| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, Strong Memorial Hospital, University of Rochester School of Medicine, Rochester, New York.
Address reprint requests to: Brent DuBeshter, MD, 125 Lattimore Road, Rochester, NY 14620; E-mail: brent_dubeshter{at}urmc.rochester.edu.
 |
ABSTRACT |
|---|
TOP
 ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: Cervical cytology was obtained in 300 patients with endometrial cancer before surgical staging, which included lymphadenectomy. Tumor grade and histology from endometrial sampling were compared with final pathology, and the risk of nodal spread in relation to cervical cytology, tumor grade, and myometrial invasion was assessed using 
2 and logistic regression analysis.
RESULTS: Endometrial cells on cervical cytology, deep myometrial invasion, and high-grade tumor were associated with 91%, 87%, and 83% of the cases with nodal spread, respectively. In patients with grade 1 tumor on biopsy, final pathology revealed grade 2 in 21%, and grade 3 in 2%. In patients with normal cervical cytology, no nodal metastases occurred with grade 1 tumor on biopsy, and no aortic metastases occurred, regardless of grade. Cervical cytology and tumor grade contributed independently to the likelihood of nodal metastases.
CONCLUSION: All patients with endometrial cancer should undergo lymphadenectomy until a reliable system is found to identify those with negligible (less than 1%) risk of nodal spread. The risk of lymph node spread in those with normal cervical cytology is low (2%). Further study of those with normal cervical cytology is needed to determine if lymphadenectomy can be omitted with grade 1 tumor on biopsy, or whether aortic lymphadenectomy is necessary regardless of grade.
Although endometrial cancer is the most common gynecologic malignancy, controversy exists regarding its optimum management.1 Over the past decade, surgical staging, including pelvic and aortic lymphadenectomy, has become more important in selecting patients for postoperative radiation therapy.2 Tumor grades, assessed by endometrial sampling and myometrial invasion determined intraoperatively, are the traditional criteria used to determine whether lymphadenectomy is necessary.3 Unfortunately, assessment of myometrial invasion by frozen section is not always accurate, discrepancies between tumor grade on biopsy and in the hysterectomy specimen occur regularly,4,5 and there is a lack of consensus regarding which patients require lymphadenectomy.
Tumor grade on final pathology may be higher than that on endometrial biopsy in 10–30% of cases.4–6 Therefore, it has been recommended that tumor grade be reassessed intraoperatively by frozen section.1 An accurate assessment of tumor grade is essential, as many authorities do not recommend lymphadenectomy for grade 1 tumors with limited myometrial invasion.7–9.
Although myometrial invasion has been assessed preoperatively by a variety of radiological procedures, intraoperative frozen section remains the gold standard for this determination. However, even frozen section is inaccurate in distinguishing superficial from deep myometrial invasion in 5–10% of cases.10–12 A reliable preoperative method to select patients who require pelvic and/or aortic lymphadenectomy would be ideal.
Although cervical cytology in patients with endometrial cancer has been shown to correlate with stage, grade, myometrial invasion, and the risk of lymph node metastases,13,14 many standard medical texts do not list abnormal cervical cytology as an indication for lymphadenectomy.7,8 In addition, the predictive value of the combination of cervical cytology and tumor grade on endometrial biopsy for nodal metastases has not been compared with myometrial invasion and tumor grade, the traditional factors used in this assessment. As a result, it is unclear currently how the results of cervical cytology should be incorporated into treatment decisions for patients with endometrial cancer.15
In this study we evaluated the role of cervical cytology, tumor grade on endometrial biopsy, and myometrial invasion in determining the risk of nodal spread in patients with endometrioid adenocarcinoma of the endometrium.
|
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The preoperative evaluation in all study patients included endometrial sampling by either biopsy or dilation and curettage, cervical cytology, chest x-ray, and standard preoperative tests. Papanicolaou smears were performed using a plastic broom and were analyzed according to conventional (nonthin prep) cytologic techniques. Patients with nonendometrioid adenocarcinoma on endometrial biopsy were excluded from further analysis, as were those who did not have preoperative cervical cytology.
Surgical treatment consisted of total abdominal hysterectomy with bilateral salpingo-oophorectomy, complete pelvic lymphadenectomy, and peritoneal washings in all cases. Aortic lymphadenectomy was usually performed for high-grade tumors or for deep myometrial invasion, but was sometimes omitted in morbidly obese patients or if unresectable upper abdominal disease was present.
