- Email: [email protected]
Clinical significance of tumor volume in endometrial cancer: A Japan–Korea cooperative study
Gynecologic Oncology 131 (2013) 294–298
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Clinical significance of tumor volume in endometrial cancer: A Japan–Korea cooperative study Yukiharu Todo a,⁎, Hyuck-Jae Choi b,⁎⁎, Sokbom Kang c, Jae-Weon Kim d,e, Joo-Hyun Nam f, Hidemichi Watari g, Akiko Tamakoshi h, Noriaki Sakuragi g a
Division of Gynecologic Oncology, National Hospital Organization, Hokkaido Cancer Center, 4-2 Kikusui, Shiroishi-ku, Sapporo 003-0804, Japan Department of Radiology, University of Ulsan, College of Medicine, Asan Medical Center, Seoul, Republic of Korea Center for Uterine Cancer, National Cancer Center, Goyang, Republic of Korea d Department of Obstetrics and Gynecology, College of Medicine, Seoul National University, Seoul, Republic of Korea e Cancer Research Institute, College of Medicine, Seoul National University, Seoul, Republic of Korea f Department of Obstetrics and Gynecology, University of Ulsan, College of Medicine, Asan Medical Center, Seoul, Republic of Korea g Department of Gynecology, Hokkaido University, Graduate School of Medicine, Japan h Department of Public Health, Hokkaido University, Graduate School of Medicine, Japan b c
H I G H L I G H T S • Tumor volume is an independent risk factor for nodal metastasis in endometrial cancer. • Tumor volume can be assessed in a preoperative setting.
a r t i c l e
i n f o
Article history: Received 29 June 2013 Accepted 7 August 2013 Available online 14 August 2013 Keywords: Endometrial cancer Lymphadenectomy Lymph node metastasis Tumor size Tumor volume Risk assessment
a b s t r a c t Objective. The aim of this study was to elucidate the significance of tumor volume as a risk factor for predicting lymph node metastasis. Methods. We applied the tumor volume index to the data that were collected for 327 Korean patients with endometrial cancer who underwent preoperative assessment including magnetic resonance imaging (MRI) and subsequent surgery including systematic lymphadenectomy. The volume index, which we previously reported in the literature, was defined as the product of maximum longitudinal diameter along the uterine axis, maximum anteroposterior diameter in a sagittal section image, and maximum horizontal diameter in a horizontal section image according to MRI data, from 425 Japanese women with endometrial cancer. Relationships between lymph node metastasis and results of preoperative examinations including volume index were analyzed by logistic regression analysis. Results. The prevalence of affected lymph nodes was 14.2%. Multivariate analysis showed that high-grade histology assessed by endometrial biopsy [odds ratio (OR); 2.9, 95% confidence interval (CI): 1.4–6.4], volume index (OR; 2.4, 95% CI: 1.1–5.3), node enlargement assessed by MRI (OR; 4.2, 95% CI: 1.4–13.2), and high serum cancer antigen (CA)125 level (OR; 3.6, 95% CI: 1.6–8.1) were significantly and independently related to lymph node metastasis. When volume index was excluded from the analysis, myoinvasion assessed by MRI was an independent risk factor for lymph node metastasis as well as high-grade histology, node enlargement, and high serum CA125 level. Conclusion. Volume index is compatible with myometrial invasion as a factor for predicting lymph node metastasis in endometrial cancer. © 2013 Elsevier Inc. All rights reserved.
Introduction The therapeutic effects of lymphadenectomy in endometrial cancer are an issue of great debate [1–3]. Pelvic lymphadenectomy does not have a therapeutic effect in low-risk endometrial cancer [1,2]; however, ⁎ Corresponding author. Fax: +81 11 811 9180. ⁎⁎ Corresponding author. E-mail addresses: [email protected] (Y. Todo), [email protected] (H.-J. Choi). 0090-8258/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ygyno.2013.08.008
systematic lymphadenectomy including para-aortic lymphadenectomy might have efficacy in selected cases of endometrial cancer, especially in high-risk cases [3]. Therefore, tailoring lymphadenectomy in endometrial cancer is now being discussed in order to maximize the therapeutic effect of surgery and minimize its invasiveness and adverse effects [4]. This will include preoperative assessment of the probability of lymph node metastasis in each patient to allocate only those with a certainty of lymph node metastasis to full lymphadenectomy. When planning a prospective clinical trial on the therapeutic significance of lymphadenectomy, an
Y. Todo et al. / Gynecologic Oncology 131 (2013) 294–298
adequate population in which low-risk patients are effectively excluded is needed to assess the full benefit of lymphadenectomy. If a population comprises a large proportion of low-risk patients, significance of lymphadenectomy would be underestimated because low-risk patients do not benefit from lymphadenectomy. In 2012, Kang et al. confirmed a low-risk group with endometrioid histology by endometrial biopsy, b50% myometrial invasion with no extension beyond the corpus and no enlarged lymph nodes by magnetic resonance imaging (MRI), and cancer antigen (CA) 125-level ≤ 35 U/ml before surgery; only 1.3% of the patients in the group had lymph node metastasis when assuming that the prevalence of lymph node metastasis is 10% in the target patient cohort [5]. Use of the Korean gynecologic oncology group (KGOG) low-risk criteria gave almost the same result as the results in two Japanese cohorts [6]. The criteria would provide suitable eligibility in a clinical study to assess the therapeutic effect of lymphadenectomy. On the other hand, Mariani et al. reported that lymphadenectomy can be omitted under the following conditions: endometrioid histology (grade 1 or 2), myometrial invasion b 50%, and tumor diameter b 2 cm [7]. Gynecologic oncologists in the United States regard these three factors as being important and have named them “Mayo criteria” [8]. A unique point of the Mayo criteria is that tumor size is used together with myometrial invasion. We have previously shown that tumor volume is a factor for predicting lymph node metastasis [9] and prognosis [10]. In the present study, we focused on the clinical significance of tumor size (volume) in relation to treatment strategy in endometrial cancer. The present study was a Japan–Korea cooperative study using a Korean cohort. We aimed to establish whether tumor volume is still a risk factor for lymph node metastasis in endometrial cancer in Korean patients.
295
a
b
c
Materials and methods Patients This study was carried out using data from 327 patients with endometrial carcinoma who had undergone preoperative endometrial biopsy, MRI, and serum CA125 assay, and extensive surgical staging during 2000 and 2008 at 3 hospitals in Korea. All patients underwent lymphadenectomy in addition to hysterectomy and bilateral salpingo-oophorectomy. Para-aortic lymphadenectomy was performed at the discretion of the surgeon. Approval by each institutional review board was separately obtained from the hospitals' ethics boards. Data evaluation Preoperative pathology was assessed using endometrial curettage or hysteroscopic biopsy. Cytologic preparation or Pipelle endometrial biopsy was not allowed. Pathologic findings revealed the presence of both endometrioid and non-endometrioid subtype (mixed subtype), and for this analysis, the cases were classified according to nonendometrioid subtype. MRI data were interpreted at Korean data center. Myoinvasion was classified using binary classification according to the revised FIGO staging system (b50% vs. ≥50%). It was evaluated in either T2-weighted MR images or gadolinium-enhanced T1-weighted images. Volume index was calculated as previously described by Todo et al. [11]. The index was defined as the product of the maximum longitudinal diameter (centimeter) along the uterine axis, the maximum anteroposterior diameter (centimeter), namely thickness, on a sagittal section image, and the maximum horizontal diameter (centimeter) on a horizontal section image (Fig. 1). Lymph node enlargement was defined as lymph node size N 1 cm in the short axis. Serum CA125 levels were determined by radioimmunoassay. All CA125 levels were reported using IU/ml. All the above-mentioned factors were evaluated in Korea. The data were collected at the Korean Gynecologic Oncology Group Center and sent to Japan.
