- Email: [email protected]
Should epithelial ovarian carcinoma metastatic to the inguinal lymph nodes be assigned stage IVB?
YGYNO-976817; No. of pages: 4; 4C: Gynecologic Oncology xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Should epithelial ovarian carcinoma metastatic to the inguinal lymph nodes be assigned stage IVB? Dimitrios Nasioudis ⁎, Eloise Chapman-Davis, Melissa K. Frey, Thomas A. Caputo, Steven S. Witkin, Kevin Holcomb Department of Obstetrics and Gynecology, Weill Cornell Medicine, New York, NY, USA
H I G H L I G H T S • In a large database, 151 women with EOC and isolated inguinal LN metastases were identified. • Their survival was similar to those with stage III disease and regional nodal involvement. • Our findings do not support the reclassification of these patients as stage IVB.
a r t i c l e
i n f o
Article history: Received 31 May 2017 Received in revised form 8 July 2017 Accepted 10 July 2017 Available online xxxx Keywords: Epithelial ovarian cancer Stage IV New FIGO classification Inguinal lymph nodes Overall survival Prognosis
a b s t r a c t Objective. According to the revised FIGO staging system women with ovarian carcinoma and inguinal lymph node (LN) metastases, formerly stage III, are now considered stage IVB. In this study we compare their survival to that of women with stage III and stage IV disease. Methods. Women diagnosed with epithelial ovarian carcinoma were drawn from the Surveillance, Epidemiology, and End Results database (2004–2013). Four groups were formed: group 1 (stage IV due to positive inguinal nodes), group 2 (stage III with positive para-aortic/pelvic nodes), group 3 (stage IV with positive distant nodes) and group 4 (stage IV with distant metastases). Overall (OS) and cancer-specific survival (CSS) rates were evaluated with the Kaplan-Meier method. The log-rank test and Cox-hazard models were employed for univariate and multivariate survival analysis. Results. A total of 11,152 women were identified. Five-year OS for women in group 1 (n = 151) was 46.3% compared to 44.9% for those in group 2 (n = 4,403) (p = 0.4), 32.9% in group 3 (n = 642) (p b 0.001) and 25.3% in group 4 (n = 5,956) (p b 0.001). After controlling for age, race and histology, group 1 had improved overall and cancer-specific mortality compared to groups 3 and 4 but not group 2. Conclusions. Ovarian cancer patients with stage IV ovarian cancer due solely to inguinal nodal metastases have similar survival as those with pelvic/para-aortic nodal involvement and improved survival compared to those harboring distant metastases. Our findings do not support the reclassification of these patients as stage IVB. © 2017 Elsevier Inc. All rights reserved.
1. Introduction Epithelial ovarian cancer (EOC) is currently the second most common, but most lethal gynecologic malignancy in the United States [1]. The FIGO staging system was developed by the International Federation of Gynecology and Obstetrics in an effort to permit comparisons of outcomes between different cancer centers [2]. It is also designed to adequately reflect disease burden and prognosis of women with EOC [2]. Recently, after almost three decades of use, due to the need for better ⁎ Corresponding author at: Department of Obstetrics and Gynecology, Weill Cornell Medicine, 1300 York Avenue, Box 35, New York, NY 10065, United States. E-mail address: [email protected] (D. Nasioudis).
prognostic discrimination, the FIGO staging system was revised and several changes were introduced [3]. The prognostic role of the revised staging system, which has a larger number of sub-stages, has already been evaluated in several retrospective cohort studies [2,4,5]. Nevertheless, unresolved classification issues remain especially for women with stage IV disease [6]. Currently, due to improved pre-operative imaging techniques and the performance of more comprehensive staging procedures, almost 1 in 5 women (12–21%) are diagnosed with stage IV disease [6]. On the other hand, women with EOC rarely present with isolated inguinal lymph node (LN) metastases, with an estimated incidence of 3–5% [6, 7–10]. Based on the revised FIGO staging system, women with inguinal LN metastasis previously considered as stage IIIC are now assigned to
http://dx.doi.org/10.1016/j.ygyno.2017.07.124 0090-8258/© 2017 Elsevier Inc. All rights reserved.
