Original article
Relevance of hepatoduodenal ligament lymph nodes in resectional surgery for gastric cancer S. L. Lee1 , H. H. Lee2 , Y. H. Ko3 , K. Y. Song2 , C. H. Park2 , H. M. Jeon2 and S. S. Kim4 1
Department of Radiology, 2 Division of Gastrointestinal Surgery, Department of Surgery, 3 Division of Medical Oncology, Department of Internal Medicine, and 4 Division of Gastroenterology, Department of Internal Medicine, College of Medicine, Catholic University of Korea, Seoul, Korea Correspondence to: Dr H. H. Lee, Division of Gastrointestinal Surgery, Department of Surgery, Catholic University of Korea, 65-1 Gumo-Dong, Uijeongbu-City, Gyenggi-Do, 480-717, Korea (e-mail: [email protected])
Background: Hepatoduodenal lymph node (HDLN) positivity is considered distant metastasis in gastric
cancer according to the seventh American Joint Committee on Cancer (AJCC) classification. In contrast, the International Union Against Cancer seventh edition and the Japanese Gastric Cancer Association both consider HDLN as a regional lymph node that can be included in the context of a curative resection. The purpose of this study was to determine whether there was justification for considering HDLN involvement as a distant metastasis for which resectional surgery could not have survival benefit. Methods: This study enrolled consecutive patients with gastric cancer having D2 or greater resections, with removal and pathological assessment of the HDLN, between 1989 and 2009. The pathological stage of all patients was determined based on the seventh AJCC criteria, with HDLN included as a regional lymph node. Results: A total of 1872 patients had their HDLN removed, of whom 68 had a metastatic lymph node in the hepatoduodenal ligament. The 5-year survival rate of these 68 patients was 30 per cent, compared with 47·7 per cent for those with stage III (P < 0·001) and 9·8 per cent for those with stage IV (P = 0·007) HDLN-negative tumours. The 5-year survival rate of 41 patients with HDLN metastasis and no evidence of distant metastasis at any other site was significantly higher than that among 120 patients with stage IV disease without HDLN metastasis (P < 0·001), whereas 5-year survival did not differ between the 41 patients with stage I–III disease with HDLN metastasis and 568 patients with stage III tumours without HDLN metastasis (P = 0·184). HDLN metastasis was not a significant factor for survival in multivariable analysis. Conclusion: It is inappropriate to include the HDLN in the distant metastatic lymph node group in gastric cancer. The seventh AJCC criteria for node grouping should be revised. Presented to the Tenth International Gastric Cancer Congress, Verona, Italy, June 2013 Paper accepted 18 December 2013 Published online 26 February 2014 in Wiley Online Library (www.bjs.co.uk). DOI: 10.1002/bjs.9438
Introduction
The seventh American Joint Committee on Cancer (AJCC) classification contains many changes in the staging of gastric cancer1,2 . The subserosal layer was upgraded to T3, and T4 was subclassified into T4a for serosal invasion and T4b for adjacent structural invasion. Criteria for the number of metastatic lymph nodes were revised, and stage N3 was subdivided into N3a and N3b. Positive peritoneal cytology was classified as M1. Although there has been some controversy regarding these classification changes3 – 9 , important changes related to the distant metastasis grouping seem to have gone unnoticed. 2014 BJS Society Ltd Published by John Wiley & Sons Ltd
The seventh AJCC classification includes hepatoduodenal, retropancreatic, mesenteric and para-aortic lymph nodes in the nodal grouping definition of distant metastasis sites for gastric cancer1 . Most surgeons would accept that there is no convincing evidence that resection of paraaortic (station 16 lymph node), retropancreatic (station 13) and mesenteric (station 14v) lymph nodes, when involved, confers significant survival benefit10 – 13 . The hepatoduodenal lymph node (HDLN; station 12), however, is still considered as a regional lymph node by the International Union Against Cancer (UICC) tumour node metastasis (TNM) seventh edition14 and the Japanese Gastric Cancer Association (JGCA)15 . Gastric cancer surgeons BJS 2014; 101: 518–522
