Sentinel Lymph Node Mapping vs Systematic Lymphadenectomy for Endometrial Cancer: Surgical Morbidity and Lymphatic Complications

Journal Pre-proof

Sentinel Lymph Node Mapping vs Systematic Lymphadenectomy for Endometrial Cancer: Surgical Morbidity and Lymphatic Complications Guilherme S. Accorsi M.D , Luiza L. Paiva M.D , Ronald Schmidt M.D , Marcelo Vieira MSc , Ricardo Reis PhD , Carlos Andrade MSc PII: DOI: Reference:

S1553-4650(19)30374-7 https://doi.org/10.1016/j.jmig.2019.07.030 JMIG 3909

To appear in:

The Journal of Minimally Invasive Gynecology

Received date: Accepted date:

23 July 2019 24 July 2019

Please cite this article as: Guilherme S. Accorsi M.D , Luiza L. Paiva M.D , Ronald Schmidt M.D , Marcelo Vieira MSc , Ricardo Reis PhD , Carlos Andrade MSc , Sentinel Lymph Node Mapping vs Systematic Lymphadenectomy for Endometrial Cancer: Surgical Morbidity and Lymphatic Complications, The Journal of Minimally Invasive Gynecology (2019), doi: https://doi.org/10.1016/j.jmig.2019.07.030

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Inc. on behalf of AAGL.

Sentinel Lymph Node Mapping vs Systematic Lymphadenectomy for Endometrial Cancer: Surgical Morbidity and Lymphatic Complications.

Authors

Guilherme S. Accorsi, M.D; Luiza L. Paiva M.D; Ronald Schmidt M.D; Marcelo Vieira, MSc; Ricardo Reis, PhD; Carlos Andrade, MSc.

Institution • 1Department Gynecologic Oncology of Barretos Cancer Hospital, Brazil.

Author for correspondence: • Guilherme Spagna Accorsi M.D • Catanduva Cancer Hospital • Rua dos Estudantes 225, Parque Iracema, Catanduva - SP, Brazil • +55(17) 3522 – 7080 • [email protected]

Word count: 2,905 Tables: 9 References: 33

Conflict of Interest:

All the authors declare that they have no conflict of interest.

Institutional review board approval:

The local ethics committee was consulted and approved this article.

1

Precis Sentinel lymph node biopsy does not increase surgical morbidity in simple hysterectomy in endometrial cancer patients.

2

Abstract Study Objective: Sentinel lymph node (SLN) mapping has been proven to accurately stage endometrial cancer (EC). However, there is a lack of studies comparing the incidence of complications between different lymph node approaches in EC. Our objective is to define the complication rates of SLN biopsy in EC patients.

Design: A retrospective cohort study Setting: A tertiary referral hospital Patients: All patients who were surgically treated for endometrial cancer form April 2013 to March 2018

Interventions: We evaluated intraoperative complications and 30-day complications using the Memorial Sloan Lettering Cancer Center’s Surgical Secondary Events Grading System, separating the patients into four groups: group I, hysterectomy (HT); group II, hysterectomy plus sentinel lymph node biopsy (HT+SLN); group III, hysterectomy plus pelvic lymphadenectomy, with or without para-aortic dissection (HT+LND); and group IV, hysterectomy plus lymphadenectomy and sentinel lymph node biopsy (HT+SLN+LND).

Measurements and Main Results: We identified a total of 250 cases. As compared with the HT group, the HT+SLN group did not show any increased risk of complications in terms of intraoperative complications (0 vs 1; p = 1.0) and 30-day complications (8 vs 7; p = 0.782), but surgical time was approximately 20 minutes longer (p = 0.016). Performing LND was associated with a greater risk of 30-day complications (hazard ratio [HR]: 3.11; 95% confidence interval [CI]: 1.62-5.98), intraoperative complications (HR: 14.25; 95% CI: 1.85-19.63), and lower-limb lymphedema (HR: 8.14; 95% CI: 1.01- 65.27).

3

Conclusion: SLN mapping does not increase morbidity in the surgical treatment of EC patients, and compared with comprehensive lymphadenectomy, it has a lower risk of complications. Our findings support the use of the SLN algorithm in EC patients

KEYWORDS: lymph node excision, lymph node dissection, lymph node biopsy, sentinel biopsy, sentinel lymph node, endometrial Neoplasm, Endometrial Carcinoma.

