Endometrial Endometrioid Adenocarcinoma: A Pathologic Analysis of 827 Consecutive Cases

Anatomic Pathology / INSTITUTIONAL REVIEW OF ENDOMETRIAL CARCINOMA

Endometrial Endometrioid Adenocarcinoma A Pathologic Analysis of 827 Consecutive Cases Sharon Nofech-Mozes, MD,1 Zeina Ghorab, MD,1 Nadia Ismiil, MD, FRCPC,1 Ida Ackerman, MD, FRCPC,2 Gillian Thomas, BSc, MD, FRCPC, FRCR(Hon),2 Lisa Barbera, BSc, MD, MPA, FRCPC,2 Al Covens, MD, FRCPC,3 and Mahmoud A. Khalifa, MD, PhD, FRCPC1 Key Words: Endometrial cancer; Pathologic features; Binary grading; Staging; Myoinvasion; Lymphovascular invasion; Cervical involvement DOI: 10.1309/UDYANQ6XTK6UUTXQ

We reviewed 827 consecutive cases of pure endometrial endometrioid adenocarcinoma (EEA) treated by hysterectomy to update the distribution of pathologic features. Tumor grade (reported in a 2tiered system), depth of myometrial invasion, presence of cervical involvement, lymphovascular invasion (LVI), and evidence of extrauterine disease were recorded. The median age at diagnosis was 62 years (range, 30-94 years). The tumor was high grade in 94 cases (11.4%), invaded into the outer half of the myometrium in 249 (30.1%), was positive for cervical involvement in 171 (20.7%), and was positive for LVI in 182 (22.0%). Lymph nodes (sampled in 85 cases) were positive in 13 (1.6%), and ovarian metastases were present in 15 cases (1.8%). High tumor grade was significantly associated with deep myometrial invasion (P < .0001), cervical involvement (P = .0065), and LVI (P < .0001). EEA manifests most commonly with low tumor grade and without deep myometrial invasion. High tumor grade is significantly associated with deep myometrial invasion, cervical involvement, and LVI.

110 110

Am J Clin Pathol 2008;129:110-114 DOI: 10.1309/UDYANQ6XTK6UUTXQ

Endometrial cancer is the most common invasive malignancy of the female genital tract in North America. The National Cancer Institute of Canada estimated that approximately 4,100 Canadian women will be newly diagnosed with cancer of the uterine corpus every year.1 The majority of these cases are endometrioid adenocarcinoma (or type I), which is the most common histologic type. Most patients seek care early because of vaginal bleeding, and the tumor is usually confined to the uterine corpus at the time of diagnosis. The majority of the data on endometrial carcinoma were obtained more than 2 decades ago and have been repeatedly quoted without significant updates.2 More recent data, however, are reported in the setting of clinical trials or for selected subsets of patients with specific risks. These data do not necessarily reflect the experience of practicing pathologists in general service today. The aim of our study was to update the data on the distribution of the pathologic features in endometrial endometrioid adenocarcinoma (EEA) and to examine the rate of lymph node sampling following the International Federation of Gynecology and Obstetrics (FIGO) guidelines for surgical staging.3 In the past 2 decades, there have been significant changes in the patterns of distribution of the pathologic features in other disease sites such as the breast, cervix, and colon.4-10 In view of these modifications, we looked for possible changes in the pathologic trends in endometrial cancer. Another aim of this study was to examine the association between tumor grade, based on the recently introduced 2-tiered (or binary) system, and established histopathologic prognosticators. The new 2-tiered grading system was © American Society for Clinical Pathology

Downloaded from http://ajcp.oxfordjournals.org/ by guest on April 2, 2016

Abstract

Anatomic Pathology / ORIGINAL ARTICLE

introduced in view of the limited reproducibility of FIGO grade 2. The binary system dichotomizes tumors into lowgrade and high-grade tumors based on the proportion of nonsquamous solid growth. It has been shown that the new grading system is more reproducible and has superior prognostic power.11,12