Fig. 1. Volume index, defined as the product of maximum longitudinal diameter (a: 3.8 cm) along uterine axis, maximum anteroposterior diameter (b: 2.5 cm) in sagittal section image, and maximum horizontal diameter (c: 3.8 cm) in horizontal section image. Volume index of upper image was 36.1.
Analysis for selecting risk factors for lymph node metastasis Logistic regression analysis was conducted to select the risk factors for lymph node metastasis in Japan. Six variables were used for this analysis: (1) tumor grade/histology assessed by endometrial biopsy; (2) myoinvasion assessed by MRI; (3) volume index assessed by MRI; (4) lymph node enlargement assessed by MRI; (5) extrauterine involvement assessed by MRI; and (6) serum CA125 level. Tumor grade/histology, volume index and serum CA125 were determined as independent risk factors for lymph node metastasis in the previous study conducted in Japan [9] (Table 1). Every variable was classified using binary classification. Histological data were classified into 2 categories (G1/G2 vs. G3/non-endometrioid). Two cases with grade 1 mucinous adenocarcinoma and one case with grade 2 mucinous adenocarcinoma were classified into the G1/G2 category. Volume index was classified into 2 categories by dividing the cut-off level previously determined by Todo et al. [9]. Volume index measurements were used to obtain a receiver operating characteristics (ROC) curve for lymph node metastasis. Based on the sensitivity and the
Table 1 Risk factors for predicting lymph node metastasis confirmed in previous study [9]. Factor
High-risk criteria
Tumor grade/histology Volume index assessed by MRI Serum CA125
Endometrioid grade 3/non-endometrioid 36 or more ≥70 U/ml for patients b 50 years of age and ≥28 U/ml for patients ≥ 50 years of age
296
Y. Todo et al. / Gynecologic Oncology 131 (2013) 294–298
specificity obtained from the ROC curve, the cut-off level was determined. Large volume index was finally defined as ≥36 [9]. CA125 level was also classified into 2 categories by dividing the cut-off level as previously described [9]. High CA125 level was defined as ≥70 U/ml for patients b 50 years of age and ≥28 U/ml for patients ≥ 50 years of age. Forty-six of 327 patients had at least 1 of the 6 variables missing. Therefore, this analysis was conducted using data of the 281 patients who had all 6 variables. Variables that achieved statistical significance in univariate analysis were subsequently included in a multivariate analysis. The statistical significance level was set at 0.05. Statistical analyses were performed with StatView J-5.0 PPC (SAS Institute, Cary, NC, USA). Results The clinical and pathological characteristics of the patients are shown in Table 2. A total of 202 patients (71.9%) were in FIGO stage (1988) I, 15 (5.3%) in stage II, 56 (19.9%) in stage III, and 8 (2.9%) in stage IV or unknown. A total of 115 patients (40.9%) had grade 1 endometrioid adenocarcinoma, 76 (27.0%) grade 2 endometrioid adenocarcinoma, 38 (13.5%) grade 3 endometrioid adenocarcinoma, 1 mucinous adenocarcinoma, 22 serous adenocarcinoma, 6 clear cell carcinoma, 10 mixed-type carcinoma, and 4 had other types of carcinoma. A total of 60 patients (21.4%) underwent pelvic lymphadenectomy alone, and 221 patients (78.6%) underwent pelvic and para-aortic lymphadenectomy. The median number of lymph nodes removed was 29, the median number of pelvic lymph nodes removed was 22, and the median number of para-aortic lymph nodes removed was 6. Consequently, a total of 40 (14.2%) had lymph node metastasis. Table 3 shows the result of logistic regression analysis. Univariate analysis revealed that all factors were significantly related to lymph node metastasis. Multivariate analysis confirmed that the following 4 factors were independent risk factors for lymph node metastasis: (1) tumor grade/histology assessed by endometrial biopsy; (2) volume index assessed by MRI; (3) lymph node enlargement assessed by MRI; (4) serum CA125 level. Myoinvasion assessed by MRI had an OR of 3.6 [95% confidence interval (CI) = 1.8–7.1] before adjusting other factors but an OR of 2.1 (95% CI = 0.9–4.7) after adjusting volume index. Eventually, it had an OR of 1.7 (95% CI = 0.7–4.0) after adjusting for tumor grade/histology, volume index, and serum CA125. Table 4 shows the prevalence of affected lymph nodes according to the number of risk factors. In terms of the assessment including 3 factors (tumor grade/histology, volume index, and serum CA125 level), the rates of lymph node metastasis were 2.7% in the no-risk-factor group, and 18.6% in the ≥1 risk factor group at the prevalence of nodal metastasis of 10% using Bayes' theorem. In terms of the assessment including 4 factors (tumor grade/histology, volume index, lymph node enlargement, and serum CA125 level), the rates of lymph node metastasis were 1.9% in the no-risk-factor group, and 19.1% in the ≥1 risk factor group at the prevalence of nodal metastasis of 10% using Bayes' theorem. Fig. 2 shows the distributions of volume index according to the two cohorts: (1) Korean cohort; present study, n = 281; and (2) Japanese cohort; previous study, n = 425 [9]. Despite these evaluations being conducted in different countries, the distributions were very similar. In the Korean cohort, 8.9% of the patients had volume indexes of 27–45 and 9.1% in the Japanese cohort had volume indexes in the same range. Less than 10% of the patients had a volume index close to 36, which has been determined as a cut-off level for predicting lymph node metastasis in endometrial cancer. Discussion Tailoring lymphadenectomy in endometrial cancer is needed in order to maximize the therapeutic effect of surgery and minimize its
Table 2 Clinical background of patients with endometrial cancer who underwent extensive surgical staging including lymphadenectomy.
