- Email: [email protected]
Clinical significance of enlarged cardiophrenic lymph nodes in advanced ovarian cancer: Implications for survival
YGYNO-976929; No. of pages: 6; 4C: Gynecologic Oncology xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Clinical significance of enlarged cardiophrenic lymph nodes in advanced ovarian cancer: Implications for survival Ismail Mert a, Amanika Kumar a, Shannon P. Sheedy b, Amy L. Weaver c, Michaela E. McGree c, Bohyun Kim b, William A. Cliby a,⁎ a b c
Department of Gynecologic Oncology, Mayo Clinic, Rochester, MN, USA Department of Diagnostic Radiology, Mayo Clinic, Rochester, MN, USA Division of Biomedical Statistics and Informatics, Rochester, MN, USA
H I G H L I G H T S • Enlarged cardiophrenic lymph nodes (CPLN) can be diagnosed with preoperative CT. • Enlarged CPLNs are associated with worse survival in patient with no gross RD. • The impact on survival associated with enlarged CPLNs is dependent on RD elsewhere.
a r t i c l e
i n f o
Article history: Received 11 September 2017 Received in revised form 21 October 2017 Accepted 23 October 2017 Available online xxxx
a b s t r a c t Objective. Advanced ovarian cancer (OC) commonly spreads to cardiophrenic lymph nodes (CPLNs), and is often visible on preoperative imaging. We investigated the prognostic significance of abnormal CPLNs in OC detected by preoperative CT scans using three different definitions. Methods. Patients undergoing primary debulking surgery for stage IIIC/IV with residual disease (RD) ≤1.0 cm and a preoperative abdominopelvic CT scan available were included. Scans were reviewed by two blinded radiologists. We characterized abnormal CPLNs using three different definitions: i) qualitative assessment score (QAS); ii) nodes N7 mm on the short axis; or, iii) nodes ≥10 mm on the short axis. We compared overall survival (OS) using the log-rank test. Results. Of the 253 patients (mean age 64.0 years), 136 had no gross residual disease (NGR) and 117 had RD. By the QAS definition, CPLNs were abnormal in 28 (11.1%) patients and removed in one case. Among patients with NGR, presence of abnormal CPLNs was associated with worse OS (median OS, 38.4 vs. 69.6 months, p = 0.08). We observed no association between abnormal CPLNs and OS among patients with RD (median OS, 37.5 vs. 28.5 months, p = 0.49). OS was significantly better in NGR group without abnormal CPLNs (median OS for NGR vs. RD, 69.6 vs. 28.5 months, p b 0.001); however, there was no difference in OS between patients with NGR versus RD when abnormal CPLNs were present (median OS, 38.4 vs. 37.5 months, p = 0.99). Lack of benefit from NGR when abnormal CPLNs were present was observed for all three definitions tested. Conclusion. Abnormal CPLNs are an important predictor of survival in advanced stage OC. Management of abnormal CPLNs should be considered in treatment planning when the goal is NGR. © 2017 Published by Elsevier Inc.
1. Introduction Surgery remains a mainstay in the treatment of ovarian cancer (OC), with surgical goals evolving over time as more information is available on risk stratification, identification of groups most or least likely to benefit from primary surgery, and identification of specific sites requiring alternative approaches. A recent statement by the Society of ⁎ Corresponding author at: Mayo Clinic, Rochester, MN 55905, USA. E-mail address: [email protected] (W.A. Cliby).
Gynecologic Oncology and American Society of Clinical Oncology reiterated the goal of resection to no gross residual disease (NGR) when possible and at a minimum to 1 cm citing that postoperative disease burden is strongly associated with survival. This shift in focus to strive for NGR has led to improved rates of NGR resections and has triggered incorporation of new surgical algorithms to achieve higher rates of NGR during upfront surgery [1–3] or triage to neoadjuvant chemotherapy (NACT) when appropriate. Although, the definition of NGR and its benefits is well studied within the abdominal compartment, the prognostic impact and the
https://doi.org/10.1016/j.ygyno.2017.10.024 0090-8258/© 2017 Published by Elsevier Inc.
Please cite this article as: I. Mert, et al., Clinical significance of enlarged cardiophrenic lymph nodes in advanced ovarian cancer: Implications for survival, Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.10.024
2
I. Mert et al. / Gynecologic Oncology xxx (2017) xxx–xxx
management of disease outside of the abdomen remain in question. Currently there is an increasing awareness of the frequency of abnormal cardiophrenic lymph nodes (CPLNs) and their association with worse survival. CPLNs (often referred to as supradiaphragmatic or pericardial lymph nodes) belong to a group of parietal thoracic lymph nodes [4] and likely drain from some portions of the diaphragm, liver, pleura, pericardium, and anterior abdominal wall; they are often considered to be a principal nodal station for the lymphatic drainage of the whole abdominal cavity [5–7]. Enlargement of CPLNs is reported in 15–28% of the OC patients using computerized tomography (CT) scan across different definitions [4,8]. Removal of enlarged or suspicious abdominal retroperitoneal lymph nodes is part of routine cytoreduction when NGR can be achieved, but management strategies of abnormal CPLNs are just beginning to emerge. An effective strategy requires the use of a definition of abnormal CPLNs that is sensitive and specific. Prader et al. showed 90% sensitivity for metastasis when ≥10 mm on short axis in preoperative CT scan was utilized in patients with OC [8]. On the other hand, Kim et al. demonstrated N7 mm on short axis showed the best diagnostic performance for predicting positive CPLNs [9]. To aid in this evolution in the management of CPLNs, we investigated the prognostic significance of abnormal CPLNs detected on preoperative CT scan. We analyzed three different definitions of abnormal CPLNs in current use and their respective outcomes. 2. Methods This is a single institution retrospective cohort study approved by the Mayo Clinic Institutional Review Board. Patients were identified from a prospectively maintained surgical database of patients who
have underwent primary debulking surgery (PDS) for stage IIIC or IV ovarian, fallopian tube, or primary peritoneal cancer from 1/2/2003– 12/30/2011. Patients who underwent surgery with palliative intent or received neoadjuvant chemotherapy, or who had denied access to their medical records for research were excluded. The inclusion criteria for this study was a preoperative abdominopelvic CT scan that included the lower chest within 90 days prior to surgery and residual disease at the end of PDS of 0 to 1 cm. Scans were reviewed by two blinded radiologists. All the nodes in the thorax were measured in two perpendicular axes. We used three definitions of abnormal CPLNs. First, we performed a qualitative assessment of the lymph nodes based on size, heterogeneity, and architecture (i.e. irregular border, round in shape, heterogeneous internal attenuation) using a 5point scale for certainty that a lesion represents a metastatic neoplasm, defined as: 1 definitely normal; 2 probably normal; 3 indeterminate; 4 probably abnormal; 5 definitely abnormal [10]. We defined a patient as having abnormal CPLNs if they had a qualitative assessment score (QAS) of 4 or 5. The other two definitions used a simple definition based on length of the short axis of the CPLN, one using N 7 mm [9] and one using ≥ 10 mm [8] based on previous findings. Presence of abnormal CPLNs seen on preoperative CT scans was not considered a criterion for stage IV disease in this study, nor was the CPLNs considered in defining residual disease (RD). No gross residual disease was defined as no visible disease at the end of PDS and gross residual disease was defined as any measurable RD less than or equal to 1 cm disease after debulking surgery. Patients with residual disease N 1 cm were excluded. Surgical complexity was defined using previously published criteria [11] and divided into three categories, low, intermediate, and high complexity surgery.
