C Burke, M Foley, P Lenehan, P Kelehan, G Flannelly
National Maternity Hospital, Dublin 2

Abstract
The surgical management of early stage endometrial carcinoma is controversial. The benefits of pelvic lymphadenectomy and administration of radiotherapy in this group have been disputed. We aimed to document the experience of stage 1 endometrial carcinoma at the National Maternity Hospital during the 10 year period 1989-1998 and to evaluate and compare clinical outcomes between retrospectively-assigned low and high-risk tumour groups. Seventy seven women were diagnosed with Stage 1 endometrial carcinoma in this period. Thirty-nine women had low-risk and 38 had high-risk tumours. Women with high-risk tumours were older and had a higher rate of lymph-vascular space invasion by tumour on histological examination. Three women (3.9%) developed disease recurrence and died of their disease; one low-risk and two high-risk tumour patients. Survival without recurrence did not differ between the two risk groups. No consistent pattern existed in surgical staging between the two risk groups. A prospectivelyassigned definition of risk would minimise variations in clinical practice by providing a basis for a more tailored approach to adjuvant treatments.

Introduction
Endometrial carcinoma is generally regarded as the least aggressive of gynaecological malignancies, yet approximately one quarter of women treated for endometrial cancer die within five years¹. Even early stage disease i.e. confined to the uterus has a reported recurrence rate of over 7 percent². Current evidence on the ideal surgical treatment of women with early stage endometrial carcinoma is conflicting. The role of lymphadenectomy is controversial^3,4. The Irish National Cancer Strategy document of 1996 states that “assessment of the extent of disease and confirmation or exclusion of local or distant spread is a most important exercise in the care of patients”. An evidencebased approach to practice is advocated in order “to ensure a consistently high-quality service, based on best practice, throughout the country”⁵.

The aim of this study is to document the experience of the management of stage 1 endometrial carcinoma at the National Maternity Hospital over the ten year period 1989-1998, with particular focus on the extent of surgery performed, the administration of adjuvant radiotherapy, disease recurrence and overall disease-free survival. In addition the study examined whether a risk allocation into low and high risk groups might form a basis for a more tailored approach to further management.

Methods
All women who had surgical treatment for stage 1 endometrial carcinoma at the National Maternity Hospital during the years 1989-1998 inclusive were identified from the oncology database. Additional information was obtained from the hospital annual reports and histopathology records with subsequent analysis of patient records. Data was collected on patient age, weight, presenting symptom and duration, co-existing medical conditions, tumour grade and stage, extent of surgery performed, administration of adjuvant radiotherapy or chemotherapy and follow-up details. A model of two risk categories was established based on disease stage, grade and histological subtype. Low-risk patients were identified as those having stage 1a or 1b, grade 1 or 2 endometrioid endometrial adenocarcinoma. Women with stage 1c tumours, poorly differentiated tumours regardless of degree of myometrial invasion, or a histological subtype other than endometrioid were allocated to the high-risk group. In some cases, disease staging was reported according the pre-1988 FIGO staging system where depth of tumour invasion was measured in thirds of the total myometrial thickness. The staging in these cases was converted to FIGO 1988 staging by classifying all tumours with middle-third myometrial invasion as 1c tumours. Any surgery additional to abdominal hysterectomy and bilateral salpingo-oophorectomy was documented as was the administration of adjuvant radiotherapy or chemotherapy.

The primary endpoint of the study was disease-free survival which was assessed up to the end of 2000. Disease-free status was established by evidence of a recent normal gynaecological examination at a follow-up clinic, or verbal communication with the patients general practitioner confirming that there had been no recent reports of vaginal bleeding.

Results
Seventy-seven women underwent surgical treatment for stage 1 endometrial cancer during the study period. Median age at presentation was 57.5 years (range 33-86 years). Presenting symptoms are shown in Table 1.

Duration between first onset of bleeding and date of admission for definitive surgery in post-menopausal women within the group was 172 days on average (range 1 day – 5 years). Total abdominal hysterectomy and bilateral salpingo-oophorectomy (TAH and BSO) was the sole surgical intervention in 66 women (85.7%). Eleven (14.3%) had additional sampling performed. Lymph node sampling was performed in 5 women (6.5%), four had pelvic lymphadenectomy (5.2%) and one (1.3%) had para-aortic lymph node sampling. Omentectomy was performed in 3 women (3.9%); two of these had mucinous tumours diagnosed by curettage. Two women (3%) had pelvic peritoneal washings taken. Vaginal cuff excision was performed in 1 case (1.3%). Tumour histology is shown in Table 2.