Age (years) Median (range) Preoperative biopsy findings Histological subtype Endometrioid G1 G2 G3 Mucinous Serous Clear Mixed Others Preoperative MRI Myometrial invasion Less than half Half or more Volume index Median (range) Less than 36 36 or more Enlarged lymph nodes (over 1 cm) No Yes Tumor spread out of uterine corpus No Yes Preoperative biomarker (CA125) Median (range) Surgicopathological findings FIGO surgical stage (1988) 1A 1B 1C 2A 2B 3A 3B 3C 4 Unknown Histological subtype Endometrioid G1 G2 G3 Grade unknown Mucinous Serous Clear Mixed Others Unknown Para-aortic lymphadenectomy No Yes Number of harvested LNs Median (range) Number of harvested PLNs Median (range) Number of harvested PANs Median (range)
Total patients (N = 281)
LNM (−) (n = 241)
LNM (+) (n = 40)
53 (23–82)
53 (23–82)
55 (39–76)
121 89 27 3 18 11 8 4
114 79 19 0 11 8 7 3
7 10 8 3 7 3 1 1
203 78
184 57
19 21
12.1 (0–989) 210 71
9.1 (0–989) 191 50
50.5 (0–710) 19 21
264 17
232 9
32 8
249 32
220 21
29 11
15.6 (0.1–1950)
13.5 (0.1–167)
39.1 (6.9–1950)
67 106 29 7 8 17 4 35 6 2
67 106 29 7 8 17 4 0 1 2
0 0 0 0 0 0 0 35 5 0
115 76 38 4 1 22 6 10 4 5
112 67 28 3 0 13 3 7 3 5
3 9 10 1 1 9 3 3 1 0
60 221
59 182
1 39
29 (4–99)
28 (4–99)
32.5 (9–77)
22 (2–69)
22 (2–69)
23 (8–55)
6 (0–35)
5 (0–31)
10 (0–35)
FIGO: International Federation of Gynecology and Obstetrics, LN: lymph node, PLN: pelvic lymph node, PAN: para-aortic lymph nodes, LNM: lymph node metastasis.
Y. Todo et al. / Gynecologic Oncology 131 (2013) 294–298 Table 3 Logistic regression analysis of factors relating to lymph node metastasis.
Korean cohort (n=281)
297
Japanease cohort (n=425)
50 Univariate analysis
Multivariate analysis
Odds ratio (95% CI)
Odds ratio (95% CI)
45 p-Value
p-Value
40
a
Tumor grade/histology G1/G2 G3/Non-endometrioid MR-myoinvasion b1/2 ≥1/2 MR-volume index Less than 36 36 or more MR-node enlargement No Yes Extrauterine involvement No Yes Serum CA125b Low High
35 1.0 4.0 (2.01–8.06)
b0.0001
2.9 (1.36–6.37)
0.0060
30 25
1.0 3.6 (1.79–7.09)
20
0.0003
15
1.0 4.2 (2.11–8.47)
b0.0001
2.4 (1.12–5.26)
0.025
1.0 6.4 (2.32–17.9)
0.0004
4.2 (1.37–13.2)
0.012
10 5 0 0-9
1.0 4.0 (1.74–9.08)
0.0011
1.0 4.7 (2.25–9.62)
b0.0001
Fig. 2. Country distributions of volume index measurements. 3.6 (1.61–8.06)
0.0018
invasiveness and adverse effects. Patients with a low risk of lymph node metastasis should not be treated with lymphadenectomy because they would suffer adverse effects including lower extremity lymphedema. However, there has been no consensus for identifying the low-risk group for nodal metastasis. In 2012, Kang et al. proposed a preoperative assessment to identify patients with a low risk of lymph node metastasis in endometrial cancer [5]. The KGOG low-risk criteria were as follows: endometrioid histology by endometrial biopsy; b 50% myometrial invasion with no extension beyond the corpus and no enlarged lymph nodes by MRI; and CA125-level ≤ 35 IU/ml before surgery. Finally, only 1.3% of the patients in the group had nodal metastasis at the assumed prevalence of 10%. The reproducibility of its performance was confirmed in a Korea–Japan cooperative study using two Japanese cohorts [6]. In addition, the predictive performances of three prediction models for identifying a low-risk group in endometrial cancer were compared [12]. The three models included criteria modified from the gynecologic oncology group (GOG) pilot study proposed by Boronow [13], criteria modified from the GOG-33 data proposed by Creasman et al. [14], and
Table 4 Prevalence of affected lymph nodes according to the number of risk factors. Number of risk factors (biopsy, volume index, CA125) 0
1
2
3
152 3.9 2.7
77 13.7 14.7
45 33.3 23.5
7 57.1 40.2
Number of risk factors (biopsy, volume index, CA125, node enlargement)
Number of patients LNM (%) Bayesian-adjusted LNM ratea
99-
Measurement of volume index
CI: confidence interval. a G1/G2: endometrioid grade 1, endometrioid grade 2, mucinous grade 2, G3: endometrioid grade 3. b Low: b70 U/ml for patients b 50 years of age and b28 U/ml for patients ≥ 50 years of age, high: ≥70 U/ml for patients b 50 years of age and ≥28 U/ml for patients ≥ 50 years of age.
Number of patients LNM (%) Bayesian-adjusted LNM ratea
9-18 18-27 27-36 36-45 45-54 54-63 63-72 72-81 81-90 90-99
0
1
2
3–4
148 2.7 1.9
76 19.7 13.9
41 29.3 20.6
16 56.3 39.6
LNM: lymph node metastasis. a Adjusted rate at the prevalence of nodal metastasis of 10% (the prevalence of affected nodes was 14.2%).