Table 1 Patient and tumor characteristics by presence of abnormal CPLNs where abnormal CPLNs defined as QAS of 4 or 5. Characteristic
Age (years), mean (SD) Primary site of disease, N (%) Ovary Peritoneal Fallopian tube FIGO stage, N (%) IIIC IV FIGO grade, N (%) 1 2 3 Unknown Histology, N (%) Non-serous Serous Preoperative CA-125 (U/mL), median (IQR) ASA score, N (%) b3 ≥3 Residual disease, N (%) 0 0.1–1 cm Surgical complexity, N (%) Low Intermediate High Adjuvant chemotherapy, N (%) No Yes Unknown Accordion grade 3+ 30-day postoperative complications, N (%) No Yes
Total
Abnormal CPLNs not present
Abnormal CPLNs present
N = 253
N = 225
N = 28
64.0 (11.0)
64.5 (11.1)
59.6 (9.3)
172 (68.0) 65 (25.7) 16 (6.3)
149 (66.2) 61 (27.1) 15 (6.7)
23 (82.1) 4 (14.3) 1 (3.6)
192 (75.9) 61 (24.1)
176 (78.2) 49 (21.8)
16 (57.1) 12 (42.9)
6 (2.4) 5 (2.0) 238 (94.1) 4 (1.6)
6 (2.7) 5 (2.2) 210 (93.3) 4 (1.8)
0 (0.0) 0 (0.0) 28 (100.0) 0 (0.0)
36 (14.2) 217 (85.8) 729 (254, 1730)
32 (14.2) 193 (85.8) 727 (235, 1744)
4 (14.3) 24 (85.7) 730 (450, 1720)
133 (52.6) 120 (47.4)
113 (50.2) 112 (49.8)
20 (71.4) 8 (28.6)
136 (53.8) 117 (46.2)
124 (55.1) 101 (44.9)
12 (42.9) 16 (57.1)
17 (6.7) 128 (50.6) 108 (42.7)
15 (6.7) 116 (51.6) 94 (41.8)
2 (7.1) 12 (42.9) 14 (50.0)
11 (4.3) 222 (87.7) 20 (7.9)
11 (4.9) 196 (87.1) 18 (8.0)
0 (0.0) 26 (92.9) 2 (7.1)
191 (75.5) 62 (24.5)
170 (75.6) 55 (24.4)
21 (75.0) 7 (25.0)
P⁎
0.03 0.23
0.01
0.99
0.99
0.28 0.03
0.22
0.68
0.71
0.95
Abbreviations: ASA, American Society of Anesthesiologists; CA-125, cancer antigen 125; FIGO, International Federation of Gynecology and Obstetrics; IQR, interquartile range; SD, standard deviation. ⁎ Chi-square or Fisher's exact p value reported for categorical variables, t-test p value reported for age, and Wilcoxon rank-sum p value reported for preoperative CA-125.
Please cite this article as: I. Mert, et al., Clinical significance of enlarged cardiophrenic lymph nodes in advanced ovarian cancer: Implications for survival, Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.10.024
I. Mert et al. / Gynecologic Oncology xxx (2017) xxx–xxx
Statistical analyses were done using SAS version 9.3 (SAS Institute, Inc.; Cary, NC). Comparisons of baseline patient and tumor characteristics between patients with and without abnormal CPLNs were evaluated using the two-sample t-test for age, Wilcoxon rank-sum for preoperative CA-125, and chi-square or Fisher's exact test for categorical variables. Overall survival following PDS was calculated using the KaplanMeier method and compared between groups using the log-rank test. Associations with overall survival were summarized using the hazard ratio (HR) and corresponding 95% confidence interval (CI) estimated from Cox proportional hazards models. All calculated p values were two-sided and p values b 0.05 were considered statistically significant.
3. Results Between January 2003 and December 2011, 617 patients underwent PDS with advanced stage disease (stage IIIC or IV). Among patients with advanced stage disease, 276 patients had complete CT images available and research authorization for review. Twenty-three of the 276 patients were excluded due to RD N1 cm leaving us with 253 patients for the analysis. We observed abnormal CPLNs with QAS of 4 or 5 in 28 (11.1%) patients. Removal of CPLNs was not routinely performed at PDS during this time frame and only one out of 28 patients had CPLNs resected at PDS.
3
Baseline patient and tumor characteristics of the cohort, overall and stratified by presence of abnormal CPLNs defined by QAS of 4 or 5, are listed in Table 1. There were 192 patients (75.9%) with stage IIIC disease and the majority had serous histology. We observed no difference in primary site of the disease, FIGO grade, histology, preoperative CA-125 level, residual disease, adjuvant chemotherapy, accordion grade 3 + 30-day postoperative complication rate and surgical complexity between the groups with and without abnormal CPLNs based on QAS (all p values N 0.05, Table 1). Women with abnormal CPLNs were younger (mean age, 59.6 vs. 64.5 years, p = 0.03), and were more likely to have stage IV disease (p = 0.01). For patients with abnormal CPLNs, stage IV disease was due to a combination of parenchymal involvement of the liver and spleen (n = 3), malignant pleural effusion or pleural based disease (n = 6), rectal wall involvement (n = 1) or organ involvement outside of the abdomen (n = 3). Among patients with abnormal CPLNs, 28.6% had an American Society of Anesthesiologists (ASA) score of ≥ 3 compared to 49.8% of women without abnormal CPLNs (p = 0.03). We sought to evaluate the impact of abnormal CPLNs on OS stratified by RD. At PDS, 136 (53.8%) patients had NGR; 12 (8.8%) of the NGR group had abnormal CPLNs. Among the patients with NGR, the percentage with high surgical complexity was higher among patients with versus without abnormal CPLNs (41.7% (5/12) vs. 28.2% (35/124)), although not statistically significant (p = 0.34). Furthermore, among
Fig. 1. Comparison of overall survival by presence/absence of residual disease and presence/absence of abnormal CPLNs (defined as QAS 4 or 5).
