Endometrial cancer is the most common gynecologic malignancy and the fourth most common malignancy in women in the United States. The histologic grade is a well-recognized prognostic factor and correlates with risk of extrauterine disease and overall outcome. The probability of pelvic node metastases in clinical stage I grades 1, 2, and 3 endometrial cancer is 2.8%, 8.7%, and 18.3%, respectively, and the reported risk for aortic node metastases is 1.6%, 4.9%, and 11.1%, respectively.1 The grade of the tumor alone is an important predictor of survival. The 5-year survival for grades 1, 2, and 3 are 87%, 75%, and 58%, respectively.2 However, these data are based on the histologic grade assigned at final pathology. In 1988, the International Federation of Gynecology and Obstetrics (FIGO) established a surgical staging system. The main intent of the staging system was to define risk groups for adjuvant treatment. Since the development of the FIGO staging system, there has not been a consensus on which patients presenting with apparent early-stage endometrial cancer should have comprehensive surgical staging.3
Most authors agree that patients with grade 1 or 2 tumor with evidence of more than 50% myometrial invasion, grade 3 disease, high-risk histologic variants (papillary serous/clear cell adenocarcinoma and sarcoma), and/or cervical extension of the tumor require full surgical staging. However, the need for a complete staging procedure in patients with grade 1 disease is debated. Although 15–25% of women with preoperative grade 1 histology will demonstrate postoperative uterine features suggesting risk for metastatic disease (Horowitz NS, Powell MA, Smith JH, Rader JS, Gibb RK, Mutch DG, et al. Staging grade 1 endometrial cancers: saving dollars and lives [abstract]. Proc Am Soc Clin Oncol 2003;22:457),4,5 the majority of these cases are not staged, and gynecologic oncologists are involved in only a minor portion of their surgeries. Data retrieved from the National Cancer Database in 1996 showed that in the United States only 30% of endometrial cancer patients are being staged, and gynecologic oncologists are involved in only 45% of their surgeries (Partridge EE, Jessup JM, Donaldson ES, Taylor PT, Randal M, Braley P, et al. 1996 Patient care evaluation study of cancer of the corpus uteri. The National Cancer Data Base, American College of Surgeons [abstract]. Gynecol Oncol 1999;72:445). Patients with grade 1 disease probably are disproportionately represented in this group of unstaged patients, with the possible mistaken assumption that they are at low risk for metastatic disease.
The objectives of this study were to 1) estimate the correlation between the presenting endometrioid grade 1 disease and the final histology and FIGO stage, 2) compare the accuracy of endometrial biopsy and dilation and curettage (D&C) in reference to the hysterectomy final pathology, 3) identify the proportion of patients who presented with grade 1 disease and demonstrated intra- and postoperative histological and clinical findings, suggesting a risk for metastatic disease, and 4) estimate the percentage of patients with preoperative grade 1 tumor for whom the staging results influenced treatment recommendations. To achieve our last and primary objective we will determine how many surgically staged patients were identified to have extrauterine disease and, therefore, were recommended to receive adjuvant therapy and how many of the patients with intrauterine high-risk features avoided teletherapy because staging did not reveal metastatic disease.
MATERIALS AND METHODS
All patients at the Arthur G. James Cancer Hospital of Ohio State University diagnosed with uterine malignancy between January 1, 1997, and July 31, 2003, were identified. All patients who did not undergo primary surgical treatment at Ohio State University were excluded from the study. Approval for this study was given by the Ohio State University Institutional Review Board. The medical records were reviewed, and a database containing all demographic, clinical, and pathological information was created using Microsoft Access 2000 (Microsoft, Redmond, WA) software. For the purpose of our study the following data were retrieved from the database: 1) the pathological grade and tumor cell type from the endometrial curetting or biopsy that lead to the initial diagnosis of malignancy; 2) intraoperative information, including type of surgery and staging procedure, estimated blood loss, and surgical complications; 3) the postoperative course, including length of hospital stay and postoperative complications; and 4) surgical-pathologic information gathered from the pathology reports, which included cytology, histology, and grade of the final tumor specimen, myometrial invasion, pelvic and para-aortic lymph node number and status, and tumor features such as lymph vascular space invasion. Lymphadenectomy was defined as 6 or more total pelvic lymph nodes removed and 3 or more paraortic lymph nodes removed. This arbitrary definition was introduced to exclude patients in whom only minimal lymph node sampling was performed. Final FIGO stage was determined according to the surgical-pathologic findings. Our attending pathologists reviewed all endometrial biopsies and curetting slides that were read by outside pathologists, and in any case of disagreement, we chose the interpretation of the Ohio State University pathologists. All surgeries were performed by 1 of 4 gynecologic oncologists.
