Therapeutic role of lymphadenectomy for cervical cancer

Authors

  • Monjri Shah MD,

    1. Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, Columbia University College of Physicians and Surgeons, New York, New York
    Search for more papers by this author
  • Sharyn N. Lewin MD,

    1. Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, Columbia University College of Physicians and Surgeons, New York, New York
    2. Herbert Irving Comprehensive Cancer Center, New York, New York
    Search for more papers by this author
  • Israel Deutsch MD,

    1. Herbert Irving Comprehensive Cancer Center, New York, New York
    2. Department of Radiation Oncology, Columbia University College of Physicians and Surgeons, New York, New York
    Search for more papers by this author
  • William M. Burke MD,

    1. Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, Columbia University College of Physicians and Surgeons, New York, New York
    2. Herbert Irving Comprehensive Cancer Center, New York, New York
    Search for more papers by this author
  • Xuming Sun MS,

    1. Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, Columbia University College of Physicians and Surgeons, New York, New York
    Search for more papers by this author
  • Thomas J. Herzog MD,

    1. Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, Columbia University College of Physicians and Surgeons, New York, New York
    2. Herbert Irving Comprehensive Cancer Center, New York, New York
    Search for more papers by this author
  • Jason D. Wright MD

    Corresponding author
    1. Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, Columbia University College of Physicians and Surgeons, New York, New York
    2. Herbert Irving Comprehensive Cancer Center, New York, New York
    • Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, Columbia University College of Physicians and Surgeons, 161 Fort Washington Ave, 8th Floor, New York, NY 10032
    Search for more papers by this author
    • Fax: (212) 305-3412


Abstract

BACKGROUND.

Despite the diagnostic value of lymphadenectomy for early-stage cervical cancer, its therapeutic role is unknown. We examined the therapeutic potential of extensive lymphadenectomy in women with early-stage cervical cancer.

METHODS.

Women with stage IA2-IIA cervical cancer who underwent radical hysterectomy with lymphadenectomy in the Surveillance, Epidemiology, and End Results (SEER) database were analyzed. Patients were stratified according to the number of nodes removed. The effect of the extent of lymphadenectomy on overall and cancer-specific survival was examined using multivariable Cox proportional hazards models. Separate analyses were performed for node positive and node negative patients.

RESULTS.

Among 5522 women, 893 (16%) had <10 nodes, 2030 (37%) had 11-20, 1487 (27%) had 21-30 nodes, and 1112 (20%) had >30 nodes removed. Black women, those >65 years of age and those diagnosed later in the study, were less likely to have 10 or more nodes removed (P < .05 for all). Among women with positive lymph nodes, a more extensive lymphadenectomy had no effect on survival (HR = 0.75; 95% CI, 0.47-1.22). For women with negative lymph nodes, a more extensive lymphadenectomy was associated with improved survival. Compared with node negative patients with less than 10 nodes removed, patients with 21-30 nodes removed were 24% (HR = 0.76; 95% CI, 0.53-1.09) less likely to die, whereas those with >30 nodes removed were 37% (HR = 0.64; 95% CI, 0.43-0.96) less likely to die from their tumors.

CONCLUSIONS.

Node negative, early-stage cervical cancer patients who undergo a more extensive lymphadenectomy have an improved survival. Cancer 2011. © 2010 American Cancer Society.

Worldwide cervical cancer is a leading cause of cancer-associated mortality in women. In the United States it's estimated that 11,270 women were diagnosed with cervical cancer in 2009 and that 4070 died from the disease.1 For women with early stage cervical cancer (stage IA2-IIA), treatment consists of either radiation or surgery. Radical hysterectomy with removal of the regional lymph is the procedure of choice for these women.2

The status of the regional lymph nodes is one of the most important prognostic factors for cervical cancer.3-5 In the Gynecologic Oncology Group's (GOG) surgical-pathology study of patients with IB cervical cancer, the 3-year disease free interval was 74% for those with positive nodes versus 86% for node negative patients.3 Other studies have suggested that survival decreases with the number of positive nodes.4

Given the prognostic significance of lymph node metastases, assessment of the regional lymph nodes is a primary goal in women with early-stage cervical cancer. Theoretically, a more thorough lymphadenectomy should increase the likelihood of detecting occult metastatic disease. This principle has been recognized for various tumors and has led to recommendations for more thorough lymphadenectomies for several cancers.6, 7 For example, for patients with stage II-III colon cancer, removal of at least 12 lymph nodes is recommended.6

The benefits of lymphadenectomy may extend beyond merely detecting metastatic disease. For several cancers a survival benefit has been noted among node negative patients who undergo a more extensive lymphadenectomy.8-16 To date, relatively little data exist to describe the possible benefit of a more extensive lymphadenectomy for early-stage cervical cancer. The goal of our study was to examine the influence of the extent of lymphadenectomy on survival for women with early-stage cervical cancer.