the Mayo clinic criteria proposed by Mariani et al. [7]. They finally showed that these three models had similar negative predictive powers and proposed a false-negative rate of b2% as an index for determining the clinical usefulness of preoperative prediction models for low risk of nodal metastasis, assuming that the prevalence of lymph node metastasis is 10% in the target patient cohort [12]. This proposal is acceptable considering the fact that false-negative rates for detecting nodal metastasis using sentinel node biopsy have been reported be 5% for breast cancer [15]. The KGOG low-risk criteria provide a versatile system in a clinical setting. However, there would be interobserver variability in results of myometrial invasion assessment by MRI. MRI-based evaluation of deep myometrial invasion in a multi-institutional cooperative study showed a sensitivity of 54% and specificity of 89%, indicating that results of previous single institutional studies might have been biased [16]. In 2000, Mariani et al. reported that there was a significant difference in prognosis for patients with low-risk endometrial cancer who had tumor sizes b 2 cm and ≥2 cm, and they showed that the latter group required lymphadenectomy or adjuvant radiotherapy [7]. They reported that lymphadenectomy can be omitted under the following conditions: (1) endometrioid histology (grade 1 or 2), myometrial invasion b 50%, and tumor diameter b 2 cm, or (2) endometrioid histology and no myometrial invasion (independent of grade and tumor diameter). Interestingly, tumor size is included in their criteria together with myometrial invasion. Tumor size has been integrated into many staging systems in various types of cancer, including head and neck cancer, breast cancer, lung cancer, renal cancer, uterine cervical cancer, vulvar cancer, uterine sarcoma, melanoma, and soft tissue sarcoma, but not in endometrial cancer [17]. It has recently been reported that tumor size successively reflects the state of metastatic progression in endometrial cancer and that the spectrum theory accounts for the disease properties of endometrial cancer [10]. In that study, tumor volume was a prognostic factor independently of advanced age, high-risk histological grade/subtype, deep myoinvasion, lymph node metastasis, and type of surgery. Hellman, who is the leader among the spectrum theory supporters in the field of breast cancer, stated that tumor size is the most important parameter indicating the ultimate likelihood of distant metastasis and the pace of metastatic dissemination [18]. One may ponder that small-sized uterine papillary serous carcinoma (UPSC) is usually associated with distant metastasis and UPSC is a systemic disease. However, lymph node metastasis rate (60.0%) in the large volume index group was higher than that (35.3%) in the small volume index group among the 22 cases of UPSC in the present study despite there being no statistical difference between the two groups. The relationship between lymph node metastasis and tumor volume in serous histology is not as conspicuous as that in
298
Y. Todo et al. / Gynecologic Oncology 131 (2013) 294–298
endometrioid histology. Nevertheless, it is possible that tumor volume in serous histology is still associated with lymph node metastasis. It is possible that tumor size will be integrated into a staging system of endometrial cancer in the future. Tumor size can be more easily measured in a preoperative setting because MRI has already become popular throughout the world. It has been suggested that volume index was an independent risk factor for lymph node metastasis in endometrial cancer in Japanese cohorts [9,11]. In the present study, the volume index was still an independent risk factor for lymph node metastasis also in a Korean cohort. We obtained the same results from two different cohorts in different countries, namely, reproducibility in regard to clinical significance of tumor volume has been validated. We do not intend to draw the conclusion that volume index is superior to MR-myoinvasion in predicting lymph node metastasis. Because myometrial invasion has been a conventional index of intrauterine tumor extent for a long time, it seems that MR-myoinvasion may be more intuitive to clinicians than volume index. In fact, many physicians in Japan routinely judge myometrial invasion using MRI in a preoperative assessment. Kang et al. showed that the proportions of patients in the low-risk group were 33% based on prediction models with tumor size and 45–56% based on prediction models without tumor size, the difference between the two models being significant [12]. However, MR-myoinvasion has a problematic issue, namely, interobserver inconsistency or variability. There would be some occasions where attending physicians have difficulty in judging myometrial invasion using MRI. In the present study, distributions of volume index values were similar despite the fact that they were evaluated in two different countries. In addition, the number of patients with a volume index close to 36, which had been determined as a cut-off value for predicting lymph node metastasis, was small. The degree of interobserver inconsistency in the volume index may be lower than we expected. We propose that volume index be used for a substitute index when it is difficult to judge myometrial invasion using MRI. Although gross tumor size would be predictive of lymph node involvement, it was not assessed in the present study. Gross tumor size would be one of the good predictors of nodal metastasis under the condition that MRI is rarely used in a preoperative setting. However, there has been no consensus regarding how to measure gross tumor size. The significance of gross tumor size should be validated in the future.
Conflict of interest statement The authors declare that they have no conflicts of interest.
References [1] ASTEC study group. Efficacy of systematic pelvic lymphadenectomy in endometrial cancer (MRC ASTEC trial): a randomized study. Lancet 2009;373:125–36. [2] Benedetti-Panici P, Basile S, Maneschi F, Alberto Lissoni A, Signorelli M, Scambia G, et al. Systematic pelvic lymphadenectomy vs no lymphadenectomy in early-stage endometrial carcinoma: randomized clinical trial. J Natl Cancer Inst 2008;100: 1707–16. [3] Todo Y, Kato H, Kaneuchi M, Watari H, Takeda M, Sakuragi N. Survival Effect of Para-aortic Lymphadenectomy in Endometrial Cancer (SEPAL Study): a retrospective cohort analysis. Lancet 2010;375:1165–72. [4] Sakuragi N. Emerging concept of tailored lymphadenectomy in endometrial cancer. J Gynecol Oncol 2012;23:210–2. [5] Kang S, Kang WD, Chung HH, Jeong DH, Seo SS, Lee JM, et al. Preoperative identification of a low-risk group for lymph node metastasis in endometrial cancer: a Korean gynecologic oncology group study. J Clin Oncol 2012;30:1329–34. [6] Kang S, Todo Y, Odagiri T, Mitamura T, Watari H, Kim JW, et al. A low-risk group for lymph node metastasis is accurately identified by Korean gynecologic oncology group criteria in two Japanese cohorts with endometrial cancer. Gynecol Oncol 2013;129:33–7. [7] Mariani A, Webb MJ, Keeney GL, Hoddock MG, Calori G, Podratz KC. Low-risk corpus cancer: is lymphadenectomy or radiotherapy necessary? Am J Obstet Gynecol 2000;182:1506–19. [8] Milam MR, Java J, Walker JL, Metzinger DS, Parker LP, Coleman RL. Nodal metastasis risk in endometrioid endometrial cancer. Obstet Gynecol 2012;119:286–92. [9] Todo Y, Okamoto K, Hayashi M, Minobe S, Nomura E, Hareyama H, et al. A validation study of a scoring system to estimate the risk of lymph node metastasis for patients with endometrial carcinoma for tailoring the indication of lymphadenectomy. Gynecol Oncol 2007;104:623–8. [10] Todo Y, Watari H, Okamoto K, Hareyama H, Minobe S, Kato H, et al. Tumor volume successively reflects the state of disease progression in endometrial cancer. Gynecol Oncol 2013;129:472–7. [11] Todo Y, Sakuragi N, Nishida R, Yamada T, Ebina Y, Yamamoto R, et al. Combined use of magnetic resonance imaging, CA 125 assay, histologic type, and histologic grade in the prediction of lymph node metastasis in endometrial carcinoma. Am J Obstet Gynecol 2003;188:1265–72. [12] Kang S, Lee JM, Lee JK, Kim JW, Cho CH, Kim SM, et al. How low is low enough? Evaluation of various risk-assessment models for lymph node metastasis in endometrial cancer: a Korean multicenter study. J Gynecol Oncol 2012;23:251–6. [13] Boronow RC. Surgical staging of endometrial cancer: evolution, evaluation, and responsible challenge: a personal perspective. Gynecol Oncol 1997;66:179–89. [14] Creasman WT, Morrow CP, Bundy BN, Homesley HD, Graham JE, Heller PB. Surgical pathologic spread patterns of endometrial cancer: a gynecologic oncology group study. Cancer 1987;60:2035–41. [15] Lyman GH, Giuliano AE, Somerfield MR, Benson III AB, Bodurka DC, Burstein HJ, et al. American Society of Clinical Oncology guideline recommendations for sentinel lymph node biopsy in early-stage breast cancer. J Clin Oncol 2005;23: 7703–20. [16] Hricak H, Rubinstein LV, Gherman GM, Karstaedt N. MR imaging evaluation of endometrial carcinoma: results of an NCI cooperative study. Radiology 1991;179:829–32. [17] Atlas TNM. Illustrated guide to the TNM/pTNM classification of malignant tumours. In: Ch Writtekind, Greene FL, Hutter RVP, Klimpfinger M, Sobin LH, editors. 5th edition 2004. Berlin, Heidelberg: Springer-Verlag; 2005. [18] Karnovsky Hellman S, Lecture Memorial. Natural history of small breast cancers. J Clin Oncol 1994;12:2229–34.