Statistical evaluation was performed using Microsoft Excel 2000 and SPSS 12 (SPSS Inc, Chicago, IL) software. Fisher exact test was used where applicable. Significance was held at the standard value of P < .05.
During the study period, 349 patients underwent surgical management for uterine malignancy in the Division of Gynecologic Oncology at Ohio State University. A total of 181 (52%) women were identified with preoperative endometrioid grade 1 disease. Table 1 shows the demographic and clinical features of this group of patients. The diagnostic sampling method was either endometrial biopsy or D&C, with a frequency of 48% and 52%, respectively. Surgical staging, denoting at least pelvic lymph node dissection, was performed in 82% of cases, while the rate for complete surgical staging, denoting pelvic and para-aortic lymphadenectomy was 66%. The mean number of dissected pelvic and para-aortic nodes was 26 (range 2–66). Omitted or incomplete lymph node dissections occurred at the discretion of the surgeon and involved either gross disease outside the pelvis or carcinomas grossly confined to the endometrium with high operative risk.
A total of 6 patients (3.2%) had severe operative complications (4 vessel injury, 1 ureteric injury, and 1 nerve injury). There were 2 perioperative mortalities (1 pulmonary embolism and 1 myocardial infarction). One of these patients was staged.
Table 2 summarizes presenting versus final pathologic grade. Nineteen percent of patients were upgraded, with 15% grade 2, 0.5% grade 3, 2.5% serous or clear cell, and 1% mixed mesodermal tumor. No cases presenting as grade 1 endometrial carcinoma were downgraded, even if the final pathology demonstrated no residual tumor, because review of the diagnostic slides demonstrated grade 1 adenocarcinoma in all cases. There was no significant difference between the accuracy of endometrial biopsy and D&C in predicting the final tumor grade (P = .6). Table 3 summarizes the final FIGO stages of patients presenting with grade 1 endometrial cancer. Interestingly, 10.5% of the grade 1 endometrial cancer patients had extrauterine spread (> IIb), 3.9% of the patients had positive nodes (IIIc), and 18% of patients with grade 1 disease had a final stage greater than stage I. Two patients with node involvement and one patient with stage IV had no high-risk intrauterine features. Using common criteria for full surgical staging—ie, a) grade 1 or 2 with > 50% myometrial invasion, b) grade 3 tumor, c) cervical extension, and d) high risk histologic variants (clear cell, papillary serous, or sarcoma)—we identified 47 patients (26%) who needed complete surgical staging (Table 4).
A total of 22 patients (12%) received chemotherapy and/or external radiation (Table 5); 17 of those patients were among the 47 patients with high-risk operative features. The other 5 patients were 1 patient with synchronous endometrioid cancer in the ovary and 4 more patients with stage IIIa (1 patient), IIIc (2 patients), and IV (1 patient) who had no intrauterine high-risk features. As a result of surgical staging, 22 patients (12.2%) with either extrauterine disease or high-risk histologic variants were treated with chemotherapy, with or without external radiation. Thirty more patients (16.6%) avoided adjuvant teletherapy and/or chemotherapy based on the staging results, 17 with deep invasion, 2 with grade 3 disease, and 11 with cervical involvement. No patients with endometrioid tumor stage Ia–IIb received adjuvant teletherapy or chemotherapy.