MATERIALS AND METHODS

Data from the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) database were analyzed. SEER is a population-based cancer registry that includes approximately 26% of the United States' population.17 SEER is made up of several geographically distinct tumor registries.18 Data from SEER 17 registries were used. All data were publicly available, de-identified, and exempt from Institutional Review Board review.

Women with stage IA2 to IIA cervical cancer with squamous, adenosquamous, and adenocarcinoma histology who underwent radical hysterectomy from 1988 to 2005 were included in the analysis. Only subjects who underwent a lymphadenectomy were included in the study. Clinical and pathologic data including age at diagnosis (<40, 40-65, >65 years), race (white, black, other), marital status, and tumor grade (1, 2, or 3) were collected. Patients in which tumor size was recorded were characterized as <2 cm, 2-4 cm, and >4 cm. Year of diagnosis was classified as 1988-1993, 1994-1999, or 2000-2005. Patients were categorized based on the geographic area of residence at the time of diagnosis: central (Detroit, Iowa, Kentucky, Louisiana, Utah), eastern (Connecticut, New Jersey, Atlanta, rural Georgia), and western (Alaska, California, Hawaii, Los Angeles, New Mexico, San Francisco, San Jose, Seattle) United States. Staging data were derived from the American Joint Cancer Committee (AJCC) staging information and recorded extent of disease codes.18 Use of adjuvant radiotherapy was recorded for each subject.

For each subject the total number of nodes removed was recorded. To explore the effect of the extent of lymphadenectomy on survival, patients were stratified by total node count into the following groups: <10 nodes, 11-20 nodes, 21-30 nodes, and >30 lymph nodes. Separate analyses were performed for lymph node positive and node negative patients.

The vital status of each patient was recorded. Survival was calculated as the number of months from cancer diagnosis to date of death. Overall and cancer-specific survival were calculated for each patient. Subjects alive at last follow-up were censored. Frequency distributions between categorical variables were compared using chi-square test. Logistic regression models were developed to describe predictors of removal of a given number of nodes (>10, >20, and >30 nodes). In the logistic regression analysis, we modeled all the categories of each variable, controlling for all other variables in the model. Cox proportional hazards models were developed to examine cancer-specific survival. In Cox proportional hazards analyses, we modeled the cancer-specific mortality hazard ratios comparing patients according to number of nodes removed, controlling for the other predictive variables. Separate Cox models were generated for node positive and node negative patients. All analyses were performed with SAS version 9.2 (SAS Institute Inc., Cary, North Carolina).

RESULTS

A total of 5522 patients who met the inclusion criteria were identified. Table 1 displays the clinical and demographic characteristics of the cohort, stratified by number of nodes removed. Overall, 893 (16%) women had <10 nodes removed, 2030 (37%) had 11-20 nodes, 1487 (27%) had 21-30 nodes, and 1112 (20%) had >30 nodes removed.

