Outcome of fertility-preserving treatment in young women with endometrial carcinomas


Dr K. Niwa, Department of Obstetrics and Gynecology, Gifu University School of Medicine, 40 Tsukasa-machi, Gifu-city 500-8705, Japan.


Objective  To evaluate the outcome of conservative treatment of young women with endometrial cancer.

Design  Observational study.

Setting  Gifu University Hospital, Japan from 1988 to 2002.

Population  Twelve women with endometrial cancer, FIGO IA estimated by MRI under 35 years.

Methods  Patients were treated with medroxyprogestreone acetate (400–600 mg/day) for 6–10 months, with endometrial curettage performed every four weeks.

Main outcome measures  Response to therapy, pregnancies and reoccurrence of disease during follow up over a 30-month period.

Results  All cases had pathological complete remissions within 6–10 months. Seven of 10 wishing to have babies conceived, and five of them were delivered of full-term babies. Eight of nine cases receiving long term follow up (over 30 months) developed recurrent disease, with four opting for hysterectomy. No patient developed distant metastases or had disease-related death.

Conclusion  Conservative therapy is feasible in carefully selected young women with endometrial cancer. Recurrence rates were high during long term observation even after pathological complete remissions. Therefore, close follow up is recommended.


Endometrial carcinoma is the disease of peri- and postmenopausal women. It is relatively rare in women under 40 years accounting for only 2.1–14.4% of all cases. These patients usually suffer from oestrogen or hormone-related disorders such as obesity, infertility, nulliparity, ovarian abnormalities and polycystic ovary syndrome.1,2 Generally, endometrial carcinoma arising in young women tend to be well differentiated and localised.1,2

The standard therapy for this malignancy is a total hysterectomy, bilateral salpingo-oophorectomy, retroperitoneal lymph node sampling and peritoneal fluid cytology. In early stage of the disease surgical cure rates are high, although child-bearing capacity is lost. Successful hormonal therapy (usually progestins) for the early endometrial carcinoma to preserve fertility has been reported.2–9 However, the dosage, duration of treatment and the follow up periods for these cases vary.

In this study, we evaluated the long term outcome of a cohort of women with endometrial carcinoma, who wished to retain their fertility.


From January 1988 to December 2002, 12 patients were recruited to the study at the Department of Obstetrics and Gynecology, Gifu University Hospital. Inclusion criteria were: (1) age younger than 40 years, (2) nulliparous, (3) confirmed endometrial adenocarcinoma with grade 1 differentiation, (4) presence of progesterone receptor, (5) normal serum CA125 level (<35 U/mL), (6) absence of myometrial invasion or extrauterine spread by vaginal ultrasound and magnetic resonance imaging (MRI), and (7) having a strong desire to preserve fertility potential. No patients had oestrogen-secreting ovarian tumours, thyroid or liver disease, or previous hormonal therapy. Using the criteria of Kurman and Norris,10 two gynaecologic cytopathologists (KN and HM) independently reviewed the original histologic slides. Progesterone receptors of the tumour in the pathological specimens were examined immunohistochemically before starting therapy.

Eligible patients were extensively counselled for the possibility of recurrence or progression. After approval from the Gifu University Hospital Ethical Committee and informed patient consent, medroxyprogestreone acetate therapy (400–600 mg/day, po) was commenced with whole wall endometrial curettage was performed once every four weeks. Treatment continued for at least six months, and for at least two months after the disappearance of pathological abnormalities. Response was assessed by the pathological specimens obtained at curettage. If no response was noted, the plan was to offer surgery. After the documentation of pathologically complete remission, the women were closely followed up, with pelvic examination, and serial serum tumour markers (CA125).


Patients' characteristics are summarised in Table 1. The age at diagnosis ranged from 23 to 34 years. Pathological diagnoses were as follows: 10 with endometrial adenocarcinoma and 2 with adenosquamous cell carcinomas, all grade 1 differentation. All progesterone receptors were positive.The duration of treatment was 6–10 months, and all 12 cases showed pathological complete remissions within this time (Table 2).

Table 1.  Patient characteristics and clinical course (N= 12).
CasesAgeHistologic typesaBMIbTreated period (months)PgR
  • a

    ADC = adenocarcinoma; G1 = well differentiated; ASC = adenosquamous carcinoma.

  • b

    BMI (body mass index): body weight (kg)/[body height (cm)]2.

131ADC, G119.56(+)
223ASC, G122.38(+)
331ASC, G121.96(+)
428ADC, G139.86(+)
528ADC, G126.86 × 3(+)
634ADC, G128.86(+)
733ADC, G120.46(+)
833ADC, G135.66(+)
924ADC, G135.010(+)
1030ADC, G117.16(+)
1133ADC, G118.36(+)
1230ADC, G120.76(+)
Table 2.  Summary of patients' outcome (N= 12).
CasesMarital statusPregnancyDeliveryaFollow up (months)Outcomeb
  • a

    LC = living children.

  • b

    C/S = caesarean section; AtH = atypical endometrial hyperplasia; NED = no evidence of disease; CEP = cyclophosphamide + epirubicin + cisplatin; PROM = premature rupture of membrane; IUFD = intrauterine fetal death.