Patients presenting with grade 1 endometrial cancer are often considered to have low-risk disease requiring only a hysterectomy–bilateral salpingo-oophorectomy (BSO) and not requiring adjuvant therapy. The first objective of our study was to estimate the correlation between the presenting endometrioid grade 1 disease and the final histology and FIGO stage. Nineteen percent of our grade 1 cases were upgraded in the final histopathology, and this finding is supported by other investigators. Daniel and Peters4 reported that pathological study of the entire uterus upgraded 15–25% of all grade 1 tumors, and Horowitz et al (Proc Am Soc Clin Oncol 2003;22:457) observed that 12% of grade 1 histologies will be upgraded in the final pathology. The discrepancy between the grade of adenocarcinoma diagnosed at presentation and the final histopathology can be attributed to a sampling error in the original diagnostic procedure (sampling versus examination of the entire uterus specimen) and a poor concordance between pathologists with respect to the FIGO grading system. We tried to limit the later intervariation by choosing the interpretation of Ohio State University pathologists when discrepancy was found with the outside pathologist and by having the initial slides and the final pathology specimens evaluated by the same pathologist at our institution.
Interestingly, 25% of the patients had deep myometrial invasion, with or without extrauterine disease (stage > Ib), and 10.5% had extrauterine and cervical disease (stage > IIb), very similar to the data reported by Peterson et al.6 The finding that 15.5% of the patients had deep (> 50%) myometrial invasion is similar to data from Creasman et al1 reporting a 12% and 10% myometrial invasion to the middle and deep third of the myometrium, respectively, in clinical stage I grade 1 patients. The evidence that staging patients with deep myometrial invasion conveys a survival advantage gives additional support to the argument that this group of patients should be surgically staged.7,8
Of special interest are the 5 patients with stage IV and 7 patients with stage IIIc, all presenting as grade 1 tumors. Three of the 5 patients with stage IV had serous differentiation, and 1 more patient had no intrauterine high-risk features. Of the 7 patients with stage IIIc (node involvement), 2 patients had no high-risk features with minimal myometrial invasion (13–13.4%) and grade 1–2 tumors on the final pathology, and 5 more patients had either deep myometrial invasion (2 patients) or a combination of deep invasion and cervical (2 patients) or adnexal involvement (1 patient). Based upon uterine pathology alone (had staging not been performed), the 3 patients (stage IIIc and IV) with no risk factor would have been observed only, and adjuvant pelvic radiation would have been recommended for at least 4 of the other 5 patients with nodal spread. The updated data coming from the Gynecologic Oncology Group (GOG) Protocol 122 trial showing survival benefit to chemotherapy compared with whole abdominal radiation in advanced endometrial cancer (Randall ME, Brunetto G, Muss H, Mannell RS, Fowler JM, Thigpen JT, et al. Whole abdominal radiotherapy versus combination doxorubicin-cisplatin chemotherapy in advanced endometrial carcinoma: a randomized phase III trial of the GOG [abstract]. Proc Am Soc Clin Oncol 2003;22:3.) emphasize even more the importance of identifying and treating systemically patients with extrauterine spread, rather than giving a nonspecific radiation to patients at risk.
The second objective of our study was to compare the accuracy of endometrial biopsy and D&C in reference to the hysterectomy final pathology. The data in the literature concerning the accuracy of these 2 sampling methods would suggest that they are equally effective in making the diagnosis of cancer.9,10 In our series no statistically significant difference was found in the distribution of the grades between the 2 sampling method, ie, biopsy and D&C (Fisher exact test P = .61). This fact should both encourage physicians to start with endometrial biopsy when evaluating postmenopausal bleeding and discourage physicians from assuming that the risk of extrauterine disease is smaller if the interpretation of grade 1 tumor was based on the results of a D&C rather than a biopsy.
The third objective of our study was to check the proportion of patients presenting with grade 1 histology who demonstrated intra- and postoperative histological and clinical findings suggesting a risk for metastatic disease. No standard criteria for triaging patients to surgical staging exist. Most would agree that risk of extrauterine disease is increased in the presence of the high-risk intrauterine features previously discussed11 and that these patients should be staged. Forty-seven of our grade 1 patients (26%) were found to have high-risk intrauterine features on final pathology. This finding is supported by other publications.4,5,12,13 In other words, 1 of every 4 patients presenting with grade 1 disease will need a retroperitoneal dissection according to either intra- or postoperative information. If the high-risk features are diagnosed intraoperatively, a lymph node dissection should be performed. If the high-risk intrauterine features are diagnosed only in the final pathology after hysterectomy and salpingo-oophorectomy (TAH+BSO) alone, the treatment options are either adjuvant postoperative teletherapy, with or without chemotherapy, or full surgical staging. Both these options carry substantial risk. The combination of surgery and radiotherapy is associated with significant morbidity in up to 12% of the patients.14–17 The alternative choice of reoperation for staging carries the associated morbidity of a major operation, whether accomplished via laparotomy or laparoscopically.18,19 Both these options further emphasize the importance of comprehensive staging in patients with high-risk uterine features.