Table 1. Association Between Number of Lymph Nodes Removed and Clinical Variables
 ≤10 Nodes11-20 Nodes21-30 Nodes>30 NodesP
No. (%)No. (%)No. (%)No. (%)
 893 (16.1)2030 (36.8)1487 (26.9)1112 (20.1) 
Age at diagnosis, y    <.001
 <40329 (36.8)811 (40.0)580 (39.0)425 (38.2) 
 40-65474 (53.1)1060 (52.2)816 (54.9)628 (56.5) 
 >6590 (10.1)159 (7.8)91 (6.1)59 (5.3) 
Race    .04
 White693 (77.6)1600 (78.8)1191 (80.1)901 (81.0) 
 Black96 (10.8)191 (9.4)115 (7.7)75 (6.7) 
 Other104 (11.7)239 (11.8)181 (12.2)136 (12.2) 
Year of diagnosis    <.0001
 1988-1993177 (19.8)457 (22.5)373 (25.1)307 (27.6) 
 1994-1999235 (26.3)625 (30.8)461 (31.0)378 (34.0) 
 2000-2005481 (53.9)948 (46.7)653 (43.9)427 (38.4) 
SEER registry    <.0001
 Central202 (22.6)464 (22.9)293 (19.7)231 (20.8) 
 Eastern192 (21.5)424 (20.9)249 (16.8)122 (11.0) 
 Western499 (55.9)1142 (56.3)945 (63.6)759 (68.3) 
Marital status    .28
 Married501 (56.1)1110 (54.7)811 (54.5)654 (58.8) 
 Single365 (40.9)858 (42.3)639 (43.0)427 (38.4) 
 Unknown27 (3.0)62 (3.1)37 (2.5)31 (2.8) 
Histology    .79
 Squamous618 (69.2)1374 (67.7)1028 (69.1)783 (70.4) 
 Adenocarcinoma207 (23.2)480 (23.7)339 (22.8)244 (21.9) 
 Adenosquamous68 (7.6)176 (8.7)120 (8.1)85 (7.6) 
Grade    .79
 185 (9.5)205 (10.1)136 (9.2)102 (9.2) 
 2323 (36.2)708 (34.9)544 (36.6)393 (35.3) 
 3306 (34.3)721 (35.5)526 (35.4)420 (37.8) 
 Unknown179 (20.0)396 (19.5)281 (18.9)197 (17.7) 
Tumor size    .004
 <2.0 cm187 (20.9)418 (20.6)294 (19.8)210 (18.9) 
 2.0-4.0 cm196 (22.0)455 (22.4)372 (25.0)290 (26.1) 
 >4.0 cm65 (7.3)164 (8.1)149 (10.0)120 (10.8) 
 Unknown445 (49.8)993 (48.9)672 (45.2)492 (44.2) 
Radiation    .77
 No634 (71.0)1475 (72.7)1072 (72.1)811 (72.9) 
 Yes259 (29.0)555 (27.3)415 (27.9)301 (27.1) 
 Unknown     
Positive Lymph nodes    .77
 No762 (85.3)1709 (84.2)1251 (84.1)926 (83.3) 
 Yes131 (14.7)320 (15.8)236 (15.9)186 (16.7) 
Stage    <.001
 IA2127 (14.2)269 (13.3)160 (10.8)97 (8.7) 
 IB1371 (41.6)886 (43.7)657 (44.2)485 (43.6) 
 IB2100 (11.2)237 (11.7)210 (14.1)170 (15.3) 
 IBNOS236 (26.4)539 (26.6)383 (25.8)297 (26.7) 
 IIA59 (6.6)99 (4.9)77 (5.2)63 (5.7) 

A multivariable logistic regression model revealed that women with stage IB1 tumors were more likely than those with other stage tumors to have 10 or more nodes removed (OR = 1.41; 95% CI, 1.06-1.88) (Table 2). In contrast, women >65 years of age—of black race, diagnosed later in the study— and those residing in the central or eastern United States were less likely to have 10 or more lymph nodes removed (P < .05 for all). Similar trends were noted when we examined removal of 30 or more nodes: women >65 years of age (OR = 0.71; 95% CI, 0.54-0.95), those residing in central (OR = 0.74; 95% CI, 0.63-0.88) or eastern (OR = 0.45; 95% CI, 0.37-0.55) U.S., and patients diagnosed between 2000-2005 (OR = 0.64; 95% CI, 0.54-0.76) were all less likely to have 30 or more nodes removed. Patients with stage IB and IIA tumors were more likely to undergo an extensive lymphadenectomy than those with stage IA2 disease.

