Total laparoscopic hysterectomy for endometrial cancer: patterns of recurrence and survival

Gynecologic Oncology 92 (2004) 789 793 www.elsevier.com/locate/ygyno Total laparoscopic hysterectomy for endometrial cancer: patterns of recurrence and survival Andreas Obermair, a,b, * Tom P. Manolitsas, c Yee Leung, a,d Ian G. Hammond, a,e and Anthony J. McCartney a,e a King Edward Memorial Hospital for Women, Subiaco, WA 6008, Australia b Royal Women s Hospital, Herston, QLD 4029, Australia c Monash Medical Centre, East Bentleigh, VIC 3204, Australia d St. John of God Hospital Murdoch, Murdoch, WA 6150, Australia e St. John of God Hospital Subiaco, Subiaco, WA 6008, Australia Received 24 March 2003 Abstract Objective. The impact of laparoscopic surgery on the patterns of recurrence and on prognosis in patients with endometrial cancer remains unclear. The objective of the current study was to evaluate the effect of the laparoscopic approach on patterns of recurrence, disease-free (DFS), and overall survival (OS) in patients with endometrial cancer. Methods. A retrospective review of patients presenting with stages 1 4 endometrial cancer who had a hysterectomy, bilateral salpingooophorectomy with or without surgical staging was performed. Patients either had a total laparoscopic hysterectomy (TLH) or a total abdominal hysterectomy (TAH). Patterns of recurrence, DFS and OS were the study endpoints. Results. The surgical intent was TLH in 226 patients (44.3%) and TAH in 284 patients (55.7%). TLH was converted to laparotomy in 11 patients. Patients for TLH were younger, heavier, and had a higher ASA score and were more likely to present with early-stage, welldifferentiated tumors and were less likely to have undergone lymphadenectomy. Median follow-up was 29.4 months. DFS and OS were adversely and independently affected by increasing age, higher stage, higher grade, and by deeper myometrial invasion, whereas the intention to treat (TLH vs. TAH) did not influence DFS or OS. Patterns of recurrence were similar in both groups and no port-site metastasis was noted in the TLH group. Conclusions. The incidence of port-site metastasis in early-stage endometrial cancer treated by TLH is low. Laparoscopic management does not seem to worsen the prognosis of patients with endometrial cancer. D 2004 Elsevier Inc. All rights reserved. Keywords: Endometrial; Cancer; Prognosis Introduction Endometrial cancer is the most common gynecologic malignancy in developed countries [1]. Standard management includes surgical removal of uterus, tubes, and ovaries and its stage is determined surgically [2]. Currently, surgery for endometrial cancer is performed through laparotomy in most centers worldwide but laparoscopic surgery has been suggested as an alternative to abdominal surgery for the treatment of early endometrial cancer [3,4]. The technique * Corresponding author. Research Gynaecological Oncology, Ned Hanlon Building, 6th Floor, Royal Women s Hospital, Herston, QLD, Australia. Fax: +61-7-3636-5289. E-mail address: andreas_obermair@health.qld.gov.au (A. Obermair). of a total laparoscopic hysterectomy (TLH) has evolved from our center and its concept has been presented previously [5]. When compared to TAH, operative risks were reduced and hospital stay was diminished, similar to what has been reported for laparoscopically assisted vaginal hysterectomy (LAVH) (6). In contrast to LAVH, TLH allows the completion of the hysterectomy entirely laparoscopically, without the need for vaginal surgery. While several studies observed reduced treatment-related morbidity, shorter hospital stay and quicker recovery of patients undergoing a laparoscopic procedure [7 10], patterns of recurrence and long-term risks for recurrence, and survival after laparoscopy for endometrial cancer are not well documented. Given the poor prognosis of patients with metastatic or recurrent disease, case reports of port-site and 0090-8258/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2003.12.001

790 A. Obermair et al. / Gynecologic Oncology 92 (2004) 789 793 local recurrences have raised concerns regarding outcomes in patients with endometrial cancer managed laparoscopically [11 14]. The purpose of this report is to compare patterns of recurrence, disease-free, and overall survival of patients who underwent a TLH with patients who had a total abdominal hysterectomy (TAH) F surgical staging for endometrial cancer. Patients and methods Patients A total of 600 patients underwent primary surgery for endometrial cancer at King Edward Memorial Hospital for Women, St. John of God Hospital Subiaco or St. John of God Hospital Murdoch, Perth, Western Australia, during the period January 1993 and December 2001. Patients with previous or