4

Introduction In 1988, the International Federation of Gynecologic and Obstetrics altered the staging of endometrial cancer (EC) from clinical to surgical-pathologic [1] based on the landmark Gynecologic Oncology Group (GOG 33) study [2]. Since then, there has been an ongoing debate about the value of extensive lymphadenectomy and its extent in EC, leading to a diversity of strategies for lymph node assessment [3-5]. Several authors reported a significant impact of extensive lymphadenectomy on overall and recurrence-free survival [6,7]. Even though the results of two randomized trials did not confirm these outcomes [8,9], lymph node involvement is one of the most important prognostic factors in EC, and patients may have a 5-year survival rate as low as 44% to 52% [10]. The risk of pelvic or para-aortic lymph node metastasis in clinical suspected stage I EC occurs in about 10% of patients, rising to roughly 20% with deep myometrial invasion or poorly differentiated tumors [2]. Only surgical staging provides precise pathologic information to correctly triage patients to adjuvant treatment, reducing the risk of inappropriate over- or undertreatment [11]. Performing extensive lymphadenectomy (LND) in EC increases the risk of perioperative complications [12 - 14]. The median operative time has been described to be 50% longer when extensive lymphadenectomy is performed [8]. In a prospective study by the Mayo Clinic compering extensive lymphadenectomy to a simple hysterectomy for low-risk endometrial cancer, the rate of 30-day complications associated with LND was reported to be 37.4% (p < 0.001) [15], and Todo et al. found the rate of lower-limb lymphedema (LLL) to be as high as 37.8% in extensive lymphadenectomy group [16]. Concern about higher morbidity of extensive lymphadenectomy is justified in patients with low-risk EC. Sentinel lymph node (SLN) mapping has emerged as an acceptable alternative to node assessment, a way to identify patients who are candidates to adjuvant therapy without the morbidity of systematic lymphadenectomy. SLN has been reported to accurately stage EC [17-19] without compromising oncologic outcomes [20], and although the rate of complications related to SLN biopsy reported in the literature is low [21], only a few studies have compared SLN morbidity to extensive lymphadenectomy [11,12, 22 - 24]. In a retrospective assessment of EC patients at the Memorial Sloan Kettering Cancer Center (MSKCC), there was a reduction of 10% in the 30-day

5

complication rate after SLN was incorporated [11]. Geppert et al. found a 14-fold decreased risk of LLL when performing SLN biopsy with a lower operative time [12]. Despite the increased use of SLN biopsy in EC, the potential reduction of complications following SLN mapping alone remains unknown. In this study, we sought to compare retrospectively the perioperative morbidity in women with EC undergoing SLN biopsy versus extensive lymphadenectomy. Furthermore, we evaluate the detection rate, sensitivity, and negative predictive value (NPV) of SLN mapping.

6

Materials and methods After

obtaining

approval

from

the

Institutional

Review

Board,

we

retrospectively reviewed the medical records of patients who were surgically treated for EC at Barretos Cancer Hospital from April 2013 to March 2018. Data retrieved from the hospital records included age, body mass index, previous surgeries, FIGO stage, American Society of Anesthesiologists (ASA) classification, surgical data (date, approach, lymph node assessment technique, complications, blood loss, length of time), 30-day complication rate, and pathologic information (histology, grade, myometrial invasion, tumor size, lymph – vascular space invasion pelvic and aortic node metastasis). Patients with synchronous tumors, two-step surgery, non–lymph node metastasis, and vaginal hysterectomy were excluded. For this analysis, patients included in this study were divided into four groups based on lymph node assessment: group I, hysterectomy (HT) only; group II, HT plus SLN mapping (HT+SLN); group III, HT plus pelvic lymphadenectomy, with or without para-aortic LND (HT+LND); and group IV, HT plus SLN and LND (HT+SLN+LND). Through the year of 2016 SLN mapping became a routine procedure for clinically aperient stage I EC, all cases were performed through cervical injection of blue dye (Patent blue - Citopharma®) at the 3 and 9 o’clock positions with 1 cc superficial and 1 cc deep, for a total of 4 cc, before starting pneumoperitoneum. Cervical injection is the standard method used in SLN mapping for EC in our institution, because it is the easiest and most convenient injection site and has been reported to have the highest bilateral detection rate [21]. After completing total hysterectomy, the uterus was sent to frozen section and, based on identification of tumor risk factors (myometrial invasion, tumor size, histology and grade) and clinical judgment pelvic and/or para-aortic lymphadenectomy were indicated for high risk case [4] Specimens were examined by expert gynecologic pathologists. SLN was initially examined by hematoxylin and eosin staining, and if the initial step was negative, the nodes were submitted to ultrastaging. Lymph nodes were considered metastatic if they presented with isolated tumor cells (small tumor cluster or single tumor cells 0.2 mm), micrometastasis (tumor cluster between 0.2 and 2mm), or macrometastasis (tumor clusters >2 mm). 7

Adverse events were classified into intraoperative and postoperative. Intraoperative morbidity was evaluated through blood loss, surgical time, and intraoperative

injuries

(urinary,

vascular,

neurologic,

and

intestinal),

while

postoperative complications were graded according to MSKCC’s Surgical Secondary Events Grading System (Supplementary table 1) within 90 days after surgery, and involved LLL [25]. All data were stored and managed trough REDCap (Research Electronic Data Capture) [26] electronic data capture tools, and analysis was performed using SPSS (version 21) statistical software. Categorical variables, as demographic and clinical characteristics, were presented with absolute frequencies and percentages, and were compered using Fisher’s test. Continuous variables were expressed as mean, median and standard deviation, and were compered through tables using Kruskall-Wallis test. Posteriorly, we used logistic regression to detect an association between complications and the technique chosen, throw the Backward method, with the purpose of exclude not significant statistical variables. To be able to calculate false-negative rate and NPV value, we used the group of patients who underwent bilateral pelvic lymphadenectomy after SLN mapping detected in both hemipelvisses. The test was considered a true positive if a least one SLN had metastatic disease, and as a false negative if the SLNs were free of metastatic disease but non-SLNs showed metastases. Sensitivity was calculated by dividing the number of true-positive patients by the total number of patients with pelvic lymph node metastases. The false-negative rate was estimated as the proportion of false-negatives among the patients with nonsentinel lymph node metastasis. NPV was determined by dividing the number of true-negatives by the number of patients with negative SLN, and

p value <0.05 was considered statistically significant.

