In the United States, endometrial cancer is the most common gynecologic malignancy and is the fourth most common type of cancer in women.1 In 2012, in the United States, more than 47,000 women were diagnosed with and more than 8,000 women died of endometrial cancer.1 Most women, approximately 70%, diagnosed with endometrial cancer have clinical stage I disease.2
In 1987, the Gynecologic Oncology Group published a surgical pathologic study examining patterns of spread among women with clinical stage I disease.3 The authors found that among women who had a disease that was clinically limited to the uterus, 9% had positive pelvic lymph nodes and 6% had positive para-aortic lymph nodes. The likelihood of having positive pelvic and/or para-aortic lymph nodes increased with both grade of disease and depth of invasion. Because of the results of this study, among others, the International Federation of Gynecology and Obstetrics (FIGO)4 adopted a surgical staging system for endometrial cancer in 1988.
Since the establishment of surgical staging as the initial step in the management of endometrial cancer, the performance of pelvic lymph node dissection (PLND) and para-aortic lymph node dissection (PALND) as part of surgical staging has become more common. In 2000, a study published by authors from the Mayo Clinic identified a group of women with endometrial adenocarcinoma who did not benefit from lymphadenectomy (LND): women with grade 1 or 2 endometrioid adenocarcinoma of the endometrium, tumor dimension of 2 cm or less, myometrial invasion of 50% or less on frozen section analysis, and no intraoperative evidence of macroscopic disease.5
The Mayo algorithm depends on accurate frozen section diagnosis performed in a timely fashion. Prospective data from the Mayo Clinic have suggested that intraoperative frozen section can reliably be used to make treatment decisions for women with endometrial cancer because clinically relevant discordance between intraoperative frozen section and final pathology occurred in only 1.3% of cases.6 However, it remains unknown whether the reliability demonstrated at the Mayo Clinic can be duplicated at other institutions.7,8
In addition to the controversy surrounding the use of frozen section and the decision to perform LND, there has also been a great deal of debate surrounding the therapeutic benefit of lymph node dissection. Two large prospective trials examined the therapeutic benefit of PLND among women with clinical stage I endometrial carcinoma and neither demonstrated improvement in overall survival (OS).9,10 Several criticisms of these studies have been made: that these included low-risk women, that postoperative treatment was not standardized, and that PALND was not included.11
Although there is literature suggesting that PALND imparts a survival advantage for women with endometrial adenocarcinoma and intermediate- to high-risk tumor types,12 the decision to stage a patient is inextricable from her underlying health status. In other words, an individual surgeon may be more likely to fully stage a woman who is not obese, diabetic, or otherwise medically compromised. We hypothesized that the observed survival benefit of PALND is a reflection of the general health of the patient rather than a therapeutic benefit of surgery. The current study was undertaken to determine whether PALND added a survival advantage in women with intermediate- to high-risk endometrial adenocarcinoma, separate from underlying medical status.
MATERIALS AND METHODS
Study approval was obtained from the University of Virginia institutional review board. Using an institutional cancer registry, we identified all women who were diagnosed with endometrial cancer from January 1, 2002, through December 31, 2009. Using the information obtained from the cancer registry, we excluded all women with nonendometrioid histology. We included all women with grade 2 or 3 endometrioid adenocarcinoma. For patients with grade 1 endometrioid adenocarcinoma, we reviewed pathology reports and operative reports if path reports were not available. Only those subjects with grade 1 histology who had a frozen section and a depth of invasion noted to be 50% or greater were included.
Each patient’s clinical information was reviewed using an institutional electronic medical record. Information abstracted included demographic information; clinical data including medical comorbidities; method of diagnosis (endometrial biopsy vs curettage); preoperative imaging obtained; the route of surgery (laparotomy, vaginal, laparoscopic, or robotic) and other surgical details; details about frozen section and final pathology; type(s) of adjuvant therapy administered; and information regarding recurrence, vital status, and cause of death as determined by ICD-10 codes on death certificates. All pathology reports were reviewed, and stage was determined on the basis of the 2009 FIGO staging system.13 Women with dual primaries or who were treated with chemotherapy or radiation before having surgery were excluded from analysis.