Hepatoduodenal lymph node metastasis in gastric cancer
519
Comparison of clinicopathological factors between patients with and without hepatoduodenal lymph node metastasis
Table 1
Age (years)* Sex M 1255 F 617 Extent of resection Total 432 Subtotal 1440 Histological type Differentiated 788 Undifferentiated 1084 Depth of invasion† T1 621 T2 276 T3 389 T4 586 Lymph node metastasis† N0 868 N1 257 N2 245 N3 502 Pathological stage† I 737 II 381 III 607 IV 147
HDLNnegative (n = 1804)
HDLNpositive (n = 68)
56·6(12·2)
57·7(11·5)
1206 (96·1) 598 (96·9)
49 (3·9) 19 (3·1)
406 (94·0) 1398 (97·1)
26 (6·0) 42 (2·9)
0·465§ 0·370
0·003
0·159 765 (97·1) 1039 (95·8)
23 (2·9) 45 (4·2)
621 (100) 266 (96·4) 377 (96·9) 540 (92·2)
0 (0) 10 (3·6) 12 (3·1) 46 (7·8)
868 (100) 255 (99·2) 240 (98·0) 441 (87·8)
0 (0) 2 (0·8) 5 (2·0) 61 (12·2)
737 (100) 379 (99·5) 568 (93·6) 120 (81·6)
0 (0) 2 (0·5) 39 (6·4) 27 (18·4)
0·8 0·6 0·4 0·2 0
< 0·001
30
60
90
120
Time after surgery (months)
< 0·001
No. at risk HDLN-negative, stage I HDLN-negative, stage II HDLN-negative, stage III HDLN-positive, all stages HDLN-negative, stage IV
737 379 568 68 120
663 306 321 22 24
583 263 224 12 8
441 212 167 11 5
347 154 136 9 3
Overall survival of patients with hepatoduodenal ligament lymph node (HDLN) metastasis (any stage) versus patients with stage I, II, III or IV cancer (seventh American Joint Committee on Cancer classification) without HDLN metastasis. P < 0·001 (all HDLN-positive versus HDLN-negative stage III), P = 0·007 (all HDLN-positive versus HDLN-negative stage IV) (log rank test)
Fig. 1
< 0·001
Values in parentheses are percentages unless indicated otherwise; *values are mean(s.d.). HDLN, hepatoduodenal lymph node. †According to the seventh edition of the American Joint Committee on Cancer classification. ‡χ2 test, except §t test.
in Korea and Japan remove the HDLN routinely during D2 dissection16 . The aim of this study was to evaluate the clinical relevance of the HDLN and how this should be classified in gastric cancer. Methods
Patients who were diagnosed with gastric adenocarcinoma and underwent gastrectomy accompanied by lymph node dissection of D2 or greater with histopathological assessment of HDLN at Uijeongbu and Seoul St Mary’s Hospitals between 1989 and 2009 were included. Those who received neoadjuvant chemotherapy or had other organ malignancies were excluded. The study was approved by the Institutional Review Board of the Ethics Committee of the College of Medicine, Catholic University of Korea (XC13RIMI0034U). Clinicopathological data, including tumour stage, were collected from the Gastric Cancer Patients Registries of Uijeongbu and Seoul St Mary’s Hospitals. The pathological stage of all enrolled patients was reclassified 2014 BJS Society Ltd Published by John Wiley & Sons Ltd
1·0
P‡
Cumulative survival rate
No. of patients (n = 1872)
HDLN-negative, stage I HDLN-negative, stage II HDLN-negative, stage III HDLN-positive, all stages HDLN-negative, stage IV
according to the seventh AJCC classification criteria, with HDLN involvement included in the number of lymph node metastases. Histological type was categorized as differentiated or undifferentiated. Poorly differentiated tubular adenocarcinoma, signet ring cell adenocarcinoma and mucinous adenocarcinoma were assigned to the undifferentiated group. Regular follow-up programmes were conducted according to standard protocols (every 3 and 6 months for advanced and early gastric cancer respectively, for the first 3 years; every 12 months thereafter) and included the determination of tumour marker levels, abdominal imaging and endoscopic examination. Overall survival rates were determined from the day of surgery until the day of death using registration data from the Korean National Statistical Office and patient records.