8

Results We identified a total of 250 women: 54 in the group I, 61 in group II, 89 in group III, and 46 in group IV. Patients and clinicopathologic characteristics are presented in Table 1. Patients who underwent the SLN (Groups II and IV) technique were more likely to have a minimally invasive surgery (laparoscopy or robotic) compere to groups I and III, and only three minimally invasive surgeries were converted to laparotomy: 1 patient in the group I and 2 patients in the group III, and all 3 were initially laparoscopic. The cohort of women who underwent LND, groups III and IV, were more likely to have type II histology (p < 0.001), greater than 50% myometrial invasion (p < 0.001), lymph vascular invasion (p = 0.019), and grade 3 tumor (p < 0.001). Surgical times were longer in the groups III and IV — 370 (80 – 600) minutes and 240 (125 – 420) minutes (p < 0.001) respectively—and the patients in group III had a higher mean blood loos than the patients from group II (100 vs 20 cc; p < 0.001). Intraoperative complications were also higher in patients who had extensive lymphadenectomy: 11.2% in the group III and 13.0% in the group IV (p = 0.005). Of the total of 16 complications during surgery, 12 (75%) were vascular, and all of them were reported only among those who underwent nodal assessment; LLL was found only in patients from group III (p = 0.001). Postoperative complications (30-day) compared among the 4 groups, I vs II vs III vs IV, were also more frequent in the groups with lymphadenectomy, respectively: 14.8% vs 11.5% vs 38.2% vs 19.6% (p < 0.01; Table 2). We additionally compared group I vs the group II, these groups showed no differences in blood loss (median: 35 vs 20 cc; p = 0.386), intraoperative complications (0 vs 1; p = 1.0), or 30-day complications (8 vs 7; p = 0.782), and no LLL was reported. However, surgical time was significantly longer in group undergoing SLN mapping (median: 135 vs 152 minutes; p = 0.016), which added approximately 20 minutes to the procedure. There was only one case of anaphylaxis due to application of blue dye during SLN mapping (Table 3). And it is relevant to emphasizing that all patients in group II were submitted to minimally invasive surgery (p < 0.001). With purpose of evaluating the surgical morbidity of LND, we defined 2 groups: no LND and LND. The first group included patients from groups I and II, and the 9

second group included patients from groups III and IV. Performing lymphadenectomy was associated with a longer surgical time (median: 150 vs 270 minutes; p < 0.001) and a higher blood loss (median: 27.5 vs 80 cc; p < 0.001). The incidence of intraoperative complications was higher (1 vs 16, p = 0.001) among extensive lymphadenectomy group, yielding an odds ratio [OR] of 14.25 (p = 0.011), and an increased risk of 30-day complications (13.0% vs 31.9%, p < 0.001; OR: 3.11, p = 0.001), especially a higher proportion of major complications (grade III, IV, and V; 5.2% vs 12.6%, p < 0.01; OR: 3.08, p = 0.023). The risk of LLL in the LND group was 8-fold higher when comparing with those who were not extensively staged (0% vs 6.7%, p < 0.01; OR: 8.14, p = 0.048; Table 4). When we analyzed the patients who were submitted to pelvic lymphadenectomy only against those for whom para-aortic LND was also performed, there were no differences in surgical approach (p = 0.295), intraoperative complications (4.1% vs 15.1%; p = 0.084), or LLL (4.1% vs 8.1%; p = 0.487). However, extended lymphadenectomy was associated with an increase in surgical time (median 210 vs 300, p < 0.001), a higher blood loss (median 50 vs 100, p = 0.029), and more 30-day complications (p = 0.013; Table 5). After univariate analysis, only extensive lymphadenectomy was a significant risk factor for intraoperative complications (0.9% vs 11.9%, p = 0.001, Supplementary Table 2). A multivariate logistic regression model was used to assess the independent influence of different variables on the probability of 30 – day complications, including age, BMI, extensive lymphadenectomy and surgical technique (Supplementary table 3). This analysis confirmed that extensive lymphadenectomy and performing surgery by laparotomy were independent risk factors for occurrence of complications. Comprehensive lymphadenectomy had a 3-fold increased risk of 30-day complications (95% confidence interval [CI]: 1.59–6.66). Patients submitted to laparotomy had an increased risk of 30-day complications (OR: 7.02; 95% CI: 2.08–23.66; Table 6) The overall and bilateral SLN detection rates were 87.9% (94/107) and 59.8% (64/107), respectively. To calculate the sensitivity, false-negative rate, and NPV of SLN, we used the data only from patients who underwent SLN mapping and LND as well as bilateral SLN mapping. There were 26 patients, of whom 5 (19%) had lymph node metastasis. SLN biopsy detected 4 of 5 metastases, yielding a sensitivity of 80%, and only one patient had lymph node metastasis with a negative SLN; the false-negative rate was 20% (1/5) and the NPV was 95.4% (21/22). 10