Cervical cytology results were obtained from the medical record and classified as normal if there were no endometrial cells present. If atypical squamous cells of undetermined significance were noted without concomitant endometrial cells, the cytology was classified as normal. However, regardless of the patient’s menopausal status, the presence of any endometrial cells, whether cytologically benign, suspicious, or malignant, was considered abnormal.
Surgical pathology reports were reviewed for tumor grade, histology, myometrial invasion, and malignant cells on peritoneal washings. Tumor grade and histology from preoperative endometrial samplings were compared with findings from the hysterectomy specimen. Pelvic and aortic node involvement or other extrauterine spread was also noted. The risk of pelvic or aortic lymph node metastases was related to the results of cervical cytology, tumor grade as assessed by endometrial sampling, and myometrial invasion determined from final pathologic review of the hysterectomy specimen, using 2 testing and logistic regression analysis. P values less than .05 were considered significant.
|
RESULTS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Descriptive data on cervical cytology, tumor grade on biopsy, myometrial invasion from final pathology, and other factors according to lymph node status are listed in Table 1
. Multivariable logistic regression for the parameters that are known preoperatively, tumor grade on biopsy and cervical cytology, revealed odds ratios for grade 3 tumor and abnormal cervical cytology that were similar (5.0 versus 5.3). Cervical cytology and tumor grade contributed independently to the likelihood of nodal metastases. Age, menopausal status, and hormone replacement therapy were not associated with nodal metastases or with abnormal cervical cytology.
|
. Overall, 23% of those with grade 1 endometrioid adenocarcinoma on biopsy had a higher grade tumor, and 5% had a virulent histology, such as papillary serous or clear cell carcinoma found in the hysterectomy specimen. Other discrepancies between tumor grade and histology are also noted in Table 2. No residual cancer was found in the hysterectomy specimen in eight (3%) of the patients.
|
, the risk of lymph node metastases by tumor grade on endometrial biopsy and myometrial invasion from final pathologic review are listed. With less than 50% myometrial invasion, only two patients had pelvic node metastases and one patient had aortic node spread. Deep myometrial invasion or high-grade tumor on biopsy (grade 2 or 3) was associated with 20 (87%) and 19 (83%) of the 23 cases with lymph node spread, respectively. As expected, the risk of lymph node metastases increased with higher grade and deep myometrial invasion.
|
. Cervical cytology was normal in 109 (36%) cases, of which two (2%) patients had pelvic node involvement. There was no pelvic or aortic lymph node metastasis with grade 1 tumor and normal cervical cytology. No aortic node metastasis occurred in any patients with normal cervical cytology, regardless of tumor grade or myometrial invasion. Cervical cytology was abnormal (presence of any endometrial cells) in 191 (64%) cases, of which 20 (11%) had pelvic node spread and 11 of 142 (8%) had aortic node metastases. The results of cervical cytology correlated with both the risk of pelvic (P < .01) and aortic (P = .03) nodal spread.
|
shows the positive and negative predictive values for lymph node metastases, either pelvic or aortic, of cervical cytology, deep myometrial invasion, and high-grade tumor, as well as the combination of tumor grade on biopsy with cytology or myometrial invasion. Cervical cytology had the greatest sensitivity and a negative predictive value of 0.98. Abnormal cervical cytology was associated with 21 (91%) of 23 cases with lymph node spread. For grade 1 tumor, the positive and negative predictive values of cervical cytology and myometrial invasion were similar.
|
|
DISCUSSION |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Unfortunately, tumor grade on endometrial biopsy frequently differs from that found in the hysterectomy specimen. In our patients with grade 1 adenocarcinoma on endometrial biopsy, 23% had higher tumor grade and 5% had virulent histologic subtypes, such as papillary serous or clear cell carcinoma, found on final pathology. Assessment of myometrial invasion by frozen section has also been shown to be inaccurate in 5–10% of cases10–12; this was not evaluated in our study.
Although the traditional criteria of tumor grade and myometrial invasion on final pathology correlate well with the risk of lymph node spread,18–20 inaccurate preoperative or intraoperative assessment of both have led to difficulty in subsequent treatment decisions. As a result, inadequately staged patients may choose a second operative procedure to determine with certainty whether there has been tumor spread.21 Because of its importance in determining subsequent therapy, several authorities currently recommend lymphadenectomy in all patients with endometrial cancer.22 In a recent study supporting this position, Takeshima et al23 found pelvic lymph node metastases in four (5%) of 83 patients with grade 1 tumor and no myometrial invasion.