Statistical analysis was performed using IBM Statistical Package for the Social Sciences statistics 19 software (Statistical Package for the Social Sciences Inc). The χ2 test, the Fisher exact test, or the t test was used for all univariate analyses. Overall survival and disease-specific survival (DSS) rates were calculated using the Kaplan-Meier method and were compared using the log-rank test.
A total of 772 women were diagnosed with endometrial adenocarcinoma at the University of Virginia between January 1, 2002, and December 31, 2009. Of these, 253 met the study inclusion criteria and were included in the analysis. The patient and tumor characteristics for the entire patient population are shown in Table 1. The mean age at the time of surgery was 63.3 years (range, 29–92). Preoperative height and weight were recorded for 207 patients (82%). The mean preoperative body mass index (BMI) was 33.3 kg/m2 (range, 17.6–70.7 kg/m2). Of the women who had preoperative BMI recorded, 125 (60.4%) were obese. Most of the patients had at least 1 medical comorbidity, the most common being obesity, hypertension, diabetes, and pulmonary disease. Most of the patients in this intermediate- to high-risk group of women had grade 2 and 3 disease (53% and 33%, respectively).
Two hundred twenty women (87%) had an open procedure. There were 174 patients who had PLND and 82 patients who had PALND. All women who had PALND also had PLND. The rate of positive nodes was 13% (23/174) for the women who had PLND and was 7% (6/82) for the women who had both PLND and PALND. Only 1.2% (1/82) of the women who had both PLND and PALND had negative pelvic but positive para-aortic nodes.
The 1 patient who had positive para-aortic nodes and negative pelvic nodes was statistically outside the norm for our patient population: she was 45 years old and had no other medical comorbidities. She underwent a laparotomy and full surgical staging, with her final pathology revealing grade 3 endometrioid adenocarcinoma of the endometrium with 3 positive para-aortic lymph nodes. She was treated with adjuvant chemotherapy and tumor-directed radiotherapy but experienced an abdominal recurrence 14 months after her primary surgery and ultimately died of the disease 2 years after her initial surgery.
The study population was divided into those who had PALND performed and those who did not. The patient characteristics of each group are shown in Table 2. The patientswho had PALND had a lower BMI (31 kg/m2 vs 35kg/m2, P = 0.013), although the mean BMI for both groups was in the obese range. The women who had PALND had lower rates of diabetes (16% vs 30%, P = 0.02), hypertension (46% vs 61%, P = 0.03), pulmonary disease (4% vs 15%, P= 0.01), and coronary artery disease (1% vs 9%, P = 0.02). The proportion of women who had an open procedure was not different between the 2 groups. No significant trend was noted between the 2 groups other than the distribution of tumor grade. Of note, there was no difference in the rates of postoperative vaginal cuff brachytherapy, external beam radiationtherapy, or administration of chemotherapy based on PALND.
A total of 30 women recurred, and the recurrence rate among the women who had PALND performed was no different than that among the women who did not have PALND performed (14.3% vs 7.5%, P value = not significant). The most common sites of recurrence were the vagina, the abdomen, and the lung. There were 4 patients (1.6%) who had evidence of recurrence in the para-aortic lymph nodes. In 2 of these women, the recurrence was isolated to the para-aortic lymph nodes, and the other 2 had widely metastatic disease. None of the 4 women who recurred in their para-aortic lymph nodes had PALND performed at the time of their initial surgery.
The OS and DSS rates were calculated using the Kaplan-Meier method and were compared between the 2 groups using the log-rank test. When the patients who had PALND were compared with those who did not, there was a significant difference in OS (P = 0.007). The 5-year OS was 96% for those who had PALND and 82% for those who did not (Fig. 1). However, there was no difference in 5-year DSS, 96% for those who had PALND and 89% (P value = not significant) for those who did not (Fig. 2).
Endometrial cancer is the fourth most common malignancy in women in the United States and is the most common gynecologic malignancy. Because it is directly related to obesity, as the obesity rates continue to climb, the rates of endometrial cancer will likely continue to increase as well. Because of the increasing incidence of the disease, the importance of optimizing the management strategy for patient safety and cost is exceedingly important.
During the last 30 years, endometrial cancer staging and management have changed significantly. Clinical staging has transformed into a surgical staging system. Surgical management has morphed from an open hysterectomy and bilateralsalpingo-oophorectomy, often via a pfannenstiel incision, to an open procedure including the removal of pelvic and para-aortic lymph nodes, to more and more frequently surgical staging via minimally invasive techniques. The question of whether LND is beneficial and for whom it is beneficial remains controversial.