Table 2. Multivariate Logistic Regression Models of Factors Associated With Extent of Lymphadenectomy
 Removal of >10 NodesRemoval of >20 NodesRemoval of >30 Nodes
Age at diagnosis, y
 <40ReferentReferentReferent
 40-650.92 (0.77-1.09)1.05 (0.94-1.18)1.04 (0.91-1.20)
 >650.65 (0.48-0.87)0.65 (0.52-0.81)0.71 (0.54-0.95)
Race
 WhiteReferentReferentReferent
 Black0.73 (0.56-0.96)0.74 (0.61-0.91)0.80 (0.62-1.03)
 Other0.91 (0.71-1.17)0.87 (0.73-1.03)0.88 (0.72-1.08)
Year of diagnosis
 1988-1993ReferentReferentReferent
 1994-19991.03 (0.82-1.29)0.87 (0.75-1.00)0.93 (0.79-1.11)
 2000-20050.71 (0.57-0.87)0.65 (0.56-0.75)0.64 (0.54-0.76)
SEER registry
 WesternReferentReferentReferent
 Eastern0.75 (0.61-0.92)0.59 (0.51-0.69)0.45 (0.37-0.55)
 Central0.80 (0.65-0.97)0.71 (0.62-0.82)0.74 (0.63-0.88)
Marital status
 MarriedReferentReferentReferent
 Single1.14 (0.97-1.34)0.97 (0.86-1.08)0.89 (0.78-1.02)
Histology
 SquamousReferentReferentReferent
 Adenocarcinoma0.94 (0.70-1.27)1.06 (0.87-1.29)1.14 (0.89-1.45)
 Adenosquamous1.00 (0.71-1.40)1.02 (0.81-1.28)1.07 (0.81-1.42)
Grade
 1ReferentReferentReferent
 20.96 (0.71-1.30)1.08 (0.88-1.32)1.04 (0.81-1.33)
 31.01 (0.74-1.38)1.06 (0.85-1.31)1.11 (0.84-1.43)
Tumor size
 <2.0 cmReferentReferentReferent
 2.0-4.0 cm1.10 (0.85-1.42)1.13 (0.95-1.34)1.12 (0.92-1.38)
 >4.0 cm1.40 (0.88-2.23)1.26 (0.92-1.71)1.22 (0.85-1.76)
Radiation
 NoReferentReferentReferent
 Yes0.81 (0.65-1.00)0.91 (0.78-1.05)0.79 (0.65-0.95)
Stage
 IA2ReferentReferentReferent
 IB11.41 (1.06-1.88)1.43 (1.16-1.77)1.44 (1.10-1.88)
 IB21.41 (0.93-2.14)1.53 (1.13-2.05)1.62 (1.12-2.33)
 IBNOS1.24 (0.94-1.63)1.39 (1.14-1.69)1.40 (1.09-1.81)
 IIA0.96 (0.64-1.44)1.33 (0.98-1.81)1.48 (1.02-2.15)

Cox proportional hazards models were constructed to examine the effect of the extent of lymphadenectomy on cancer-specific survival (Table 3). When the entire cohort was examined, those patients who had >30 nodes removed were 29% (HR = 0.71; 95% CI, 0.53-0.96) less likely to die from cervical cancer than patients with 10 or fewer nodes removed. Among women with metastasis to the pelvic lymph nodes there was no association between extent of lymphadenectomy and survival. In contrast, for node negative patients there was a correlation between extent of lymphadenectomy and cancer-specific survival. Node negative patients with >30 nodes removed were 36% (HR = 0.64; 95% CI, 0.43-0.96) less likely to die from their cancer than patients with 10 or fewer nodes removed.

