Cancer of the uterine corpus is the most common gynecologic malignancy in industrialized nations with an incidence of 43,470 and an estimated death rate of 7,950 annually in the United States.1,2 A dualistic model for carcinogenesis in endometrial cancer has been accepted worldwide.3–5 The majority of endometrial carcinomas are classified as type I carcinomas and are related to unopposed estrogenic stimulation resulting from obesity or exogenous hormone use. Type I endometrioid endometrial carcinomas occur in women at a median of 60 years of age, originate from hyperplastic endometrium, and generally have a good prognosis. In contrast, type II carcinomas include serous and clear cell histology, are unrelated to estrogenic stimulation, and occur in relatively older women. The majority of type II endometrial carcinomas arise in a background of atrophic endometrium and generally have a poor prognosis.6,7
Distinct carcinogenic pathways underlie the observed clinical differences between type I and II endometrial cancer. Type I carcinomas are characterized by diploid tumors, expression of estrogen and progesterone receptors, PTEN alterations, microsatellite instability, and mutations of KRas and CTNNB1. Type II carcinomas on the contrary, are often aneuploid, and show overexpression of p53 and Her2/neu.5–9
Approximately 20% of cases do not fit within the dualistic model described and include patients with endometrioid endometrial carcinoma associated with recurrence and poor clinical outcome.10,11 It has been suggested that a third endometrial cancer entity exists: endometrioid carcinomas arising in atrophic endometrium. This purported “type III endometrial cancer” is hypothesized to be associated with clinical outcomes intermediate between type I and type II lesions.12 Furthermore, carcinogenesis in this group may occur through distinct mechanisms.
The aim of the current study was to estimate the clinical relevance of this hypothesized third type of endometrial carcinoma. Therefore, we compared clinical and pathologic features in patients with hyperplastic and atrophic background endometrium in a large series of patients with grade 1 endometrioid endometrial carcinoma. Second, we analyzed the prognostic effect of the background endometrium.
MATERIALS AND METHODS
The Dutch nationwide network and registry of histo-and cytopathology (Pathologisch Anatomisch Lan-delijk Geautomatiseerd Archief) was used to search for all patients diagnosed and surgically treated with at least hysterectomy and bilateral oophorectomy at the Radboud University Nijmegen, Medical Centre and the Canisius-Wilhelmina Hospital, Nijmegen, The Netherlands, for primary grade 1 endometrioid endometrial carcinoma from January 1999 through December 2009. Subsequently, all patients diagnosed with primary grade 1 endometrioid endometrial carcinoma and surgically treated with at least hysterectomy and bilateral oophorectomy in the Mayo Clinic, Rochester, Minnesota, from January 2002 through December 2008 were included in the study. A total of 572 patients with grade 1 endometrioid endometrial carcinoma were identified. This number included 143 patients from Nijmegen and 429 patients from the Mayo Clinic. Clinical data were abstracted from patient records. Age, menopausal state, body mass index (BMI, calculated as weight (kg)/[height (m)]2), parity, personal medical history, treatment, stage of disease, date of recurrence of disease, date of death, and the cause of death were registered. Stage of disease was based on the 2009 International Federation of Gynecology and Obstetrics (FIGO) staging system.13 Patients with a personal history of any malignancy or any synchronous primary malignancy were excluded from further analyses. Follow-up data were extracted from the medical charts. In case of incomplete follow-up data, the patient or the physician was contacted. The median time of follow-up was 50 months (range 0–128 months).
The slides of the primary carcinoma and the surrounding background endometrium of all patients were retrieved from the pathology archives and reviewed. Review was done systematically including the following items: the histologic type, tumor grade, depth of myometrial invasion, the presence of lymphovascular space invasion, and the nature of the background endometrium.14 Review was performed in every hospital separately by an independent experienced pathologist (J.B., S.B., D.V.), who was unaware of the results of the original pathology reports or the clinical outcome of the patients. There was no systematic review between the centers. In case of doubt about the diagnosis or in case of discrepancy with the original pathology report, a second review was performed by another experienced pathologist from the concerning hospital and consensus about the diagnosis was made. Background endometrium was grouped in eight categories: simple hyperplasia, simple atypical hyperplasia, complex hyperplasia, complex atypical hyperplasia, disordered proliferative endometrium, atrophic endometrium, normal proliferative endometrium, and determination of the background endometrium not possible.
