and periodical pelvic ultrasound and/or hysteroscopy [33] should be recommended. Consolidation hysterectomy when completing child-bearing often is advocated by many investigators. Recurrence rates after FST using HT after initial response are 22-67 % [24•]. A case report of a 31-year-old grade 1, clinical stage I EC patient developed occult (asymptomatic) myometrial invasion with normal endometrium has been reported [50]. Therefore, routine pelvic MRI should be considered for follow-up of EC patients who determine to preserve their uterus after fertility-preservation is not a concern. In our hospital, we routinely recommend consolidation hysterectomy but preserving bilateral ovary is allowed if the patients receiving hysterectomy are in a state of sustained remission. However, we had two cases of late recurrences. One patient had left ovarian recurrence at 90 months after consolidation hysterectomy and bilateral salpingectomy with ovarian preservation at 47 months from FST. The other was treated with standard hysterectomy and bilateral salpingo-oophorectomy with lymph node dissection at the first relapse in the uterus (181 months after initial FST). The latter suffered a second relapse in the upper abdomen at 25 months after definitive surgery and a third relapse in the left lung 23 months later.

Investigations on Molecular Profiles Related to Outcomes

It is well known that response to progestin is closely related to progesterone receptor in clinical trials involving advanced and recurrent EC [7, 8]. Yamazawa et al. reported that the probability of complete response to MPA was related to positive expression of PgR in the setting of FST in young EC patients [30]. A multicenter trial investigated the efficacy of letrozole (an aromatase inhibitor) for advanced and recurrent EC, and the response had no correlation with biological markers analyzed [51]. Other in vitro investigations have not found consistent markers for reliable prediction to outcome of various HT for FST purpose [5259].

Conclusions

When fertility preservation is not a concern, standard surgical staging and tailored adjuvant therapy regardless of age should be practiced. Current evidence cannot justify reliable selection criteria for ovarian preservation at the time of definitive surgery. High remission rates with subsequent pregnancies are seen in clinical stage IA without myometrial invasion and grade 1 EC of young women using oral high-dose progestins as FST, yet late recurrences and death should be cautioned. Intrauterine progestins and aromatase inhibitors have limited data at present. Further investigations of molecular profiles for selecting candidates for preserving ovary or FST and whether to do a consolidation hysterectomy are necessary.