Cervical cancer is the second most common type of cancer in women worldwide, with an estimated age standardized incidence rate of 15.2 per 100,000 women and a mortality rate of 7.8 per 100,000 women. In the European Union, it is the seventh most common type of cancer in women with an age standardized incidence rate of 9.0 per 100,000 women and a mortality rate of 3.0 per 100,000 women.1 Cervical cancer is clinically staged according to definitions set by the International Federation of Gynecology and Obstetrics (FIGO).2 As opposed to the staging of other gynecological tumors, lymph node metastases are not included in the staging of cervical cancer. However, it is important to assess the lymph node status in this disease because it is a negative prognostic factor for survival3 and it determines the choice of initial therapy, as well as the need for adjuvant treatment.4
Cervical cancer is known to spread to the pelvic lymphatic system via the first draining lymph node, the sentinel lymph node (SN).5 If this node is tumor free, then the other draining lymph nodes (non-SNs [nSNs]) are assumed not to contain tumor. Currently, the criterion standard for assessing the nodal status in cervical cancer is systematic pelvic lymph node dissection (LND).6 Such an extensive lymphadenectomy leads to lymphocyst formation in approximately 20% and to lymphedema in approximately 10% of the patients with FIGO stage IB to IIA disease.7–9 Morbidity rates have decreased because of implementation of laparoscopic lymphadenectomy.10 To further minimize these complications, the SN biopsy is currently being evaluated for adoption as the standard of care in early-stage cervical cancer.11,12 This procedure entails detection and excision of the SN after submucosal injection of a radioisotope tracer and/or blue dye around the primary tumor.13 Essential in this procedure is the optimal histopathologic evaluation of the SN by serial sectioning and immunohistochemistry.14,15 This technique assures reliable detection of low-volume disease (LVD; micrometastasis [0.2-2 mm] or isolated tumor cells [ITC; <0.2 mm]).16 Compared with pelvic LND, SN biopsy increases the detection rate of metastases up to 2.8-fold.17
Before pelvic LND can be abandoned in favor of performing SN biopsy alone, the sensitivity of the latter to detect metastases needs to be similar (or higher) than that of pelvic LND. Furthermore, it needs to be clarified whether the assessment of nodal status by pelvic LND, as well as its extent, is solely diagnostic or whether it also affects survival.18–20 With SN ultrastaging, information is now provided on LVD that hitherto was not available. This new information can be used to study the effect of extended surgical treatment of pelvic lymph nodes on the outcome of the disease.
Therefore, in this study, we aimed to clarify whether the extent of pelvic LND affects survival in patients with a negative SN, in patients with LVD, and in patients with macrometastasis in the SN.
MATERIALS AND METHODS
Our study population consisted of 645 patients from 8 centers (Ostrava and Prague in Czech Republic, Amsterdam and Utrecht in The Netherlands, New York in the United States, Paris and Toulouse in France, and Krakow in Poland). In this study population of 645 patients, we previously described the clinical significance of micrometastasis in the lymph nodes and the false-negative rate of 2.8% (whole group) and 1.3% for patients with optimal bilateral mapping.21,22 Patients with FIGO stage IA to IIB cervical cancer of squamous, adeno, or adenosquamous histologic type without clinical or radiologic signs of lymphadenopathy were included. In patients where no SN ultrastaging was performed and/or survival end points were not adequately documented, they were excluded from the study.