Materials and Methods

The median age at diagnosis was 62 years (range, 30-94 years). In our cohort, 315 cases (38.1%) had their surgical treatment in-house, and 512 cases (61.9%) were referred to the Toronto-Sunnybrook Regional Cancer Centre, Toronto, Canada, for postoperative consultation after receiving their surgical treatment in nonacademic community hospitals. At the time of diagnosis, only 43 patients (5.2%) were premenopausal. Lymph nodes were sampled in 85 patients (10.3%). Based on the binary grading system, 733 cases (88.6%) had low-grade tumors, and in the remaining 94 cases (11.4%), the tumor was high grade. In 643 cases (77.8%), the tumor was confined to the uterine corpus, with no evidence of serosal or cervical involvement and no extrauterine disease. The distribution of the various pathologic features is summarized in ❚Table 1❚. There was a significant association between tumor grade and deep myometrial invasion (P < .0001). Only 26.5% of low-grade tumors had deep myometrial invasion, whereas 59% of the high-grade tumors demonstrated it. Cervical involvement was seen in 171 cases (20.7%). There was a significant association between tumor grade and cervical involvement (P = .0065). LVI was noted in 182 (22.0%) of the cases and showed a significant association with tumor grade (P < .0001). We compared the characteristics of consultation cases referred from community hospitals with those of in-house cases ❚Table 2❚. The referred cases had a slightly lower proportion of high-grade tumors and a lower proportion of ovarian and lymph node metastases. The distribution rates for age, cervical involvement, deep myometrial invasion, and LVI were comparable. The rate of lymph node sampling was almost 10 times higher in the in-house cases. In the entire cohort, lymph nodes were sampled in only 85 cases (10.3%). Lymph node sampling was carried out in 2.5% of the consultation cases (operated on by general gynecologists) and in

❚Table 1❚ Distribution of Pathologic Features According to Tumor Grade* Variable Myometrial invasion Absent Inner half Outer half Cervical involvement Absent Glandular Stromal Lymphovascular invasion Absent Present *

Low Grade (n = 733)

High Grade (n = 94)

Total (n = 827)

152 (20.7) 387 (52.8) 194 (26.5)

2 (2) 37 (39) 55 (59)

154 (18.6) 424 (51.3) 249 (30.1)

592 (80.8) 60 (8.2) 81 (11.1)

64 (68) 9 (10) 21 (22)

656 (79.3) 69 (8.3) 102 (12.3)

602 (82.2) 131 (17.9)

43 (46) 51 (54)

645 (78.0) 182 (22.0)

Data are given as number (percentage).

Am J Clin Pathol 2008;129:110-114

© American Society for Clinical Pathology 111

DOI: 10.1309/UDYANQ6XTK6UUTXQ

111 111

Downloaded from http://ajcp.oxfordjournals.org/ by guest on April 2, 2016

Ethics approval for this study was obtained from the Toronto Academic Health Sciences Council Research Ethics Board. We carried out an institutional review that consisted of 827 consecutive cases of endometrial carcinoma treated primarily by hysterectomy and referred to Sunnybrook Health Sciences Centre, Toronto, Canada, between July 1999 and June 2004. The Department of Anatomic Pathology accessions and reviews about 30,000 surgical specimens per annum, including specimens from patients undergoing surgery in-house and from cases referred from community centers, representing all risk groups. Only cases with the pure endometrioid type of adenocarcinoma were included in the analysis. A group of 4 pathologists with special expertise in gynecologic pathology reviewed all cases (S.N.M., Z.G., N.I., and M.A.K.). According to our gross specimen examination protocol for in-house cases, the entire endometrium and adnexa are routinely submitted for microscopic evaluation. In contrast, the gross specimen examination of referred cases varies according to local standards, with a minimum of 4 representative sections of the endomyometrium per case. The following characteristics were recorded: (1) Tumor grade was recorded based on the binary system, with the low-grade category including tumors with up to 50% nonsquamous solid areas, corresponding to FIGO grades 1 and 2 together and the high-grade category for tumors with more than 50% nonsquamous solid areas, corresponding to FIGO grade 3. (2) Staging was based on the extent of disease in the submitted surgical specimen. However, because a relatively small proportion of cases was formally staged, it was noted that this did not necessarily correspond to the FIGO surgical staging system. (3) Deep myometrial invasion was defined as involvement of the outer half of the myometrium. (4) Cervical involvement was further qualified as glandular or stromal invasion. (5) Lymphovascular invasion (LVI) was defined by the presence of malignant cells within endothelial-lined spaces on H&E-stained sections without immunohistochemical studies. (6) Lymph node status was recorded when available. Statistical analysis of the results was performed by using the χ2 test as done by the SAS 8.2 (SAS Institute, Cary, NC) system for comparison of proportions between the groups.