Discussion In this retrospective cohort of 250 EC patients, SLN biopsy did not increase the risk of intraoperative complications or 30-day complications, did not cause LLL, and added only approximately 20 minutes to the surgical time compared to simple hysterectomy. To our knowledge, this is the first study to objectively report the morbidity of SLN compared with lymphadenectomy and simple hysterectomy. This study has some drawback that must be considered. Although this is not a small sample of endometrial cancer patients, the retrospective nature of our cohort and its results have limited evidence. Further, some may argue that cervical injection alone may not be appropriate for EC, and that additional site of injection or the association of other tracer might increase the detection rate, but as already mentioned, cervical injection of blue dye is simple, reproducible and has an efficient detection rate. According to the guideline of the National Comprehensive Cancer Network (NCCN version 3.2019), extensive lymphadenectomy continues to be an important aspect of surgical staging, since it provides relevant information to tailor adjuvant therapy, and considers the use of SLN in low or high-risk histologies [27]. The most recent consensus of three societies (European Society of Gynecologic Oncology, European Society for Radiotherapy and Oncology, and European Society of Medical Oncology) does not specify the method of surgical staging in EC. [28] Patients who are node positive benefit from chemotherapy compared with those who do not receive chemotherapy; this highlights the importance of pathological assessment of lymph node metastasis [29]. However, LND was associated with an increase in morbidity [13]. In our study, we found that in group III, surgical time and blood loss were higher than in any other group, in accordance with previous reports [9,14]. Our data show that this same group presented the highest incidence of morbidity: a 38.2% incidence of 30-day complications and a 7.0% incidence of intraoperative complications. Dowdy et al. performed a prospective analysis of LND and found a 37.5% incidence of 30-day complications [15]. Although the incidence of LLL in our study was not as high as 37.8%, as previously reported [16], we identified an incidence of 10% in group III (p < 0.01), most likely because we evaluated patients until 90 days after the surgery. Compared with HT alone, extensive lymphadenectomy is associated with an 11

increase in morbidity. The ASTEC trial and the Italian collaborative trial reported a considerable increase in lymphatic complications (relative risk [RR]: 3.73; 95% CI: 1.04–13.27) and surgical morbidity (RR: 8.39; 95% CI: 4.06–17.33) [8,9]. Our results support that finding, in that extensive lymphadenectomy was associated with a 14-fold increase in intraoperative complications and a higher risk of 30-day complications. Regarding LLL, Yost et al. identified lymphadenectomy as an independent risk factor (RR: 2.04; 95% CI: 1.39–2.99) [30]. Most EC patients presented in the initial stage as IA G1/2, with a low risk of lymph node metastasis (4–5%) and a 5-year overall survival rate of 97% [15]. In those patients, extensive lymphadenectomy is unnecessary and would result in more harm than good. Despite the fact that they are heavily criticized for nonstandardization of the lymphadenectomy and adjuvant therapy, two randomized clinical studies did indeed not find any therapeutic value in complete lymphadenectomy [8,9]. In a prospective series, Mariani [31] reported no benefit from extensive lymphadenectomy in a low-risk group of patients (grade 1 or 2 endometrioid tumor with a tumor diameter of 2 cm or less and a myometrial invasion of 50% or less). In this subgroup, no lymph node metastases were detected, and they presented a 5-year cancer-specific survival of 100%. With the problems exposed above, SLN appears to be a potential solution, a way to proper identify lymphatic disease with less morbidity. A prospective multiinstitutional trial, the SENTI-ENDO study, concluded that SLN may be an alternative to LND in low-risk EC patients, as a result of 98% NPV [17]. The MSKCC retrospectively evaluated the application of an algorithm, where upon selective lymphadenectomy would be performed in the unmapped hemi-pelvis, and they improved their NPV to 99.8%. [18]. According to these previous reports, we found an NPV of 95.4%. The addition of SLN mapping appears to improve the detection of lymph node metastases in apparent uterine-confined EC, and a three-fold increase in the detection of metastases has been reported [32,33]. Leitao and colleagues [11] prospectively evaluated the incorporation of a previously published algorithm [18] and reported a decrease in surgical morbidity without compromising the detection of IIIC disease, suggesting that the number of patients submitted to LND could be safely reduced by incorporating SLN mapping. Compared with LND, SLN presented a reduction in surgical time (152.5 vs 370.0 minutes, p < 0.001) and blood loss (20 vs 100 cc, p < 0.001), as reported in a retrospective analysis [20]. In our cohort of patients, there was no LLL described in the 12

group II, representing an 8-fold risk reduction; the same reduction was described by Geppert et al. in a prospective assessment of SLN vs LND [12]. SLN patients were more likely to be submitted to a minimally invasive approach. The NCCN guideline defines minimally invasive approach as standard care for apparent uterine-confined disease, duo to a lower rate of surgical morbidity, without compromise in oncologic outcome [27]. Furthermore, we analyzed the specific morbidity that SLN adds to hysterectomy. Besides the increase of approximately 20 minutes in surgical time, performing SLN biopsy does not increase surgical morbidity, as there was no statistical difference in intraoperative complications, blood loss, or 30-day complications and no case of LLL was found. These comparisons have not yet been reported in the literature. Even though the morbidity of LND has been extensively reported, our findings confirm that extensive lymphadenectomy has a higher risk of complications and increases blood loss and surgical time. On the other hand, SLN did not increase surgical complications. Given that data from the literature verify the accuracy of SLN mapping and that international guidelines propose it as a possible staging technique, we strongly recommend the use of an SLN algorithm to manage low-risk patients with EC.