Please cite this article as: I. Mert, et al., Clinical significance of enlarged cardiophrenic lymph nodes in advanced ovarian cancer: Implications for survival, Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.10.024
4
I. Mert et al. / Gynecologic Oncology xxx (2017) xxx–xxx
the patients with NGR, OS tended to be worse for patients with abnormal CPLNs, specifically, the median OS was 38.4 months in patients with abnormal CPLNs versus 69.6 months in patients without abnormal CPLNs (p = 0.08, Fig. 1a). The unadjusted hazard ratio (HR) for the association between presence of abnormal CPLNs and OS among patients with NGR was 1.84 (95% CI: 0.91–3.72) and after adjusting for stage, the adjusted HR was 1.61 (95% CI: 0.77–3.35). Of 117 patients with gross RD 0.1–1.0 cm, 16 (13.7%) had abnormal CPLNs. Among the patients with gross RD, we observed no significant difference in OS between patients with versus without abnormal CPLNs (median OS, 37.5 vs. 28.5 months, p = 0.49, Fig. 1b). In the absence of abnormal CPLNs, NGR was significantly associated with improved OS (median OS, 69.6 vs. 28.5 months, respectively, NGR group vs. RD group; p b 0.001; Fig. 1c). However, in the presence of abnormal CPLNs, the survival benefit of NGR was diminished (median OS, 38.4 versus 37.5 months, respectively, NGR group vs. RD group: p = 0.99; Fig. 1d). We evaluated two published criteria for abnormal CPLNs such that a patient was defined as having abnormal CPLNs if the length of the short axis of any CPLN was a) N7 mm or b) ≥10 mm. Upon using N7 mm measured in short axis to define abnormal CPLNs, 53 (20.9%) patients had abnormal CPLNs. When patients with abnormal CPLNs based on N7 mm on the short axis were compared to patients without abnormal
CPLNs, there was no difference in age, primary site of disease, FIGO stage, FIGO grade, histology, ASA score, or surgical complexity scores, adjuvant chemotherapy and accordion grade 3+ 30-day postoperative complication rate but a significant difference was observed in preoperative CA-125 levels and residual disease (Supplementary Table 1). Looking at the association with survival starting with patients with NGR, median OS was 69.6 months in patients with no abnormal CPLNs compared to 35.5 months in patients with abnormal CPLNs (p = 0.18, Fig. 2a). Among patients with gross RD 0.1–1.0 cm, similar to the QAS definition, presence of abnormal CPLNs was not associated with OS (median OS, 31.7 months for patients without abnormal CPLNs vs. 24.5 months for patients with abnormal CPLNs; p = 0.88, Fig. 2b). Among patients without abnormal CPLNs, patients resected to NGR had significantly better OS compared to those with gross RD 0.1– 1.0 cm (median OS, 69.6 vs. 31.7 months, respectively; p b 0.001; Fig. 2c). However, among patients with abnormal CPLNs, for patients resected to NGR compared to patients with gross RD 0.1–1.0 cm the difference in OS did not reach statistical significance (median OS, 35.5 versus 24.5 months, respectively; p = 0.16; Fig. 2d). Finally, when abnormal CPLNs were defined as ≥10 mm on the short axis, only 20 (7.9%) patients met criteria for abnormal CPLNs and none of the patient or tumor characteristics were different between patients with or without abnormal CPLNs (Supplementary Table 2). NGR was
Fig. 2. Comparison of overall survival by presence/absence of residual disease and presence/absence of abnormal CPLNs (defined as N7 mm on the short axis).
Please cite this article as: I. Mert, et al., Clinical significance of enlarged cardiophrenic lymph nodes in advanced ovarian cancer: Implications for survival, Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.10.024
I. Mert et al. / Gynecologic Oncology xxx (2017) xxx–xxx
significantly associated with improved OS among patients without abnormal CPLNs (median OS for NGR vs. gross RD, 61.5 vs. 29.0 months; p b 0.001; Fig. 3c). However, the number of patients with abnormal CPLNs was too few to meaningfully compare the OS within this stratum (Fig. 3d).
4. Discussion In the present study, we add to the growing literature collectively suggesting that the presence of abnormal CPLNs worsens survival in advanced OC. Specifically, we demonstrate a trend toward worse survival in OC patients with NGR after PDS when abnormal CPLNs are present (versus absent) using a QAS of 4 or 5 to define abnormal CPLNs (HR 1.84, 95% CI 0.91–3.72). This effect was not seen among those with residual disease up to 1 cm which is consistent with the current paradigm around RD and outcomes. These findings imply that abnormal CPLNs are surrogates for worse disease and have important implications for treatment planning when identified on preoperative imaging. Among patients without abnormal CPLNs, NGR was significantly associated with improved OS compared to gross RD. However, when abnormal CPLNs were present the OS was similar in those with NGR versus gross RD.
5
The optimal size cut-off for pathologic CPLNs is unknown. Several studies demonstrate no difference in outcomes when using a cut-off of 5 mm [4,12] while others have used N7 mm [9] or ≥ 10 mm [8]. We used predefined criteria published in large prospective radiologic studies of patients undergoing PDS. When we used N 7 or ≥10 mm on short axis for the definition of abnormal CPLNs without considering a qualitative score, patients with NGR and without abnormal CPLNs had better survival outcomes compared to patients with NGR and with abnormal CPLNs consistent with those previous reports. Furthermore, in addition to enlargement, morphologic evaluation of the node is also a valuable parameter for the assessment of these nodes. Reproducibility of these criteria will need to be tested in other studies. Little is known regarding benefit of lymphadenectomy outside of the abdominal cavity in patients with advanced stage OC. The Lymphadenectomy in Ovarian Neoplasms (LION) trial was designed to assess pelvic and paraaortic lymphadenectomy in advanced stage OC after intraabdominal complete debulking has been completed and did not show any survival benefit with systemic lymphadenectomy [13]. On the other hand, enlarged or abnormal retroperitoneal lymph node removal is standard in the surgical management of advanced stage OC. At this time, how abnormal CPLNs portend worse OS is not clear. We observed a clinically, strong association between the presence of abnormal/enlarged CPLNs and OS only in patients with NGR disease across
Fig. 3. Comparison of overall survival by presence/absence of residual disease and presence/absence of abnormal CPLNs (defined as ≥10 mm on the short axis).