Results

Nofech-Mozes et al / INSTITUTIONAL REVIEW OF ENDOMETRIAL CARCINOMA

❚Table 2❚ Characteristics of In-House Cases With Surgery Done by Gynecologic Oncologists and Referred Cases With Surgery Done by General Gynecologists*

Median age (range), y High-grade tumor Deep myometrial invasion Cervical involvement Lymphovascular invasion Ovarian metastases Lymphadenectomy Positive lymph nodes *

In-House Cases (n = 315)

Referred Cases (n = 512)

61 (33-93) 45 (14.3) 86 (27.3) 63 (20.0) 76 (24.1) 9 (2.9) 72 (22.9) 9 (2.9)

62 (30-94) 49 (9.6) 163 (31.8) 108 (21.1) 106 (20.7) 6 (1.2) 13 (2.5) 4 (0.8)

Data are given as number (percentage) unless otherwise indicated.

Discussion This study is an updated summary of the distribution of various pathologic features in 827 consecutive cases of EEA, of which 38.1% were in-house cases and 61.9% were referred cases from community general gynecologists. We presented a comprehensive analysis of the pathologic characteristics, which included the entire cohort of EEA in our practice. The pathologic features of referred cases were reviewed by pathologists with special expertise in gynecologic pathology to ensure diagnostic accuracy, consistency, and completeness of pathology reports as described earlier.13 The differences in the proportion of high-grade tumors and ovarian and lymph node metastases between in-house and consultation cases reflect the nature of our practice in which cases with a preoperative diagnosis of high-grade tumor are often referred to be operated on by gynecologist oncologists. The data in the present study are different from the recent data on EEA, which describe selected subgroups, mostly in 112 112

Am J Clin Pathol 2008;129:110-114 DOI: 10.1309/UDYANQ6XTK6UUTXQ

© American Society for Clinical Pathology

Downloaded from http://ajcp.oxfordjournals.org/ by guest on April 2, 2016

22.9% of the in-house cases (operated on by gynecologic oncologists). As expected, the rate of lymph node sampling was much higher in high-grade tumors (30/94 [32%]) than in low-grade tumors (55/733 [7.5%]). It is interesting that the proportion of nodal metastases found was similar in the lowgrade (8/55 [15%]) and high-grade (5/30 [17%]) tumors. When we examined the clinical scenarios that prompted the lymph node sampling in our 72 in-house cases, we found that 37 cases had the diagnosis of high-grade or nonendometrioid histologic features in preoperative tissue samples. In 19 cases, the uterus was assessed intraoperatively, and deep myometrial invasion or cervical involvement was confirmed. In 11 cases, cervical involvement was suspected in the preoperative physical examination. Intraoperatively, grossly enlarged lymph nodes were evident in 2 cases, and 2 cases showed evidence of advanced disease.

the setting of clinical trials. Unlike in other disease sites such as breast, cervix, and colon, for which screening programs have been implemented, the distribution of histopathologic parameters of EEA has not changed since the 1970s. Endometrial carcinoma still manifests most commonly with a low histologic grade, and, in approximately 70% of patients, the tumor does not invade the outer half of the myometrium. The actual proportion of very early stage EEA and low-grade tumors may be even slightly higher owing to the possible selection bias in the population referred to our cancer center for treatment. Our current study confirmed a significant association between tumor grade and deep myometrial invasion, cervical involvement, and LVI, even when using a binary grading system. The age at diagnosis is in line with that previously reported.2,14,15 A comprehensive review of endometrial cancer in Norway reported by Abeler et al2 in 1992 summarized data collected in the 1970s. This study included 1,740 cases treated with hysterectomy and showed that approximately a quarter of these cases had high-grade tumors, which is more than twice the 11.4% observed in our study. However, the Norwegian study included cases with nonendometrioid histologic features, which may account for this difference. Another study, by Chan et al,14 included 12,333 cases of surgically staged EEA collected from the Surveillance, Epidemiology, and End Results database and also found that one quarter of patients had high-grade tumors. Unlike our study, Chan et al14 did not include a centralized pathology review and had a selection bias by excluding low-risk patients because the study exclusively analyzed surgically staged cases. The rates of deep myometrial invasion, cervical involvement, and LVI ❚Table 3❚ reported by Abeler et al2 are quite similar to our rates. Inoue et al15 studied 238 cases of endometrial cancer, of which only 4 had nonendometrioid histologic features, treated primarily by hysterectomy between 1976 and 1993. The distribution of pathologic characteristics found in their report was comparable to the distribution found in the present study (Table 3).2,15,16 Although since 1988 FIGO requires surgical staging that includes pelvic and para-aortic lymphadenectomy in all cases of endometrial cancer, the procedure is still controversial and has not been universally applied in clinical practice. More specifically, the role of comprehensive surgical staging, including pelvic and para-aortic lymphadenectomy for all patients, has been questioned.14,17-22 The rate of lymph node sampling varies in different countries and even in North America between different centers. The rate of lymph node sampling in our study was 10.3%, which is in line with the recently reported rate of surgical staging in the province of Ontario, Canada.23 This figure is substantially lower than the rate of lymph node sampling in the United States where, according to Surveillance,