Therapeutic Procedures and Pathologic Evaluation
Radioisotope and blue dye were injected preoperatively and intraoperatively, respectively, around the primary tumor. Sentinel lymph nodes were detected at laparoscopy or laparotomy by visual inspection and a gamma probe. Fresh frozen analysis of the excised SN with subsequent paraffin embedding and pathologic ultrastaging was performed as previously described.21 The total surgical specimen was palpated for lymph nodes. Nodes were dissected out and were counted manually. If additional smaller nodes were seen on microscopy, they were added to the total. All SN negative for metastasis on the initial routine section stained by hematoxylin and eosin (H&E) were further examined according to the pathologic ultrastaging protocol of the respective institutions. The entire node was cut at regular intervals, which varied at individual centers between 150 and 500 μm; in 7 of the 8 centers and 98% of the patients, the intervals measured 250 μm or less. Three consecutive sections (5-μm thick) were obtained at each level. The first slide was stained with H&E, whereas the second was used for immunohistochemical staining for cytokeratin. Pelvic nSNs were processed identically in all institutions by single section of each node examined by a routine H&E staining.
Lymph node involvement was defined as ITC or clusters (smaller than 0.2 mm in greatest diameter), micrometastasis (smaller than 2 mm in greatest diameter), or macrometastasis (equal to or larger than 2 mm).16
After the SN biopsy, all patients underwent full pelvic LND. Consequently, simple hysterectomy (N = 3), radical hysterectomy (N = 532), simple trachelectomy (N = 22), or radical trachelectomy (N = 88) was performed. The surgical specimens of the latter procedures were evaluated according to common histopathologic practice. Adjuvant therapy (radiotherapy, chemotherapy, or both) was administered according to national or institutional guidelines in 213 (33.0%) of the 645 patients. Adjuvant therapy was administered to 116 (85.3%) of patients with macrometastasis, 38 (82.6%) patients with micrometastasis, 13 (52%) patients with ITC, and 46 (10.5%) patients with negative lymph nodes (Table 1). In the patients with negative SN, 8 (7.8%) of the patients with 16 or less nodes and 52 (14.7%) of the patients with more than 16 nodes removed received adjuvant therapy (P = 0.062). Among all patients with LVD in the SN, 5 (45.5%) patients with 16 or less nodes and 67 (80.7%) of the patients with more than 16 nodes removed received adjuvant therapy (P = 0.026). In patients with macrometastasis in the SN, 13 (76.5%) of the patients with 16 or less nodes and 68 (87.2%) of the patients with more than 16 nodes removed received adjuvant therapy (P = 0.283).
Standard summary statistics were used to describe primary data, that is, frequency tables and median supplied with fifth to 95th percentile range. Maximum likelihood (ML) and chi-square (χ2) testing was performed to compare categorical variables, and Kruskal-Wallis followed by Mann-Whitney U testing was applied for mutual comparisons of variants in continuous variables. Kaplan-Meier survival probability estimates with log-rank testing were used to describe and compare variants in time-to-event end points, that is, overall survival (OS) and relapse-free survival (RFS). Time-to-event end point was calculated from time of surgery. We were not able to correct for start and duration of adjuvant therapy because of unavailability. Univariate and multivariate proportional hazards Cox regression models were applied to quantify the association of potential risk factors and survival. First, estimates of hazards ratio (HR; with 95% confidence intervals [CIs]) were tested using Wald χ2 test. Subsequently, parameters with potential risk power (P < 0.10 in univariate Cox regression) were subjected to stepwise selection algorithm in multivariate Cox regression. For all statistical tests, a 2-tailed P value of less than 0.05 was considered significant. Statistical power to detect differences within groups was limited, mainly in the stratified analysis.
Characteristics of Patients and Tumors
Patient and tumor characteristics were stratified according to the result of the SN ultrastaging (Table 1). With increasing FIGO stage, there was a significant increase in the size of SN metastases (ML-χ2, P < 0.001). Similarly, vaginal and parametrial involvement and metastasis in the nSNs were positively associated with the size of metastases in SN (P < 0.001). No significant association was found with age, histologic subtype, or the presence of lymphovascular space invasion (LVSI).
Factors Associated With Lymph Node Involvement and Survival
The number of positive nSNs was a significant predictor for the development of recurrence and the risk of death (Table 2), both as a continuous variable and when analyzed in categories. The HR for recurrence and death was 8.80 (95% CI, 3.10–24.96) and 8.89 (95% CI, 2.02–39.37), respectively, if more than 5 lymph nodes were positive.