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Clinical significance of tumor volume in endometrial cancer: A Japan–Korea cooperative study Yukiharu Todo a,⁎, Hyuck-Jae Choi b,⁎⁎, Sokbom Kang c, Jae-Weon Kim d,e, Joo-Hyun Nam f, Hidemichi Watari g, Akiko Tamakoshi h, Noriaki Sakuragi g a
Division of Gynecologic Oncology, National Hospital Organization, Hokkaido Cancer Center, 4-2 Kikusui, Shiroishi-ku, Sapporo 003-0804, Japan Department of Radiology, University of Ulsan, College of Medicine, Asan Medical Center, Seoul, Republic of Korea Center for Uterine Cancer, National Cancer Center, Goyang, Republic of Korea d Department of Obstetrics and Gynecology, College of Medicine, Seoul National University, Seoul, Republic of Korea e Cancer Research Institute, College of Medicine, Seoul National University, Seoul, Republic of Korea f Department of Obstetrics and Gynecology, University of Ulsan, College of Medicine, Asan Medical Center, Seoul, Republic of Korea g Department of Gynecology, Hokkaido University, Graduate School of Medicine, Japan h Department of Public Health, Hokkaido University, Graduate School of Medicine, Japan b c
H I G H L I G H T S • Tumor volume is an independent risk factor for nodal metastasis in endometrial cancer. • Tumor volume can be assessed in a preoperative setting.
a r t i c l e
i n f o
Article history: Received 29 June 2013 Accepted 7 August 2013 Available online 14 August 2013 Keywords: Endometrial cancer Lymphadenectomy Lymph node metastasis Tumor size Tumor volume Risk assessment
a b s t r a c t Objective. The aim of this study was to elucidate the significance of tumor volume as a risk factor for predicting lymph node metastasis. Methods. We applied the tumor volume index to the data that were collected for 327 Korean patients with endometrial cancer who underwent preoperative assessment including magnetic resonance imaging (MRI) and subsequent surgery including systematic lymphadenectomy. The volume index, which we previously reported in the literature, was defined as the product of maximum longitudinal diameter along the uterine axis, maximum anteroposterior diameter in a sagittal section image, and maximum horizontal diameter in a horizontal section image according to MRI data, from 425 Japanese women with endometrial cancer. Relationships between lymph node metastasis and results of preoperative examinations including volume index were analyzed by logistic regression analysis. Results. The prevalence of affected lymph nodes was 14.2%. Multivariate analysis showed that high-grade histology assessed by endometrial biopsy [odds ratio (OR); 2.9, 95% confidence interval (CI): 1.4–6.4], volume index (OR; 2.4, 95% CI: 1.1–5.3), node enlargement assessed by MRI (OR; 4.2, 95% CI: 1.4–13.2), and high serum cancer antigen (CA)125 level (OR; 3.6, 95% CI: 1.6–8.1) were significantly and independently related to lymph node metastasis. When volume index was excluded from the analysis, myoinvasion assessed by MRI was an independent risk factor for lymph node metastasis as well as high-grade histology, node enlargement, and high serum CA125 level. Conclusion. Volume index is compatible with myometrial invasion as a factor for predicting lymph node metastasis in endometrial cancer. © 2013 Elsevier Inc. All rights reserved.
Introduction The therapeutic effects of lymphadenectomy in endometrial cancer are an issue of great debate [1–3]. Pelvic lymphadenectomy does not have a therapeutic effect in low-risk endometrial cancer [1,2]; however, ⁎ Corresponding author. Fax: +81 11 811 9180. ⁎⁎ Corresponding author. E-mail addresses: [email protected] (Y. Todo), [email protected] (H.-J. Choi). 0090-8258/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ygyno.2013.08.008
systematic lymphadenectomy including para-aortic lymphadenectomy might have efficacy in selected cases of endometrial cancer, especially in high-risk cases [3]. Therefore, tailoring lymphadenectomy in endometrial cancer is now being discussed in order to maximize the therapeutic effect of surgery and minimize its invasiveness and adverse effects [4]. This will include preoperative assessment of the probability of lymph node metastasis in each patient to allocate only those with a certainty of lymph node metastasis to full lymphadenectomy. When planning a prospective clinical trial on the therapeutic significance of lymphadenectomy, an
Y. Todo et al. / Gynecologic Oncology 131 (2013) 294–298
adequate population in which low-risk patients are effectively excluded is needed to assess the full benefit of lymphadenectomy. If a population comprises a large proportion of low-risk patients, significance of lymphadenectomy would be underestimated because low-risk patients do not benefit from lymphadenectomy. In 2012, Kang et al. confirmed a low-risk group with endometrioid histology by endometrial biopsy, b50% myometrial invasion with no extension beyond the corpus and no enlarged lymph nodes by magnetic resonance imaging (MRI), and cancer antigen (CA) 125-level ≤ 35 U/ml before surgery; only 1.3% of the patients in the group had lymph node metastasis when assuming that the prevalence of lymph node metastasis is 10% in the target patient cohort [5]. Use of the Korean gynecologic oncology group (KGOG) low-risk criteria gave almost the same result as the results in two Japanese cohorts [6]. The criteria would provide suitable eligibility in a clinical study to assess the therapeutic effect of lymphadenectomy. On the other hand, Mariani et al. reported that lymphadenectomy can be omitted under the following conditions: endometrioid histology (grade 1 or 2), myometrial invasion b 50%, and tumor diameter b 2 cm [7]. Gynecologic oncologists in the United States regard these three factors as being important and have named them “Mayo criteria” [8]. A unique point of the Mayo criteria is that tumor size is used together with myometrial invasion. We have previously shown that tumor volume is a factor for predicting lymph node metastasis [9] and prognosis [10]. In the present study, we focused on the clinical significance of tumor size (volume) in relation to treatment strategy in endometrial cancer. The present study was a Japan–Korea cooperative study using a Korean cohort. We aimed to establish whether tumor volume is still a risk factor for lymph node metastasis in endometrial cancer in Korean patients.
295
a
b
c
Materials and methods Patients This study was carried out using data from 327 patients with endometrial carcinoma who had undergone preoperative endometrial biopsy, MRI, and serum CA125 assay, and extensive surgical staging during 2000 and 2008 at 3 hospitals in Korea. All patients underwent lymphadenectomy in addition to hysterectomy and bilateral salpingo-oophorectomy. Para-aortic lymphadenectomy was performed at the discretion of the surgeon. Approval by each institutional review board was separately obtained from the hospitals' ethics boards. Data evaluation Preoperative pathology was assessed using endometrial curettage or hysteroscopic biopsy. Cytologic preparation or Pipelle endometrial biopsy was not allowed. Pathologic findings revealed the presence of both endometrioid and non-endometrioid subtype (mixed subtype), and for this analysis, the cases were classified according to nonendometrioid subtype. MRI data were interpreted at Korean data center. Myoinvasion was classified using binary classification according to the revised FIGO staging system (b50% vs. ≥50%). It was evaluated in either T2-weighted MR images or gadolinium-enhanced T1-weighted images. Volume index was calculated as previously described by Todo et al. [11]. The index was defined as the product of the maximum longitudinal diameter (centimeter) along the uterine axis, the maximum anteroposterior diameter (centimeter), namely thickness, on a sagittal section image, and the maximum horizontal diameter (centimeter) on a horizontal section image (Fig. 1). Lymph node enlargement was defined as lymph node size N 1 cm in the short axis. Serum CA125 levels were determined by radioimmunoassay. All CA125 levels were reported using IU/ml. All the above-mentioned factors were evaluated in Korea. The data were collected at the Korean Gynecologic Oncology Group Center and sent to Japan.