Anatomic Pathology / ORIGINAL ARTICLE

❚Table 3❚ Distribution of Pathologic Characteristics in the Present Study and Other Studies

Present study Abeler et al2 Inoue et al15 Lachance et al16

No. of Cases

Mean Age (y)

No. of Cases With Lymph Node Sampling

High-Grade Tumor (%)

Deep Myometrial Cervical Invasion (%)

Involvement (%)

LVI (%)

Positive Lymph Nodes (%)

827 1,740 238 341

62.7 62 NA 60

85 NA 210 335

12 26* 13 28*

30.1 26 30 NA

20.7 23 18 NA

22.0 23 34 NA

16 NA 12 NA

LVI, lymphovascular invasion; NA, not available. * Included nonendometrioid types.

From the 1Department of Anatomic Pathology, Sunnybrook Health Sciences Centre; 2Radiation Oncology, and 3Division of Gynecologic Oncology, Toronto-Sunnybrook Regional Cancer Centre, Toronto, Canada. Presented in part at the United States and Canadian Academy of Pathology Annual Meeting; March 2007; San Diego, CA. Address reprint requests to Dr Khalifa: Dept of Anatomic Pathology, Sunnybrook Health Sciences Centre, 2075 Bayview Ave, Room E-416, Toronto, ON, Canada M4N 3M5.

References 1. Canadian Cancer Statistics 2007. Available at http://www.cancer.ca/vgn/images/portal/cit_86751114/36/15/ 1816216925cw_2007stats_en.pdf. Accessed October 23, 2007.

2. Abeler VM, Kjorstad KE, Berle E. Carcinoma of the endometrium in Norway: a histopathological and prognostic survey of a total population. Int J Gynecol Cancer. 1992;2:9-22. 3. International Federation of Gynecology and Obstetrics. Corpus cancer staging [news]. Int J Gynecol Obstet. 1989;28:189-190. 4. Cress RD, Morris C, Ellison GL, et al. Secular changes in colorectal cancer incidence by subsite, stage at diagnosis, and race/ethnicity, 1992-2001. Cancer. 2006;107(5 suppl):1142-1152. 5. Shen Y, Yang Y, Inoue LY, et al. Role of detection method in predicting breast cancer survival: analysis of randomized screening trials. J Natl Cancer Inst. 2005;97:1195-1203. 6. Berry DA, Cronin KA, Plevritis SK, et al. Effect of screening and adjuvant therapy on mortality from breast cancer. N Engl J Med. 2005;353:1784-1792. 7. Wang SS, Sherman ME, Hildesheim A, et al. Cervical adenocarcinoma and squamous cell carcinoma incidence trends among white women and black women in the United States for 1976-2000. Cancer. 2004;100:1035-1044. 8. Kotake K, Honjo S, Sugihara K, et al. Changes in colorectal cancer during a 20-year period: an extended report from the multi-institutional registry of large bowel cancer, Japan. Dis Colon Rectum. 2003;46(10 suppl):S32-S43. 9. Faivre-Finn C, Bouvier-Benhamiche AM, Phelip JM, et al. Colon cancer in France: evidence for improvement in management and survival. Gut. 2002;51:60-64. 10. Cress RD, Morris CR, Wolfe BM. Cancer of the colon and rectum in California: trends in incidence by race/ethnicity, stage, and subsite. Prev Med. 2000;31:447-453. 11. Alkushi A, Abdul-Rahman ZH, Lim P, et al. Description of a novel system for grading of endometrial carcinoma and comparison with existing grading systems. Am J Surg Pathol. 2005;29:295-304. 12. Scholten AN, Smit VT, Beerman H, et al. Prognostic significance and interobserver variability of histologic grading systems for endometrial carcinoma. Cancer. 2004;100:764-772. 13. Khalifa MA, Dodge J, Covens A, et al. Slide review in gynecologic oncology ensures completeness of reporting and diagnostic accuracy. Gynecol Oncol. 2003;90:425-430. 14. Chan JK, Cheung MK, Huh WK, et al. Therapeutic role of lymph node resection in endometrioid corpus cancer: a study of 12,333 patients. Cancer. 2006;107:1823-1830. 15. Inoue Y, Obata K, Abe K, et al. The prognostic significance of vascular invasion by endometrial carcinoma. Cancer. 1996;78:1447-1451. 16. Lachance JA, Everett EN, Greer B, et al. The effect of age on clinical/pathologic features, surgical morbidity, and outcome in patients with endometrial cancer. Gynecol Oncol. 2006;101:470-475.