Clinical Impact of the Number of Removed Pelvic Lymph Nodes
Receiver operating characteristic analysis was performed to establish a cutoff for number of removed lymph nodes (Table 3). On the basis of these small numbers in our study, a cut of 17 is defined (P = 0.043). We confirmed this finding by logistic regression analysis.
To relate the number of removed nodes to outcome, Kaplan-Meier analysis was performed with OS and RFS as end points. No statistically significant difference in RFS or OS in relation to the number of nSNs removed was observed among patients with FIGO stage IA to IB1 disease (Fig. 1A and B). However, in patients with FIGO stage IB2 to II, both RFS (P = 0.032) and OS (P = 0.014) were significantly better in patients with more than 16 nodes removed (Fig. 1C and D) than in patients with 16 or less nSNs removed.
The previously mentioned findings were tested using univariate and multivariate Cox proportional hazards regression analysis to exclude a possible confounding effect of other parameters (Table 4). Both models confirmed that removing more than 16 nSNs significantly reduced the risk of recurrence and the risk of death in patients with FIGO stage IB2 to II disease. Adjuvant treatment was used as a covariate in multivariate models, but no significant multivariate-adjusted effect on the time-to-event end points was found.
To determine whether removing more than 16 nSNs has a similar effect on survival in patients with or without metastatic lymph nodes, we stratified patients for SN status (Fig. 2). Three categories were defined; these are as follows: SN negative (n = 456), LVD (including ITC and micrometastasis, n = 94), and macrometastasis (n = 95). We showed that only among patients with LVD in the SN that the OS was significantly better (P = 0.046) if more than 16 nSNs were removed. No statistically significant differences were observed among patients with negative SN or macrometastasis in the SN. There were too few patients with LVD to further stratify the results according to ITC or micrometastases.
Of the 94 patients with LVD in the SN, 71 (75.5%) had no metastasis detected in the nSNs. The Kaplan-Meier analysis was repeated for this subpopulation of women (Fig. 3) and showed a trend toward better OS in women with more than 16 nSNs removed (P = 0.055) than in patients in whom less than 16 nSNs were removed.
In this multicenter cohort study, we studied 645 patients who had undergone an SN biopsy with pathologic ultrastaging and subsequent pelvic LND. This is the largest multicenter retrospective study of its kind to date, which provided sufficient numbers to analyze the effect of LND after SN biopsy in the subset of patients with LVD. This study is limited because of its retrospective study design. There was no randomization for number of removed nodes, treatment was not randomized, there were small differences in sentinel node analysis techniques within centers, and some interesting data (location of nodes) were not available.
We showed that known risk factors (FIGO stage, vaginal involvement, and parametrial involvement) were not only related to the occurrence of lymph node metastasis but also to the size of metastases. Furthermore, in this study, we confirmed the finding that the number of positive nSNs is associated with survival and recurrence.3,23,24
For all stages, besides having a diagnostic value, we can conclude that systematic pelvic lymphadenectomy performed in addition to SN was only associated with better survival for patients with LVD in the SN. Removing more than 16 nSNs led to a better survival in patients with LVD in the SN.
To assess whether this effect was due to increased detection of lymph node metastases or this was a true therapeutic effect, we performed survival analysis in patients with LVD in the SN and negative nSNs. We detected a trend toward better OS in the latter group (P = 0.055). This possible therapeutic effect could be explained by removing additional LVD in nSNs, which is not detected by routine pathologic assessment. Unfortunately, we were unable to stratify for both FIGO stage and SN status because of the limited number of events.