Please cite this article as: D. Nasioudis, et al., Should epithelial ovarian carcinoma metastatic to the inguinal lymph nodes be assigned stage IVB?, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.124
2
D. Nasioudis et al. / Gynecologic Oncology xxx (2017) xxx–xxx
stage IVB [3]. The prognostic significance of this reclassification has yet to be examined. The aim of this retrospective population-based study was to evaluate the prognosis of women with stage IV EOC due to isolated inguinal LN metastases following cancer-directed surgery and compare it to that of women with stage III disease and pelvic and/or paraaortic LN and those with stage IV disease due to other distant metastases. 2. Material and methods A cohort of women diagnosed between 2004 and 2013 with stage III or stage IV disease (as provided by the SEER database) was drawn from the National Cancer Institute's Surveillance Epidemiology, and End Results (SEER) database [11]. In the present study, data deriving from 18 cancer registries were included (Detroit, Iowa, Kentucky, Louisiana, Utah, Connecticut, New Jersey, Atlanta, Rural and Greater Georgia, Alaska, California, Hawaii, Los Angeles, New Mexico, San Francisco, San Jose, Seattle), as released on April 2016. The SEER database covers approximately 27.8% of the total US population based on the 2010 census and incorporates high-quality data of patients with primary malignant tumors. The ICD-O-3/WHO 2008 site specific code “C.569/ovary” and the ICD-O-3 histology codes as grouped by the International Association for Research on Cancer (IARC) were employed to identify all women with epithelial ovarian cancer [12]. Patient data are de-identified and available to the public for research purposes. Demographic, clinicopathological and survival data were extracted using the “case-listing” option. Information regarding the administration of chemotherapy is not captured in the SEER database. Only women with active follow-up, stage III or IV disease, who underwent cancer-directed surgery (CDS), as assessed by the site-specific surgery codes, were eligible for inclusion in the present study. Based on the information available at the Collaborative Staging (CS) coding fields (CS lymph nodes and CS mets), four groups were formed. Group 1 included women with inguinal LN metastasis and absence of any other distant metastases formerly classified as stage IIIC, group 2 included women with aortic and/or pelvic LN metastases and absence of distant metastases (stage III), group 3 included women with metastases to distant LNs only and group 4 included women with metastases to other distant sites (excluding distant LNs). Women with distant metastases NOS (not otherwise specified) or distant metastases to both distant sites and distant lymph nodes were excluded. Median and 5-year Overall (OS) and Cancer-specific survival (CSS) rates were calculated after generation of Kaplan-Meier curves and comparisons between the different groups were made with the log-rank test. For the estimation of CSS, only those with one tumor or the first of multiple primary tumors were included; women who died from causes other than ovarian cancer were censored. In addition, Cox proportional hazard models were constructed to control for confounders such as patient age, race and tumor histology. Frequency of distribution of categorical and continuous variables was compared with the chisquare test and Mann-Whitney U test, respectively. All statistical analyses were performed with the SPSS v.24 statistical package and the alpha level of statistical significance was set at 0.05. 3. Results A total of 11,152 women met the inclusion criteria. Groups 1 and 2 included 151 (1.4%) and 4403 (39.5%) women, while groups 3 and 4 included 642 (5.8%) and 5956 (53.4%) women respectively. The majority of women included in group 2 had intraperitoneal disease (stage IIIC). Median patient age was 61 years and the majority were White women (84.9%). Women in group 2 were younger (median 59 years) compared to those in group 1 (median 63 years, p b 0.001), group 3 (median 62 years, p b 0.001) and group 4 (median 63 years, p b 0.001). Most tumors were of serous histology (71.%) and based on available information median tumor size was 8.2 cm (n = 7445, range 0.1–
85 cm), while the majority were grade III/IV (84.7%, n = 7687). There were no statistically differences between the different groups in regard to patient race (p = 0.19) and tumor histology (p = 0.25). However, women in group 1 had smaller tumors (n = 108, median size 7 cm) compared to those in group 2 (n = 3332, median size 9 cm) (p b 0.001). Rates of optimal debulking (defined as gross residual tumor nodules of 1 cm or less in size) were 90% (36/40) and 90.4% (1131/ 1251) in groups 1 and 2, respectively, compared to 83% (117/141) and 81.1% (936/1154) in groups 3 and 4, respectively (p b 0.001). According to the reverse Kaplan-Meier method, median follow-up of the cohort was 57 months (95% CI: 55.56, 58.44). Median OS for women in group 1 was 55 months (95% CI: 37.4, 72.6) versus 53 months (95% CI: 50.4, 55.6) for those in group 2 (p = 0.41), 39 months (95% CI: 35.4, 42.6) for those in group 3 (p = 0.001) and 30 months (95% CI: 28.8, 31.2) for those in group 4 (p b 0.001) (Fig. 1). Similarly, CSS for women in group 1 was 66 months versus 59 months (95% CI: 55.0, 63.0) for those in group 2 (p = 0.34), 42 months (95% CI: 37.3, 46.7) for those in group 3 (p b 0.001) and 32 months (95% CI: 30.7, 33.3) for those in group 4 (p b 0.001)(Fig. 2). OS and CSS for women in group 3 was superior to that of women in group 4 (p b 0.001) but inferior to those in group 2 (p b 0.001). Table 1 summarizes 5-year and median OS, CSS in each study group. According to the multivariate Cox-proportional hazard models, after controlling for age (b65 versus ≥ 65 years), race (White versus nonWhite) and histology (serous versus non-serous), women in group 1 had better overall and cancer-specific mortality compared with those in group 3 (HR:0.66, 95% CI: 0.5, 0.85 and HR:0.62, 95% CI:0.46, 0.84 respectively) and group 4 (HR: 0.5, 95% CI: 0.39, 0.64 and HR:0.46, 95% CI: 0.35, 0.61 respectively). However there was no statistically significant difference in overall and cancer-specific mortality between women in group 1 compared to those in group 2. In addition, women in group 3 had better overall and cancer-specific mortality compared to those in group 4 (HR:0.76, 95% CI: 0.69, 0.85 and HR:0.74, 95% CI: 0.66, 0.84 respectively) but worse than those in group 2 (HR: 1.31, 95% CI: 1.18, 1.47 and HR: 1.35, 95% CI: 1.19, 1.52 respectively). Table 2 summarizes the results of the multivariate survival analysis. 4. Discussion In the present study we investigated the prognostic significance of inguinal lymph node metastases in women with EOC. These women comprise a very small group but had improved survival compared to other stage IV patients, harboring distant metastases. While, the main pattern of EOC spread is direct tumor cell seeding within the abdominal cavity, lymphatic dissemination is also likely, especially for serous tumors [13]. Metastases to inguinal LNs (while uncommon) can be observed, presumably via lymphatic drainage through the round ligament of the uterus [13]. The prognostic significance of isolated metastases to inguinal LNs has never been investigated in the past. In our study, women with stage IVB disease due to inguinal LN metastasis had similar prognosis to those with pelvic or para- aortic lymph node involvement, even after controlling for tumor histology, patient age and race. Moreover, in our cohort women with stage IV disease solely due to metastases to distant LNs had superior OS and CSS compared to those with distant metastasis to other sites (excluding inguinal LNs). Similarly to our results, Suh et al. (2015) reported that women with stage IV disease assigned on the basis of metastases to supraclavicular LNs have better overall survival than other stage IV cases and were also less likely to present with multiple metastatic lesions [5]. In the recent revision of the FIGO staging system for ovarian cancer, several changes were introduced [3]. Stage IC has been further subdivided into IC1, IC2, IC3 while stage IIC was abolished. In addition, stage III has been restructured and tumors previously considered as stage IIIC on the basis of metastases to pelvic or para-aortic lymph nodes may now correspond to stage IIIA (if only positive lymph nodes are present) or IIIB (if macroscopic extrapelvic metastasis of size
Please cite this article as: D. Nasioudis, et al., Should epithelial ovarian carcinoma metastatic to the inguinal lymph nodes be assigned stage IVB?, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.124
D. Nasioudis et al. / Gynecologic Oncology xxx (2017) xxx–xxx
3
Fig. 1. Overall Survival of women with epithelial ovarian cancer stratified by study group (n = 11,152, p b 0.001 from log-rank).