Hyperplasia was defined as a proliferation of glands with an increase in gland:stroma ratio of 3:1 and a variety of abnormal architectural patterns.15 Cytologic atypia was defined as enlarged, rounded, polymorphic nuclei with loss of polarity, prominent nucleoli, chromatin clumping, and an increased nuclear to cytoplasmic ratio.6 Hyperplasia was categorized according to the World Health Organization classification system for hyperplasia, which is based on the study of Kurman and colleagues.16 The endometrium was considered disordered proliferative when some of the present glands in postmenopausal women showed proliferative activity and the gland-to-stroma ratio was slightly increased but did not meet the hyperplasia criterion of 3:1.6,15
Atrophic endometrium was defined as shallow endometrium with a thin basalis and with a few tubular glands lined by inactive epithelium.17 In case of focal hyperplasia or focal disordered proliferative endometrium, the background endometrium was diagnosed as atrophic when more than 50% of the background endometrium was atrophic. In these cases the subclassification “mixed atrophy” was used, whereas cases without any proliferative glands were classified as “pure atrophy.”
In premenopausal women, proliferative endometrium was defined as widely spread, sometimes tortuous, tubular glands that showed mitotic activity and abundant stroma.17 In some cases the tumor covered the entire endometrial cavity, so the nature of the background endometrium could not be determined.
The median number of slides available per patient in the cases with hyperplasia was four (range 1–23). For the patients with atrophic endometrium, the median number of available slides was four as well (range 2–10).
Comparison was made between the group with atrophic and the group with hyperplastic background endometrium. The atrophy group consisted of patients with mixed and pure atrophy. The hyperplasia group consisted of patients with simple hyperplasia, simple hyperplasia with atypia, complex hyperplasia, complex hyperplasia with atypia, and disordered proliferative endometrium. When normal proliferative endometrium was found in premenopausal patients, or when no background endometrium could be found in the endometrial cavity, patients were excluded from analyses. Subanalyses were performed comparing pure atrophic background endometrium with mixed atrophic background endometrium. Differences in clinical and pathologic parameters between the group of patients with atrophic and hyperplastic endometrium were tested for statistical significance using the Pearson's χ2 test or the Fisher's exact test and the Mann–Whitney test. The P values presented are two-sided, and associations were considered significant if the P value was <.05. Survival techniques were used to study the progression-free survival. Progression-free survival was calculated from the date of surgery until the date of recurrence. In case of no recurrence, the date of last contact or death was used for censoring. The prognostic effect of the patient and tumor characteristics age, BMI, FIGO stage, peritoneal cytology, lymph node involvement, cervical involvement, lymphovascular space invasion, myometrial invasion, diameter of the tumor, and background endometrium were analyzed by using univariable and multivariable Cox proportional hazards models. The forward stepwise method was used for selection procedures for multivariable Cox proportional hazards models. These results were expressed as hazard ratios with their 95% confidence intervals. Statistical analyses were performed using the software package SPSS 18.0 for Microsoft Windows.
For the cohort from the Mayo Clinic, the study protocol was approved by the institutional review board of the Mayo Foundation; in accordance with the Minnesota Statute for Use of Medical Information in Research, only those patients who consented to the use of their medical records were included. For the cohort from the Radboud University Nijmegen Medical Centre and the Canisius-Wilhelmina Hospital, the Research Ethics Committee of the Radboud University Nijmegen Medical Centre declared that the study protocol is in accordance with the applicable rules concerning the review of research ethics committees and informed consent.
In total, 572 patients with grade 1 endometrioid endometrial carcinoma were identified. Thirty-two patients were excluded because of a personal history of other malignancy. After revision, 13 cases were diagnosed with differentiation grade 2 and were excluded. A total of 527 patients with grade 1 endometrioid endometrial carcinoma remained for analyses. Demographic and histopathologic characteristics of the 527 patients are shown in Table 1. As expected, the majority of patients were stage I and demonstrated favorable histologic characteristics. Fifty-nine patients received adjuvant therapy. In 21 patients recurrence of disease occurred, and only seven patients died as a consequence of the disease. This results in a 5-year progression-free survival rate of 96% and a 5-year overall survival rate of 99%. Note that lymph node dissection was omitted in a large proportion of the cohort. For the 405 patients from the Mayo Clinic, omission of lymph node dissection in 185 patients occurred as per protocol for patients with less than 50% myometrial invasion and tumor diameter less than 2 cm as previously described.18 For the 122 patients from Nijmegen, lymph node dissection was omitted in 116 cases without clinical suspicion of FIGO stage II or more as recommended by the Dutch guidelines for endometrioid endometrial cancer treatment.19
The type of background endometrium diagnosed in the total cohort is shown in Table 2. Some derivation of hyperplasia was present in 387 patients (73%), whereas atrophic endometrium was diagnosed in 88 patients (17%). In 25 patients (5%) there was extensive growth of the tumor in the endometrial cavity causing the entire endometrium to be substituted by the carcinoma. In these patients, no background endometrium could be identified and were therefore excluded from analyses. Furthermore, in 27 patients (5%) normal premenopausal, proliferative endometrium was found. Because the hormonal-regulated status of the premenopausal endometrium is not comparable to the nonstimulated status of the postmenopausal endometrium, these patients were also excluded from further analyses. An example of endometrioid endometrial carcinoma with atrophic background endometrium and endometrioid endometrial carcinoma with hyperplastic background endometrium is shown in Figure 1.