Previous studies have shown a therapeutic impact of LND in patients with cervical cancer who underwent full pelvic LND without SN biopsy.25,26 Excising at least 15 lymph nodes was associated with better survival (P = 0.01) compared with patients in whom less than 15 lymph nodes were removed.27 Whether this effect on survival is related to lymph node status has thus far only been investigated by 2 research groups who have provided conflicting data. Kenter et al18 showed that completing LND resulted in a longer DFS in 63 patients with lymph node–positive cervical cancer. In addition, they showed a longer DFS in 136 patients with lymph node positive, whereas completing LND in 331 patients with lymph node negative had no effect on survival.19 Contrastingly, Shah et al20 analyzed the SEER database and showed no effect of completing LND in 873 patients with lymph node–positive FIGO stage IA2 to IIA cervical cancer. However, survival was improved in 4648 patients with lymph node negative with more than 20 lymph nodes removed.
Our finding that completion of LND showed better survival in case of LVD and not in patients with LND negative seems to contradict the outcome of the SEER study. However, SN biopsy, as used in our analysis, provides a more sensitive procedure to detect metastases than what was used in the SEER study. The beneficial effect of full LND in patients with node negative might therefore be due to treatment of LVD, who were analyzed in the SEER study as patients with node negative. In contrast with 2 previous studies18,19 and in line with the SEER data, we could not find evidence that performing LND showed better survival in patients with macrometastatic lymph nodes. This difference may be explained by the fact that we excluded patients with evidently involved nodes, whereas the 2 studies that found a beneficial effect of LND also included bulky enlarged nodes.25,26
Whether our results warrant clinical implementation of full LND as a second operation after the sentinel node procedure should be further validated, preferably in a randomized controlled trial. In such prospective study, it should be assessed whether patients in whom an SN procedure is performed and who are found to have LVD do indeed benefit from additional lymphadenectomy and/or radiotherapy, as this retrospective cohort analysis suggests.
The authors thank Jonas van de Lande (VU University Medical Center in Amsterdam, The Netherlands, currently Kennemergasthuis, Haarlem), Jan Lacheta (General University Hospital in Prague, Czech Republic), and Anne-Claire Sans (Institut Claudius Regaud in Toulouse, France) for data acquisition.
1. Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. GLOBOCAN 2008 v1.2, Cancer Incidence and Mortality Worldwide: IARC Cancer Base No.10 [Internet].Lyon, France: International Agency for Research on Cancer; 2010.Available at: http://globocan.iarc.fr
, accessed on 21/november/. http://globocan.iarc.fr
. 2011. Ref Type: Internet Communication.
2. Pecorelli S. Revised FIGO staging for carcinoma of the vulva, cervix, and endometrium. Int J Gynaecol Obstet. 2010; 108: 176.
3. Macdonald OK, Chen J, Dodson M, et al. Prognostic significance of histology and positive lymph node involvement following radical hysterectomy in carcinoma of the cervix. Am J Clin Oncol. 2009; 32: 411–416.
4. Moore DH. Cervical cancer. Obstet Gynecol. 2006; 107: 1152–1161.
5. Benedetti-Panici P, Maneschi F, Scambia G, et al. Lymphatic spread of cervical cancer: an anatomical and pathological study based on 225 radical hysterectomies with systematic pelvic and aortic lymphadenectomy. Gynecol Oncol. 1996; 62: 19–24.
6. Cibula D, Abu-Rustum NR. Pelvic lymphadenectomy in cervical cancer–surgical anatomy and proposal for a new classification system. Gynecol Oncol. 2010; 116: 33–37.
7. Conte M, Panici PB, Guariglia L, et al. Pelvic lymphocele following radical para-aortic and pelvic lymphadenectomy for cervical carcinoma: incidence rate and percutaneous management. Obstet Gynecol. 1990; 76: 268–271.
8. Magrina JF. Primary surgery for stage IB-IIA cervical cancer, including short-term and long-term morbidity and treatment in pregnancy. J Natl Cancer Inst Monogr. 1996; 21: 53–59.
9. Ohba Y, Todo Y, Kobayashi N, et al. Risk factors for lower-limb lymphedema after surgery for cervical cancer. Int J Clin Oncol. 2011; 16: 238–243.