Please cite this article as: I. Mert, et al., Clinical significance of enlarged cardiophrenic lymph nodes in advanced ovarian cancer: Implications for survival, Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.10.024
6
I. Mert et al. / Gynecologic Oncology xxx (2017) xxx–xxx
all definitions. This might be due to higher number of stage IV disease (66.7% stage IV if abnormal CPLNs were present based on QAS 4 or 5 versus 17.7% if not, p b 0.001), though high surgical complexity was not different in the NGR group when abnormal CPLNs were present versus not (41.7% vs. 28.2%, respectively, p = 0.65). Worse survival associated with abnormal CPLNs might be due to tumor biology that enables some ovarian cancer cells a better vascular permeation and metastasis skills and cause a spread in other compartments of the body such as chest but this would require further molecular experiments to confirm. Removing abnormal CPLNs has not routinely been advocated or discussed during PDS; therefore, there is limited data examining the morbidity associated with surgical resection during abdominal surgery. Several surgical techniques have been described to remove CPLNs. Lim et al. removed CPLNs using video-assisted thoracic surgery (VATS) through a 2 cm incision made in the fifth intercostal space at the anterior axillary line [14]. Complete cytoreduction was achieved in all patients and median time required for VATS was 65 min and median estimated blood loss was 20 mL. Median number of harvested lymph nodes was three [14]. In a series of 11 patients, transdiaphragmatic approach for resection of CPLNs was performed [15]. With this technique, mobilization of liver enabled full exposure of the diaphragm which was incised in order to explore the pleural cavity and CPLNs [15]. The average surgery time was 385 min and average blood loss was 564 mL, median number of postoperative hospital days was 7, and overall postoperative surgical morbidity associated with removal of abnormal CPLNs was low [15]. The most common complication was small pleural effusion detected on chest X-ray followed by small apical pneumothorax, comparable to other procedures involving the diaphragm during PDS [15]. Cowan et al. showed % 7 complication rate associated with excision of the CPLNs and removal of CPLNs did not delay chemotherapy [16]. Prader and Yoo et al. also removed abnormal CPLNs through the diaphragm after mobilization of the liver and there were no major complications observed associated with the removal of CPLNs [8,17]. These results showed that in expert hands, resection of abnormal CPLNs carries minor risk to the patients with minimal morbidity and mortality. Our study has limitations commonly associated with retrospective studies. Given the rarity of abnormal CPLNs in OC, statistical power was limited for some categories. During the years of study we did not routinely remove CPLNs, so did not have pathologic verification of the involvement of the abnormal nodes, relying instead on existing studies for cut-off values of pathologic nodes. There were some imbalances with respect to poor prognostic factors in QAS and short axis N 7 mm definitions but the same trend toward worse survival associated with presence of enlarged/abnormal CPLNs was seen in NGR group across all definitions including ≥10 mm short axis model in which these prognostic factors were comparable (HRs of 1.84, 1.51 and 1.73 in Figs. 1a, 2a, and 3a, respectively). Strengths include using patients treated at a single institution with a relatively standard approach to advanced OC and two radiologists experienced in gynecologic imaging evaluated all the CT scans. Prospective studies with larger sample size will be needed to elucidate the clinical significance enlarged/abnormal CPLNs. In conclusion, presence of abnormal CPLNs appears to have an important effect on OS in patients with advanced OC. This is supported by the observed impact on survival in the cohort of patient with no other gross residual disease remaining (HR = 1.61 after adjusting for stage). Our results show that despite achieving NGR in the abdomen, patients with abnormal/ enlarged CPLNs had worse survival compared the patient without abnormal CPLNs. Furthermore, the median OS of women with abnormal CPLNs and NGR was similar to patients who
had RD 0.1–1.0 cm after PDS surgery. It is not clear if resection of CPLNs will improve survival when NGR is reached. However, given the minimal morbidity of the procedure and recent data suggesting ideal management of lymph nodes in the retroperitoneum should be to remove enlarged nodes, this seems reasonable. Collectively these findings support a clinical trial to investigate feasibility, safety, and impact on outcome of treatment of women with abnormal CPLNs. Preoperative imaging with CT scan is an easy and widely utilized modality for the management of OC and may guide gynecologic oncologists in approaching to patients with advanced stage OC with abnormal CPLNs. All authors declare no conflict of interest. Supplementary data to this article can be found online at https://doi. org/10.1016/j.ygyno.2017.10.024.
References [1] D.S. Chi, E.L. Eisenhauer, O. Zivanovic, Y. Sonoda, N.R. Abu-Rustum, D.A. Levine, et al., Improved progression-free and overall survival in advanced ovarian cancer as a result of a change in surgical paradigm, Gynecol. Oncol. 114 (2009) 26–31. [2] P. Harter, Z.M. Muallem, C. Buhrmann, D. Lorenz, C. Kaub, R. Hils, et al., Impact of a structured quality management program on surgical outcome in primary advanced ovarian cancer, Gynecol. Oncol. 121 (2011) 615–619. [3] S. Wallace, A. Kumar, M. Mc Gree, A. Weaver, A. Mariani, C. Langstraat, et al., Efforts at maximal cytoreduction improve survival in ovarian cancer patients, even when complete gross resection is not feasible, Gynecol. Oncol. 145 (1) (2017) 21–26. [4] B.J. Holloway, M.E. Gore, R.P. A'Hern, C. Parsons, The significance of paracardiac lymph node enlargement in ovarian cancer, Clin. Radiol. 52 (1997) 692–697. [5] G.B. Feldman, Lymphatic obstruction in carcinomatous ascites, Cancer Res. 35 (1975) 325–332. [6] M.F. Abu-Hijleh, O.A. Habbal, S.T. Moqattash, The role of the diaphragm in lymphatic absorption from the peritoneal cavity, J. Anat. 186 (Pt 3) (1995) 453–467. [7] J. Hynninen, A. Auranen, O. Carpen, K. Dean, M. Seppanen, J. Kemppainen, et al., FDG PET/CT in staging of advanced epithelial ovarian cancer: frequency of supradiaphragmatic lymph node metastasis challenges the traditional pattern of disease spread, Gynecol. Oncol. 126 (2012) 64–68. [8] S. Prader, P. Harter, C. Grimm, A. Traut, K.U. Waltering, P.F. Alesina, et al., Surgical management of cardiophrenic lymph nodes in patients with advanced ovarian cancer, Gynecol. Oncol. 141 (2016) 271–275. [9] T.H. Kim, M.C. Lim, S.I. Kim, S.S. Seo, S.H. Kim, S.Y. Park, Preoperative prediction of cardiophrenic lymph node metastasis in advanced ovarian cancer using computed tomography, Ann. Surg. Oncol. 23 (2016) 1302–1308. [10] R.S. Suidan, P.T. Ramirez, D.M. Sarasohn, J.B. Teitcher, S. Mironov, R.B. Iyer, et al., A multicenter prospective trial evaluating the ability of preoperative computed tomography scan and serum CA-125 to predict suboptimal cytoreduction at primary debulking surgery for advanced ovarian, fallopian tube, and peritoneal cancer, Gynecol. Oncol. 134 (2014) 455–461. [11] D.S. Chi, F. Musa, F. Dao, O. Zivanovic, Y. Sonoda, M.M. Leitao, et al., An analysis of patients with bulky advanced stage ovarian, tubal, and peritoneal carcinoma treated with primary debulking surgery (PDS) during an identical time period as the randomized EORTC-NCIC trial of PDS vs neoadjuvant chemotherapy (NACT), Gynecol. Oncol. 124 (2012) 10–14. [12] V. Kolev, S. Mironov, O. Mironov, N. Ishill, C.S. Moskowitz, G.J. Gardner, et al., Prognostic significance of supradiaphragmatic lymphadenopathy identified on preoperative computed tomography scan in patients undergoing primary cytoreduction for advanced epithelial ovarian cancer, Int. J. Gynecol. Cancer 20 (2010) 979–984. [13] https://clinicaltrials.gov/ct2/show/NCT00712218. [14] M.C. Lim, H.S. Lee, D.C. Jung, J.Y. Choi, S.S. Seo, S.Y. Park, Pathological diagnosis and cytoreduction of cardiophrenic lymph node and pleural metastasis in ovarian cancer patients using video-assisted thoracic surgery, Ann. Surg. Oncol. 16 (2009) 1990–1996. [15] C.J. LaFargue, B.T. Sawyer, R.E. Bristow, Short-term morbidity in transdiaphragmatic cardiophrenic lymph node resection for advanced stage gynecologic cancer, Gynecol. Oncol. Rep. 17 (2016) 33–37. [16] R.A. Cowan, J. Tseng, V. Murthy, R. Srivastava, K.C. Long Roche, O. Zivanovic, et al., Feasibility, safety and clinical outcomes of cardiophrenic lymph node resection in advanced ovarian cancer, Gynecol. Oncol. 147 (2) (2017) 262–266. [17] H.J. Yoo, M.C. Lim, Y.J. Song, Y.S. Jung, S.H. Kim, C.W. Yoo, et al., Transabdominal cardiophrenic lymph node dissection (CPLND) via incised diaphragm replace conventional video-assisted thoracic surgery for cytoreductive surgery in advanced ovarian cancer, Gynecol. Oncol. 129 (2013) 341–345.