≤ 2 cm is also present) [3]. FIGO IV has been further subdivided into stage IVA and IVB which include women with malignant pleural effusion and those with other distant metastases respectively [3]. In addition women with positive inguinal lymph nodes, abdominal wall metastases and transmural or mucosal bowel involvement are now considered as stage IVB disease regardless of initial intra-abdominal tumor spread [3]. Similar to our results other authors have questioned the prognostic value of the revised definition and sub-classification of stage IV disease. More specifically, several studies have failed to demonstrate any
prognostic significance of the stage IV sub-classification into stage IVA and IVB. Rosendahl et al. (2016) reported similar 5-year overall survival rates (13%) for women with stage IVA (n = 149) and stage IVB (n = 613) disease [2]. In their retrospective study, Paik et al. (2015) did not detect any difference in overall and progression free survival between the two substages after controlling for patient age, cell type, grade and residual disease [4]. Atavesen et al. (2016) prospectively compared the overall survival of 102 women with stage IVA to that of 138 women with stage IVB and did not find any statistically significant differences (median OS was 25 and 28 months respectively, as well [14].
Fig. 2. Cancer-specific survival of women with epithelial ovarian cancer stratified by study group (n = 9679, p b 0.001 from log-rank).
Please cite this article as: D. Nasioudis, et al., Should epithelial ovarian carcinoma metastatic to the inguinal lymph nodes be assigned stage IVB?, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.124
4
D. Nasioudis et al. / Gynecologic Oncology xxx (2017) xxx–xxx
Table 1 Median and 5-year (yr) Overall (OS) (p b 0.001, from log-rank) and Cancer-specific survival (CSS) (p b 0.001 from log-rank) of women with stage III/IV epithelial ovarian cancer stratified by study group. OS
Group 1 (stage IV inguinal LNs) Group 2 (stage III aortic/pelvic LNs) Group 3 (stage IV distant LNs) Group 4 (stage IV other distant metastases)
CSS
5-yr/median
5-yr/median
n = 151/46.3%/55 months n = 4403/44.3%/53 months n = 642/32.9%/39 months n = 5956/25.1%/30 months
n = 130/50.4%/66 months n = 3786/48.9%/59 months n = 561/36.3%/42 months n = 5202/27.2%/32 months
Another retrospective study examined the prognosis of 86 women who presented solely with abdominal wall metastases (currently classified as stage IVB disease) and questioned the rationale of upstaging these patients [15]. Authors reported that their overall survival was superior to that of women with distant metastases at other sites (median OS: 58 versus 25 months) [15]. Undoubtedly, there is eminent need of high-quality prospective studies that will clarify the inconsistencies between FIGO staging guidelines and clinical outcomes. As previously outlined, attributing the diagnosis of stage IV disease to a sub-cohort of women who in fact may have a better prognosis can result in an increased emotional burden [15]. For these women access to experimental treatment options or extended surgical procedures may be also limited. A major strength of the present study is that data derive from a population-based multi-institutional database minimizing selection bias. In addition all women were diagnosed with EOC between 2004 and 2013, thus their management was based on current guidelines and may have been less heterogenous. However, several limitations should be noted. Firstly, due to the lack of central pathology review, possible tumor misclassifications could not be excluded. While most women should have received platinum-based adjuvant chemotherapy as outlined in major guidelines, information on its administration and composition was not available. Moreover, we could not assess the extent of surgical staging and residual disease following debulking surgery for all patients. A higher likelihood of achieving R0 may account for the observed survival benefit of women with isolated metastases to inguinal or distant LNs, compared to women with distant metastases at other sites. Moreover, no information was available on the imaging modalities employed in the pre-operative assessment of the patients. Also, while all women underwent a surgical staging procedure, histopathological confirmation of inguinal LN and distant metastases was not available for all cases. Lastly due to the lack of information on the administration of
Table 2 Cox-proportional hazard models of overall and cancer-specific mortality of women with stage III/IV epithelial ovarian cancer.