Tumour was well differentiated in 30 women (39.0%), moderately differentiated in 32 (41.6%) and poorly differentiated in 15 (19.5%). Stage of disease was 1a in 9 women (11.7%), 1b in 33 (42.5%) and 1c in 35 women (45.8%). Lymph-vascular space (LVS) involvement was commented upon in 34 cases (44.1%) and was present in 13 of these (38%). Thirty eight women (49.3%) had tumours which we classified as high-risk, with 39 women (50.7%) having low-risk tumours. Table 3 describes high risk tumours in our cohort. Details regarding number, age, and adjuvant treatments according to tumour risk group are seen in Table 4.

Seventeen women (22.1%) had radiotherapy treatment, thirteen (76%) of whom had high-risk tumours. One woman received primary radiotherapy prior to surgery and sixteen women had radiotherapy following surgery. Radiotherapy was withheld in two women because of poor performance status (patients aged 84 and 77 respectively). Three women (3.9%) received chemotherapy as part of their treatment regime: mucinous cystadenocarcinoma was present in two cases and nodular metastasis of a papillary endometrioid adenocarcinoma in one case.

Follow-up data was obtained for 65 patients (84%) Of these, details of at least 3 years follow-up – the time by which most recurrences occur⁶ – was achieved in 55 patients (85%). We were unable to determine follow-up in 6 (16%) of women with high-risk tumours and 5 (13%) low-risk patient; reasons for this included lack of general practitioner details, change of general practitioner and follow-up in a different unit.

Disease recurrence was documented in the three women who died of disease (4.6%). There were two vault recurrences and one bone recurrence. Two of these women had high-risk tumours and one woman had a low-risk tumour. None of these women had been administered adjuvant radiotherapy. Ten women (15.3%)included in the study have died since their surgery.Three women (4.6%) have died of disease and seven (10.8%) of causes unrelated to their cancer.There was no difference in disease-free survival when low risk patients were compared with high risk patients (log rank = 0.1).

Discussion
Two of the principal issues in the treatment of endometrial cancer relate to the effectiveness of surgical staging and adjuvant radiotherapy for high-risk early stage disease. Pelvic and paraaortic lymphadenectomy are regarded as “indispensable” interventions for the treatment of endometrial cancer by certain authors¹. Twenty-two para-aortic dissections were required to potentially benefit 1 patient in another study⁷. The PORTEC study⁴ which analysed stage 1 disease showed that patients receiving neither lymphadenectomy nor radiotherapy had worse actuarial 5- year survival rates compared with those who received both of these treatment modalities. Mariani et al⁸ propose that women with FIGO Grade I/II endometrial carcinoma with a greatest tumour diameter of less than 2cm, myometrial invasion of less than 50% and no intra-operative evidence of macroscopic disease can be treated optimally with hysterectomy alone.

Adjuvant radiotherapy in early stage “high intermediate risk” endometrial adenocarcinoma has been shown to decrease risk of disease recurrence⁹. High-risk stage 1 endometrial cancer patients treated with post-operative radiotherapy have a similar overall five-year survival rate to untreated low-risk patients¹⁰. Results of the ASTEC (A Study in the Treatment of Endometrial Cancer) trial are hoped to clarify the benefits of pelvic lymphadenectomy and adjuvant radiotherapy in high-risk stage 1 endometrial cancer¹¹ Early unpublished results indicate that pelvic lymphadenecomy does not confer a survival benefit and radiotherapy data is still awaited.

Increasing age is a negative prognostic factor in endometrial cancer¹². This finding was confirmed in our study with a median patient of 53 and 64.5 years in the low and high-risk groups respectively. Importantly though, over thirty per cent of women in our study were premenopausal. This finding highlights the need for imaging and endometrial biopsy in women over the age of 40 who present with new-onset menorrhagia or other menstrual irregularities.

The practice of full surgical staging in women with endometrial cancer did not seem to be consistent in our patient cohort. This is most likely because the putative benefits of pelvic lymphadenecomy were not widely accepted during the timeframe of this study.

Lymph-vascular space (LVS) involvement is an independent prognostic factor for recurrence and death from disease in women with endometrial carcinoma^13,14. Vascular space invasion carries a relative risk of death from disease of 1.5¹⁵. In our study cohort, comment upon the presence or absence of LVS invasion was recorded in 44% of women. High-risk tumours had LVS involvement in 52.4% compared with 15.4% of low-risk tumours. We did not include LVS invasion as a criterion for inclusion in our high-risk category. Had we done so, it may have changed our risk group profiles.