Statistical analysis Differences between groups were analysed using the t test for continuous variables and the χ2 test or Fisher’s exact test for proportions. Survival analysis was done using the Kaplan–Meier method with a log rank test for univariable www.bjs.co.uk
BJS 2014; 101: 518–522
520
S. L. Lee, H. H. Lee, Y. H. Ko, K. Y. Song, C. H. Park, H. M. Jeon and S. S. Kim
HDLN-negative, stage I HDLN-negative, stage II HDLN-negative, stage III HDLN-positive, stage I–III
HDLN-positive, stage I–III HDLN-negative, stage IV 1·0 Cumulaive survival rate
Cumulaive survival rate
1·0 0·8 0·6 0·4 0·2 0
30
60
90
120
0·8 0·6 0·4 0·2 0
Time after surgery (months) No. at risk HDLN-positive, stage I–III 41 HDLN-negative, stage IV 120
a
18 24
11 8
11 5
30
60
90
120
Time after surgery (months) 9 3
HDLN-positive stage I–III versus HDLN-negative stage IV
No. at risk HDLN-negative, stage I 737 HDLN-negative, stage II 379 HDLN-negative, stage III 568 HDLN-positive, stage I–III 41
b
663 306 321 18
583 263 224 11
441 212 167 11
347 154 136 9
HDLN-positive stage I–III versus HDLN-negative stages I, II and III
Overall survival of patients with stage I–III cancer and hepatoduodenal ligament lymph node (HDLN) metastasis versus a patients with stage IV disease without HDLN metastasis and b patients with stage I, II or III disease without HDLN metastasis. a P < 0·001; b P = 0·184 ( HDLN-positive versus HDLN-negative stage III) (log rank test)
Fig. 2
analysis. Multivariable analysis for survival was carried out using a Cox proportional hazards model with the backward logistic regression method. Statistical analyses were performed with SPSS version 13.0 software (IBM, Armonk, New York, USA). P < 0·050 was considered to indicate statistical significance. Table 2
Results
The mean(s.d.) follow-up time for the 1872 enrolled patients was 93·1(72·7) (range 0·3–267·4) months. Of these, 68 (3·6 per cent) had a HDLN metastasis. The characteristics of patients with and without HDLN
Multivariable logistic regression analysis of prognostic factors for survival
Age (≥ 60 versus < 60 years) Sex (M versus F) Extent of resection (total versus subtotal) Histology (undifferentiated versus differentiated) Depth of invasion T2 versus T1 T3 versus T1 T4 versus T1 Lymph node metastasis N1 versus N0 N2 versus N0 N3 versus N0 Distant metastasis (M1 versus M0) HDLN (positive versus negative) Station 8a lymph node (positive versus negative) Station 9 lymph node (positive versus negative) Station 13 lymph node (positive versus negative) Station 14v lymph node (positive versus negative) Station 16 lymph node (positive versus negative)
Coefficient
s.e.