13

Author Contributions

Carlos A. and Guilherme A. conceived and developed the presented idea and performed the primary analysis. Luiza P. provided the data acquisition. All authors discussed the results, participated in data interpretation and contributed to the final manuscript

14

References 1. Shepherd JH. Revised FIGO staging for gynaecological cancer. Br J Obstet Gynaecol. 1989 Aug;96(8):889–92. 2. 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 Oct 15;60(8 Suppl):2035–41. 3. Sala P, Morotti M, Menada MV, Cannavino E, Maffeo I, Abete L, et al. Intraoperative frozen section risk assessment accurately tailors the surgical staging in patients affected by early-stage endometrial cancer: the application of 2 different risk algorithms. Int J Gynecol Cancer. 2014 Jul;24(6):1021–6. 4. Mariani A, Dowdy SC, Cliby WA, Gostout BS, Jones MB, Wilson TO, et al. Prospective assessment of lymphatic dissemination in endometrial cancer: a paradigm shift in surgical staging. Gynecol Oncol. 2008 Apr;109(1):11–8. 5. Bodurtha Smith AJ, Fader AN, Tanner EJ. Sentinel lymph node assessment in endometrial cancer: a systematic review and meta-analysis. Am J Obstet Gynecol. 2017;216(5):459-476.e10. 6. Todo Y, Kato H, Kaneuchi M, Watari H, Takeda M, Sakuragi N. Survival effect of para-aortic lymphadenectomy in endometrial cancer (SEPAL study): a retrospective cohort analysis. Lancet. 2010 Apr 3;375(9721):1165–72. 7. Chan JK, Cheung MK, Huh WK, Osann K, Husain A, Teng NN, et al. Therapeutic role of lymph node resection in endometrioid corpus cancer: a study of 12,333 patients. Cancer. 2006 Oct 15;107(8):1823–30. 8. ASTEC study group, Kitchener H, Swart AMC, Qian Q, Amos C, Parmar MKB. Efficacy of systematic pelvic lymphadenectomy in endometrial cancer (MRC ASTEC trial): a randomised study. Lancet. 2009 Jan 10;373(9658):125– 36. 9. Benedetti Panici P, Basile S, Maneschi F, Alberto Lissoni A, Signorelli M, Scambia G, et al. Systematic pelvic lymphadenectomy vs. no lymphadenectomy in early-stage endometrial carcinoma: randomized clinical trial. J Natl Cancer Inst. 2008 Dec 3;100(23):1707–16. 10. Partridge EE, Shingleton HM, Menck HR. The National Cancer Data Base report on endometrial cancer. J Surg Oncol. 1996 Feb;61(2):111–23. 15

11. Leitao MM, Khoury-Collado F, Gardner G, Sonoda Y, Brown CL, Alektiar KM, et al. Impact of incorporating an algorithm that utilizes sentinel lymph node mapping during minimally invasive procedures on the detection of stage IIIC endometrial cancer. Gynecol Oncol. 2013 Apr;129(1):38–41. 12. Geppert B, Lönnerfors C, Bollino M, Persson J. Sentinel lymph node biopsy in endometrial cancer-Feasibility, safety and lymphatic complications. Gynecol Oncol. 2018;148(3):491–8. 13. Frost JA, Webster KE, Bryant A, Morrison J. Lymphadenectomy for the management of endometrial cancer. Cochrane Database of Systematic Reviews [Internet]. 2015 [cited 2019 Jan 9];(9). Available from: https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD007585.pub3/ abstract 14. Coronado PJ, Rychlik A, Martínez-Maestre MA, Baquedano L, Fasero M, García-Arreza A, et al. Role of lymphadenectomy in intermediate-risk endometrial cancer: a matched-pair study. J Gynecol Oncol. 2018 Jan;29(1):e1. 15. Dowdy SC, Borah BJ, Bakkum-Gamez JN, Weaver AL, McGree ME, Haas LR, et al. Prospective assessment of survival, morbidity, and cost associated with lymphadenectomy in low-risk endometrial cancer. Gynecol Oncol. 2012 Oct;127(1):5–10. 16. Todo Y, Yamamoto R, Minobe S, Suzuki Y, Takeshi U, Nakatani M, et al. Risk factors for postoperative lower-extremity lymphedema in endometrial cancer survivors who had treatment including lymphadenectomy. Gynecol Oncol. 2010 Oct;119(1):60–4. 17. Ballester M, Dubernard G, Lécuru F, Heitz D, Mathevet P, Marret H, et al. Detection rate and diagnostic accuracy of sentinel-node biopsy in early stage endometrial cancer: a prospective multicentre study (SENTI-ENDO). Lancet Oncol. 2011 May;12(5):469–76. 18. Barlin JN, Khoury-Collado F, Kim CH, Leitao MM, Chi DS, Sonoda Y, et al. The importance of applying a sentinel lymph node mapping algorithm in endometrial cancer staging: beyond removal of blue nodes. Gynecol Oncol. 2012 Jun;125(3):531–5. 19. Rossi EC, Kowalski LD, Scalici J, Cantrell L, Schuler K, Hanna RK, et al. A comparison of sentinel lymph node biopsy to lymphadenectomy for