10. Chen Y, Xu H, Li Y, et al. The outcome of laparoscopic radical hysterectomy and lymphadenectomy for cervical cancer: a prospective analysis of 295 patients. Ann Surg Oncol. 2008; 15: 2847–2855.
11. Buist MR, Pijpers RJ, van Lingen A, et al. Laparoscopic detection of sentinel lymph nodes followed by lymph node dissection
in patients with early stage cervical cancer. Gynecol Oncol. 2003; 90: 290–296.
12. Verheijen RH, Pijpers R, van Diest PJ, et al. Sentinel node detection in cervical cancer. Obstet Gynecol. 2000; 96: 135–138.
13. van de Lande J, Torrenga B, Raijmakers PG, et al. Sentinel lymph node
detection in early stage uterine cervix carcinoma: a systematic review. Gynecol Oncol. 2007; 106: 604–613.
14. Altgassen C, Hertel H, Brandstädt A, et al. Multicenter validation study of the sentinel lymph node
concept in cervical cancer: AGO Study Group. J Clin Oncol. 2008; 26: 2943–2951.
15. Gortzak-Uzan L, Jimenez W, Nofech-Mozes S, et al. Sentinel lymph node
biopsy vs. pelvic lymphadenectomy in early stage cervical cancer: is it time to change the gold standard? Gynecol Oncol. 2010; 116: 28–32.
16. Hermanek P, Hutter RV, Sobin LH, et al. International Union Against Cancer. Classification of isolated tumor cells
. Cancer. 1999; 86: 2668–2673.
17. Vicus D, Covens A. Role of sentinel lymph node
biopsy in cervical cancer: pro. Int J Gynecol Cancer. 2010; 20: S34–S36.
18. Kenter GG, Hellebrekers BW, Zwinderman KH, et al. The case for completing the lymphadenectomy when positive lymph nodes are found during radical hysterectomy for cervical carcinoma. Acta Obstet Gynecol Scand. 2000; 79: 72–76.
19. Pieterse QD, Kenter GG, Gaarenstroom KN, et al. The number of pelvic lymph nodes in the quality control and prognosis of radical hysterectomy for the treatment of cervical cancer. Eur J Surg Oncol. 2007; 33: 216–221.
20. Shah M, Lewin SN, Deutsch I, et al. Therapeutic role of lymphadenectomy for cervical cancer. Cancer. 2011; 117: 310–317.
21. Cibula D, Abu-Rustum NR, Dusek L, et al. Prognostic significance of low volume sentinel lymph node
disease in early-stage cervical cancer. Gynecol Oncol. 2012; 124: 496–501.
22. Cibula D, Abu-Rustum NR, Dusek L, et al. Bilateral ultrastaging of sentinel lymph node
in cervical cancer: lowering the false-negative rate and improving the detection of micrometastasis
. Gynecol Oncol. 2012; 127: 462–466.
23. Inoue T, Morita K. The prognostic significance of number of positive nodes in cervical carcinoma stages IB, IIA, and IIB. Cancer. 1990; 65: 1923–1927.
24. van Gorp T, Kruse AJ, Slangen BF, et al. Lymph node density as a surrogate marker for positive lymph nodes. Br J Cancer. 2011; 104: 221–222.
25. Kjorstad KE, Kolbenstvedt A, Strickert T. The value of complete lymphadenectomy in radical treatment of cancer of the cervix, stage IB. Cancer. 1984; 54: 2215–2219.
26. Marnitz S, Köhler C, Roth C, et al. Is there a benefit of pretreatment laparoscopic transperitoneal surgical staging in patients with advanced cervical cancer? Gynecol Oncol. 2005; 99: 536–544.
27. Rossi PJ, Horowitz IR, Johnstone PA, et al. Lymphadenectomy for patients with cervical cancer: is it of value? J Surg Oncol. 2009; 100: 404–406.