There is little doubt that the addition of para-aortic node dissection adds both operative time and morbidity to the surgical procedure, regardless of the surgical approach.14,15 Laparotomy requires extension of the incision above the umbilicus to adequately dissect the para-aortic lymph nodes, leading to more postoperative morbidity for patients who are often obese and at risk for postoperative wound infection, wound hernia, and prolonged immobilization. Para-aortic node dissection via minimally invasive techniques requires an extra robotic arm in the case of robotic surgery and increased operating time as well as surgical expertise in performing this technique.16,17
In our series, only 1 patient (1.2%) who had both PLND and PALND had negative pelvic but positive para-aortic nodes. Thus, in this group of patients, pelvic nodes are a very accurate measure of disease in the para-aortic nodes. This finding has also been corroborated by larger retrospective studies in the past.18 Unlike in the past decades, currently, chemotherapy is the standard treatment for women with positive nodes and can be expected to sterilize microscopic disease in the para-aortic nodes.19,20 Thus, it is difficult in the subset of patients considered in this study to justify PALND for all patients. In addition, given the multiple variables that influence the performance of para-aortic node dissection, most important of which are surgical technique, surgical skill, and patient comorbidities, a full cost analysis is not possible, but given the very low risk for isolated para-aortic node metastasis and the high likelihood that patients will die with the disease rather than of the disease, the performance of para-aortic node dissection in all intermediate- to high-risk patients as defined by this study is unwarranted.
The current study demonstrates that among the women with intermediate- to high-risk endometrioid adenocarcinoma, there is an improved OS among the women who had PALND compared with those who did not, but this finding is misleading because there is no difference in DSS. In fact, the women who had PALND were less obese and less likely to have medical comorbidities. It is also likely that the women who were medically more fit received more aggressive adjuvant therapy (outside the scope of this study but worth considering). We therefore suggest that the improved OS is not due to the therapeutic benefit of PALND but rather due to the better underlying health status of those women who had PALND.
There are several limitations of this study, the most obvious being its retrospective nature at a single institution. Certainly, prospective assessment of this question would be ideal but would also require a study so large that cost would likely be prohibitive. In addition to obtaining a sample size large enough to examine DSS, one would also have to control for the rates of comorbidities.
Recently, Clements and colleagues21 published a cost analysis of the Mayo approach to lymph node dissection triage. Clements and colleagues21 used a decision analysis model to compare 2 strategies for the surgical management of women with newly diagnosed endometrial cancer: (1) routine LND independent of intraoperative risk factors or (2) selective LND for women with high- or intermediate-risk tumors based on intraoperative assessment. The study concluded that a strategy of selective LND based on intraoperative risk factors for managing patients with endometrial cancer was less cost-effective than routine LND. This study mirrors our own findings: in practice, the length of time to obtain frozen section results and the accuracy of those results with respect to final pathology do not justify the cost.
On the basis of this cost-analysis data, the following question is raised: should all patients with intermediate- or high-risk endometrial adenocarcinoma undergo PLND, but because the risk for positive para-aortic nodes is extraordinarily low in the setting of negative pelvic nodes (<2%),18 because women with positive pelvic nodes will likely receive adjuvant chemotherapy that theoretically would sterilize microscopically positive para-aortic nodes, and because most women will die of comorbid medical factors rather than recurrent disease,22 should routine para-aortic node dissection not be performed?
The data collected in our study in combination with the cost-effectiveness data recently published by Clements et al21 suggest the following: the optimal management strategy for this group of women is PLND for all and PALND only for those with obvious metastatic disease.
1. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012. CA Cancer J Clin. 2012; 62: 10–29.
2. Creasman WT, Odicino F, Maisonneuve P, et al. Carcinoma of the corpus uteri. FIGO 26th Annual Report on the Results of Treatment in Gynecological Cancer. Int J Gynaecol Obstet. 2006; 95 (suppl 1): S105–S143.
3. Creasman WT, Morrow CP, Bundy BN, et al. Surgical pathologic spread patterns of endometrial cancer. A Gynecologic Oncology Group Study. Cancer. 1987; 60: 2035–2041.