concurrent malignancy (n = 77) and patients who had a vaginal hysterectomy (n = 13) were excluded from this series. No other exclusion criteria applied otherwise. Analysis is based on the data of 510 patients. All operations were performed by one of three gynecologic oncologists or by the Fellow in Gynecologic Oncology. Cases were identified from the West Australian Gynecologic Cancer Service database and cross-checked with the clinical coding database of the medical records departments and the database of histopathologic diagnoses from the Department of Pathology. Staging and grading were determined according to the 1988 FIGO criteria [2]. Preoperatively, patients were scored according to the criteria of the American Society of Anaesthesiologists [15]. Treatment All patients either had a laparotomy with TAH or laparoscopy with TLH. Patients were allocated to either laparoscopy or laparotomy at the surgeon s preference. Patients considered for TLH had a CT scan of the pelvis and abdomen. Those with a uterine size larger than 10 weeks or evidence of extrauterine disease were excluded. Patients with grade 3 tumors or with non-endometrioid histologic type were originally treated with laparotomy but from 1997 onwards were considered eligible for TLH. Depending on the intraoperative assessment by frozen section examination, a surgical staging, which included peritoneal washings and pelvic/aortic lymphadenectomy, was performed or omitted. All patients with grade 2 tumors invading into the outer half of the myometrium, all patients with grade 3 tumors, patients with tumors extending into the uterine cervix, and patients with tumors of non-endometrioid histologic type were staged surgically. A lymph node dissection was to the surgeon s discretion if patients had a grade 1 tumor confined to the endometrium or a grade 1 or grade 2 tumor limited to the inner half of the myometrium at frozen section. Surgical staging was also omitted for safety reasons in case of advanced age, coexisting medical morbidity and insufficient exposure due to obesity. Patients who should have undergone surgical staging but in whom a surgical staging was not performed at the initial operation were restaged surgically or they were offered radiotherapy to the whole pelvis. The technique of TLH has been described previously [5]. In principle, every step of the TLH duplicates the technique of the open procedure. Laparoscopic staplers were used initially to secure the ovarian and the uterine vessels but staples were replaced by bipolar diathermy from 1996. A silicone tube (Tyco Healthcare, Australia) is inserted transvaginally to display the vaginal fornices and the uterine vessels. While the pneumoperitoneum is maintained by the tube, the vagina is divided over the edge of the tube and the uterus. The pelvic and aortic lymph nodes are removed through the tube before suturing the vaginal vault. Statistical analysis Comparison of age, body weight, number of previous laparotomies, and ASA score was performed using t test, and Chi-square tests were used to compare the distribution of frequencies. Disease-free survival (DFS) was defined as the period from surgery until the date of first recurrence. Overall survival (OS) was defined as the time from surgery to death. Data on patients who were alive were censored at the last follow-up visit. Univariate analysis of PFS was performed as outlined by Kaplan and Meier [16] and survival curves were compared using the log-rank test [17]. Multivariate Cox models were calculated to identify independent prognostic factors for disease-free survival (DFS) and overall survival (OS). P values are the result of two-sided tests and P < 0.05 was considered to indicate a statistically significant difference. Statistical analysis was done using SPSS (Version 9.0, SPSS, INC, Chicago, IL). Results Analysis is based on data from 510 patients. The surgical intent was laparoscopy in 226 patients (44.3%) and laparotomy in 284 patients (55.7%). The laparoscopic procedure was converted to laparotomy in 11 patients (4.8%) due to failed access associated with severe adhesions (n = 6, 2.6%) and to control significant hemorrhage (n = 5, 2.2%). Overall, the median age was 63.3 years (range, 26 94 years) and the median weight was 75.0 kg (range, 41 to 170 kg). Two hundred and fifty-seven patients (50.4%) had at least one previous laparotomy and the median number of previous laparotomies was 1 (range 0 6). The preoperative ASA score was 1, 2, and 3 in 107 (21.0%), 263 (51.6%), and 101 (19.8%) patients, respectively. Detailed patients characteristics are given in Table 1. When compared to patients in the laparotomy group, patients in the laparoscopy group were younger, heavier, and had a higher ASA score.