Fig. 1. Volume index, defined as the product of maximum longitudinal diameter (a: 3.8 cm) along uterine axis, maximum anteroposterior diameter (b: 2.5 cm) in sagittal section image, and maximum horizontal diameter (c: 3.8 cm) in horizontal section image. Volume index of upper image was 36.1.
Analysis for selecting risk factors for lymph node metastasis Logistic regression analysis was conducted to select the risk factors for lymph node metastasis in Japan. Six variables were used for this analysis: (1) tumor grade/histology assessed by endometrial biopsy; (2) myoinvasion assessed by MRI; (3) volume index assessed by MRI; (4) lymph node enlargement assessed by MRI; (5) extrauterine involvement assessed by MRI; and (6) serum CA125 level. Tumor grade/histology, volume index and serum CA125 were determined as independent risk factors for lymph node metastasis in the previous study conducted in Japan [9] (Table 1). Every variable was classified using binary classification. Histological data were classified into 2 categories (G1/G2 vs. G3/non-endometrioid). Two cases with grade 1 mucinous adenocarcinoma and one case with grade 2 mucinous adenocarcinoma were classified into the G1/G2 category. Volume index was classified into 2 categories by dividing the cut-off level previously determined by Todo et al. [9]. Volume index measurements were used to obtain a receiver operating characteristics (ROC) curve for lymph node metastasis. Based on the sensitivity and the
Table 1 Risk factors for predicting lymph node metastasis confirmed in previous study [9]. Factor
High-risk criteria
Tumor grade/histology Volume index assessed by MRI Serum CA125
Endometrioid grade 3/non-endometrioid 36 or more ≥70 U/ml for patients b 50 years of age and ≥28 U/ml for patients ≥ 50 years of age
296
Y. Todo et al. / Gynecologic Oncology 131 (2013) 294–298
specificity obtained from the ROC curve, the cut-off level was determined. Large volume index was finally defined as ≥36 [9]. CA125 level was also classified into 2 categories by dividing the cut-off level as previously described [9]. High CA125 level was defined as ≥70 U/ml for patients b 50 years of age and ≥28 U/ml for patients ≥ 50 years of age. Forty-six of 327 patients had at least 1 of the 6 variables missing. Therefore, this analysis was conducted using data of the 281 patients who had all 6 variables. Variables that achieved statistical significance in univariate analysis were subsequently included in a multivariate analysis. The statistical significance level was set at 0.05. Statistical analyses were performed with StatView J-5.0 PPC (SAS Institute, Cary, NC, USA). Results The clinical and pathological characteristics of the patients are shown in Table 2. A total of 202 patients (71.9%) were in FIGO stage (1988) I, 15 (5.3%) in stage II, 56 (19.9%) in stage III, and 8 (2.9%) in stage IV or unknown. A total of 115 patients (40.9%) had grade 1 endometrioid adenocarcinoma, 76 (27.0%) grade 2 endometrioid adenocarcinoma, 38 (13.5%) grade 3 endometrioid adenocarcinoma, 1 mucinous adenocarcinoma, 22 serous adenocarcinoma, 6 clear cell carcinoma, 10 mixed-type carcinoma, and 4 had other types of carcinoma. A total of 60 patients (21.4%) underwent pelvic lymphadenectomy alone, and 221 patients (78.6%) underwent pelvic and para-aortic lymphadenectomy. The median number of lymph nodes removed was 29, the median number of pelvic lymph nodes removed was 22, and the median number of para-aortic lymph nodes removed was 6. Consequently, a total of 40 (14.2%) had lymph node metastasis. Table 3 shows the result of logistic regression analysis. Univariate analysis revealed that all factors were significantly related to lymph node metastasis. Multivariate analysis confirmed that the following 4 factors were independent risk factors for lymph node metastasis: (1) tumor grade/histology assessed by endometrial biopsy; (2) volume index assessed by MRI; (3) lymph node enlargement assessed by MRI; (4) serum CA125 level. Myoinvasion assessed by MRI had an OR of 3.6 [95% confidence interval (CI) = 1.8–7.1] before adjusting other factors but an OR of 2.1 (95% CI = 0.9–4.7) after adjusting volume index. Eventually, it had an OR of 1.7 (95% CI = 0.7–4.0) after adjusting for tumor grade/histology, volume index, and serum CA125. Table 4 shows the prevalence of affected lymph nodes according to the number of risk factors. In terms of the assessment including 3 factors (tumor grade/histology, volume index, and serum CA125 level), the rates of lymph node metastasis were 2.7% in the no-risk-factor group, and 18.6% in the ≥1 risk factor group at the prevalence of nodal metastasis of 10% using Bayes' theorem. In terms of the assessment including 4 factors (tumor grade/histology, volume index, lymph node enlargement, and serum CA125 level), the rates of lymph node metastasis were 1.9% in the no-risk-factor group, and 19.1% in the ≥1 risk factor group at the prevalence of nodal metastasis of 10% using Bayes' theorem. Fig. 2 shows the distributions of volume index according to the two cohorts: (1) Korean cohort; present study, n = 281; and (2) Japanese cohort; previous study, n = 425 [9]. Despite these evaluations being conducted in different countries, the distributions were very similar. In the Korean cohort, 8.9% of the patients had volume indexes of 27–45 and 9.1% in the Japanese cohort had volume indexes in the same range. Less than 10% of the patients had a volume index close to 36, which has been determined as a cut-off level for predicting lymph node metastasis in endometrial cancer. Discussion Tailoring lymphadenectomy in endometrial cancer is needed in order to maximize the therapeutic effect of surgery and minimize its
Table 2 Clinical background of patients with endometrial cancer who underwent extensive surgical staging including lymphadenectomy.