Hazard ratio
P
0·475 0·224 0·065 0·048
0·077 0·084 0·088 0·081
1·61 (1·38, 1·87) 1·25 (1·06, 1·47) 1·07 (0·90, 1·27) 1·05 (0·90, 1·23)
< 0·001 0·008 0·461 0·552
0·358 0·820 1·320
0·166 0·149 0·148
1·43 (1·03, 1·98) 2·27 (1·70, 3·05) 3·74 (2·80, 5·00)
0·031 < 0·001 < 0·001
0·120 0·435 1·032 0·987 0·029 0·615 0·138 0·639 0·208 0·824
0·145 0·137 0·125 0·112 0·164 0·346 0·120 0·277 0·107 0·205
1·13 (0·84, 1·50) 1·55 (1·18, 2·02) 2·81 (2·20, 3·59) 2·66 (2·14, 3·31) 1·03 (0·75, 1·42) 1·85 (0·94, 3·64) 1·15 (0·91, 1·45) 1·89 (1·10, 3·26) 1·23 (1·00, 1·52) 2·28 (1·52, 3·41)
0·409 0·002 < 0·001 < 0·001 0·860 0·075 0·248 0·021 0·051 < 0·001
Values in parentheses are 95 per cent confidence intervals. HDLN, hepatoduodenal lymph node.
2014 BJS Society Ltd Published by John Wiley & Sons Ltd
www.bjs.co.uk
BJS 2014; 101: 518–522
Hepatoduodenal lymph node metastasis in gastric cancer
metastasis are shown in Table 1. HDLN metastasis was found more frequently in patients who underwent total gastrectomy. The proportion of patients with HDLN metastasis was higher with more advanced tumours. Sixtyone (90 per cent) of 68 patients with metastasis in the HDLN had N3 disease. The 5-year survival rate for the 68 patients with HDLN metastasis was 30 per cent, which was intermediate between and significantly different from 5-year survival rates for patients with stage III (47·7 per cent; P < 0·001) and stage IV (9·8 per cent, P = 0·007) gastric cancer without HDLN metastasis (Fig. 1). The 5-year survival rate of 41 patients with HDLN metastasis, after excluding 27 patients with definite distant metastasis, was analysed to clarify the intrinsic effect of HDLN metastasis on survival. These patients had a significantly better 5-year survival rate than the 120 patients with stage IV disease without HDLN metastasis (42 versus 9·8 per cent respectively; P < 0·001) (Fig. 2a). The 5-year survival rate did not differ significantly between the 41 patients with stage I–III disease with HDLN metastasis and 568 patients who had stage III tumours without HDLN metastasis (42 versus 47·7 per cent respectively; P = 0·184) (Fig. 2b). Although overall 5-year survival differed significantly according to HDLN involvement (P < 0·001), HDLN metastasis was not an independent prognostic factor in the multivariable analysis (Table 2). Discussion
HDLN metastasis was associated with advanced gastric cancer. The 5-year survival rate of 41 patients with HDLN metastasis and no evidence of distant metastasis was similar to that of patients with stage III tumours without HDLN metastasis. Their survival was significantly better than that of patients with stage IV disease without HDLN metastasis. These results indicate that HDLN metastasis does not carry the same prognosis as established distant metastasis, as implied by the seventh AJCC classification system. The inclusion of patients with a positive HDLN in AJCC stage IV creates apparently improved survival for this group compared with that for other classification systems, where the 5-year survival rate is usually below 10 per cent4,17 . The removal of these patients with HDLN-positive disease from those otherwise classified as having AJCC stage III disease has little effect on stage-specific survival as patients who are HDLN-positive have a 5-year survival rate similar to that of patients with AJCC stage III disease, although clearly the new staging system must reduce the proportion 2014 BJS Society Ltd Published by John Wiley & Sons Ltd
521