Group Group 1 (inguinal LNs) Group 2 (aortic/pelvic LNs) Group 3 (distant LNs) Group 4 (distant metastases) Histology Serous Non-serous Race White Non-white Age (years) b65 ≥65
OS (n = 11,152) HR (95% CI)
CSS (n = 9661) HR (95% CI)
Referent 1.16 (0.91, 1.49) 1.52 (1.17, 1.98) 2.0 (1.56. 2.55)
Referent 1.2 (0.9, 1.59) 1.61 (1.19, 2.18) 2.17 (1.64, 2.88)
Referent 1.2 (1.14, 1.27)
Referent 1.17 (1.1, 1.24)
Referent 1.08 (1.01, 1.16)
Referent 1.05 (0.97, 1.14)
Referent 1.48 (1.41, 1.55)
Referent 1.38 (1.31, 1.46)
chemotherapy, we could not exclude women who received neoadjuvant chemotherapy. Due to similar survival rates between women in group 1 and group 2, we did not perform a sub-analysis of women with both inguinal and regional LN metastases. After examining a retrospective cohort drawn from a populationbased database we demonstrated that women with EOC and inguinal LN metastases have similar survival following CDS compared to those with pelvic and para-aortic node involvement and improved survival compared to those harboring distant metastases. Our findings do not support the reclassification of these women as stage IVB. Future studies are greatly warranted to further clarify the prognosis of women with EOC and inguinal LN metastases. Given the very low incidence of inguinal LN metastases, a multi-institutional collaboration could appropriately address this controversy. Conflicts of interest The authors report no conflicts of interest. References [1] D. Jelovac, D.K. Armstrong, Recent progress in the diagnosis and treatment of ovarian cancer, CA Cancer J. Clin. 61 (3) (2011 May–Jun) 183–203. [2] M. Rosendahl, C.K. Høgdall, B.J. Mosgaard, Restaging and survival analysis of 4036 ovarian cancer patients according to the 2013 FIGO classification for ovarian, fallopian tube, and primary peritoneal cancer, Int. J. Gynecol. Cancer 26 (4) (2016 May) 680–687. [3] D.G. Mutch, J. Prat, 2014 FIGO staging for ovarian, fallopian tube and peritoneal cancer, Gynecol. Oncol. 133 (3) (2014 Jun) 401–404. [4] E.S. Paik, Y.Y. Lee, E.J. Lee, et al., Survival analysis of revised 2013 FIGO staging classification of epithelial ovarian cancer and comparison with previous FIGO staging classification, Obstet Gynecol. Sci. 58 (2) (2015 Mar) 124–130. [5] D.H. Suh, T.H. Kim, J.W. Kim, et al., Improvements to the FIGO staging for ovarian cancer: reconsideration of lymphatic spread and intraoperative tumor rupture, J. Gynecol. Oncol. 24 (4) (2013) 352–358. [6] B. Ataseven, L.M. Chiva, P. Harter, A. Gonzalez-Martin, A. du Bois, FIGO stage IV epithelial ovarian, fallopian tube and peritoneal cancer revisited, Gynecol. Oncol. 142 (3) (2016 Sep) 597–607. [7] A.L. Oei, J.A. de Hullu, J.M. Grefte, B.P. Havenith, An enlarged groin node as first manifestation of a malignancy: don't forget the ovaries, Eur. J. Obstet. Gynecol. Reprod. Biol. 138 (2) (2008 Jun) 240–242. [8] D. Ang, K.Y. Ng, H.K. Tan, A.Y. Chung, B.S. Yew, V.K. Lee, Ovarian carcinoma presenting with isolated contralateral inguinal lymph node metastasis: a case report, Ann. Acad. Med. Singap. 36 (6) (2007 Jun) 427–430. [9] H.S. Scholz, S. Lax, K.F. Tamussino, E. Petru, Inguinal lymph node metastasis as the only manifestation of lymphatic spread in ovarian cancer: a case report, Gynecol. Oncol. 75 (3) (1999 Dec) 517–518. [10] X.J. Yang, F.Y. Zheng, Y.S. Xu, R.Y. Ou, Ovarian cancer initially presenting with isolated ipsilateral superficial inguinal lymph node metastasis: a case study and review of the literature, J Ovarian Res. 7 (2014 Feb 10) 20. [11] https://seer.cancer.gov/data/. [12] Chapter 4: Histological Groups, Cancer Incidence in Five Continents, IARC scientific publications No164, 2014 2014, p. 84. [13] M. Kleppe, A.C. Kraima, R.F. Kruitwagen, et al., Understanding lymphatic drainage pathways of the ovaries to predict sites for sentinel nodes in ovarian cancer, Int. J. Gynecol. Cancer 25 (8) (2015 Oct) 1405–1414. [14] B. Ataseven, P. Harter, C. Grimm, et al., The revised 2014 FIGO staging system for epithelial ovarian cancer: is a subclassification into FIGO stage IVA and IVB justified? Gynecol. Oncol. 142 (2) (2016 Aug) 243–247. [15] B. Ataseven, A. du Bois, P. Harter, et al., Impact of Abdominal Wall metastases on prognosis in epithelial ovarian cancer, Int. J. Gynecol. Cancer 26 (9) (2016 Nov) 1594–1600.