16

endometrial cancer staging (FIRES trial): a multicentre, prospective, cohort study. Lancet Oncol. 2017;18(3):384–92. 20. Zahl Eriksson AG, Ducie J, Ali N, McGree ME, Weaver AL, Bogani G, et al. Comparison of a sentinel lymph node and a selective lymphadenectomy algorithm in patients with endometrioid endometrial carcinoma and limited myometrial invasion. Gynecol Oncol. 2016 Mar;140(3):394–9. 21. Bodurtha Smith AJ, Fader AN, Tanner EJ. Sentinel lymph node assessment in endometrial cancer: a systematic review and meta-analysis. Am J Obstet Gynecol. 2017;216(5):459-476.e10. 22. Papadia A, Imboden S, Siegenthaler F, Gasparri ML, Mohr S, Lanz S, et al. Laparoscopic Indocyanine Green Sentinel Lymph Node Mapping in Endometrial Cancer. Ann Surg Oncol. 2016;23(7):2206–11. 23. Paley PJ, Veljovich DS, Press JZ, Isacson C, Pizer E, Shah C. A prospective investigation of fluorescence imaging to detect sentinel lymph nodes at roboticassisted endometrial cancer staging. Am J Obstet Gynecol. 2016;215(1):117.e1-7. 24. Ducie JA, Eriksson AGZ, Ali N, McGree ME, Weaver AL, Bogani G, et al. Comparison of a sentinel lymph node mapping algorithm and comprehensive lymphadenectomy in the detection of stage IIIC endometrial carcinoma at higher risk for nodal disease. Gynecol Oncol. 2017;147(3):541–8. 25. Strong VE, Selby LV, Sovel M, Disa JJ, Hoskins W, Dematteo R, et al. Development and assessment of Memorial Sloan Kettering Cancer Center’s Surgical Secondary Events grading system. Ann Surg Oncol. 2015 Apr;22(4):1061–7. 26. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap), a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009 Apr;42(2):377–81. 27. NCCN, Clinical practice guidelines in oncology, Uterine Neoplasms, 2019 Available from: https://www.nccn.org/professionals/physician_gls/pdf/uterine.pdf 


17

28. Colombo N, Creutzberg C, Amant F, Bosse T, González-Martín A, Ledermann J, et al. ESMO-ESGO-ESTRO Consensus Conference on Endometrial Cancer. Int J Gynecol Cancer. 2016 Jan;26(1):2–30. 29. Boer SM de, Powell ME, Mileshkin L, Katsaros D, Bessette P, Haie-Meder C, et al. Adjuvant chemoradiotherapy versus radiotherapy alone for women with high-risk endometrial cancer (PORTEC-3): final results of an international, open-label, multicentre, randomised, phase 3 trial. The Lancet Oncology. 2018 Mar 1;19(3):295–309. 30. Yost KJ, Cheville AL, Al-Hilli MM, Mariani A, Barrette BA, McGree ME, et al. Lymphedema After Surgery for Endometrial Cancer: Prevalence, Risk Factors, and Quality of Life. Obstet Gynecol. 2014 Aug;124(2 0 1):307–15. 31. Mariani A, Webb MJ, Keeney GL, Haddock MG, Calori G, Podratz KC. Lowrisk corpus cancer: is lymphadenectomy or radiotherapy necessary? Am J Obstet Gynecol. 2000 Jun;182(6):1506–19. 32. Raimond E, Ballester M, Hudry D, Bendifallah S, Daraï E, Graesslin O, et al. Impact of sentinel lymph node biopsy on the therapeutic management of earlystage endometrial cancer: Results of a retrospective multicenter study. Gynecologic Oncology. 2014 Jun 1;133(3):506–11. 33. Holloway RW, Gupta S, Stavitzski NM, Zhu X, Takimoto EL, Gubbi A, et al. Sentinel lymph node mapping with staging lymphadenectomy for patients with endometrial cancer increases the detection of metastasis. Gynecol Oncol. 2016 May;141(2):206–10.