Age (years) Median (range) Preoperative biopsy findings Histological subtype Endometrioid G1 G2 G3 Mucinous Serous Clear Mixed Others Preoperative MRI Myometrial invasion Less than half Half or more Volume index Median (range) Less than 36 36 or more Enlarged lymph nodes (over 1 cm) No Yes Tumor spread out of uterine corpus No Yes Preoperative biomarker (CA125) Median (range) Surgicopathological findings FIGO surgical stage (1988) 1A 1B 1C 2A 2B 3A 3B 3C 4 Unknown Histological subtype Endometrioid G1 G2 G3 Grade unknown Mucinous Serous Clear Mixed Others Unknown Para-aortic lymphadenectomy No Yes Number of harvested LNs Median (range) Number of harvested PLNs Median (range) Number of harvested PANs Median (range)
Total patients (N = 281)
LNM (−) (n = 241)
LNM (+) (n = 40)
53 (23–82)
53 (23–82)
55 (39–76)
121 89 27 3 18 11 8 4
114 79 19 0 11 8 7 3
7 10 8 3 7 3 1 1
203 78
184 57
19 21
12.1 (0–989) 210 71
9.1 (0–989) 191 50
50.5 (0–710) 19 21
264 17
232 9
32 8
249 32
220 21
29 11
15.6 (0.1–1950)
13.5 (0.1–167)
39.1 (6.9–1950)
67 106 29 7 8 17 4 35 6 2
67 106 29 7 8 17 4 0 1 2
0 0 0 0 0 0 0 35 5 0
115 76 38 4 1 22 6 10 4 5
112 67 28 3 0 13 3 7 3 5
3 9 10 1 1 9 3 3 1 0
60 221
59 182
1 39
29 (4–99)
28 (4–99)
32.5 (9–77)
22 (2–69)
22 (2–69)
23 (8–55)
6 (0–35)
5 (0–31)
10 (0–35)
FIGO: International Federation of Gynecology and Obstetrics, LN: lymph node, PLN: pelvic lymph node, PAN: para-aortic lymph nodes, LNM: lymph node metastasis.
Y. Todo et al. / Gynecologic Oncology 131 (2013) 294–298 Table 3 Logistic regression analysis of factors relating to lymph node metastasis.
Korean cohort (n=281)
297
Japanease cohort (n=425)
50 Univariate analysis
Multivariate analysis
Odds ratio (95% CI)
Odds ratio (95% CI)
45 p-Value
p-Value
40
a
Tumor grade/histology G1/G2 G3/Non-endometrioid MR-myoinvasion b1/2 ≥1/2 MR-volume index Less than 36 36 or more MR-node enlargement No Yes Extrauterine involvement No Yes Serum CA125b Low High
35 1.0 4.0 (2.01–8.06)
b0.0001
2.9 (1.36–6.37)
0.0060
30 25
1.0 3.6 (1.79–7.09)
20
0.0003
15
1.0 4.2 (2.11–8.47)
b0.0001
2.4 (1.12–5.26)
0.025
1.0 6.4 (2.32–17.9)
0.0004
4.2 (1.37–13.2)
0.012
10 5 0 0-9
1.0 4.0 (1.74–9.08)
0.0011
1.0 4.7 (2.25–9.62)
b0.0001
Fig. 2. Country distributions of volume index measurements. 3.6 (1.61–8.06)
0.0018
invasiveness and adverse effects. Patients with a low risk of lymph node metastasis should not be treated with lymphadenectomy because they would suffer adverse effects including lower extremity lymphedema. However, there has been no consensus for identifying the low-risk group for nodal metastasis. In 2012, Kang et al. proposed a preoperative assessment to identify patients with a low risk of lymph node metastasis in endometrial cancer [5]. The KGOG low-risk criteria were as follows: endometrioid histology by endometrial biopsy; b 50% myometrial invasion with no extension beyond the corpus and no enlarged lymph nodes by MRI; and CA125-level ≤ 35 IU/ml before surgery. Finally, only 1.3% of the patients in the group had nodal metastasis at the assumed prevalence of 10%. The reproducibility of its performance was confirmed in a Korea–Japan cooperative study using two Japanese cohorts [6]. In addition, the predictive performances of three prediction models for identifying a low-risk group in endometrial cancer were compared [12]. The three models included criteria modified from the gynecologic oncology group (GOG) pilot study proposed by Boronow [13], criteria modified from the GOG-33 data proposed by Creasman et al. [14], and
Table 4 Prevalence of affected lymph nodes according to the number of risk factors. Number of risk factors (biopsy, volume index, CA125) 0
1
2
3
152 3.9 2.7
77 13.7 14.7
45 33.3 23.5
7 57.1 40.2
Number of risk factors (biopsy, volume index, CA125, node enlargement)
Number of patients LNM (%) Bayesian-adjusted LNM ratea
99-
Measurement of volume index
CI: confidence interval. a G1/G2: endometrioid grade 1, endometrioid grade 2, mucinous grade 2, G3: endometrioid grade 3. b Low: b70 U/ml for patients b 50 years of age and b28 U/ml for patients ≥ 50 years of age, high: ≥70 U/ml for patients b 50 years of age and ≥28 U/ml for patients ≥ 50 years of age.
Number of patients LNM (%) Bayesian-adjusted LNM ratea
9-18 18-27 27-36 36-45 45-54 54-63 63-72 72-81 81-90 90-99
0
1
2
3–4
148 2.7 1.9
76 19.7 13.9
41 29.3 20.6
16 56.3 39.6
LNM: lymph node metastasis. a Adjusted rate at the prevalence of nodal metastasis of 10% (the prevalence of affected nodes was 14.2%).