Please cite this article as: D. Nasioudis, et al., Should epithelial ovarian carcinoma metastatic to the inguinal lymph nodes be assigned stage IVB?, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.124
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Should epithelial ovarian carcinoma metastatic to the inguinal lymph nodes be assigned stage IVB? Dimitrios Nasioudis ⁎, Eloise Chapman-Davis, Melissa K. Frey, Thomas A. Caputo, Steven S. Witkin, Kevin Holcomb Department of Obstetrics and Gynecology, Weill Cornell Medicine, New York, NY, USA
H I G H L I G H T S • In a large database, 151 women with EOC and isolated inguinal LN metastases were identified. • Their survival was similar to those with stage III disease and regional nodal involvement. • Our findings do not support the reclassification of these patients as stage IVB.
a r t i c l e
i n f o
Article history: Received 31 May 2017 Received in revised form 8 July 2017 Accepted 10 July 2017 Available online xxxx Keywords: Epithelial ovarian cancer Stage IV New FIGO classification Inguinal lymph nodes Overall survival Prognosis
a b s t r a c t Objective. According to the revised FIGO staging system women with ovarian carcinoma and inguinal lymph node (LN) metastases, formerly stage III, are now considered stage IVB. In this study we compare their survival to that of women with stage III and stage IV disease. Methods. Women diagnosed with epithelial ovarian carcinoma were drawn from the Surveillance, Epidemiology, and End Results database (2004–2013). Four groups were formed: group 1 (stage IV due to positive inguinal nodes), group 2 (stage III with positive para-aortic/pelvic nodes), group 3 (stage IV with positive distant nodes) and group 4 (stage IV with distant metastases). Overall (OS) and cancer-specific survival (CSS) rates were evaluated with the Kaplan-Meier method. The log-rank test and Cox-hazard models were employed for univariate and multivariate survival analysis. Results. A total of 11,152 women were identified. Five-year OS for women in group 1 (n = 151) was 46.3% compared to 44.9% for those in group 2 (n = 4,403) (p = 0.4), 32.9% in group 3 (n = 642) (p b 0.001) and 25.3% in group 4 (n = 5,956) (p b 0.001). After controlling for age, race and histology, group 1 had improved overall and cancer-specific mortality compared to groups 3 and 4 but not group 2. Conclusions. Ovarian cancer patients with stage IV ovarian cancer due solely to inguinal nodal metastases have similar survival as those with pelvic/para-aortic nodal involvement and improved survival compared to those harboring distant metastases. Our findings do not support the reclassification of these patients as stage IVB. © 2017 Elsevier Inc. All rights reserved.
1. Introduction Epithelial ovarian cancer (EOC) is currently the second most common, but most lethal gynecologic malignancy in the United States [1]. The FIGO staging system was developed by the International Federation of Gynecology and Obstetrics in an effort to permit comparisons of outcomes between different cancer centers [2]. It is also designed to adequately reflect disease burden and prognosis of women with EOC [2]. Recently, after almost three decades of use, due to the need for better ⁎ Corresponding author at: Department of Obstetrics and Gynecology, Weill Cornell Medicine, 1300 York Avenue, Box 35, New York, NY 10065, United States. E-mail address: [email protected] (D. Nasioudis).
prognostic discrimination, the FIGO staging system was revised and several changes were introduced [3]. The prognostic role of the revised staging system, which has a larger number of sub-stages, has already been evaluated in several retrospective cohort studies [2,4,5]. Nevertheless, unresolved classification issues remain especially for women with stage IV disease [6]. Currently, due to improved pre-operative imaging techniques and the performance of more comprehensive staging procedures, almost 1 in 5 women (12–21%) are diagnosed with stage IV disease [6]. On the other hand, women with EOC rarely present with isolated inguinal lymph node (LN) metastases, with an estimated incidence of 3–5% [6, 7–10]. Based on the revised FIGO staging system, women with inguinal LN metastasis previously considered as stage IIIC are now assigned to
http://dx.doi.org/10.1016/j.ygyno.2017.07.124 0090-8258/© 2017 Elsevier Inc. All rights reserved.