Gross inspection and frozen section of the hysterectomy specimen were proposed by many physicians as a helpful tool for differentiating between the patients at high risk and those at low risk for extrauterine disease and preventing an unnecessary staging procedure. Data concerning the accuracy of intraoperative assessment of myometrial invasion by gross inspection and frozen section can be found in the literature and was not one of our objectives. It is, however, important to note that a recent publication from the M. D. Anderson Cancer Center by Frumovitz et al20 concluded that the combination of intraoperative frozen section analysis for histologic grade and depth of myometrial invasion does not correlate well with final pathologic grade and stage, although contradicted data exist.21
The surgical management of endometrial cancer not only affects postoperative treatment, but it also has a financial impact. The various staging and postoperative treatment strategies that exist will effect the total cost of care. Barnes et al22 compared the economic impact of 3 different approaches to the treatment of endometrial cancer patients. The first approach was comprehensive surgical staging of all patients, with adjuvant treatment reserved for documented cases of extrauterine disease; the second approach was TAH+BSO with lymph node dissection reserved for cases of myometrial invasion, followed by adjuvant radiation based on the presence of uterine risk factors; and the third approach was TAH+BSO, followed by intraoperative pathologic assessment of the uterus and consultation with a surgical oncologist for comprehensive staging. The first approach, comprehensive surgical staging with adjuvant treatment reserved for cases with extrauterine disease, yielded the lowest charges per patient and was 20–26% less expensive than the 2 other approaches. Orr et al23 reviewed 444 patients with stage I endometrial cancer who underwent comprehensive surgical staging followed by brachytherapy. The authors noted that an additional $3.8 million dollars would have charged if adjuvant radiation therapy had been used in this patient population. Fanning et al24 reported a cost saving of 39% for treatment of early-stage endometrial cancer with lymphadenectomy and vaginal brachytherapy compared with selective lymphadenectomy and radiation treatment.
The fourth objective of our study was to estimate the numbers of patients with preoperative grade 1 tumor who were recommended to either receive or avoid adjuvant treatment based on the staging procedure results. Based on the staging results, 22 patients with either extrauterine disease or high-risk histological variants received chemotherapy and or teletherapy. Thirty more patients with high-risk features (17 with deep myometrial invasion, 2 with grade 3 disease, and 11 with cervical involvement) avoided adjuvant teletherapy. All together, surgical staging significantly impacted postoperative treatment decisions in 29% of the patients presenting initially with grade 1 disease. Even when excluding patients with high-risk histological variants (clear cell/serous/sarcoma) and adnexal involvement, who would have received some type of adjuvant therapy without comprehensive staging, we are still left with 38 patients (21%) who benefited from the staging procedure.
The decision not to give adjuvant therapy to patients with high-risk features and negative staging is based on the increasing number of retrospective studies showing that, in completely surgically staged patients, in the absence of proven extrauterine spread or adverse histologic cell type, there is no survival advantage for adjunctive teletherapy.25–29 Support for these studies came from the results of a prospective randomized study of surgically staged (stage I) patients with intermediate risk factors (GOG 99) who did not demonstrate a survival advantage with the addition of adjuvant teletherapy.30 In addition, the risk for recurrence in surgically staged patients with endometrial carcinoma confined to the uterine corpus is small, and the majority of these recurrences can be salvaged with radiation therapy.29,31
Two of the objections to routine surgical staging are the additional time required and the risks of the retroperitoneal dissection. Reviewing the data shows that, if the staging procedure is performed at the time of the primary operation by appropriately trained surgeons, the procedure can be performed without significantly increasing patient morbidity.32,33 The mean additional time is 30–45 minutes, and the main risk is persistent lymphocyst formation, which may require surgical drainage. Less common complications include major vessel, ureteric, and nerve injuries, which are reported to occur in 1–3% of the cases. In our series of surgically high-risk patients, 3.2% of the staged patients had severe surgical complications, and there were 2 perioperative mortalities (1 pulmonary embolism, 1 myocardial infarction). One of these patients was staged. The mean hospital stay was 3.8 days and was not significantly different between the staged and the unstaged groups.