Results of analyses of atrophic background endometrium (n=88) and hyperplastic background endometrium (n=387) with clinical and pathologic characteristics are shown in Table 3. There were significant associations between atrophic endometrium and older age (P<.01) and lower BMI (P<.01). Furthermore, patients with atrophic endometrium were more likely to have advanced-stage disease (P<.01), malignant cells in peritoneal cytology (P=.01), lymph node metastases (P=.01), cervical involvement (P=.03), lymphovascular space invasion (P<.01), and deep myometrial invasion (P<.01). Note that atrophic endometrium was present in all patients with metastatic lymph nodes. No differences were found between the atrophic and hyperplastic group with respect to hypertension, diabetes mellitus, and tumor diameter. When analyses were limited to patients with pure atrophic endometrium compared with patients with mixed atrophic endometrium, a significant association was found only with pure atrophic endometrium and deep myometrial invasion (data shown in Appendix 1, available online at http://links.lww.com/AOG/A304).
In a Cox proportional hazard model, clinicopathologic variables including pre-existing endometrium were analyzed for their association with progression-free survival. In univariable analyses age, BMI, FIGO stage, malignant cells in peritoneal cytology, lymph node metastases, lymphovascular space invasion, myometrial invasion, and atrophic endometrium were significantly associated with progression-free survival (Table 4). No associations were found between progression-free survival and cervical involvement and diameter of the tumor. In multivariable analyses, only older age, advanced FIGO stage, and the presence of atrophic endometrium were independent predictors of progression-free survival (Table 5). Furthermore, when comparing pure atrophic endometrium with mixed atrophic endometrium, no significant difference in progression-free survival was found (data shown in Appendix 2, available online at http://links.lww.com/AOG/A304). In addition, a Kaplan-Meier survival analysis with a log-rank test was made to show differences in survival in patients with atrophic and hyperplastic background endometrium (Fig. 2).
In this study, the presence of atrophic endometrium in patients with grade 1 endometrioid endometrial carcinoma was associated with predictors of poor clinical outcome including high FIGO stage, lymphovascular space invasion, and deep myometrial invasion. Furthermore, atrophic endometrium was significantly associated with poor progression-free survival in multivariable analyses.
The association with deep myometrial invasion and lymphovascular space invasion indicates more aggressive behavior of the tumors with atrophic background endometrium. The expression of this aggressive behavior can be found in the correlation with high FIGO stage, malignant cells in peritoneal cytology, and a poor progression-free survival.
Atrophy has been associated with poor survival previously.12 However, in the study by Sivridis and colleagues, endometrioid endometrial carcinomas with tumor grades 1, 2, and 3 were included. The tumor grade is one of the most important predictors of recurrent disease in endometrioid endometrial carcinoma patients.20,21 Grade 3 endometrioid endometrial carcinomas are a distinct, biologically more aggressive subtype, showing p53 expression in 17-57% of the cases.22,23 It has been suggested that grade 3 endometrioid endometrial carcinoma is a type II endometrial cancer; furthermore, the diagnosis of serous carcinoma compared with grade 3 endometrioid endometrial carcinoma is subject to a great deal of variability among pathologists. In addition, grade 3 endometrioid endometrial carcinomas are more often seen in a background of atrophic endometrium, whereas grade 1 endometrioid endometrial carcinomas more often have a background of hyperplastic endometrium.7,24 Therefore, the difference in survival found by Sivridis and colleagues could be partly confounded by the inclusion of grade 3 tumors. In the current study, atrophic endometrium remained a significant predictor of poor progression-free survival with multivariable analyses in a group limited to patients with grade 1 endometrioid endometrial carcinoma.