Please cite this article as: D. Nasioudis, et al., Should epithelial ovarian carcinoma metastatic to the inguinal lymph nodes be assigned stage IVB?, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.124
2
D. Nasioudis et al. / Gynecologic Oncology xxx (2017) xxx–xxx
stage IVB [3]. The prognostic significance of this reclassification has yet to be examined. The aim of this retrospective population-based study was to evaluate the prognosis of women with stage IV EOC due to isolated inguinal LN metastases following cancer-directed surgery and compare it to that of women with stage III disease and pelvic and/or paraaortic LN and those with stage IV disease due to other distant metastases. 2. Material and methods A cohort of women diagnosed between 2004 and 2013 with stage III or stage IV disease (as provided by the SEER database) was drawn from the National Cancer Institute's Surveillance Epidemiology, and End Results (SEER) database [11]. In the present study, data deriving from 18 cancer registries were included (Detroit, Iowa, Kentucky, Louisiana, Utah, Connecticut, New Jersey, Atlanta, Rural and Greater Georgia, Alaska, California, Hawaii, Los Angeles, New Mexico, San Francisco, San Jose, Seattle), as released on April 2016. The SEER database covers approximately 27.8% of the total US population based on the 2010 census and incorporates high-quality data of patients with primary malignant tumors. The ICD-O-3/WHO 2008 site specific code “C.569/ovary” and the ICD-O-3 histology codes as grouped by the International Association for Research on Cancer (IARC) were employed to identify all women with epithelial ovarian cancer [12]. Patient data are de-identified and available to the public for research purposes. Demographic, clinicopathological and survival data were extracted using the “case-listing” option. Information regarding the administration of chemotherapy is not captured in the SEER database. Only women with active follow-up, stage III or IV disease, who underwent cancer-directed surgery (CDS), as assessed by the site-specific surgery codes, were eligible for inclusion in the present study. Based on the information available at the Collaborative Staging (CS) coding fields (CS lymph nodes and CS mets), four groups were formed. Group 1 included women with inguinal LN metastasis and absence of any other distant metastases formerly classified as stage IIIC, group 2 included women with aortic and/or pelvic LN metastases and absence of distant metastases (stage III), group 3 included women with metastases to distant LNs only and group 4 included women with metastases to other distant sites (excluding distant LNs). Women with distant metastases NOS (not otherwise specified) or distant metastases to both distant sites and distant lymph nodes were excluded. Median and 5-year Overall (OS) and Cancer-specific survival (CSS) rates were calculated after generation of Kaplan-Meier curves and comparisons between the different groups were made with the log-rank test. For the estimation of CSS, only those with one tumor or the first of multiple primary tumors were included; women who died from causes other than ovarian cancer were censored. In addition, Cox proportional hazard models were constructed to control for confounders such as patient age, race and tumor histology. Frequency of distribution of categorical and continuous variables was compared with the chisquare test and Mann-Whitney U test, respectively. All statistical analyses were performed with the SPSS v.24 statistical package and the alpha level of statistical significance was set at 0.05. 3. Results A total of 11,152 women met the inclusion criteria. Groups 1 and 2 included 151 (1.4%) and 4403 (39.5%) women, while groups 3 and 4 included 642 (5.8%) and 5956 (53.4%) women respectively. The majority of women included in group 2 had intraperitoneal disease (stage IIIC). Median patient age was 61 years and the majority were White women (84.9%). Women in group 2 were younger (median 59 years) compared to those in group 1 (median 63 years, p b 0.001), group 3 (median 62 years, p b 0.001) and group 4 (median 63 years, p b 0.001). Most tumors were of serous histology (71.%) and based on available information median tumor size was 8.2 cm (n = 7445, range 0.1–
85 cm), while the majority were grade III/IV (84.7%, n = 7687). There were no statistically differences between the different groups in regard to patient race (p = 0.19) and tumor histology (p = 0.25). However, women in group 1 had smaller tumors (n = 108, median size 7 cm) compared to those in group 2 (n = 3332, median size 9 cm) (p b 0.001). Rates of optimal debulking (defined as gross residual tumor nodules of 1 cm or less in size) were 90% (36/40) and 90.4% (1131/ 1251) in groups 1 and 2, respectively, compared to 83% (117/141) and 81.1% (936/1154) in groups 3 and 4, respectively (p b 0.001). According to the reverse Kaplan-Meier method, median follow-up of the cohort was 57 months (95% CI: 55.56, 58.44). Median OS for women in group 1 was 55 months (95% CI: 37.4, 72.6) versus 53 months (95% CI: 50.4, 55.6) for those in group 2 (p = 0.41), 39 months (95% CI: 35.4, 42.6) for those in group 3 (p = 0.001) and 30 months (95% CI: 28.8, 31.2) for those in group 4 (p b 0.001) (Fig. 1). Similarly, CSS for women in group 1 was 66 months versus 59 months (95% CI: 55.0, 63.0) for those in group 2 (p = 0.34), 42 months (95% CI: 37.3, 46.7) for those in group 3 (p b 0.001) and 32 months (95% CI: 30.7, 33.3) for those in group 4 (p b 0.001)(Fig. 2). OS and CSS for women in group 3 was superior to that of women in group 4 (p b 0.001) but inferior to those in group 2 (p b 0.001). Table 1 summarizes 5-year and median OS, CSS in each study group. According to the multivariate Cox-proportional hazard models, after controlling for age (b65 versus ≥ 65 years), race (White versus nonWhite) and histology (serous versus non-serous), women in group 1 had better overall and cancer-specific mortality compared with those in group 3 (HR:0.66, 95% CI: 0.5, 0.85 and HR:0.62, 95% CI:0.46, 0.84 respectively) and group 4 (HR: 0.5, 95% CI: 0.39, 0.64 and HR:0.46, 95% CI: 0.35, 0.61 respectively). However there was no statistically significant difference in overall and cancer-specific mortality between women in group 1 compared to those in group 2. In addition, women in group 3 had better overall and cancer-specific mortality compared to those in group 4 (HR:0.76, 95% CI: 0.69, 0.85 and HR:0.74, 95% CI: 0.66, 0.84 respectively) but worse than those in group 2 (HR: 1.31, 95% CI: 1.18, 1.47 and HR: 1.35, 95% CI: 1.19, 1.52 respectively). Table 2 summarizes the results of the multivariate survival analysis. 4. Discussion In the present study we investigated the prognostic significance of inguinal lymph node metastases in women with EOC. These women comprise a very small group but had improved survival compared to other stage IV patients, harboring distant metastases. While, the main pattern of EOC spread is direct tumor cell seeding within the abdominal cavity, lymphatic dissemination is also likely, especially for serous tumors [13]. Metastases to inguinal LNs (while uncommon) can be observed, presumably via lymphatic drainage through the round ligament of the uterus [13]. The prognostic significance of isolated metastases to inguinal LNs has never been investigated in the past. In our study, women with stage IVB disease due to inguinal LN metastasis had similar prognosis to those with pelvic or para- aortic lymph node involvement, even after controlling for tumor histology, patient age and race. Moreover, in our cohort women with stage IV disease solely due to metastases to distant LNs had superior OS and CSS compared to those with distant metastasis to other sites (excluding inguinal LNs). Similarly to our results, Suh et al. (2015) reported that women with stage IV disease assigned on the basis of metastases to supraclavicular LNs have better overall survival than other stage IV cases and were also less likely to present with multiple metastatic lesions [5]. In the recent revision of the FIGO staging system for ovarian cancer, several changes were introduced [3]. Stage IC has been further subdivided into IC1, IC2, IC3 while stage IIC was abolished. In addition, stage III has been restructured and tumors previously considered as stage IIIC on the basis of metastases to pelvic or para-aortic lymph nodes may now correspond to stage IIIA (if only positive lymph nodes are present) or IIIB (if macroscopic extrapelvic metastasis of size
Please cite this article as: D. Nasioudis, et al., Should epithelial ovarian carcinoma metastatic to the inguinal lymph nodes be assigned stage IVB?, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.124
D. Nasioudis et al. / Gynecologic Oncology xxx (2017) xxx–xxx
3
Fig. 1. Overall Survival of women with epithelial ovarian cancer stratified by study group (n = 11,152, p b 0.001 from log-rank).