The morbidity of lymphadenectomy must be taken into account when recommending this procedure to all patients with endometrial cancer. Fortunately, it appears to be minimal. Larson et al24 found that rates of transfusion, febrile morbidity, postoperative complications, and mortality were similar for patients with endometrial cancer treated with or without lymphadenectomy. Even elderly patients (over 70) have not been shown to have a significant increase in morbidity associated with lymphadenectomy.25 Any increase in morbidity attributable to lymphadenectomy would have to be balanced against the morbidity associated with either pelvic irradiation or surgical restaging that are usually recommended to those inadequately staged.
It may be possible to avoid lymphadenectomy in patients with negligible (less than 1%) risk of lymph node spread. We found no lymph node metastases, either pelvic or aortic, among 66 patients with normal cytology and grade 1 tumor on biopsy. Based on an analysis of this finding, we are reasonably certain that the risk of lymph node metastases in this group is less than 5% (P = .03). However, a study with approximately 460 patients with normal Papanicolaou smears and grade 1 tumor would be required to show a risk of less than 1%. If these findings are confirmed in a larger study, many patients would not require lymphadenectomy. In our study population, the need for this procedure would have been eliminated in 22% of patients.
The results of this study also suggest that aortic lymphadenectomy may not be necessary in some patients where pelvic lymphadenectomy is warranted. No aortic node metastases occurred among 74 patients with normal cervical cytology, regardless of tumor grade. A larger study is necessary before aortic lymphadenectomy can be safely omitted in those with normal cytology. However, in a recent publication26 it was noted that the risk of aortic node spread was less than 1% in patients with grade 1 tumor and less than one third myometrial invasion.
Shedding of atypical or malignant cells on cervical cytology has been shown to correlate with myometrial invasion, grade, stage, and the risk of lymph node metastases.13,14 We found the presence of any endometrial cells, whether benign or malignant on cervical cytology, to be a powerful predictor of lymph node spread, comparable to high-grade tumor. Based on multivariable logistic regression, both cervical cytology and tumor grade on biopsy, the only two parameters easily assessed preoperatively, contributed independently to the likelihood of lymph node metastases.
Although there is a clear relationship between shedding of tumor cells on cervical cytology and various poor prognostic factors such as histology, grade, and nodal spread, the underlying pathophysiology has not been thoroughly investigated. In reviewing Larson et al’s and our data, 15 of 17 (88%) patients with clear cell or papillary serous endometrial cancer, known for their virulence, shed malignant cells on cervical cytology.13,14 We are currently investigating cell adhesion molecules such as CD44 and cadherin to determine their relationship to positive cervical cytology and other poor prognostic factors.
Lymphadenectomy is warranted in all patients with endometrial cancer until a reliable system is found to identify those with negligible (less than 1%) risk of nodal spread. The current practice of selecting patients for lymphadenectomy based on tumor grade and myometrial invasion is suboptimal because of the frequent discrepancies between final pathology and the preoperative or intraoperative assessment of these factors. By assessing cervical cytology and tumor grade on biopsy, it may be possible to avoid lymphadenectomy entirely for those with grade 1 tumor, and aortic lymphadenectomy regardless of grade, if cervical cytology is normal.
|
Footnotes |
|---|
Received July 8, 2002. Received in revised form August 26, 2002. Accepted September 5, 2002.
|
REFERENCES |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Horowitz NS, Peters WA, Smith MR, Drescher CW, Atwood MA, Mate TP. Adjuvant high dose vaginal brachytherapy as treatment of stage I and II endometrial carcinoma. Obstet Gynecol 2002;99:235–40.
3. Burke TW, Eifel PJ, Muggia FM. Cancer of the uterine body. In: DeVita VT Jr, Hellman S, Rosenberg SA, eds. Cancer principles & practice of oncology. 5th ed. Philadelphia: Lippincott Raven Publishers, 1997:1478–99.