Please cite this article as: I. Mert, et al., Clinical significance of enlarged cardiophrenic lymph nodes in advanced ovarian cancer: Implications for survival, Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.10.024
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Clinical significance of enlarged cardiophrenic lymph nodes in advanced ovarian cancer: Implications for survival Ismail Mert a, Amanika Kumar a, Shannon P. Sheedy b, Amy L. Weaver c, Michaela E. McGree c, Bohyun Kim b, William A. Cliby a,⁎ a b c
Department of Gynecologic Oncology, Mayo Clinic, Rochester, MN, USA Department of Diagnostic Radiology, Mayo Clinic, Rochester, MN, USA Division of Biomedical Statistics and Informatics, Rochester, MN, USA
H I G H L I G H T S • Enlarged cardiophrenic lymph nodes (CPLN) can be diagnosed with preoperative CT. • Enlarged CPLNs are associated with worse survival in patient with no gross RD. • The impact on survival associated with enlarged CPLNs is dependent on RD elsewhere.
a r t i c l e
i n f o
Article history: Received 11 September 2017 Received in revised form 21 October 2017 Accepted 23 October 2017 Available online xxxx
a b s t r a c t Objective. Advanced ovarian cancer (OC) commonly spreads to cardiophrenic lymph nodes (CPLNs), and is often visible on preoperative imaging. We investigated the prognostic significance of abnormal CPLNs in OC detected by preoperative CT scans using three different definitions. Methods. Patients undergoing primary debulking surgery for stage IIIC/IV with residual disease (RD) ≤1.0 cm and a preoperative abdominopelvic CT scan available were included. Scans were reviewed by two blinded radiologists. We characterized abnormal CPLNs using three different definitions: i) qualitative assessment score (QAS); ii) nodes N7 mm on the short axis; or, iii) nodes ≥10 mm on the short axis. We compared overall survival (OS) using the log-rank test. Results. Of the 253 patients (mean age 64.0 years), 136 had no gross residual disease (NGR) and 117 had RD. By the QAS definition, CPLNs were abnormal in 28 (11.1%) patients and removed in one case. Among patients with NGR, presence of abnormal CPLNs was associated with worse OS (median OS, 38.4 vs. 69.6 months, p = 0.08). We observed no association between abnormal CPLNs and OS among patients with RD (median OS, 37.5 vs. 28.5 months, p = 0.49). OS was significantly better in NGR group without abnormal CPLNs (median OS for NGR vs. RD, 69.6 vs. 28.5 months, p b 0.001); however, there was no difference in OS between patients with NGR versus RD when abnormal CPLNs were present (median OS, 38.4 vs. 37.5 months, p = 0.99). Lack of benefit from NGR when abnormal CPLNs were present was observed for all three definitions tested. Conclusion. Abnormal CPLNs are an important predictor of survival in advanced stage OC. Management of abnormal CPLNs should be considered in treatment planning when the goal is NGR. © 2017 Published by Elsevier Inc.
1. Introduction Surgery remains a mainstay in the treatment of ovarian cancer (OC), with surgical goals evolving over time as more information is available on risk stratification, identification of groups most or least likely to benefit from primary surgery, and identification of specific sites requiring alternative approaches. A recent statement by the Society of ⁎ Corresponding author at: Mayo Clinic, Rochester, MN 55905, USA. E-mail address: [email protected] (W.A. Cliby).
Gynecologic Oncology and American Society of Clinical Oncology reiterated the goal of resection to no gross residual disease (NGR) when possible and at a minimum to 1 cm citing that postoperative disease burden is strongly associated with survival. This shift in focus to strive for NGR has led to improved rates of NGR resections and has triggered incorporation of new surgical algorithms to achieve higher rates of NGR during upfront surgery [1–3] or triage to neoadjuvant chemotherapy (NACT) when appropriate. Although, the definition of NGR and its benefits is well studied within the abdominal compartment, the prognostic impact and the
https://doi.org/10.1016/j.ygyno.2017.10.024 0090-8258/© 2017 Published by Elsevier Inc.
Please cite this article as: I. Mert, et al., Clinical significance of enlarged cardiophrenic lymph nodes in advanced ovarian cancer: Implications for survival, Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.10.024
2
I. Mert et al. / Gynecologic Oncology xxx (2017) xxx–xxx
management of disease outside of the abdomen remain in question. Currently there is an increasing awareness of the frequency of abnormal cardiophrenic lymph nodes (CPLNs) and their association with worse survival. CPLNs (often referred to as supradiaphragmatic or pericardial lymph nodes) belong to a group of parietal thoracic lymph nodes [4] and likely drain from some portions of the diaphragm, liver, pleura, pericardium, and anterior abdominal wall; they are often considered to be a principal nodal station for the lymphatic drainage of the whole abdominal cavity [5–7]. Enlargement of CPLNs is reported in 15–28% of the OC patients using computerized tomography (CT) scan across different definitions [4,8]. Removal of enlarged or suspicious abdominal retroperitoneal lymph nodes is part of routine cytoreduction when NGR can be achieved, but management strategies of abnormal CPLNs are just beginning to emerge. An effective strategy requires the use of a definition of abnormal CPLNs that is sensitive and specific. Prader et al. showed 90% sensitivity for metastasis when ≥10 mm on short axis in preoperative CT scan was utilized in patients with OC [8]. On the other hand, Kim et al. demonstrated N7 mm on short axis showed the best diagnostic performance for predicting positive CPLNs [9]. To aid in this evolution in the management of CPLNs, we investigated the prognostic significance of abnormal CPLNs detected on preoperative CT scan. We analyzed three different definitions of abnormal CPLNs in current use and their respective outcomes. 2. Methods This is a single institution retrospective cohort study approved by the Mayo Clinic Institutional Review Board. Patients were identified from a prospectively maintained surgical database of patients who
have underwent primary debulking surgery (PDS) for stage IIIC or IV ovarian, fallopian tube, or primary peritoneal cancer from 1/2/2003– 12/30/2011. Patients who underwent surgery with palliative intent or received neoadjuvant chemotherapy, or who had denied access to their medical records for research were excluded. The inclusion criteria for this study was a preoperative abdominopelvic CT scan that included the lower chest within 90 days prior to surgery and residual disease at the end of PDS of 0 to 1 cm. Scans were reviewed by two blinded radiologists. All the nodes in the thorax were measured in two perpendicular axes. We used three definitions of abnormal CPLNs. First, we performed a qualitative assessment of the lymph nodes based on size, heterogeneity, and architecture (i.e. irregular border, round in shape, heterogeneous internal attenuation) using a 5point scale for certainty that a lesion represents a metastatic neoplasm, defined as: 1 definitely normal; 2 probably normal; 3 indeterminate; 4 probably abnormal; 5 definitely abnormal [10]. We defined a patient as having abnormal CPLNs if they had a qualitative assessment score (QAS) of 4 or 5. The other two definitions used a simple definition based on length of the short axis of the CPLN, one using N 7 mm [9] and one using ≥ 10 mm [8] based on previous findings. Presence of abnormal CPLNs seen on preoperative CT scans was not considered a criterion for stage IV disease in this study, nor was the CPLNs considered in defining residual disease (RD). No gross residual disease was defined as no visible disease at the end of PDS and gross residual disease was defined as any measurable RD less than or equal to 1 cm disease after debulking surgery. Patients with residual disease N 1 cm were excluded. Surgical complexity was defined using previously published criteria [11] and divided into three categories, low, intermediate, and high complexity surgery.