of patients ascribed to stage III disease at the expense of an increase in stage IV disease. A Cox proportional hazards regression model demonstrated that HDLN metastasis was not an independent prognostic factor, as most patients with HDLN metastasis were in the N3 group. Although lymph node metastasis was a significant prognostic factor, the N1 category of lymph node metastasis had no significant survival value. Controversy exists regarding the seventh AJCC classification system for lymph node metastasis6,18 . Many studies7,8,19 have commented on the changes in node status criteria compared with the sixth AJCC classification criteria. Previous studies have demonstrated that the seventh AJCC classification is poor at discriminating between N1 and N2 categories3,20 , and this is confirmed by the present results. The HDLN has three components based on lymph node location: lymph nodes along the proper hepatic artery (12a), along the bile duct (12b) and along the portal vein (12p)15 . Even though station 12 is classified as a regional lymph node in the JGCA classification system, D2 dissection during a distal and total gastrectomy includes only station 12a according to the JGCA treatment guidelines16 . Although the three components may have differing degrees of importance in terms of prognosis, the present study could not evaluate these nodes individually. HDLN metastasis in gastric cancer is not associated with the same survival as other distant metastases included in the seventh AJCC classification. HDLN should be appropriately recognized as a regional lymph node. Survival comparisons between centres in the East and West will continue to be hampered by variations in staging systems. Acknowledgements
This study was supported by a grant from the National Research Foundation of Korea (no. 2012R1A1A1043576). Disclosure: The authors declare no conflict of interest. References 1 Edge SB, Byrd DR, Compton CC, Fritz AG, Greene FL, Trotti A. American Joint Committee on Cancer (AJCC) Cancer Staging Manual (7th edn). Springer: New York, 2010. 2 Edge SB, Compton CC. The American Joint Committee on Cancer: the 7th edition of the AJCC cancer staging manual and the future of TNM. Ann Surg Oncol 2010; 17: 1471–1474. 3 Jung H, Lee HH, Song KY, Jeon HM, Park CH. Validation of the seventh edition of the American Joint Committee on Cancer TNM staging system for gastric cancer. Cancer 2011; 117: 2371–2378. 4 Marrelli D, Morgagni P, de Manzoni G, Coniglio A, Marchet A, Saragoni L et al.; Italian Research Group for
www.bjs.co.uk
BJS 2014; 101: 518–522
522
5
6
7
8
9
10
11
12
S. L. Lee, H. H. Lee, Y. H. Ko, K. Y. Song, C. H. Park, H. M. Jeon and S. S. Kim
Gastric Cancer (IRGGC). Prognostic value of the 7th AJCC/UICC TNM classification of noncardia gastric cancer: analysis of a large series from specialized Western centers. Ann Surg 2012; 255: 486–491. Dikken JL, van de Velde CJ, Gonen M, Verheij M, Brennan MF, Coit DG. The New American Joint Committee on Cancer/International Union Against Cancer staging system for adenocarcinoma of the stomach: increased complexity without clear improvement in predictive accuracy. Ann Surg Oncol 2012; 19: 2443–2451. Kim SH, Ha TK, Kwon SJ. Evaluation of the 7th AJCC TNM staging system in point of lymph node classification. J Gastric Cancer 2011; 11: 94–100. Fang WL, Huang KH, Chen JH, Lo SS, Hsieh MC, Shen KH et al. Comparison of the survival difference between AJCC 6th and 7th editions for gastric cancer patients. World J Surg 2011; 35: 2723–2729. Chae S, Lee A, Lee JH. The effectiveness of the new (7th) UICC N classification in the prognosis evaluation of gastric cancer patients: a comparative study between the 5th/6th and 7th UICC N classification. Gastric Cancer 2011; 14: 166–171. Mezhir JJ, Shah MA, Jacks LM, Brennan MF, Coit DG, Strong VE. Positive peritoneal cytology in patients with gastric cancer: natural history and outcome of 291 patients. Ann Surg Oncol 2010; 17: 3173–3180. Tokunaga M, Ohyama S, Hiki N, Fukunaga T, Aikou S, Yamaguchi T. Can superextended lymph node dissection be justified for gastric cancer with pathologically positive para-aortic lymph nodes? Ann Surg Oncol 2010; 17: 2031–2036. An JY, Pak KH, Inaba K, Cheong JH, Hyung WJ, Noh SH. Relevance of lymph node metastasis along the superior mesenteric vein in gastric cancer. Br J Surg 2011; 98: 667–672. Masuda TA, Sakaguchi Y, Toh Y, Aoki Y, Harimoto N,