≤ 2 cm is also present) [3]. FIGO IV has been further subdivided into stage IVA and IVB which include women with malignant pleural effusion and those with other distant metastases respectively [3]. In addition women with positive inguinal lymph nodes, abdominal wall metastases and transmural or mucosal bowel involvement are now considered as stage IVB disease regardless of initial intra-abdominal tumor spread [3]. Similar to our results other authors have questioned the prognostic value of the revised definition and sub-classification of stage IV disease. More specifically, several studies have failed to demonstrate any
prognostic significance of the stage IV sub-classification into stage IVA and IVB. Rosendahl et al. (2016) reported similar 5-year overall survival rates (13%) for women with stage IVA (n = 149) and stage IVB (n = 613) disease [2]. In their retrospective study, Paik et al. (2015) did not detect any difference in overall and progression free survival between the two substages after controlling for patient age, cell type, grade and residual disease [4]. Atavesen et al. (2016) prospectively compared the overall survival of 102 women with stage IVA to that of 138 women with stage IVB and did not find any statistically significant differences (median OS was 25 and 28 months respectively, as well [14].
Fig. 2. Cancer-specific survival of women with epithelial ovarian cancer stratified by study group (n = 9679, p b 0.001 from log-rank).
Please cite this article as: D. Nasioudis, et al., Should epithelial ovarian carcinoma metastatic to the inguinal lymph nodes be assigned stage IVB?, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.124
4
D. Nasioudis et al. / Gynecologic Oncology xxx (2017) xxx–xxx
Table 1 Median and 5-year (yr) Overall (OS) (p b 0.001, from log-rank) and Cancer-specific survival (CSS) (p b 0.001 from log-rank) of women with stage III/IV epithelial ovarian cancer stratified by study group. OS
Group 1 (stage IV inguinal LNs) Group 2 (stage III aortic/pelvic LNs) Group 3 (stage IV distant LNs) Group 4 (stage IV other distant metastases)
CSS
5-yr/median
5-yr/median
n = 151/46.3%/55 months n = 4403/44.3%/53 months n = 642/32.9%/39 months n = 5956/25.1%/30 months
n = 130/50.4%/66 months n = 3786/48.9%/59 months n = 561/36.3%/42 months n = 5202/27.2%/32 months
Another retrospective study examined the prognosis of 86 women who presented solely with abdominal wall metastases (currently classified as stage IVB disease) and questioned the rationale of upstaging these patients [15]. Authors reported that their overall survival was superior to that of women with distant metastases at other sites (median OS: 58 versus 25 months) [15]. Undoubtedly, there is eminent need of high-quality prospective studies that will clarify the inconsistencies between FIGO staging guidelines and clinical outcomes. As previously outlined, attributing the diagnosis of stage IV disease to a sub-cohort of women who in fact may have a better prognosis can result in an increased emotional burden [15]. For these women access to experimental treatment options or extended surgical procedures may be also limited. A major strength of the present study is that data derive from a population-based multi-institutional database minimizing selection bias. In addition all women were diagnosed with EOC between 2004 and 2013, thus their management was based on current guidelines and may have been less heterogenous. However, several limitations should be noted. Firstly, due to the lack of central pathology review, possible tumor misclassifications could not be excluded. While most women should have received platinum-based adjuvant chemotherapy as outlined in major guidelines, information on its administration and composition was not available. Moreover, we could not assess the extent of surgical staging and residual disease following debulking surgery for all patients. A higher likelihood of achieving R0 may account for the observed survival benefit of women with isolated metastases to inguinal or distant LNs, compared to women with distant metastases at other sites. Moreover, no information was available on the imaging modalities employed in the pre-operative assessment of the patients. Also, while all women underwent a surgical staging procedure, histopathological confirmation of inguinal LN and distant metastases was not available for all cases. Lastly due to the lack of information on the administration of
Table 2 Cox-proportional hazard models of overall and cancer-specific mortality of women with stage III/IV epithelial ovarian cancer.