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. Oakley G, Nahhas WA. Endometrial adenocarcinoma: Therapeutic impact of preoperative histopathologic examination of endometrial tissue. Eur J Gynecol Oncol 1989; 10:255–60.[Medline]
6. Larson DM, Johnson KK, Broste SK, Krawisz BR, Kresl JJ. Comparison of D&C and office endometrial biopsy in predicting final histopathologic grade in endometrial cancer. Obstet Gynecol 1995;86:38–42.[Abstract]
7. Disaia PJ, Creasman WT. Clinical gynecologic oncology. 6th ed. St. Louis: Mosby, Inc., 2002.
8. Hacker NF. Uterine cancer. In: Berek JS, Hacker NF, eds. Practical gynecologic oncology. 3rd ed. Philadelphia: Lippincott Williams & Wilkins, 2000:407–55.
9. Herbst AL. Neoplastic diseases of the uterus. In: Stenchever MA, Droegemuller W, Herbst AL, Mishell DR. Comprehensive gynecology. 4th ed. St. Louis: Mosby, Inc., 2001.
10. Fanning J, Tsukada Y, Piver MS. Intraoperative frozen section diagnosis of depth of myometrial invasion in endometrial adenocarcinoma. Gynecol Oncol 1990;37:47–50.[Medline]
11. Goff BA, Rice LW. Assessment of depth of myometrial invasion in endometrial adenocarcinoma. Gynecol Oncol 1990;38:46–48.[Medline]
12. Doering DL, Barnhill DR, Weiser EB, Burke TW, Wood-ward JE, Park RC. Intraoperative evaluation of depth of myometrial invasion in stage I endometrial adenocarcinoma. Obstet Gynecol 1989;74:930–3.
13. DuBeshter B, Warshal DP, Angel C, Dvoretsky PM, Lin JY, Raubertas RF. Endometrial carcinoma: The relevance of cervical cytology. Obstet Gynecol 1991;77:458–62.
14. Larson DM, Johnson KK, Reyes CN Jr, Broste SK. Prognostic significance of malignant cervical cytology in patients with endometrial cancer. Obstet Gynecol 1994; 84:399–403.
15. Fukuda K, Mori M, Uchiyama M, Iwai K, Iwasaka T, Sugimori H, et al. Preoperative cervical cytology in endometrial carcinoma and its clinicopathologic relevance. Gynecol Oncol 1999;72:271–2.[Medline]
16. Anderson JM, Stea B, Hallum AV, Rogoff E, Childers J High-dose-rate postoperative vaginal cuff irradiation alone for stage IB and IC endometrial cancer. Int J Radiat Oncol Biol Phys 2000;46:417–25.[Medline]
17. Larson DM, Broste SK, Krawisz BR. Surgery without radiotherapy for primary treatment of endometrial cancer. Obstet Gynecol 1998;91:355–9.[Abstract]
18. Creasman WT, Morrow CP, Bundy BN, Homesley HD, Graham JE, Heller PB. Surgical pathologic spread patterns of endometrial cancer. Cancer 1987;60:2035–41.[Medline]
19. Boronow RC, Morrow CP, Creasman WT, Disaia PJ, Silverberg SG, Miller A, et al. Surgical staging in endometrial cancer: Clinical-pathologic findings of a prospective study. Obstet Gynecol 1984;63:825–32.
20. Lurain JR. Uterine cancer. In: Berek JS, ed. Novak’s gynecology. 13th ed. Philadelphia: Lippincott Williams & Wilkins, 2002:1143–96.
21. 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.[Medline]
22. Orr JW Jr, Roland PY, Leichter D, Orr PF. Endometrial cancer: Is surgical staging necessary? Curr Opin Oncol 2001;13:408–12.[Medline]
23. Takeshima N, Hirai Y, Tanaka N, Yamawaki T, Yamauchi K, Hasumi K. Pelvic lymph node metastasis in endometrial cancer with no myometrial invasion. Obstet Gynecol 1996;88:280–2.[Abstract]
24. Larson DM, Johnson K, Olson KA. Pelvic and paraaortic lymphadenectomy for surgical staging of endometrial cancer: Morbidity and mortality. Obstet Gynecol 1992;79: 998–1001.
25. Susini R, Susini T, Ferrandina G, Poerio A, Margariti PA, Carminati R, et al. Systematic pelvic and aortic lymphadenectomy in elderly gynecologic oncologic patients. Cancer 2001;92:2562–8.[Medline]
26. DuBeshter B. Endometrial cancer: Predictive value of cervical cytology. Gynecol Oncol 1999;72:271–2.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