Table 1 Patient and tumor characteristics by presence of abnormal CPLNs where abnormal CPLNs defined as QAS of 4 or 5. Characteristic
Age (years), mean (SD) Primary site of disease, N (%) Ovary Peritoneal Fallopian tube FIGO stage, N (%) IIIC IV FIGO grade, N (%) 1 2 3 Unknown Histology, N (%) Non-serous Serous Preoperative CA-125 (U/mL), median (IQR) ASA score, N (%) b3 ≥3 Residual disease, N (%) 0 0.1–1 cm Surgical complexity, N (%) Low Intermediate High Adjuvant chemotherapy, N (%) No Yes Unknown Accordion grade 3+ 30-day postoperative complications, N (%) No Yes
Total
Abnormal CPLNs not present
Abnormal CPLNs present
N = 253
N = 225
N = 28
64.0 (11.0)
64.5 (11.1)
59.6 (9.3)
172 (68.0) 65 (25.7) 16 (6.3)
149 (66.2) 61 (27.1) 15 (6.7)
23 (82.1) 4 (14.3) 1 (3.6)
192 (75.9) 61 (24.1)
176 (78.2) 49 (21.8)
16 (57.1) 12 (42.9)
6 (2.4) 5 (2.0) 238 (94.1) 4 (1.6)
6 (2.7) 5 (2.2) 210 (93.3) 4 (1.8)
0 (0.0) 0 (0.0) 28 (100.0) 0 (0.0)
36 (14.2) 217 (85.8) 729 (254, 1730)
32 (14.2) 193 (85.8) 727 (235, 1744)
4 (14.3) 24 (85.7) 730 (450, 1720)
133 (52.6) 120 (47.4)
113 (50.2) 112 (49.8)
20 (71.4) 8 (28.6)
136 (53.8) 117 (46.2)
124 (55.1) 101 (44.9)
12 (42.9) 16 (57.1)
17 (6.7) 128 (50.6) 108 (42.7)
15 (6.7) 116 (51.6) 94 (41.8)
2 (7.1) 12 (42.9) 14 (50.0)
11 (4.3) 222 (87.7) 20 (7.9)
11 (4.9) 196 (87.1) 18 (8.0)
0 (0.0) 26 (92.9) 2 (7.1)
191 (75.5) 62 (24.5)
170 (75.6) 55 (24.4)
21 (75.0) 7 (25.0)
P⁎
0.03 0.23
0.01
0.99
0.99
0.28 0.03
0.22
0.68
0.71
0.95
Abbreviations: ASA, American Society of Anesthesiologists; CA-125, cancer antigen 125; FIGO, International Federation of Gynecology and Obstetrics; IQR, interquartile range; SD, standard deviation. ⁎ Chi-square or Fisher's exact p value reported for categorical variables, t-test p value reported for age, and Wilcoxon rank-sum p value reported for preoperative CA-125.
Please cite this article as: I. Mert, et al., Clinical significance of enlarged cardiophrenic lymph nodes in advanced ovarian cancer: Implications for survival, Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.10.024
I. Mert et al. / Gynecologic Oncology xxx (2017) xxx–xxx
Statistical analyses were done using SAS version 9.3 (SAS Institute, Inc.; Cary, NC). Comparisons of baseline patient and tumor characteristics between patients with and without abnormal CPLNs were evaluated using the two-sample t-test for age, Wilcoxon rank-sum for preoperative CA-125, and chi-square or Fisher's exact test for categorical variables. Overall survival following PDS was calculated using the KaplanMeier method and compared between groups using the log-rank test. Associations with overall survival were summarized using the hazard ratio (HR) and corresponding 95% confidence interval (CI) estimated from Cox proportional hazards models. All calculated p values were two-sided and p values b 0.05 were considered statistically significant.
3. Results Between January 2003 and December 2011, 617 patients underwent PDS with advanced stage disease (stage IIIC or IV). Among patients with advanced stage disease, 276 patients had complete CT images available and research authorization for review. Twenty-three of the 276 patients were excluded due to RD N1 cm leaving us with 253 patients for the analysis. We observed abnormal CPLNs with QAS of 4 or 5 in 28 (11.1%) patients. Removal of CPLNs was not routinely performed at PDS during this time frame and only one out of 28 patients had CPLNs resected at PDS.
3
Baseline patient and tumor characteristics of the cohort, overall and stratified by presence of abnormal CPLNs defined by QAS of 4 or 5, are listed in Table 1. There were 192 patients (75.9%) with stage IIIC disease and the majority had serous histology. We observed no difference in primary site of the disease, FIGO grade, histology, preoperative CA-125 level, residual disease, adjuvant chemotherapy, accordion grade 3 + 30-day postoperative complication rate and surgical complexity between the groups with and without abnormal CPLNs based on QAS (all p values N 0.05, Table 1). Women with abnormal CPLNs were younger (mean age, 59.6 vs. 64.5 years, p = 0.03), and were more likely to have stage IV disease (p = 0.01). For patients with abnormal CPLNs, stage IV disease was due to a combination of parenchymal involvement of the liver and spleen (n = 3), malignant pleural effusion or pleural based disease (n = 6), rectal wall involvement (n = 1) or organ involvement outside of the abdomen (n = 3). Among patients with abnormal CPLNs, 28.6% had an American Society of Anesthesiologists (ASA) score of ≥ 3 compared to 49.8% of women without abnormal CPLNs (p = 0.03). We sought to evaluate the impact of abnormal CPLNs on OS stratified by RD. At PDS, 136 (53.8%) patients had NGR; 12 (8.8%) of the NGR group had abnormal CPLNs. Among the patients with NGR, the percentage with high surgical complexity was higher among patients with versus without abnormal CPLNs (41.7% (5/12) vs. 28.2% (35/124)), although not statistically significant (p = 0.34). Furthermore, among
Fig. 1. Comparison of overall survival by presence/absence of residual disease and presence/absence of abnormal CPLNs (defined as QAS 4 or 5).