A. Obermair et al. / Gynecologic Oncology 92 (2004) 789 793 791 Table 1 Patients characteristics TLH (n = 226, 44.3%) TAH (n = 248, 55.7%) Age [years] (mean F SD) 61.7 F 12.1 64.5 F 11.8 0.008 a Weight [kg] (mean F SD) 82.6 F 24.4 74.2 F 19.6 <0.0001 a Number of previous laparotomies 0.65 F 0.8 0.80 F 0.9 0.061 a (mean F SD) ASA score (mean F SD) 2.06 F 0.6 1.92 F 0.6 0.020 a Stage [n (%)] IA 58 (25.7) 50 (17.6) IB 128 (56.6) 118 (41.5) IC 24 (10.6) 43 (15.1) IIA 2 (0.9) 5 (1.8) IIB 6 (2.7) 17 (6.0) III 5 (2.4) 39 (13.8) IV 3 (1.3) 12 (4.2) Grade [n, (%)] 1 124 (54.9) 110 (38.9) 2 74 (32.7) 92 (32.5) 3 28 (12.4) 81 (28.6) Histologic type [n, (%)] endometrioid 210 (94.6) 235 (82.7) adenosquamous 3 (1.4) 17 (6.0) serous papillary 5 (2.3) 28 (9.9) clear cell 2 (0.9) 4 (1.4) others 2 (0.9) 0 Myometrial invasion [n, (%)] none 58 (25.7) 53 (18.7) inner half 134 (59.3) 146 (51.4) outer half 34 (15.0) 85 (29.9) Lymphadenectomy performed [n, (%)] No 133 (58.8) 113 (39.8) Yes 93 (41.2) 171 (60.2) Number of nodes retrieved 9.4 F 12.9 16.8 F 17.0 <0.0001 a (mean F SD) Lymph node involvement Nodes negative 56 (96.6) 96 (78.7) Nodes positive 2 (3.4) 26 (21.3) a t test. Chi-square test. P value Patients in the laparoscopy group were more likely to present with stage 1A or 1B, and with grade 1, endometrioid tumors confined to the inner half of the myometrium (Table 1). According to our indications for surgical staging, patients in the laparoscopy group were less likely to require a full surgical staging, the number of nodes retrieved was less than in the TLH group and only two patients (3.4%) in this group were found to have positive nodes (Table 1). Within a median observation period of 29.4 months (95% confidence interval, 25.7 to 33.2 months), 46 patients (9.0%) developed recurrent disease. Overall, the DFS was 85.5% at 60 months and the median disease-free survival was 11.7 months (Q1 to Q3, 8.3 to 21.9 months) for patients who developed recurrent disease. Nine of 226 patients (4.0%) in the laparoscopy group and 37 of 248 patients (14.9%) in the laparotomy group developed recurrent disease. The patterns of recurrence were similar in both groups (Table 2). No port-site recurrence was detected in the laparoscopic group. At the time of analysis, 408 patients (80.0%) were alive without disease, 16 patients (3.1%) were alive with cancer, 70 patients (13.7%) died and 16 patients (3.1%) were lost to follow-up. The overall survival for all patients was 77.5% at 60 months. Disease-free and overall survival were adversely and independently affected by increasing age, higher stage, higher grade, and by deeper myometrial invasion, whereas the histologic type and the Table 2 Numbers of patients with recurrence Vault Pelvis Abdomen Distant Multiple Total TLH 1 0 4 3 1 9 TAH 9 7 5 15 1 37 Total 10 7 9 18 2 46 Intention to treat was either a total laparoscopic hysterectomy (TLH, n = 226) or a total abdominal hysterectomy (TAH, n = 248). No port site recurrence was noted in the laparoscopy group.