1++2 × LC1381st C/S at 18 months from diagnosis; 2nd C/S at 30 months and relapse (AtH) at 81 months, finally hysterectomy; NED
2++84Miscarriage at 29 months from diagnosis; relapse at 46 months, conservative therapy; NED
3++1 × LC641st C/S 22 months from diagnosis; relapse at 40 months, finally hysterectomy; for 6 months; NED
4+60Relapse at 30 months from diagnosis, conservative therapy for 6 months; NED
5++1 × LC54Relapse at 24 months, conservative therapy with systemic CAP for six-course; 1st delivery at 38 months from diagnosis; NED
6+50Relapse at 41 months, finally hysterectomy; NED
848Relapse at 32 months, finally hysterectomy; NED
9++1 × LC42Relapse at 24 months, conservative therapy for six-course; 1st delivery at 30 months; NED
10++1 × LC30NED
12++241st pregnancy PROM, IUFD (22 weeks) at 12 months from diagnosis; 2nd pregnancy 24 weeks (now) at 24 months from diagnosis

Regarding the patients' marital status at diagnosis, 10 were married and 2 were single. In those desiring a pregnancy, the recommendation was to defer this until a three-month disease- and treatment-free interval was achieved. When ovulation was not observed for three months after treatment, clomiphene citrate was administered. Of 10 married patients, seven conceived, and five of them delivered six full-term babies (one set of male twins). The other two pregnancies resulted in a spontaneous miscarriage at five weeks of gestation and IUED at 22 weeks due to intrauterine bacterial infection causing premature rupture of the membranes. This latter patient is now pregnant.

Eight of nine cases with long term follow up (over 30 months after the start of the treatment) developed recurrent disease, with four opting for surgery. One of these patients had metastatic disease in one ovary (Case 2). Adjuvant chemotherapy in the form of six cycles of cyclophosphamide (350 mg/m2), therarubicin (30 mg/m2) and cisplatin (50 mg/m2) was given.

Of the four patients with recurrence desiring fertility preservation, three had repeated conservative therapy. One patient who had reoccurrence of endometrial carcinoma on three occasions refused a hysterectomy and was treated with six cycles of cyclophosphamide (350 mg/m2), therarubicin (30 mg/m2) and cisplatin (50 mg/m2) with medroxyprogestreone acetate (400 mg/day) and whole wall curettage once every four weeks. She eventually had pathologically complete remission, conceived and delivered of a full-term baby. None of the 12 cases in this study developed distant metastasis or had tumour-related deaths (Figs 1 and 2).

Figure 1.

Changes of the endometrial lesions on the MRI (T2) in Case 4. (A) Before therapy, a thickened endometrial lesion is seen. (B) The thickened lesion disappeared six months after the start of the therapy. Case number is described in Tables 1 and 2.

Figure 2.

Pathological changes of endometrial lesions in Case 1. (A) Well-differentiated adenocarcinoma obtained by endometrial curettage before therapy. (B) Degenerative atypical endometrial lesion obtained by endometrial curettage two months after the start of the therapy. (C) Atrophic endometrium five months after the start of the therapy. (H&E; magnification, ×85). Case number is described in Tables 1 and 2.


Standard therapy of endometrial carcinoma (stage I) is a total abdominal hysterectomy with bilateral salpingo-oophorectomy. In general, endometrial carcinoma with premenopausal women has an excellent prognosis.11 There is a therapeutic dilemma regarding the conservative management in young women who strongly wish to preserve their fertility. However, repeated endometrial curettage and medroxyprogestreone acetate therapy may be considered12 and this regimen has been proven effective for well-differentiated endometrial adenocarcinoma.8,9

Although such conservative treatment is feasible, many problems remain unresolved. The appropriate pretreatment evaluation, duration and dosage of progestin therapy, necessity of hysterectomy after child-bearing and oophorectomy at the time of hysterectomy are some examples. For the pretreatment evaluation, the present authors have used pelvic ultrasound, MRI, whole wall endometrial curettage specimens and serum CA125 levels, in order to eliminate the tumour spread or associated ovarian cancers. This approach does accept that MRI remains outwith the staging criteria in the FIGO staging manual, although is one imaging method with diagnostic potential.13

Regarding the appropriate dose of progestin a GOG study (for advanced or recurrent endometrial carcinomas) showed that low dose oral medroxyprogestreone acetate (200 mg/day) was more effective than the high dose (1000 mg/day).14 It is possible that for conservative therapy as in this study, a dose of 200–400 mg/day may suffice, with the advantage of reducing the risk of vascular events.

In the present study, all 12 tumours had a complete response to therapy, with subsequent term deliveries in five women. However, in those with >30 months follow up, most (eight of nine) developed recurrent disease. Two of four patients opting for a hysterectomy had had full-term babies. In one case, ovarian metastases were present. The risk of adnexal metastases is reported to be approximately 5%.15 When a hysterectomy was considered necessary, the authors advised bilateral oophorectomy at the time of hysterectomy. Ultimately, the decision to undergo oophorectomy is with the patient and may depend on her age, cancer risk and the risk–benefit of oestrogen replacement therapy.

The above conservative therapy is an option to treat young women who desire fertility preservation, although recurrences are common and careful follow up is mandatory.

Accepted 25 May 2004