Adjuvant radiotherapy was administered to just over one-fifth of women (21%). Although the majority (76%) of women who received adjuvant radiotherapy were in the high-risk group, half of women with high-risk tumours (19/38) did not receive adjuvant radiotherapy. This could be due to the necessary upstaging to 1c of all middle-third uterine for the purposes of this study, well-differentiated tumour histology and poor performance status of some women. There is no obvious reason why six women in our retrospectively-assigned high-risk group did not receive adjuvant radiotherapy.

Three women (4.6%) died of disease, all within three years of initial treatment. Two women were in the high-risk tumour group and one was low-risk. Despite the fact that half of women with retrospectively-assigned high-risk uterine tumours did not receive adjuvant radiotherapy, disease-free survival was no different from women in the low-risk group.

The risk categorisation used for endometrial cancers in our study was applied retrospectively and this is naturally subject to criticism. However based on our findings which showed no clear pattern in terms of additional surgery for women with high-risk tumours, we suggest that units intending to practice surgical staging for endometrial cancer could prospectively define high and low-risk category tumour groups taking into account tumour subtype and differentiation from uterine curettings together with intra-operative visual inspection of depth of myometrial penetration of tumour. This correlates well with histologicallydocumented invasion particularly of well-differentiated tumours¹⁶. Although our study deals with relatively small numbers, is limited to a single unit and is retrospective in nature, it does document practice and analyse survival in women diagnosed with endometrial cancer, the poor cousin of gynaecological malignancies.

1. Volume 25 of the FIGO Annual Report on the Results of Treatment in Gynaecological Cancer. Int J Gynecol Obstet 2003;83:Supp 1
2. Rosenberg P, Blom R, Hogberg T, Simonsen E. Death rate and recurrence pattern among 841 clinical stage 1 endometrial cancer patients with special reference to uterine papillary serous carcinoma. Gynecol Oncol 1993;51:311-5.
3. Yokoyama Y, Maruyama H, Sato S, Saito Y. Indispensibility of pelvic and paraaortic lymphadenectomy in endometrial cancers. Gynecol Oncol 1997;64:411-7.
4. Creutzberg CL, van Putten WL, Joper PC, Lybeert ML et al. Surgery and postoperative radiotherapy verus surgery alone for patients with stage 1 endometrial carcinoma: multicentre randomised trial. PORTEC Study Group. Post Operative Radiation Therapy in Endometrial Carcinoma. Lancet 2000;355:1404-11.
5. Cancer Services in Ireland – a National Strategy. Department of Health and Children publication, November 1996.
6. Aalders J, Abeler V, Koldtad P. Recurrent adenocarcinoma of the endometrium: a clinical and histopathological study of 379 patients. Gynecol Oncol 1984;17:85-103
7. Faught W, Krepart GV, Lotoki R, Heywood M. Should selective paraaortic lymphadenectomy be part of surgical staging for endometrial cancer? Gynecol Oncol 1994;55:51-5.
8. Mariani A, Webb MJ, Keeney GL, Haddock MG et al. Low-risk corpus cancer: is lymphadenectomy or radiotherapy necessary? Am J Obstet Gynecol 2000;182:1506-19.
9. Keys HM, Roberts JA, Brunetto VL, Bell JG et al. A phase III trial of surgery with or without adjunctive external pelvic radiation therapy in intermediate risk endometrial adenocarcinoma: a Gynecologic Oncology Group study. Gynecol Oncol 2004;92:744-51.
10. ukura V, Markulin-Grgic L, Santek F, Valetic J. Postoperative radiotherapy in stage 1 endometrial adenocarcinoma. Eur J Gynecol Oncol 2004;25:343-6.
11. ASTEC: A trial of lymphadenectomy and external beam radiotherapy in endometrial cancer. Sponsored and co-ordinated by the medical research council (MRC). Unpublished.
12. Aziz H, Hussain F, Edelman S, Cirrone J et al. Age and race as prognostic factors in endometrial cancer. Am J Clin Oncol 1996;19:595-600.
13. Aalders J, Abeler V, Kolstad P, Onsrud M. Postoperative external irradiation and prognostic parameters in stage 1 endometrial carcinoma. Obstet Gynecol 1980; 56:419-24.
14. Feltmate CM, Duska LR, Chang Y, McIntyre JF et al. Predictors of recurrence in surgical stage II endometrial adenocarcinoma. Gynecol Oncol 1999;73:407-11.
15. Zaino RJ, Jurman RJ, Diana KL, Morrow CP. Pathologic models to predict outcome for women with endometrial adenocarcinoma: the importance of the distinction between surgical stage and clinical stage– a Gynecologic Oncology Group study. Cancer 1996;15:1115-21.
16. Goff BA, Rice LW. Assessment of depth of myometrial invasion in endometrial adenocarcinoma. Gynecol Oncol 1990;38:46-8.