Group Group 1 (inguinal LNs) Group 2 (aortic/pelvic LNs) Group 3 (distant LNs) Group 4 (distant metastases) Histology Serous Non-serous Race White Non-white Age (years) b65 ≥65
OS (n = 11,152) HR (95% CI)
CSS (n = 9661) HR (95% CI)
Referent 1.16 (0.91, 1.49) 1.52 (1.17, 1.98) 2.0 (1.56. 2.55)
Referent 1.2 (0.9, 1.59) 1.61 (1.19, 2.18) 2.17 (1.64, 2.88)
Referent 1.2 (1.14, 1.27)
Referent 1.17 (1.1, 1.24)
Referent 1.08 (1.01, 1.16)
Referent 1.05 (0.97, 1.14)
Referent 1.48 (1.41, 1.55)
Referent 1.38 (1.31, 1.46)
chemotherapy, we could not exclude women who received neoadjuvant chemotherapy. Due to similar survival rates between women in group 1 and group 2, we did not perform a sub-analysis of women with both inguinal and regional LN metastases. After examining a retrospective cohort drawn from a populationbased database we demonstrated that women with EOC and inguinal LN metastases have similar survival following CDS compared to those with pelvic and para-aortic node involvement and improved survival compared to those harboring distant metastases. Our findings do not support the reclassification of these women as stage IVB. Future studies are greatly warranted to further clarify the prognosis of women with EOC and inguinal LN metastases. Given the very low incidence of inguinal LN metastases, a multi-institutional collaboration could appropriately address this controversy. Conflicts of interest The authors report no conflicts of interest. References [1] D. Jelovac, D.K. Armstrong, Recent progress in the diagnosis and treatment of ovarian cancer, CA Cancer J. Clin. 61 (3) (2011 May–Jun) 183–203. [2] M. Rosendahl, C.K. Høgdall, B.J. Mosgaard, Restaging and survival analysis of 4036 ovarian cancer patients according to the 2013 FIGO classification for ovarian, fallopian tube, and primary peritoneal cancer, Int. J. Gynecol. Cancer 26 (4) (2016 May) 680–687. [3] D.G. Mutch, J. Prat, 2014 FIGO staging for ovarian, fallopian tube and peritoneal cancer, Gynecol. Oncol. 133 (3) (2014 Jun) 401–404. [4] E.S. Paik, Y.Y. Lee, E.J. Lee, et al., Survival analysis of revised 2013 FIGO staging classification of epithelial ovarian cancer and comparison with previous FIGO staging classification, Obstet Gynecol. Sci. 58 (2) (2015 Mar) 124–130. [5] D.H. Suh, T.H. Kim, J.W. Kim, et al., Improvements to the FIGO staging for ovarian cancer: reconsideration of lymphatic spread and intraoperative tumor rupture, J. Gynecol. Oncol. 24 (4) (2013) 352–358. [6] B. Ataseven, L.M. Chiva, P. Harter, A. Gonzalez-Martin, A. du Bois, FIGO stage IV epithelial ovarian, fallopian tube and peritoneal cancer revisited, Gynecol. Oncol. 142 (3) (2016 Sep) 597–607. [7] A.L. Oei, J.A. de Hullu, J.M. Grefte, B.P. Havenith, An enlarged groin node as first manifestation of a malignancy: don't forget the ovaries, Eur. J. Obstet. Gynecol. Reprod. Biol. 138 (2) (2008 Jun) 240–242. [8] D. Ang, K.Y. Ng, H.K. Tan, A.Y. Chung, B.S. Yew, V.K. Lee, Ovarian carcinoma presenting with isolated contralateral inguinal lymph node metastasis: a case report, Ann. Acad. Med. Singap. 36 (6) (2007 Jun) 427–430. [9] H.S. Scholz, S. Lax, K.F. Tamussino, E. Petru, Inguinal lymph node metastasis as the only manifestation of lymphatic spread in ovarian cancer: a case report, Gynecol. Oncol. 75 (3) (1999 Dec) 517–518. [10] X.J. Yang, F.Y. Zheng, Y.S. Xu, R.Y. Ou, Ovarian cancer initially presenting with isolated ipsilateral superficial inguinal lymph node metastasis: a case study and review of the literature, J Ovarian Res. 7 (2014 Feb 10) 20. [11] https://seer.cancer.gov/data/. [12] Chapter 4: Histological Groups, Cancer Incidence in Five Continents, IARC scientific publications No164, 2014 2014, p. 84. [13] M. Kleppe, A.C. Kraima, R.F. Kruitwagen, et al., Understanding lymphatic drainage pathways of the ovaries to predict sites for sentinel nodes in ovarian cancer, Int. J. Gynecol. Cancer 25 (8) (2015 Oct) 1405–1414. [14] B. Ataseven, P. Harter, C. Grimm, et al., The revised 2014 FIGO staging system for epithelial ovarian cancer: is a subclassification into FIGO stage IVA and IVB justified? Gynecol. Oncol. 142 (2) (2016 Aug) 243–247. [15] B. Ataseven, A. du Bois, P. Harter, et al., Impact of Abdominal Wall metastases on prognosis in epithelial ovarian cancer, Int. J. Gynecol. Cancer 26 (9) (2016 Nov) 1594–1600.
Please cite this article as: D. Nasioudis, et al., Should epithelial ovarian carcinoma metastatic to the inguinal lymph nodes be assigned stage IVB?, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.124