The dualistic model for oncogenesis in patients with endometrial cancer is not applicable for approximately 20% of individual cases who present with advanced disease or recur despite the absence of risk factors. These cases with less favorable clinical outcomes may be represented by the cohort we describe here, namely patients with grade 1 endometrioid endometrial carcinoma and an atrophic background endometrium. In our study, 17% of the patients had atrophic endometrium, a similar proportion of patients who do not fit into the dualistic model.10−12 These grade 1 endometrioid endometrial carcinomas may not follow the progression model for type I tumors with unopposed estrogenic stimulation resulting in hyperplasia followed by transition to endometrioid carcinoma.6,15,25 The fact that we find grade 1 endometrioid endometrial carcinomas in atrophic endometrium, the correlation of atrophic endometrium with predictors of poor clinical outcome, and the correlation of atrophy with poor progression-free survival indicates that carcinogenesis in these cases may occur through distinct mechanisms.
Many molecular markers have been identified to emphasize the difference between type I and type II endometrial cancer. Type I tumors are characterized by estrogen and progesterone receptor expression, microsatellite instability, PTEN alterations, and mutations of KRAS and CTNNB1, whereas the majority of type II carcinomas have p53 mutations and Her-2/neu amplifications.8,9 The considerable number of molecular changes identified in type I endometrial carcinomas make it less likely that one pathway will fit all individual cases. Future endometrial cancer research should compare the immunohistochemical and molecular appearance of patients with endometrial cancer with atrophic and hyperplastic background endometrium.
The majority of postmenopausal women have atrophic endometrium.26 However, although hyperplastic endometrium is generally a result of diffuse estrogenic stimulation of the entire endometrium, focal proliferation and hyperplasia have been described.15 Furthermore, weakly proliferative endometrium has been reported in disease-free postmenopausal women in half of the cases in one study.27 These results are comparable with the results found in our study, 50% of the patients with carcinoma with atrophic endometrium show proliferation to some extent in the endometrium. The arbitrarily chosen cutoff point of more than 50% atrophic endometrium was used to categorize the background endometrium showing these ambiguous features of atrophy with focal proliferation. When comparing patients with mixed atrophic background endometrium compared with patients with pure atrophic background endometrium, few differences in clinical and pathologic characteristics were found, and no difference in progression-free survival was found, indicating that both groups are comparable.
Proliferative endometrium is present in premeno-pausal women in the proliferative phase of the menstrual cycle. Proliferative endometrium is stimulated by estrogens, but it is not caused by unopposed estrogen excess.17 This condition is not comparable either to the nonstimulated atrophic endometrium or to unopposed stimulated, hyperplastic endometrium.6 Therefore, patients with proliferative premenopausal endometrium were excluded from analyses.
This is a multicenter study including a large number of patients. Although it is retrospective with the inherent limitations of selection bias and missing data, we were able to collect complete clinical data in the vast majority of the patients. All histologic slides were reviewed by three pathologists separately in the relating hospitals where the patients were treated. Criteria for the diagnosis of atrophy, hyperplasia, or disordered proliferative endometrium were set clearly before the start of the study and were followed systematically by all pathologists.
In conclusion, we found atrophic background endometrium to be an independent prognostic factor in this large series of patients with grade 1 endometrioid endometrial carcinoma. Endometrioid endometrial carcinoma with a background of atrophic endometrium may not follow the progression model for type I tumors, which normally arise in a background of hyperplasia. The nature of the background endometrium should be mentioned in every pathology report because it provides important prognostic information. Future research should investigate possible differences in carcinogenic pathways in these patients with atrophy and endometrioid endometrial carcinoma.
3. Bokhman JV. Two pathogenetic types of endometrial carci noma. Gynecol Oncol 1983;15:10–7.
4. Sherman ME, Bur ME, Kurman RJ. p53 in endometrial cancer and its putative precursors: evidence for diverse pathways of tumorigenesis. Hum Pathol 1995;26:1268–74.
5. Sherman ME. Theories of endometrial carcinogenesis: a multidisciplinary approach. Mod Pathol 2000;13:295–308.
6. Ellenson HL, Ronnett BM, Kurman RJ. Precursor lesions of endometrial carcinoma. In: Kurman RJ, Ellenson HL, Ronnett BM, editors. Blaustein's pathology of the female genital tract. 6th ed. New York (NY): Springer; 2011. p. 359–91.
7. Ellenson HL, Ronnett BM, Soslow RA, Zaino RJ, Kurman RJ. Endometrial carcinoma. In: Kurman RJ, Ronnett BM, Ellen son HL, editors. Blaustein's pathology of the female genital tract. 6th ed. New York (NY): Springer; 2011. p. 393–452.