2014 BJS Society Ltd Published by John Wiley & Sons Ltd
13
14
15
16
17
18
19
20
Taomoto J et al. Clinical characteristics of gastric cancer with metastasis to the lymph node along the superior mesenteric vein (14v). Dig Surg 2008; 25: 351–358. Tokunaga M, Ohyama S, Hiki N, Fukunaga T, Inoue H, Yamada K et al. Therapeutic value of lymph node dissection in advanced gastric cancer with macroscopic duodenum invasion: is the posterior pancreatic head lymph node dissection beneficial? Ann Surg Oncol 2009; 16: 1241–1246. Sobin LH, Gospodarowicz MK, Wittekind C. International Union Against Cancer (UICC) TNM Classification of Malignant Tumors (7th edn). Wiley-Blackwell: Oxford, 2010. Japanese Gastric Cancer Association. Japanese classification of gastric carcinoma: 3rd English edition. Gastric Cancer 2011; 14: 101–112. Japanese Gastric Cancer Association. Japanese gastric cancer treatment guidelines 2010 (ver. 3). Gastric Cancer 2011; 14: 113–123. Kim DH, Oh CA, Oh SJ, Choi MG, Noh JH, Sohn TS et al. Validation of seventh edition AJCC gastric cancer staging modifications. J Surg Oncol 2012; 105: 26–30. Sun Z, Wang ZN, Zhu Z, Xu YY, Xu Y, Huang BJ et al. Evaluation of the seventh edition of American Joint Committee on Cancer TNM staging system for gastric cancer: results from a Chinese monoinstitutional study. Ann Surg Oncol 2012; 19: 1918–1927. Ahn HS, Lee HJ, Hahn S, Kim WH, Lee KU, Sano T et al. Evaluation of the seventh American Joint Committee on Cancer/International Union Against Cancer Classification of gastric adenocarcinoma in comparison with the sixth classification. Cancer 2010; 116: 5592–5598. Lee HH, Song KY, Park CH, Jeon HM. Undifferentiated-type gastric adenocarcinoma: prognostic impact of three histological types. World J Surg Oncol 2012; 10: 254.
www.bjs.co.uk
BJS 2014; 101: 518–522
Relevance of hepatoduodenal ligament lymph nodes in resectional surgery for gastric cancer S. L. Lee1 , H. H. Lee2 , Y. H. Ko3 , K. Y. Song2 , C. H. Park2 , H. M. Jeon2 and S. S. Kim4 1
Department of Radiology, 2 Division of Gastrointestinal Surgery, Department of Surgery, 3 Division of Medical Oncology, Department of Internal Medicine, and 4 Division of Gastroenterology, Department of Internal Medicine, College of Medicine, Catholic University of Korea, Seoul, Korea Correspondence to: Dr H. H. Lee, Division of Gastrointestinal Surgery, Department of Surgery, Catholic University of Korea, 65-1 Gumo-Dong, Uijeongbu-City, Gyenggi-Do, 480-717, Korea (e-mail: [email protected])
Background: Hepatoduodenal lymph node (HDLN) positivity is considered distant metastasis in gastric
cancer according to the seventh American Joint Committee on Cancer (AJCC) classification. In contrast, the International Union Against Cancer seventh edition and the Japanese Gastric Cancer Association both consider HDLN as a regional lymph node that can be included in the context of a curative resection. The purpose of this study was to determine whether there was justification for considering HDLN involvement as a distant metastasis for which resectional surgery could not have survival benefit. Methods: This study enrolled consecutive patients with gastric cancer having D2 or greater