Please cite this article as: I. Mert, et al., Clinical significance of enlarged cardiophrenic lymph nodes in advanced ovarian cancer: Implications for survival, Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.10.024
4
I. Mert et al. / Gynecologic Oncology xxx (2017) xxx–xxx
the patients with NGR, OS tended to be worse for patients with abnormal CPLNs, specifically, the median OS was 38.4 months in patients with abnormal CPLNs versus 69.6 months in patients without abnormal CPLNs (p = 0.08, Fig. 1a). The unadjusted hazard ratio (HR) for the association between presence of abnormal CPLNs and OS among patients with NGR was 1.84 (95% CI: 0.91–3.72) and after adjusting for stage, the adjusted HR was 1.61 (95% CI: 0.77–3.35). Of 117 patients with gross RD 0.1–1.0 cm, 16 (13.7%) had abnormal CPLNs. Among the patients with gross RD, we observed no significant difference in OS between patients with versus without abnormal CPLNs (median OS, 37.5 vs. 28.5 months, p = 0.49, Fig. 1b). In the absence of abnormal CPLNs, NGR was significantly associated with improved OS (median OS, 69.6 vs. 28.5 months, respectively, NGR group vs. RD group; p b 0.001; Fig. 1c). However, in the presence of abnormal CPLNs, the survival benefit of NGR was diminished (median OS, 38.4 versus 37.5 months, respectively, NGR group vs. RD group: p = 0.99; Fig. 1d). We evaluated two published criteria for abnormal CPLNs such that a patient was defined as having abnormal CPLNs if the length of the short axis of any CPLN was a) N7 mm or b) ≥10 mm. Upon using N7 mm measured in short axis to define abnormal CPLNs, 53 (20.9%) patients had abnormal CPLNs. When patients with abnormal CPLNs based on N7 mm on the short axis were compared to patients without abnormal
CPLNs, there was no difference in age, primary site of disease, FIGO stage, FIGO grade, histology, ASA score, or surgical complexity scores, adjuvant chemotherapy and accordion grade 3+ 30-day postoperative complication rate but a significant difference was observed in preoperative CA-125 levels and residual disease (Supplementary Table 1). Looking at the association with survival starting with patients with NGR, median OS was 69.6 months in patients with no abnormal CPLNs compared to 35.5 months in patients with abnormal CPLNs (p = 0.18, Fig. 2a). Among patients with gross RD 0.1–1.0 cm, similar to the QAS definition, presence of abnormal CPLNs was not associated with OS (median OS, 31.7 months for patients without abnormal CPLNs vs. 24.5 months for patients with abnormal CPLNs; p = 0.88, Fig. 2b). Among patients without abnormal CPLNs, patients resected to NGR had significantly better OS compared to those with gross RD 0.1– 1.0 cm (median OS, 69.6 vs. 31.7 months, respectively; p b 0.001; Fig. 2c). However, among patients with abnormal CPLNs, for patients resected to NGR compared to patients with gross RD 0.1–1.0 cm the difference in OS did not reach statistical significance (median OS, 35.5 versus 24.5 months, respectively; p = 0.16; Fig. 2d). Finally, when abnormal CPLNs were defined as ≥10 mm on the short axis, only 20 (7.9%) patients met criteria for abnormal CPLNs and none of the patient or tumor characteristics were different between patients with or without abnormal CPLNs (Supplementary Table 2). NGR was
Fig. 2. Comparison of overall survival by presence/absence of residual disease and presence/absence of abnormal CPLNs (defined as N7 mm on the short axis).
Please cite this article as: I. Mert, et al., Clinical significance of enlarged cardiophrenic lymph nodes in advanced ovarian cancer: Implications for survival, Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.10.024
I. Mert et al. / Gynecologic Oncology xxx (2017) xxx–xxx
significantly associated with improved OS among patients without abnormal CPLNs (median OS for NGR vs. gross RD, 61.5 vs. 29.0 months; p b 0.001; Fig. 3c). However, the number of patients with abnormal CPLNs was too few to meaningfully compare the OS within this stratum (Fig. 3d).
4. Discussion In the present study, we add to the growing literature collectively suggesting that the presence of abnormal CPLNs worsens survival in advanced OC. Specifically, we demonstrate a trend toward worse survival in OC patients with NGR after PDS when abnormal CPLNs are present (versus absent) using a QAS of 4 or 5 to define abnormal CPLNs (HR 1.84, 95% CI 0.91–3.72). This effect was not seen among those with residual disease up to 1 cm which is consistent with the current paradigm around RD and outcomes. These findings imply that abnormal CPLNs are surrogates for worse disease and have important implications for treatment planning when identified on preoperative imaging. Among patients without abnormal CPLNs, NGR was significantly associated with improved OS compared to gross RD. However, when abnormal CPLNs were present the OS was similar in those with NGR versus gross RD.
5
The optimal size cut-off for pathologic CPLNs is unknown. Several studies demonstrate no difference in outcomes when using a cut-off of 5 mm [4,12] while others have used N7 mm [9] or ≥ 10 mm [8]. We used predefined criteria published in large prospective radiologic studies of patients undergoing PDS. When we used N 7 or ≥10 mm on short axis for the definition of abnormal CPLNs without considering a qualitative score, patients with NGR and without abnormal CPLNs had better survival outcomes compared to patients with NGR and with abnormal CPLNs consistent with those previous reports. Furthermore, in addition to enlargement, morphologic evaluation of the node is also a valuable parameter for the assessment of these nodes. Reproducibility of these criteria will need to be tested in other studies. Little is known regarding benefit of lymphadenectomy outside of the abdominal cavity in patients with advanced stage OC. The Lymphadenectomy in Ovarian Neoplasms (LION) trial was designed to assess pelvic and paraaortic lymphadenectomy in advanced stage OC after intraabdominal complete debulking has been completed and did not show any survival benefit with systemic lymphadenectomy [13]. On the other hand, enlarged or abnormal retroperitoneal lymph node removal is standard in the surgical management of advanced stage OC. At this time, how abnormal CPLNs portend worse OS is not clear. We observed a clinically, strong association between the presence of abnormal/enlarged CPLNs and OS only in patients with NGR disease across
Fig. 3. Comparison of overall survival by presence/absence of residual disease and presence/absence of abnormal CPLNs (defined as ≥10 mm on the short axis).
Please cite this article as: I. Mert, et al., Clinical significance of enlarged cardiophrenic lymph nodes in advanced ovarian cancer: Implications for survival, Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.10.024
6
I. Mert et al. / Gynecologic Oncology xxx (2017) xxx–xxx
all definitions. This might be due to higher number of stage IV disease (66.7% stage IV if abnormal CPLNs were present based on QAS 4 or 5 versus 17.7% if not, p b 0.001), though high surgical complexity was not different in the NGR group when abnormal CPLNs were present versus not (41.7% vs. 28.2%, respectively, p = 0.65). Worse survival associated with abnormal CPLNs might be due to tumor biology that enables some ovarian cancer cells a better vascular permeation and metastasis skills and cause a spread in other compartments of the body such as chest but this would require further molecular experiments to confirm. Removing abnormal CPLNs has not routinely been advocated or discussed during PDS; therefore, there is limited data examining the morbidity associated with surgical resection during abdominal surgery. Several surgical techniques have been described to remove CPLNs. Lim et al. removed CPLNs using video-assisted thoracic surgery (VATS) through a 2 cm incision made in the fifth intercostal space at the anterior axillary line [14]. Complete cytoreduction was achieved in all patients and median time required for VATS was 65 min and median estimated blood loss was 20 mL. Median number of harvested lymph nodes was three [14]. In a series of 11 patients, transdiaphragmatic approach for resection of CPLNs was performed [15]. With this technique, mobilization of liver enabled full exposure of the diaphragm which was incised in order to explore the pleural cavity and CPLNs [15]. The average surgery time was 385 min and average blood loss was 564 mL, median number of postoperative hospital days was 7, and overall postoperative surgical morbidity associated with removal of abnormal CPLNs was low [15]. The most common complication was small pleural effusion detected on chest X-ray followed by small apical pneumothorax, comparable to other procedures involving the diaphragm during PDS [15]. Cowan et al. showed % 7 complication rate associated with excision of the CPLNs and removal of CPLNs did not delay chemotherapy [16]. Prader and Yoo et al. also removed abnormal CPLNs through the diaphragm after mobilization of the liver and there were no major complications observed associated with the removal of CPLNs [8,17]. These results showed that in expert hands, resection of abnormal CPLNs carries minor risk to the patients with minimal morbidity and mortality. Our study has limitations commonly associated with retrospective studies. Given the rarity of abnormal CPLNs in OC, statistical power was limited for some categories. During the years of study we did not routinely remove CPLNs, so did not have pathologic verification of the involvement of the abnormal nodes, relying instead on existing studies for cut-off values of pathologic nodes. There were some imbalances with respect to poor prognostic factors in QAS and short axis N 7 mm definitions but the same trend toward worse survival associated with presence of enlarged/abnormal CPLNs was seen in NGR group across all definitions including ≥10 mm short axis model in which these prognostic factors were comparable (HRs of 1.84, 1.51 and 1.73 in Figs. 1a, 2a, and 3a, respectively). Strengths include using patients treated at a single institution with a relatively standard approach to advanced OC and two radiologists experienced in gynecologic imaging evaluated all the CT scans. Prospective studies with larger sample size will be needed to elucidate the clinical significance enlarged/abnormal CPLNs. In conclusion, presence of abnormal CPLNs appears to have an important effect on OS in patients with advanced OC. This is supported by the observed impact on survival in the cohort of patient with no other gross residual disease remaining (HR = 1.61 after adjusting for stage). Our results show that despite achieving NGR in the abdomen, patients with abnormal/ enlarged CPLNs had worse survival compared the patient without abnormal CPLNs. Furthermore, the median OS of women with abnormal CPLNs and NGR was similar to patients who
had RD 0.1–1.0 cm after PDS surgery. It is not clear if resection of CPLNs will improve survival when NGR is reached. However, given the minimal morbidity of the procedure and recent data suggesting ideal management of lymph nodes in the retroperitoneum should be to remove enlarged nodes, this seems reasonable. Collectively these findings support a clinical trial to investigate feasibility, safety, and impact on outcome of treatment of women with abnormal CPLNs. Preoperative imaging with CT scan is an easy and widely utilized modality for the management of OC and may guide gynecologic oncologists in approaching to patients with advanced stage OC with abnormal CPLNs. All authors declare no conflict of interest. Supplementary data to this article can be found online at https://doi. org/10.1016/j.ygyno.2017.10.024.