8. Markova I, Pilka R, Duskova M, Zapletalova J, Kudela M. Prognostic significance of clinic pathological and selected immunohistochemical factors in endometrial cancer [in Czech]. Ceska Gynekol 2010;75:193–9.
9. Prat J, Gallardo A, Cuatrecasas M, Catasus L. Endometrial carcinoma: pathology and genetics. Pathology 2007;39:72–87.
10. Engelsen IB, Akslen LA, Salvesen HB. Biologic markers in endometrial cancer treatment. APMIS 2009; 117:693–707.
11. Lax SF. Molecular genetic changes in epithelial, stromal and mixed neoplasms of the endometrium. Pathology 2007;39: 46–54.
12. Sivridis E, Fox H, Buckley CH. Endometrial carcinoma: two or three entities? Int J Gynecol Cancer 1998;8:183–8.
13. Pecorelli S. Revised FIGO staging for carcinoma of the vulva, cervix, and endometrium. Int J Gynaecol Obstet 2009;105:103–4.
14. Silverberg SG, Mutter GL, Kurman RJ, Kubik-Huch RA, Nogales F, Tavassoli FA. Tumors of the uterine corpus: epithelial tumors and related lesions. In: Tavassoli FA, Stratton FR, editors. WHO classification of tumors: pathology and genetics of tumors of the breast and female genital organs. 5th ed. Geneva (Switzerland): World Health Organization; 2003. p. 221–32.
15. Mills AM, Longacre TA. Endometrial hyperplasia. Semin Diagn Pathol 2010;27:199–214.
16. Kurman RJ, Kaminski PF, Norris HJ. The behavior of endometrial hyperplasia. A long-term study of ‘untreated’ hyperplasia in 170 patients. Cancer 1985;56:403–12.
17. McCluggage WG. Benign Diseases of the Endometrium. In: Kurman RJ, Ellenson LH, Ronnett BM, editors. Blaustein's pathology of the female genital tract. 6th ed: New York (NY): Springer; 2011. p. 305–58.
18. Mariani A, Dowdy SC, Cliby WA, Gostout BS, Jones MB, Wilson TO, et al.. Prospective assessment of lymphatic dissem ination in endometrial cancer: a paradigm shift in surgical staging. Gynecol Oncol 2008;109:11–8.
19. Oncoline: cancer clinical practice guidelines. Available at: www.oncoline.nl
. 2011. Retrieved January 15, 2012.
20. Creutzberg CL, van Putten WL, Warlam-Rodenhuis CC, van den Bergh AC, de Winter KA, Koper PC, et al.. Outcome of high-risk stage IC, grade 3, compared with stage I endometrial carcinoma patients: the Postoperative Radiation Therapy in Endometrial Carcinoma Trial. J Clin Oncol 2004;22:1234–41.
21. Fujimoto T, Nanjyo H, Fukuda J, Nakamura A, Mizunuma H, Yaegashi N, et al.. Endometrioid uterine cancer: histopathological risk factors of local and distant recurrence. Gynecol Oncol 2009;112:342–7.
22. Halperin R, Zehavi S, Habler L, Hadas E, Bukovsky I, Schneider D. Comparative immunohistochemical study of endometrioid and serous papillary carcinoma of endometrium. Eur J Gynaecol Oncol 2001;22:122–6.
23. Lax SF, Kendall B, Tashiro H, Slebos RJ, Hedrick L. The frequency of p53, K-ras mutations, and microsatellite instability differs in uterine endometrioid and serous carcinoma: evidence of distinct molecular genetic pathways. Cancer 2000;15:88:814–24.
24. Voss MA, Ganesan R, Ludeman L, McCarthy K, Gornall R, Schaller G, et al.. Should grade 3 endometrioid endometrial carcinoma be considered a type 2 cancer—a clinical and pathological evaluation. Gynecol Oncol 2012;124:15–20.
25. Kurman RJ, McConnell TG. Precursors of endometrial and ovarian carcinoma. Virchows Arch 2010;456:l–12.
26. Gol K, Saracoglu F, Ekici A, Sahin I. Endometrial patterns and endocrinologic characteristics of asymptomatic menopausal women. Gynecol Endocrinol 2001;15:63–7.
27. Sivridis E, Giatromanolaki A. Proliferative activity in post- menopausal endometrium: the lurking potential for giving rise to an endometrial adenocarcinoma. J Clin Pathol 2004;57: 840–4.
Supplemental Digital Content
© 2012 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.