4. International Federation of Gynecology and Obstetrics. Announcements: FIGO stages- 1988 revision. Gynecol Oncol. 1989; 35: 125–127.
5. Mariani A, Webb MJ, Keeney GL, et al. Low-risk corpus cancer: is lymphadenectomy or radiotherapy necessary? Am J Obstet Gynecol. 2000; 182: 1506–1519.
6. Kumar S, Medeiros F, Dowdy SC, et al. A prospective assessment of the reliability of frozen section to direct intraoperative decision making in endometrial cancer. Gynecol Oncol. 2012; 127: 525–531.
7. Case AS, Rocconi RP, Straughn JM Jr, et al. A prospective blinded evaluation of the accuracy of frozen section for the surgical management of endometrial cancer. Obstet Gynecol. 2006; 108: 1375–1379.
8. Quinlivan JA, Petersen RW, Nicklin JL. Accuracy of frozen section for the operative management of endometrial cancer. BJOG. 2001; 108: 798–803.
9. Benedetti Panici P, Basile S, Maneschi F, et al. Systematic pelvic lymphadenectomy vs. no lymphadenectomy in early-stage endometrial carcinoma: randomized clinical trial. J Natl Cancer Inst. 2008; 100: 1707–1716.
10. ASTEC Study Group; Kitchener H, Swart AM, Qian Q, et al. Efficacy of systematic pelvic lymphadenectomy in endometrial cancer (MRC ASTEC trial): a randomised study. Lancet. 2009; 373: 125–136.
11. Uccella S, Podratz KC, Aletti GD, et al. Re: systematic pelvic lymphadenectomy vs no lymphadenectomy in early-stage endometrial carcinoma: randomized clinical trial. J Natl Cancer Inst. 2009; 101: 897–898.
12. Todo Y, Kato H, Kaneuchi M, et al. Survival effect of para-aortic lymphadenectomy in endometrial cancer (SEPAL study): a retrospective cohort analysis. Lancet. 2010; 375: 1165–1172.
13. Creasman W. Revised FIGO staging for carcinoma of the endometrium. Int J Gynaecol Obstet. 2009; 105: 109.
14. Kodama J, Seki N, Ojima Y, et al. Risk factors for early and late postoperative complications of patients with endometrial cancer. Eur J Obstet Gynecol Reprod Biol. 2006; 124: 222–226.
15. Possover M, Krause N, Plaul K, et al. Laparoscopic para-aortic and pelvic lymphadenectomy: experience with 150 patients and review of the literature. Gynecol Oncol. 1998; 71: 19–28.
16. Seamon LG, Fowler JM, Richardson DL, et al. A detailed analysis of the learning curve: robotic hysterectomy and pelvic-aortic lymphadenectomy for endometrial cancer. Gynecol Oncol. 2009; 114: 162–167.
17. Scribner DR Jr, Walker JL, Johnson GA, et al. Laparoscopic pelvic and paraaortic lymph node dissection: analysis of the first 100 cases. Gynecol Oncol. 2001; 82: 498–503.
18. Mariani A, Keeney GL, Aletti G, et al. Endometrial carcinoma: paraaortic dissemination. Gynecol Oncol. 2004; 92: 833–838.
19. Randall ME, Spirtos NM, Dvoretsky P. Whole abdominal radiotherapy versus combination chemotherapy with doxorubicin and cisplatin in advanced endometrial carcinoma (phase III): Gynecologic Oncology Group Study No. 122. J Natl Cancer Inst Monogr. 1995; 19: 13–15.
20. Vale CL, Tierney J, Bull SJ, et al. Chemotherapy for advanced, recurrent or metastatic endometrial carcinoma. Cochrane Database Syst Rev. 2012; 8: CD003915.
21. Clements AE, Tierney BJ, Cohn DE, et al. Is selective lymphadenectomy more cost-effective than routine lymphadenectomy in patients with endometrial cancer? Gynecol Oncol. 2013; 128: 166–170.
22. Ward KK, Shah NR, Saenz CC, et al. Cardiovascular disease is the leading cause of death among endometrial cancer patients. Gynecol Oncol. 2012; 126: 176–179.
Keywords:© 2014 by the International Gynecologic Cancer Society and the European Society of Gynaecological Oncology.
Intermediate- to high-risk; Endometrial adenocarcinoma; Para-aortic lymph node