References [1] D.S. Chi, E.L. Eisenhauer, O. Zivanovic, Y. Sonoda, N.R. Abu-Rustum, D.A. Levine, et al., Improved progression-free and overall survival in advanced ovarian cancer as a result of a change in surgical paradigm, Gynecol. Oncol. 114 (2009) 26–31. [2] P. Harter, Z.M. Muallem, C. Buhrmann, D. Lorenz, C. Kaub, R. Hils, et al., Impact of a structured quality management program on surgical outcome in primary advanced ovarian cancer, Gynecol. Oncol. 121 (2011) 615–619. [3] S. Wallace, A. Kumar, M. Mc Gree, A. Weaver, A. Mariani, C. Langstraat, et al., Efforts at maximal cytoreduction improve survival in ovarian cancer patients, even when complete gross resection is not feasible, Gynecol. Oncol. 145 (1) (2017) 21–26. [4] B.J. Holloway, M.E. Gore, R.P. A'Hern, C. Parsons, The significance of paracardiac lymph node enlargement in ovarian cancer, Clin. Radiol. 52 (1997) 692–697. [5] G.B. Feldman, Lymphatic obstruction in carcinomatous ascites, Cancer Res. 35 (1975) 325–332. [6] M.F. Abu-Hijleh, O.A. Habbal, S.T. Moqattash, The role of the diaphragm in lymphatic absorption from the peritoneal cavity, J. Anat. 186 (Pt 3) (1995) 453–467. [7] J. Hynninen, A. Auranen, O. Carpen, K. Dean, M. Seppanen, J. Kemppainen, et al., FDG PET/CT in staging of advanced epithelial ovarian cancer: frequency of supradiaphragmatic lymph node metastasis challenges the traditional pattern of disease spread, Gynecol. Oncol. 126 (2012) 64–68. [8] S. Prader, P. Harter, C. Grimm, A. Traut, K.U. Waltering, P.F. Alesina, et al., Surgical management of cardiophrenic lymph nodes in patients with advanced ovarian cancer, Gynecol. Oncol. 141 (2016) 271–275. [9] T.H. Kim, M.C. Lim, S.I. Kim, S.S. Seo, S.H. Kim, S.Y. Park, Preoperative prediction of cardiophrenic lymph node metastasis in advanced ovarian cancer using computed tomography, Ann. Surg. Oncol. 23 (2016) 1302–1308. [10] R.S. Suidan, P.T. Ramirez, D.M. Sarasohn, J.B. Teitcher, S. Mironov, R.B. Iyer, et al., A multicenter prospective trial evaluating the ability of preoperative computed tomography scan and serum CA-125 to predict suboptimal cytoreduction at primary debulking surgery for advanced ovarian, fallopian tube, and peritoneal cancer, Gynecol. Oncol. 134 (2014) 455–461. [11] D.S. Chi, F. Musa, F. Dao, O. Zivanovic, Y. Sonoda, M.M. Leitao, et al., An analysis of patients with bulky advanced stage ovarian, tubal, and peritoneal carcinoma treated with primary debulking surgery (PDS) during an identical time period as the randomized EORTC-NCIC trial of PDS vs neoadjuvant chemotherapy (NACT), Gynecol. Oncol. 124 (2012) 10–14. [12] V. Kolev, S. Mironov, O. Mironov, N. Ishill, C.S. Moskowitz, G.J. Gardner, et al., Prognostic significance of supradiaphragmatic lymphadenopathy identified on preoperative computed tomography scan in patients undergoing primary cytoreduction for advanced epithelial ovarian cancer, Int. J. Gynecol. Cancer 20 (2010) 979–984. [13] https://clinicaltrials.gov/ct2/show/NCT00712218. [14] M.C. Lim, H.S. Lee, D.C. Jung, J.Y. Choi, S.S. Seo, S.Y. Park, Pathological diagnosis and cytoreduction of cardiophrenic lymph node and pleural metastasis in ovarian cancer patients using video-assisted thoracic surgery, Ann. Surg. Oncol. 16 (2009) 1990–1996. [15] C.J. LaFargue, B.T. Sawyer, R.E. Bristow, Short-term morbidity in transdiaphragmatic cardiophrenic lymph node resection for advanced stage gynecologic cancer, Gynecol. Oncol. Rep. 17 (2016) 33–37. [16] R.A. Cowan, J. Tseng, V. Murthy, R. Srivastava, K.C. Long Roche, O. Zivanovic, et al., Feasibility, safety and clinical outcomes of cardiophrenic lymph node resection in advanced ovarian cancer, Gynecol. Oncol. 147 (2) (2017) 262–266. [17] H.J. Yoo, M.C. Lim, Y.J. Song, Y.S. Jung, S.H. Kim, C.W. Yoo, et al., Transabdominal cardiophrenic lymph node dissection (CPLND) via incised diaphragm replace conventional video-assisted thoracic surgery for cytoreductive surgery in advanced ovarian cancer, Gynecol. Oncol. 129 (2013) 341–345.
Please cite this article as: I. Mert, et al., Clinical significance of enlarged cardiophrenic lymph nodes in advanced ovarian cancer: Implications for survival, Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.10.024