resections, with removal and pathological assessment of the HDLN, between 1989 and 2009. The pathological stage of all patients was determined based on the seventh AJCC criteria, with HDLN included as a regional lymph node. Results: A total of 1872 patients had their HDLN removed, of whom 68 had a metastatic lymph node in the hepatoduodenal ligament. The 5-year survival rate of these 68 patients was 30 per cent, compared with 47·7 per cent for those with stage III (P < 0·001) and 9·8 per cent for those with stage IV (P = 0·007) HDLN-negative tumours. The 5-year survival rate of 41 patients with HDLN metastasis and no evidence of distant metastasis at any other site was significantly higher than that among 120 patients with stage IV disease without HDLN metastasis (P < 0·001), whereas 5-year survival did not differ between the 41 patients with stage I–III disease with HDLN metastasis and 568 patients with stage III tumours without HDLN metastasis (P = 0·184). HDLN metastasis was not a significant factor for survival in multivariable analysis. Conclusion: It is inappropriate to include the HDLN in the distant metastatic lymph node group in gastric cancer. The seventh AJCC criteria for node grouping should be revised. Presented to the Tenth International Gastric Cancer Congress, Verona, Italy, June 2013 Paper accepted 18 December 2013 Published online 26 February 2014 in Wiley Online Library (www.bjs.co.uk). DOI: 10.1002/bjs.9438
Introduction
The seventh American Joint Committee on Cancer (AJCC) classification contains many changes in the staging of gastric cancer1,2 . The subserosal layer was upgraded to T3, and T4 was subclassified into T4a for serosal invasion and T4b for adjacent structural invasion. Criteria for the number of metastatic lymph nodes were revised, and stage N3 was subdivided into N3a and N3b. Positive peritoneal cytology was classified as M1. Although there has been some controversy regarding these classification changes3 – 9 , important changes related to the distant metastasis grouping seem to have gone unnoticed. 2014 BJS Society Ltd Published by John Wiley & Sons Ltd
The seventh AJCC classification includes hepatoduodenal, retropancreatic, mesenteric and para-aortic lymph nodes in the nodal grouping definition of distant metastasis sites for gastric cancer1 . Most surgeons would accept that there is no convincing evidence that resection of paraaortic (station 16 lymph node), retropancreatic (station 13) and mesenteric (station 14v) lymph nodes, when involved, confers significant survival benefit10 – 13 . The hepatoduodenal lymph node (HDLN; station 12), however, is still considered as a regional lymph node by the International Union Against Cancer (UICC) tumour node metastasis (TNM) seventh edition14 and the Japanese Gastric Cancer Association (JGCA)15 . Gastric cancer surgeons BJS 2014; 101: 518–522
Hepatoduodenal lymph node metastasis in gastric cancer
519
Comparison of clinicopathological factors between patients with and without hepatoduodenal lymph node metastasis
Table 1
Age (years)* Sex M 1255 F 617 Extent of resection Total 432 Subtotal 1440 Histological type Differentiated 788 Undifferentiated 1084 Depth of invasion† T1 621 T2 276 T3 389 T4 586 Lymph node metastasis† N0 868 N1 257 N2 245 N3 502 Pathological stage† I 737 II 381 III 607 IV 147
HDLNnegative (n = 1804)
HDLNpositive (n = 68)
56·6(12·2)
57·7(11·5)
1206 (96·1) 598 (96·9)
49 (3·9) 19 (3·1)
406 (94·0) 1398 (97·1)
26 (6·0) 42 (2·9)
0·465§ 0·370
0·003
0·159 765 (97·1) 1039 (95·8)
23 (2·9) 45 (4·2)
621 (100) 266 (96·4) 377 (96·9) 540 (92·2)
0 (0) 10 (3·6) 12 (3·1) 46 (7·8)
868 (100) 255 (99·2) 240 (98·0) 441 (87·8)
0 (0) 2 (0·8) 5 (2·0) 61 (12·2)