In conclusion, more than a quarter of the patients presenting with preoperative grade 1 endometrial cancer will have a clinically significant risk of ultimately being discovered to have high-risk intrauterine features, and more than 10% will have extrauterine disease. Surgical staging in patients presenting with grade 1 endometrial cancer significantly impacts postoperative treatment decisions in 21–29% of the patients. Patients presenting with apparent early-stage endometrial cancer, even preoperative grade 1 cancer, should therefore be counseled regarding the likelihood of extrauterine disease, the ability to diagnose such spread during surgery, the surgical treatment options (comprehensive staging or a different approach), and the possible consequences of each treatment approach. Based on this report, our recommendation is for all patients with endometrial cancer, including patients with grade 1 tumor, to be surgically staged with pelvic and para-aortic lymph node dissection, if feasible.
1. Creasman WT, Morrow CP, Bundy BN, Homesley HD, Graham JE, Heller PB. Surgical pathologic spread patterns of endometrial cancer. A Gynecologic Oncology Group Study. Cancer 1987;60(suppl):2035–41.
2. Abeler VM, Kjorstad KE. Endometrial adenocarcinoma in Norway: a study of a total population. Cancer 1991;67:3093–103.
3. Orr JW Jr, Roland PY, Leichter D, Orr PF. Endometrial cancer: is surgical staging necessary? Curr Opin Oncol 2001;13:408–12.
4. Daniel AG, Peters WA. Accuracy of office and operating room curettage in the grading of endometrial carcinoma. Obstet Gynecol 1988;71:612–4.
5. Obermair A, Geramou M, Gucer F, Denison U, Graf AH, Kapshammer E, et al. Endometrial cancer: accuracy of the finding of a well differentiated tumor at dilatation and curettage compared to the findings at subsequent hysterectomy. Int J Gynecol Cancer 1999;9:383–6.
6. Petersen RW, Quinlivan JA, Casper GR, Nicklin JL. Endometrial adenocarcinoma–presenting pathology is a poor guide to surgical management. Aust N Z J Obstet Gynaecol 2000;40:191–4.
7. Orr JW Jr, Orr PF, Taylor PT. Surgical staging endometrial cancer. Clin Obstet Gynecol 1996;39:656–68.
8. Kilgore LC, Partridge EE, Alvarez RD, Austin JM, Shingleton HM, Noojin F, et al. Adenocarcinoma of the endometrium: survival comparisons of patients with and without pelvic node sampling. Gynecol Oncol 1995;56:29–33.
9. Epstein E, Skoog L, Valentin L. Comparison of Endorette and dilatation and curettage for sampling of the endometrium in women with postmenopausal bleeding. Acta Obstet Gynecol Scand 2001;80:959–64.
10. Dijkhuizen FP, Mol BW, Brolmann HA, Heintz AP. The accuracy of endometrial sampling in the diagnosis of patients with endometrial carcinoma and hyperplasia: a meta-analysis. Cancer 2000;89:1765–72.
11. Morrow CP, Curtin JP. Management of uterine neoplasia. In: Morrow CP, Curtin JP, editors. Gynecologic cancer surgery. New York (NY): Churchill Livingston; 1996. p. 569–626.
12. Zorlu CG, Kuscu E, Ergun Y, Aydogdu T, Cobanoglu O, Erdas O. Intraoperative evaluation of prognostic factors in stage I endometrial cancer by frozen section: how reliable? Acta Obstet Gynecol Scand 1993;72:382–5.
13. Kucera E, Kainz C, Reinthaller A, Sliutz G, Leodolter S, Kucera H, et al. Accuracy of intraoperative frozen-section diagnosis in stage I endometrial adenocarcinoma. Gynecol Obstet Invest 2000;49:62–6.
14. Larson DM, Copeland LJ, Gallager HS, Kong JP, Wharton JT, Stringer CA. Stage II endometrial carcinoma: results and complications of a combined radiotherapeutic-surgical approach. Cancer 1988;61:1528–34.
15. Creutzberg CL, van Putten WL, Koper PC, Lybeert ML, Jobsen JJ, Warlam-Rodenhuis CC, et al. The morbidity of treatment for patients with stage I endometrial cancer: results from a randomized trial. Int J Radiat Oncol Biol Phys 2001;51:1246–55.