Table 3. Cox Proportional Hazards Models of Factors Associated With Cancer-Specific Survival for Women With Early Stage Cervical Cancer
 All Patients n = 5522Lymph Node Positive n = 873Lymph Node Negative n = 4648
Lymph nodes removed
 ≤10ReferentReferentReferent
 11-200.78 (0.60-1.01)0.71 (0.46-1.10)0.79 (0.56-1.10)
 21-300.83 (0.63-1.10)0.89 (0.57-1.39)0.76 (0.53-1.09)
 >300.71 (0.53-0.96)0.75 (0.47-1.22)0.64 (0.43-0.96)
Age at diagnosis, y
 <40ReferentReferentReferent
 40-650.89 (0.73-1.09)0.92 (0.67-1.25)0.90 (0.70-1.17)
 >651.53 (1.12-2.10)1.30 (0.76-2.20)1.64 (1.10-2.44)
Race
 WhiteReferentReferentReferent
 Black1.57 (1.19-2.09)1.44 (0.92-2.25)1.66 (1.15-2.40)
 Other1.16 (0.88-1.54)1.42 (0.93-2.17)1.00 (0.68-1.45)
Year of diagnosis
 1988-1993ReferentReferentReferent
 1994-19990.85 (0.68-1.06)0.89 (0.62-1.28)0.89 (0.67-1.19)
 2000-20050.70 (0.54-0.90)0.70 (0.47-1.04)0.73 (0.52-1.02)
SEER registry
 WesternReferentReferentReferent
 Eastern1.00 (0.77-1.29)0.91 (0.73-11.61)0.98 (0.70-1.36)
 Central1.09 (0.86-1.38)1.08 (0.73-1.61)1.11 (0.82-1.50)
Marital status
 MarriedReferentReferentReferent
 Single0.96 (0.79-1.16)1.00 (0.74-1.35)0.88 (0.69-1.13)
Histology
 SquamousReferentReferentReferent
 Adenocarcinoma1.64 (1.30-2.07)2.37 (1.66-3.39)1.31 (0.96-1.79)
 Adenosquamous1.62 (1.23-2.14)1.76 (1.16-2.69)1.49 (1.02-2.16)
Grade
 1ReferentReferentReferent
 22.17 (1.28-3.67)1.19 (0.52-2.71)2.84 (1.41-5.71)
 33.17 (1.87-5.37)2.11 (0.94-4.72)3.65 (1.81-7.37)
Tumor size
 <2.0 cmReferentReferentReferent
 2.0-4.0 cm1.46 (1.06-2.00)1.25 (0.71-2.20)1.57 (1.07-2.31)
 >4.0 cm1.56 (1.01-2.43)1.68 (0.83-3.40)1.40 (0.78-2.52)
Radiation
 NoReferentReferentReferent
 Yes1.80 (1.43-2.26)0.95 (0.65-1.39)2.12 (1.62-2.76)
Stage
 IA2ReferentReferentReferent
 IB13.00 (1.45-6.19)2.39 (0.31-18.12)2.17 (0.96-4.92)
 IB24.98 (2.34-10.62)2.68 (0.34-21.01)4.81 (2.00-11.60)
 IBNOS2.71 (1.39-5.29)2.01 (0.26-15.37)2.55 (1.22-5.32)
 IIA5.56 (2.68-11.50)4.53 (0.59-34.87)3.83 (1.66-8.85)

Table 4 displays the 5-year survival of node negative patients stratified by stage and number of nodes removed. Five-year survival ranged from 94%-99% for stage IA2 patients, 89%-93% for IB tumors, and 78%-85% for those with stage IIA neoplasms. We examined the association between number of lymph nodes removed and number of nodal metastases identified (Fig. 1). Removal of up to 30 nodes identifies 79% of women with nodal metastases, whereas resection of up to 35 nodes identifies 85% of patients with nodal disease. The yield of detecting nodal metastases decreases after removal of 35 nodes.

Figure 1.

Percentage of patients with positive lymph nodes detected from the cumulative number of lymph nodes removed.

Table 4. Five-Year Survival Among Node Negative Patients Stratified by Stage and Number of Lymph Nodes Removed
 ≤10 Nodes11-20 Nodes21-30 Nodes>30 Nodes
No. (%)95% CINo. (%)95% CINo. (%)95% CINo. (%)95% CI
  1. CI indicates confidence interval.

IA2124 (94)88-97263 (99)96-100159 (98)94-9993 (99)92-100
IB603 (89)86-921383 (90)88-921042 (90)88-92792 (93)91-95
IIA35 (79)58-90%63 (85)71-9350 (85)70-9341 (80)63-90

DISCUSSION

Our findings suggest that lymphadenectomy may have a therapeutic role in women with early-stage cervical cancer. Among women with stage IA2-IIA tumors we noted that survival was improved in those who underwent a more extensive lymph node dissection. These effects were most pronounced among node negative women in whom cancer-related mortality was reduced by more than 35% when greater than 30 nodes were removed.

The prognostic importance of the extent of lymphadenectomy in patients with pathologically negative lymph nodes has been demonstrated for several solid tumors.8-16 In an analysis of 442 patients with nonsmall cell lung cancer, removal of an increased number of lymph nodes was associated with improved survival.11 Likewise, a systematic review of nodal evaluation in patients with colon cancer reported that the number of nodes removed was associated with survival for stage II and III colon cancer.9 Removal of a higher number of nodes may increase the chance of detecting metastatic disease or, alternatively, allow for the removal of micrometastases not readily detectable by standard histologic methods. In our cohort removal of a higher number of lymph nodes improved survival for women with negative lymph nodes, but had no apparent effect on survival in patients with pathologically positive nodes.