18

Table 1 Parameter

Group I 54 (21.6%) 61 (35-89) 31.8 (21.9-51)

Age, years BMI ASA 1 2 3 4

6 (11.5%) 36 (69.2%) 8 (15.4%) 2 (3.8%)

Clinicopathologic characteristics of patients Group II Group III 61 (24.4%) 89 (35.6%) 60 (44-87) 62 (31-80) 33 (21.4-48.3) 30.4 (18.0-46.3) 7 (11.5%) 42 (68.9%) 12(19.7%) 0 (0.0%)

Group IV 46 (18.4%) 63 (46-77) 29.3 (22.2-41.3)

12 (13.6%) 67 (76.1%) 8 (9.1%) 1 (1.1%)

p value 0.152 0.019 0.410

7 (15.2%) 26 (78.3%) 3 (6.5%) 0 (0.0%)

Technique Laparotomy 5 (9.3%) 0 (0.0%) 10 (11.2%) 0 (0.0%) Laparoscopy 48 (88.9%) 45 (73.8%) 71 (79.8%) 30 (65.2%) Converted* 1 (1.9%) 0 (0.0%) 2 (2.2%) 0 (0.0%) Robotic 0 (0.0%) 16 (26.2%) 6 (6.7%) 16 (34.8%) Histology Endometrioid 43 (26.4%) 49 (30.1%) 43 (26.4%) 28 (17.2%) Serous 5 (14.3%) 3 (8.6%) 20 (57.1%) 7 (20.0%) Carcinosarcoma 2 (10.0%) 1 (5.0%) 16 (80.0%) 1 (5.0%) Clear cell 1 (16.1%) 1 (16.1%) 3 (50%) 1 (16.7%) Other 3 (12.5%) 7 (29.2%) 5 (20.8%) 9 (37.5%) Tumor invasion <50% 38 (27.1%) 52 (37.1%) 32 (22.8%) 18 (13%) >50% 5 (8.3%) 7 (11.7%) 26 (43.3%) 22 (36.7%) II 11 (22.4) 1 (2.0%) 31 (63.3%) 6 (12.2%) LVI 9 (16.7%) 14 (23.0%) 35 (39.3%) 15 (32.6%) Tumor grade 1 23 (42.6%) 21 (38.9%) 5 (9.3%) 5 (9.3%) 2 19 (18.4%) 29 (28.2%) 29 (28.2%) 26 (25.2%) 3 10 (12.2%) 9 (11.0%) 49 (59.8%) 14 (17.1%) BMI = body mass index LVI = lymph vascular invasion Group I = hysterectomy plus salpingo-oophorectomy Group II = hysterectomy plus salpingo-oophorectomy and sentinel lymph node Group III = hysterectomy plus salpingo-oophorectomy and lymph node dissection Group IV = hysterectomy plus salpingo-oophorectomy, sentinel lymph node and lymph node dissection *All surgeries were initially laparoscopic

19

<0.001

< 0.001

<0.001

0.019 <0.001

Table 2 Parameter Surgical time (minutes) Blood loss (mL) Intraoperative complications No Vascular Urinary Neurological Intestinal All 30-day complications No I II III IV V Yes LLL No Yes

Group I 54 (21.6%)

Group II 61 (24.4%)

Surgical morbidity Group III 89 (35.6%)

Group IV 46 (18.4%)

135 (50-270) 35 (0-500)

152 (60-300) 20 (0-500)

370 (80-600) 100(0-2300)

240 (125-420) 45 (0-500)

54 (100%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%)

60 (98.4%) 0 (0.0%) 0 (0.0%) 1 (1.6%) 0 (0.0%) 1 (1.6%)

80 (88.8%) 6 (6.7%) 1 (1.1%) 1 (1.1%) 2 (2.2%) 10 (11.2%)

40 (87%) 6(13%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 6 (13%)

46 (85.3%) 2 (3.7%) 2 (3.7%) 0 (0.0%) 1 (1.9%) 3 (5.6%) 8 (14.8%)

54 (88.5%) 5 (8.2%) 0 (0.0%) 0 (0.0%) 2 (3.3%) 0 (0.0%) 7 (11.5%)

55 (61.8%) 16 (18.0%) 9 (10.1%) 5 (5.6%) 2 (2.2%) 2 (2.2%) 34 (38.2%)

37 (80.4%) 1 (2.2%) 0 (0.0%) 5 (10.9%) 2 (4.3%) 2 (4.3%) 9 (19.6%)

54 (100%) 0 (0.0%)

61 (100%) 0 (0.0%)

80 (89.9%) 9 (10.1%)

46 (100%) 0 (0.0%)

Group I= hysterectomy plus salpingo-oophorectomy Group II = hysterectomy plus salpingo-oophorectomy and sentinel lymph node Group III = hysterectomy plus salpingo-oophorectomy and lymph node dissection Group IV = hysterectomy plus salpingo-oophorectomy, sentinel lymph node and lymph node dissection LLL = lower-limb lymphedema

20

p value <0.001 <0.001 0.015

0.005 <0.01

<0.01 0.001

Table 3 Parameter

Surgical morbidity Group I 54 (47.0%)

Age BMI ASA

Group II 61 (53.0%)

61 (35-89) 31.8 (21.9-51)

60 (44-87) 33 (44.0-87)

6 (11.5%) 36 (89.2%) 8 (15.4%) 2 (3.8%)

7 (11.5%) 42 (68.9%) 12 (19.7%) 0 (0.0%)

Laparotomy Laparoscopy Converted* Robotic

5 (9.3%) 48 (88.9%) 1 (1.9%) 0 (0.0%) 135 (50.0-270.0) 35 (0.0-500.0)