792 A. Obermair et al. / Gynecologic Oncology 92 (2004) 789 793 Table 3 Multivariate Cox models on disease-free and overall survival Disease-free survival Overall survival OR CI P value OR CI P value Age (continuous) 1.06 1.02, 1.09 0.016 1.05 1.03, 1.08 <0.0001 Stage (1 vs. 2 vs. 2.4 1.7, 3.6 <0.0001 2.4 1.7, 3.2 <0.0001 3 vs. 4) Grade (G1 vs. 1.5 0.9, 2.5 0.084 1.6 1.1, 2.5 0.012 G2 vs. G3) Myometrial 1.2 0.6, 2.0 0.620 2.5 1.5, 4.2 <0.0001 invasion (none vs. inner half vs. outer half) Intention to treat 1.3 0.5, 2.9 0.560 0.7 0.4, 1.3 0.274 (TLH vs. TAH) Histologic type 1.0 0.4, 2.4 0.897 0.8 0.4, 1.5 0.801 (endometrioid vs. others) Lymph node 2.2 0.9, 5.0 0.051 1.2 0.7, 2.3 0.471 dissection Radiotherapy 1.7 0.8, 3.5 0.150 0.7 0.4, 1.1 0.675 OR, odds ratio; CI, 95%-confidence interval. intention to treat (laparoscopy vs. laparotomy) did not influence the overall survival probability (Table 3). Discussion Laparoscopic approaches are increasingly adopted by gynecologists in the United States [18]. Following the first the laparoscopic pelvic and aortic lymph node dissection for endometrial cancer in 1992 by Childers and Surwit [19], an increasing number of gynecologic oncologists employ laparoscopic techniques and the results upon treatment-related morbidity, postoperative pain, length of hospital stay, and recovery to normal activity associated with laparoscopic surgery are encouraging [4,6 10]. Currently, a Gynecologic Oncology Group trial compares the laparoscopic with the abdominal with some 2500 patients being enrolled (LAP2) [3]. In our center, a new laparoscopic technique was developed, which allows the completion of the hysterectomy entirely laparoscopically (TLH) without the need of vaginal surgery [5]. Unlike LAVH, the technique of TLH does not require vaginal descent, as is often absent in nulliparous patients and it also saves operating time because the vaginal component of the procedure is omitted. The procedure has been demonstrated to be feasible and safe and its morbidity is low compared to TAH [6]. While several publications have addressed treatmentrelated morbidity and short-term outcome, only limited information is available upon the impact of laparoscopic hysterectomy on disease-free and overall survival in endometrial cancer. Case reports documented the possibility of port-site metastasis and raised concerns about the safety of laparoscopy in gynecologic oncology [11 14]. However, the incidence of port-site metastasis following laparoscopy for endometrial cancer is not known and metastasis at the site of abdominal entry has also been described after TAH for endometrial cancer [20]. In the present series, the median follow up is 29 months. Compared to recent years, TLH was less frequently performed in the early years. Eltabakh [21] reviewed a series of patients with clinical stage 1 endometrial cancer who either had a LAVH (n = 100) or a TAH (n = 86) [19]. Women who underwent laparoscopy and those who underwent laparotomy had similar 2-year and 5-year disease-free and overall survival rates. There was no apparent difference with regard to the sites of recurrence between both groups and no port-site metastasis was noted in the laparoscopy group. In a European series comparing laparoscopic (n = 177) with open surgery (n = 44) for endometrial cancer, similar disease-free survival rates were observed [22]. The only prospective, randomized study compared patients who had a LAVH (n = 37) with patients who had a TAH (n =33)[8]. Overall survival and recurrence-free survival were similar in both groups. However, this study was designed to compare treatment-related morbidity rather than prognosis [8]. In the present study of 510 patients, we compared patterns of recurrence and diseasefree and overall survival of endometrial cancer patients who had a TLH (n = 226) or a TAH (n = 284). After adjustment for age, stage, grade of differentiation, depth of myometrial invasion, and histologic type, the two surgical techniques (TLH vs. TAH) demonstrated similar patterns of recurrence (Table 2) and similar probabilities for disease-free and overall survival (Table 3). Considering the fact that none of the 226 patients in whom a TLH was attempted and abandoned (n = 11) or who underwent a TLH successfully (n = 215) developed a port-site recurrence suggests that the incidence of port-site metastasis is low. The limitations of this study due to its retrospective design are acknowledged. In the early days of TLH, only patients with well-differentiated tumors were eligible for the procedure. Patients with undifferentiated tumors (G3) and with non-endometrioid histologic type were considered not eligible for TLH until 1997. Patients with a uterine size larger than 10 weeks or with evidence of extrauterine disease are