16. Torrisi JR, Barnes WA, Popescu G, Whitfield G, Barter J, Lewandowski G, et al. Postoperative adjuvant external-beam radiotherapy in surgical stage I endometrial carcinoma. Cancer 1989;64:1414–7.
17. Potish RA, Dusenbery KE. Enteric morbidity of postoperative pelvic external beam and brachytherapy for uterine cancer. Int J Radiat Oncol Biol Phys 1990;18:1005–10.
18. Childers JM, Spirtos NM, Brainard P, Surwit EA. Laparoscopic staging of the patient with incompletely staged early adenocarcinoma of the endometrium. Obstet Gynecol 1994;83:597–600.
19. Eltabbakh GH, Shamonki MI, Moody JM, Garafano LL. Laparoscopy as the primary modality for the treatment of women with endometrial carcinoma. Cancer 2001;91:378–87.
20. Frumovitz M, Slomovitz BM, Singh DK, Broaddus RR, Abrams J, Sun CC, et al. Frozen section analyses as predictors of lymphatic spread in patients with early-stage uterine cancer. J Am Coll Surg 2004;199:388–93.
21. Quinlivan JA, Petersen RW, Nicklin JL. Accuracy of frozen section for the operative management of endometrial cancer. BJOG 2001;108:798–803.
22. Barnes MN, Roland PY, Straughn M, Kilgore LC, Alvarez RD, Partridge EE. Comparison of treatment strategies for endometrial adenocarcinoma: analysis of financial impact. Gynecol Oncol 1999;74:443–7.
23. Orr JW Jr, Holimon JL, Orr PF. Stage I corpus cancer: is teletherapy necessary? Am J Obstet Gynecol 1997;176:777–88.
24. Fanning J, Hoffman ML, Andrews SJ, Harrah AW, Feldmeier JJ. Cost-effectiveness analysis of the treatment for intermediate risk endometrial cancer: postoperative brachytherapy vs observation. Gynecol Oncol 2004;93:632–6.
25. Orr JW. Surgical staging of endometrial cancer: does the patient benefit? Gynecol Oncol 1998;71:335–9.
26. Larson DM, Broste SK, Krawisz BR. Surgery without radiotherapy for primary treatment of endometrial cancer. Obstet Gynecol 1998;91:355–9.
27. Mohan DS, Samuels MA, Selim MA, Shalodi AD, Ellis RJ, Samuels JR, et al. Long-term outcomes of therapeutic pelvic lymphadenectomy for stage I endometrial adenocarcinoma. Gynecol Oncol 1998;70:165–71.
28. Lybeert ML, van Putten WL, Brolmann HA, Coebergh JW. Postoperative radiotherapy for endometrial carcinoma. Stage I. Wide variation in referral patterns but no effect on long-term survival in a retrospective study in the southeast Netherlands. Eur J Cancer 1998;34:586–90.
29. Straughn JM Jr, Huh WK, Kelly FJ, Leath CA 3rd, Kleinberg MJ, Hyde J Jr, et al. Conservative management of stage I endometrial carcinoma after surgical staging. Gynecol Oncol 2002;84:194–200.
30. Keys HM, Roberts JA, Brunetto VL, Zaino RJ, Spirtos NM, Bloss JD, et al. A phase III trial of surgery with or without adjunctive external pelvic radiation therapy in intermediate risk endometrial adenocarcinome: a Gynecologic Oncology Group study. Gynecol Oncol 2004;92:744–51.
31. Ackerman I, Malone S, Thomas G, Franssen E, Balogh J, Dembo A. Endometrial carcinoma–relative effectiveness of adjuvant irradiation vs therapy reserved for relapse. Gynecol Oncol 1996;60:177–83.
32. Homesley HD, Kadar N, Barrett RJ, Lentz SS. Selective pelvic and periaortic lymphadenectomy does not increase morbidity in surgical staging of endometrial carcinoma. Am J Obstet Gynecol 1992;167:1225–30.
33. Larson DM, Johnson K, Olson KA. Pelvic and para-aortic lymphadenectomy for surgical staging of endometrial cancer: morbidity and mortality. Obstet Gynecol 1992;79:998–1001.