0 (0.0%) 45 (73.8%) 0 (0.0%) 16 (26.2%) 152 (60.0-300.0) 20 (0.0-500.0)

No Vascular Urinary Neurological Intestinal All

54 (100%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%)

60 (98.4%) 0 (0.0%) 0 (0.0%) 1 (1.6%) 0 (0.0%) 1 (1.6%)

No I II III IV V Yes

46 (85.3%) 2 (3.7%) 2 (3.7%) 0 (0.0%) 1 (1.9%) 3 (5.6%) 8 (14.8%)

54 (88.5%) 5 (8.2%) 0 (0.0%) 0 (0.0%) 2 (3.3%) 0 (0.0%) 7 (11.5%)

No Yes

54 (100%) 0 (0.0%)

61 (100%) 0 (0.0%)

1 2 3 4 Technique

Surgical time (minutes) Blood loss (mL) Intraoperative complications

30-day complications

LLL

BMI = body mass index LLL = lower-limb lymphedema Group I = hysterectomy plus salpingo-oophorectomy Group II= hysterectomy plus salpingo-oophorectomy and sentinel lymph node *The surgery was initially laparoscopic

21

Table 4 No LND 115 (46%)

Parameter Age BMI ASA

60 (35-89) 32.4 (21.4-51) 1 2 3 4

13 (11.5%) 78 (69.0%) 20 (17.7%) 2 (1.8%)

Technique Laparotomy Laparoscopy Converted* Robotic Surgical time (minutes) Blood loss (mL) Intraoperative complications No Vascular Urinary Neurological Intestinal Total 30-day complications No I II III IV V Yes Major complications (III, IV or V) LLL No Yes

5 (4.3%) 93 (80.9%) 1 (0.9%) 16 (13.9%) 150.0 (50-300) 27.5 (0-5000) 114 (99.1%) 0 (0.0%) 0 (0.0%) 1 (0.9%) 0 (0.0%) 1 (0.9%) 100 (87%) 7 (6.1%) 2 (1.7%) 0 (0.0%) 3 (2.6%) 3 (2.6%) 15 (13%) 6 (5.2%) 115 (100%) 0 (0.0%)

Surgical morbidity LND 135 (54%) p value 63 (31-80) 0.035 29.85 (18.0-46.3) 0.003 0.101 19 (14.2%) 103 (76.9%) 11 (8.2%) 1 (0.7%) 0.645 10 (7.4%) 101 (74.8%) 2 (1.5%) 22 (16.3%) 270.0 (80-600) <0.001 80.0 (0-2300) <0.001 0.002 119 (88.1%) 12 (8.9%) 1 (0.75%) 1 (0.75%) 2 (1.5%) 16 (11.9%) 0.001 0.001 92 (68.1%) 17 (12.6%) 9 (6.7%) 10 (7.4%) 3 (2.2%) 4 (3%) 43 (31.9%) <0.001 17 (12.6%) <0.01 <0.01 126 93.3%) 9 (6.7%)

BMI = body mass index ASA = American Society of Anesthesiologists

22

Hazard ratio vs No LND (95% CI)

p value

14.25 (1.85-19.63)

0.011

3.11 (1.62-5.98) 3.08 (1.16-8.14) 8.14 (1.01-65.27)

0.001 0.023 0.048

No LND = combination of the group I and group II LND = combination of the group III and group IV LLL = lower-limb lymphedema *All surgeries were initially laparoscopic

23

Table 5

Surgical morbidity

Parameter

PE LND 49 (36.3%)

Age BMI ASA 1 2 3 4

PE/PAo LND 86 (63.7%)

67 (45-80) 32.2 (19.8-41)

61 (31-77) 29.0 (18-46.3)

3 (6.1%) 40 (81.6%) 5 (10.2%) 1 (2.0%)

16 (18.8%) 63 (74.1%) 6 (7.1%) 0 (0.0%)

1 (2.0%) 39 (79.6%) 1 (2.0%) 8 (16.3%) 210 (80-480) 50 (0.0-2000)

9 (10.5%) 62 (72.1%) 1(1.2%) 14 (16.3%) 300 (120-600) 100 (0.0-2300)

47 (95.9%) 2 (4.1%)

73 (84.9%) 13 (15.1%)

40 (81.6%) 9 (18.4%)

52 (60.5%) 34 (39.5%)

47 (95.9%) 2 (4.1%)

79 (91.9%) 7 (8.1%)

p value 0.005 0.002 0.078

Technique Laparotomy Laparoscopy Converted* Robotic Surgical time (minutes) Blood loss (mL) Intraoperative complications No Yes 30-day complications No Yes LLL No Yes

0.295

<0.001 0.029 0.084

0.013

0.487

PE LND = pelvic lymph node dissection PE PAo LND = pelvic and para-aortic lymph node dissection LLL = lower-limb lymphedema *All surgeries were initially laparoscopic

24

Table 6

Multivariate analysis: 30-day complications, LND vs no LND

Age

OR 0.99

95% CI 0.96-1.0.3

p value 0.936

BMI LND Technique

1.04 3.24 7.02

0.98-1.09 1.59-6.63 2.08-23.66

0.127 0.001 0.002

BMI: body mass index

25