Am J Clin Pathol 2008;129:110-114

© American Society for Clinical Pathology 113

DOI: 10.1309/UDYANQ6XTK6UUTXQ

113 113

Downloaded from http://ajcp.oxfordjournals.org/ by guest on April 2, 2016

Epidemiology, and End Results data, node dissection was performed in 32.7% of cases of endometrial cancer in 1992.24 In a patterns-of-care study, Roland et al25 reported only a 26% histologically confirmed lymph node assessment in patients operated on by a nongynecologic oncologist compared with 83% of patients when surgery was performed by a gynecologic oncologist. In our in-house patients, operated on by gynecologic oncologists, lymph node sampling was part of the primary surgical treatment in 22.9% of the cases. It is noteworthy that there was a change in our institutional guidelines regarding lymph node sampling within the study period, with an increased rate of nodal sampling in the 2004 cases compared with earlier cases. The similar proportion of positive nodes in low- (15%) and high-grade (17%) tumors in our study reflects the skewed population of patients with low-grade tumors who were selected for the procedure based on the presence of additional high-risk features, compared with patients with high-grade tumors who received lymph node sampling even in the absence of adjunct high-risk features. In Canada, routine lymph node sampling is still limited and surgical staging is more likely to occur in academic institutions when highgrade histologic features are diagnosed preoperatively.

Nofech-Mozes et al / INSTITUTIONAL REVIEW OF ENDOMETRIAL CARCINOMA

17. Aalders JG, Thomas G. Endometrial cancer: revisiting the importance of pelvic and para aortic lymph nodes. Gynecol Oncol. 2007;104:222-231. 18. Kirby TO, Leath CA III, Kilgore LC. Surgical staging in endometrial cancer. Oncology (Huntingt). 2006;20:45-50. 19. Greven KM, Corn BW. Endometrial cancer. Curr Probl Cancer. 1997;21:65-127. 20. Zaino RJ, Kurman RJ, Diana KL, et al. 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;77:1115-1121. 21. Creasman WT, Morrow CP, Bundy BN, et al. Surgical pathologic spread patterns of endometrial cancer: a Gynecologic Oncology Group study. Cancer. 1987;60 (8 suppl):2035-2041.

22. Boronow RC, Morrow CP, Creasman WT, et al. Surgical staging in endometrial cancer: clinical-pathologic findings of a prospective study. Obstet Gynecol. 1984;63:825-832. 23. Kwon JS, Carey MS, Cook EF, et al. Patterns of practice and outcomes in intermediate- and high-risk stage I and II endometrial cancer: a population-based study. Int J Gynecol Cancer. 2007;17:433-440. 24. Partridge EE, Shingleton HM, Menck HR. The National Cancer Data Base report on endometrial cancer. J Surg Oncol. 1996;61:111-123. 25. Roland PY, Kelly FJ, Kulwicki CY, et al. The benefits of a gynecologic oncologist: a pattern of care study for endometrial cancer treatment. Gynecol Oncol. 2004;93:125-130.

Downloaded from http://ajcp.oxfordjournals.org/ by guest on April 2, 2016

114 114

Am J Clin Pathol 2008;129:110-114 DOI: 10.1309/UDYANQ6XTK6UUTXQ

© American Society for Clinical Pathology