the Mayo clinic criteria proposed by Mariani et al. [7]. They finally showed that these three models had similar negative predictive powers and proposed a false-negative rate of b2% as an index for determining the clinical usefulness of preoperative prediction models for low risk of nodal metastasis, assuming that the prevalence of lymph node metastasis is 10% in the target patient cohort [12]. This proposal is acceptable considering the fact that false-negative rates for detecting nodal metastasis using sentinel node biopsy have been reported be 5% for breast cancer [15]. The KGOG low-risk criteria provide a versatile system in a clinical setting. However, there would be interobserver variability in results of myometrial invasion assessment by MRI. MRI-based evaluation of deep myometrial invasion in a multi-institutional cooperative study showed a sensitivity of 54% and specificity of 89%, indicating that results of previous single institutional studies might have been biased [16]. In 2000, Mariani et al. reported that there was a significant difference in prognosis for patients with low-risk endometrial cancer who had tumor sizes b 2 cm and ≥2 cm, and they showed that the latter group required lymphadenectomy or adjuvant radiotherapy [7]. They reported that lymphadenectomy can be omitted under the following conditions: (1) endometrioid histology (grade 1 or 2), myometrial invasion b 50%, and tumor diameter b 2 cm, or (2) endometrioid histology and no myometrial invasion (independent of grade and tumor diameter). Interestingly, tumor size is included in their criteria together with myometrial invasion. Tumor size has been integrated into many staging systems in various types of cancer, including head and neck cancer, breast cancer, lung cancer, renal cancer, uterine cervical cancer, vulvar cancer, uterine sarcoma, melanoma, and soft tissue sarcoma, but not in endometrial cancer [17]. It has recently been reported that tumor size successively reflects the state of metastatic progression in endometrial cancer and that the spectrum theory accounts for the disease properties of endometrial cancer [10]. In that study, tumor volume was a prognostic factor independently of advanced age, high-risk histological grade/subtype, deep myoinvasion, lymph node metastasis, and type of surgery. Hellman, who is the leader among the spectrum theory supporters in the field of breast cancer, stated that tumor size is the most important parameter indicating the ultimate likelihood of distant metastasis and the pace of metastatic dissemination [18]. One may ponder that small-sized uterine papillary serous carcinoma (UPSC) is usually associated with distant metastasis and UPSC is a systemic disease. However, lymph node metastasis rate (60.0%) in the large volume index group was higher than that (35.3%) in the small volume index group among the 22 cases of UPSC in the present study despite there being no statistical difference between the two groups. The relationship between lymph node metastasis and tumor volume in serous histology is not as conspicuous as that in
298
Y. Todo et al. / Gynecologic Oncology 131 (2013) 294–298
endometrioid histology. Nevertheless, it is possible that tumor volume in serous histology is still associated with lymph node metastasis. It is possible that tumor size will be integrated into a staging system of endometrial cancer in the future. Tumor size can be more easily measured in a preoperative setting because MRI has already become popular throughout the world. It has been suggested that volume index was an independent risk factor for lymph node metastasis in endometrial cancer in Japanese cohorts [9,11]. In the present study, the volume index was still an independent risk factor for lymph node metastasis also in a Korean cohort. We obtained the same results from two different cohorts in different countries, namely, reproducibility in regard to clinical significance of tumor volume has been validated. We do not intend to draw the conclusion that volume index is superior to MR-myoinvasion in predicting lymph node metastasis. Because myometrial invasion has been a conventional index of intrauterine tumor extent for a long time, it seems that MR-myoinvasion may be more intuitive to clinicians than volume index. In fact, many physicians in Japan routinely judge myometrial invasion using MRI in a preoperative assessment. Kang et al. showed that the proportions of patients in the low-risk group were 33% based on prediction models with tumor size and 45–56% based on prediction models without tumor size, the difference between the two models being significant [12]. However, MR-myoinvasion has a problematic issue, namely, interobserver inconsistency or variability. There would be some occasions where attending physicians have difficulty in judging myometrial invasion using MRI. In the present study, distributions of volume index values were similar despite the fact that they were evaluated in two different countries. In addition, the number of patients with a volume index close to 36, which had been determined as a cut-off value for predicting lymph node metastasis, was small. The degree of interobserver inconsistency in the volume index may be lower than we expected. We propose that volume index be used for a substitute index when it is difficult to judge myometrial invasion using MRI. Although gross tumor size would be predictive of lymph node involvement, it was not assessed in the present study. Gross tumor size would be one of the good predictors of nodal metastasis under the condition that MRI is rarely used in a preoperative setting. However, there has been no consensus regarding how to measure gross tumor size. The significance of gross tumor size should be validated in the future.
Conflict of interest statement The authors declare that they have no conflicts of interest.
References [1] ASTEC study group. Efficacy of systematic pelvic lymphadenectomy in endometrial cancer (MRC ASTEC trial): a randomized study. Lancet 2009;373:125–36. [2] Benedetti-Panici P, Basile S, Maneschi F, Alberto Lissoni A, Signorelli M, Scambia G, et al. Systematic pelvic lymphadenectomy vs no lymphadenectomy in early-stage endometrial carcinoma: randomized clinical trial. J Natl Cancer Inst 2008;100: 1707–16. [3] Todo Y, Kato H, Kaneuchi M, Watari H, Takeda M, Sakuragi N. Survival Effect of Para-aortic Lymphadenectomy in Endometrial Cancer (SEPAL Study): a retrospective cohort analysis. Lancet 2010;375:1165–72. [4] Sakuragi N. Emerging concept of tailored lymphadenectomy in endometrial cancer. J Gynecol Oncol 2012;23:210–2. [5] Kang S, Kang WD, Chung HH, Jeong DH, Seo SS, Lee JM, et al. Preoperative identification of a low-risk group for lymph node metastasis in endometrial cancer: a Korean gynecologic oncology group study. J Clin Oncol 2012;30:1329–34. [6] Kang S, Todo Y, Odagiri T, Mitamura T, Watari H, Kim JW, et al. A low-risk group for lymph node metastasis is accurately identified by Korean gynecologic oncology group criteria in two Japanese cohorts with endometrial cancer. Gynecol Oncol 2013;129:33–7. [7] Mariani A, Webb MJ, Keeney GL, Hoddock MG, Calori G, Podratz KC. Low-risk corpus cancer: is lymphadenectomy or radiotherapy necessary? Am J Obstet Gynecol 2000;182:1506–19. [8] Milam MR, Java J, Walker JL, Metzinger DS, Parker LP, Coleman RL. Nodal metastasis risk in endometrioid endometrial cancer. Obstet Gynecol 2012;119:286–92. [9] Todo Y, Okamoto K, Hayashi M, Minobe S, Nomura E, Hareyama H, et al. A validation study of a scoring system to estimate the risk of lymph node metastasis for patients with endometrial carcinoma for tailoring the indication of lymphadenectomy. Gynecol Oncol 2007;104:623–8. [10] Todo Y, Watari H, Okamoto K, Hareyama H, Minobe S, Kato H, et al. Tumor volume successively reflects the state of disease progression in endometrial cancer. Gynecol Oncol 2013;129:472–7. [11] Todo Y, Sakuragi N, Nishida R, Yamada T, Ebina Y, Yamamoto R, et al. Combined use of magnetic resonance imaging, CA 125 assay, histologic type, and histologic grade in the prediction of lymph node metastasis in endometrial carcinoma. Am J Obstet Gynecol 2003;188:1265–72. [12] Kang S, Lee JM, Lee JK, Kim JW, Cho CH, Kim SM, et al. How low is low enough? Evaluation of various risk-assessment models for lymph node metastasis in endometrial cancer: a Korean multicenter study. J Gynecol Oncol 2012;23:251–6. [13] Boronow RC. Surgical staging of endometrial cancer: evolution, evaluation, and responsible challenge: a personal perspective. Gynecol Oncol 1997;66:179–89. [14] Creasman WT, Morrow CP, Bundy BN, Homesley HD, Graham JE, Heller PB. Surgical pathologic spread patterns of endometrial cancer: a gynecologic oncology group study. Cancer 1987;60:2035–41. [15] Lyman GH, Giuliano AE, Somerfield MR, Benson III AB, Bodurka DC, Burstein HJ, et al. American Society of Clinical Oncology guideline recommendations for sentinel lymph node biopsy in early-stage breast cancer. J Clin Oncol 2005;23: 7703–20. [16] Hricak H, Rubinstein LV, Gherman GM, Karstaedt N. MR imaging evaluation of endometrial carcinoma: results of an NCI cooperative study. Radiology 1991;179:829–32. [17] Atlas TNM. Illustrated guide to the TNM/pTNM classification of malignant tumours. In: Ch Writtekind, Greene FL, Hutter RVP, Klimpfinger M, Sobin LH, editors. 5th edition 2004. Berlin, Heidelberg: Springer-Verlag; 2005. [18] Karnovsky Hellman S, Lecture Memorial. Natural history of small breast cancers. J Clin Oncol 1994;12:2229–34.