still treated by laparotomy rather than by a laparoscopic approach. Therefore, there is an overrepresentation of low-risk tumors (stage 1A or 1B, grade 1 tumors of endometrioid, invasion limited < inner half of the myometrium) in the laparoscopic group of this series. This is also reflected in the low number of lymph node dissections and in the lower number of positive nodes in the laparoscopic group. However, when adjusted for all these parameters, no difference in disease-free or overall survival could be found. While our study design does not allow rejecting an influence of the laparoscopic approach on disease-free and overall survival, this paper suggests that such a difference is unlikely. Considering the lower morbidity of TLH compared to abdominal surgery [6], the present data justify a prospec-

A. Obermair et al. / Gynecologic Oncology 92 (2004) 789 793 793 tive randomized trial comparing laparoscopy with laparotomy. This trial must enroll a sufficiently high number of patients to determine a prognostic impact of the surgical approach on the prognosis of patients with endometrial cancer. References [1] Jemal A, Thomas A, Murray T, Thun M. Cancer statistical 2002. CA Cancer J Clin 2002;52:23 47. [2] International federation of Gynecology and Obstetrics. Annual report on the results of treatment in gynecologic cancer. Int J Gynecol Obstet 1989;28:189 90. [3] Laparoscopic Surgery or Standard Surgery in Treating Patients With Endometrial Cancer or Cancer of the Uterus (http://clinicaltrials.gov/ ct/gui/action/searchaction;jsessionid=9fccb6259e7a410e39f E489A33D907DD?term=lap2). [4] Scribner Jr DR, Walker JL, Johnson GA, McMeekin SD, Gold MA, Mannel RS. Surgical management of early-stage endometrial cancer in the elderly: is laparoscopy feasible? Gynecol Oncol 2001;83: 563 8. [5] McCartney AJ, Johnson N. Using a vaginal tube to separate the uterus from the vagina during laparoscopic hysterectomy. Obstet Gynecol 1995;85:293 6. [6] Manolitsas TP, McCartney AJ. Total laparoscopic hysterectomy in the management of endometrial carcinoma. J Am Assoc Gynecol Laparosc 2002;9:54 62. [7] Magrina JF, Mutone NF, Weaver AL, Magtibay PM, Fowler RS, Cornella JC. Laparoscopic lymphadenectomy and vaginal or laparoscopic hysterectomy with bilateral salpingo-oophorectomy for endometrial cancer: morbidity and survival. Am J Obstet Gynecol 1999; 181:376 81. [8] Malur S, Possover M, Michels W, Schneider A. Laparoscopic-assisted vaginal versus abdominal surgery in patients with endometrial cancer A prospective randomized trial. Gynecol Oncol 2001;80: 239 44. [9] Gemignani ML, Curtin JP, Zelmanovich J, Patel DA, Venkatraman E, Barakat RR. Laparoscopic-assisted vaginal hysterectomy for endometrial cancer: clinical outcomes and hospital charges. Gynecol Oncol 1999;73:5 11. [10] Härkki-Siren P, Sjöberg J, Toivonen J, Tiitinen A. Clinical outcome and tissue trauma after laparoscopic and abdominal hysterectomy: a randomized controlled study. Acta Obstet Gynecol Scand 2000;79: 866 71. [11] Muntz HG, Goff BA, Madsen BL, Yon JL. Port-site recurrence after laparoscopic surgery for endometrial carcinoma. Obstet Gynecol 1999;93:807 9. [12] Faught W, Fung K, Fung M. Port site recurrences following laparoscopically managed early stage endometrial cancer. Int J Gynecol Cancer 1999;9:256 8. [13] Sonoda Y, Zerbe M, Smith A, Lin O, Barakat RR, Hoskins WJ. High incidence of positive peritoneal cytology in low-risk endometrial cancer treated by laparoscopically assisted vaginal hysterectomy. Gynecol Oncol 2002;80:378 82. [14] Chu CS, Randall TC, Bandera CA, Rubin SC. Vaginal cuff recurrence of endometrial cancer treated by laparoscopic-assisted vaginal hysterectomy. Gynecol Oncol 2003;88:62 5. [15] Committee on Records, American Society for Anaesthesiology. New classification of physical status. Anesthesiology 1963;24:111. [16] Kaplan EL, Meier P. Non-parametric estimation from incomplete observations. J Am Stat Assoc 1985;53:457 81. [17] Mantel N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 1966;50: 163 70. [18] Farquhar CM, Steiner CA. Hysterectomy rates in the United States 1990 1997. Obstet Gynecol 2002;99:229 34. [19] Childers JM, Surwit EA. Combined laparoscopic and vaginal surgery for the management of two cases of stage 1 endometrial cancer. Gynecol Oncol 1992;45:46 51. [20] Khalil AM, Chammas F, Kaspar HJ, Shamseddine AI, Seoud MA. Case report: endometrial cancer implanting in the laparotomy scar. Eur J Gynaecol Oncol 1998;19:408 9. [21] Eltabakh GH. Analysis of survival after laparoscopy in women with endometrial cancer. Cancer 2002;95:1894 901. [22] Holub Z, Jabor A, Bartos P, Eim J, Urbanek S, Pivovarnikova R. Laparoscopic surgery for endometrial cancer: long-term results of a multicentric study. Eur J Gynaecol Oncol 2002;23:305 10.