The potential therapeutic role of lymphadenectomy for women with early-stage cervical cancer has received relatively little attention. In a single institution review of 467 patients with early-stage cervical cancer, Pieterse and coworkers found no association between the number of nodes removed and survival for node negative patients, but noted improved survival in node positive patients who had a more thorough lymphadenectomy.19 In a SEER analysis of stage I-IV cervical cancer, survival was improved for patients who had more nodes removed. These findings were true for all stages, and in both node positive and node negative women. We limited our analysis to a much more homogenous group of patients, those with stage I-II tumors who underwent definitive surgical management. The report from Rossi and colleagues included a diverse group of patients, including those who underwent surgery for definitive treatment and advanced stage patients undergoing staging evaluation.19

Given the morbidity of pelvic and para-aortic lymphadenectomy, there has been interest in less invasive modalities for detecting nodal disease in women with early-stage tumors.2 Anatomic imaging modalities, such as computed tomography and magnetic resonance imaging, are only moderately sensitive and specific for the detection of nodal disease.20 Positron emission tomography is highly specific for nodal disease, but is only moderately sensitive, particularly for patients with early-stage tumors and small volume nodal metastases.21-23 More recently, interest has focused on sentinel node dissection to minimize the late effects of lymphadenectomy.24, 25 A multicenter European trial of sentinel lymphadenectomy noted an overall sensitivity of only 77%, lower than the predefined noninferiority cutoff.25 The possible therapeutic benefits of an extensive lymphadenectomy must be weighed carefully against the morbidity of the procedure.

We noted several disparities in performance of an extensive lymphadenectomy. In our cohort black women were 27% less likely than their white counterparts to have 10 or more nodes removed. Several studies have documented that black women are less likely to receive aggressive primary therapy.26-28 Our group previously reported that black women were 32% less likely than white patients to undergo radical hysterectomy.26 Similarly, there appeared to be a trend toward performing a less thorough lymphadenectomy in the later years of study. Women treated after 2000 were 29% less likely than patients treated before 1993 to have more than 10 nodes dissected. While the reasons for this changing pattern of care are not intuitively clear, detection of smaller tumors and a trend toward less radical surgery may have influenced these findings.

We recognize several important limitations in our analysis. As with any observational study, it is impossible to determine patient and physician preferences that may have impacted the extent of lymphadenectomy performed. SEER allows for stratification of important confounding variables including age, but lacks data on medical comorbidities. It is likely that women with obvious metastatic nodal disease detected intraoperatively were more likely to undergo selective nodal sampling or debulking, as opposed to a more thorough staging lymphadenectomy. To overcome this limitation we performed separate analyses for node positive and node negative patients. During the years of our study, the paradigm for adjuvant treatment of node positive and high-risk early stage cervical cancer shifted from radiation to combination chemoradiation.29 SEER lacks data on use of chemotherapy, which may have impacted our survival results. Finally, SEER lacks data on the patterns, location, and treatment of recurrences. As such, our analysis was limited to survival only.

Our study raises several important questions for the management of women with early-stage cervical cancer. Given the survival advantage we noted, the decreased morbidity of limited nodal sampling and use of noninvasive imaging technologies in lieu of lymphadenectomy must be weighed carefully against the possible survival benefit that a more thorough lymphadenectomy confers. Second, how many lymph nodes define an adequate nodal dissection for cervical cancer? Nodal counts represent an important measure of surgical quality for several tumor types. The American Society of Clinical Oncology (ASCO) recommends removal and examination of 12 or more lymph nodes for patients with colon cancer, whereas the GOG suggests removal of 10 or more nodes for patients with endometrial cancer.6, 7 It is difficult to draw definitive conclusions on how many nodes constitute a complete lymphadenectomy from our data. We did note that the percentage of patients with positive nodes detected followed a steep curve from removal of 10 to 25 nodes.

In conclusion, our findings suggest that the extent of lymphadenectomy performed for women with early-stage cervical cancer influences survival. Among node-negative women, survival is improved when a greater number of lymph nodes are removed. Further work is required to determine the number of nodes needed to confer this survival advantage and make evidence-based recommendations on what defines an adequate lymphadenectomy.

CONFLICT OF INTEREST DISCLOSURES

The authors made no disclosures.

Ancillary