737 (100) 379 (99·5) 568 (93·6) 120 (81·6)
0 (0) 2 (0·5) 39 (6·4) 27 (18·4)
0·8 0·6 0·4 0·2 0
< 0·001
30
60
90
120
Time after surgery (months)
< 0·001
No. at risk HDLN-negative, stage I HDLN-negative, stage II HDLN-negative, stage III HDLN-positive, all stages HDLN-negative, stage IV
737 379 568 68 120
663 306 321 22 24
583 263 224 12 8
441 212 167 11 5
347 154 136 9 3
Overall survival of patients with hepatoduodenal ligament lymph node (HDLN) metastasis (any stage) versus patients with stage I, II, III or IV cancer (seventh American Joint Committee on Cancer classification) without HDLN metastasis. P < 0·001 (all HDLN-positive versus HDLN-negative stage III), P = 0·007 (all HDLN-positive versus HDLN-negative stage IV) (log rank test)
Fig. 1
< 0·001
Values in parentheses are percentages unless indicated otherwise; *values are mean(s.d.). HDLN, hepatoduodenal lymph node. †According to the seventh edition of the American Joint Committee on Cancer classification. ‡χ2 test, except §t test.
in Korea and Japan remove the HDLN routinely during D2 dissection16 . The aim of this study was to evaluate the clinical relevance of the HDLN and how this should be classified in gastric cancer. Methods
Patients who were diagnosed with gastric adenocarcinoma and underwent gastrectomy accompanied by lymph node dissection of D2 or greater with histopathological assessment of HDLN at Uijeongbu and Seoul St Mary’s Hospitals between 1989 and 2009 were included. Those who received neoadjuvant chemotherapy or had other organ malignancies were excluded. The study was approved by the Institutional Review Board of the Ethics Committee of the College of Medicine, Catholic University of Korea (XC13RIMI0034U). Clinicopathological data, including tumour stage, were collected from the Gastric Cancer Patients Registries of Uijeongbu and Seoul St Mary’s Hospitals. The pathological stage of all enrolled patients was reclassified 2014 BJS Society Ltd Published by John Wiley & Sons Ltd
1·0
P‡
Cumulative survival rate
No. of patients (n = 1872)
HDLN-negative, stage I HDLN-negative, stage II HDLN-negative, stage III HDLN-positive, all stages HDLN-negative, stage IV
according to the seventh AJCC classification criteria, with HDLN involvement included in the number of lymph node metastases. Histological type was categorized as differentiated or undifferentiated. Poorly differentiated tubular adenocarcinoma, signet ring cell adenocarcinoma and mucinous adenocarcinoma were assigned to the undifferentiated group. Regular follow-up programmes were conducted according to standard protocols (every 3 and 6 months for advanced and early gastric cancer respectively, for the first 3 years; every 12 months thereafter) and included the determination of tumour marker levels, abdominal imaging and endoscopic examination. Overall survival rates were determined from the day of surgery until the day of death using registration data from the Korean National Statistical Office and patient records.
Statistical analysis Differences between groups were analysed using the t test for continuous variables and the χ2 test or Fisher’s exact test for proportions. Survival analysis was done using the Kaplan–Meier method with a log rank test for univariable www.bjs.co.uk
BJS 2014; 101: 518–522
520
S. L. Lee, H. H. Lee, Y. H. Ko, K. Y. Song, C. H. Park, H. M. Jeon and S. S. Kim
HDLN-negative, stage I HDLN-negative, stage II HDLN-negative, stage III HDLN-positive, stage I–III
HDLN-positive, stage I–III HDLN-negative, stage IV 1·0 Cumulaive survival rate
Cumulaive survival rate
1·0 0·8 0·6 0·4 0·2 0
30
60
90
120
0·8 0·6 0·4 0·2 0
Time after surgery (months) No. at risk HDLN-positive, stage I–III 41 HDLN-negative, stage IV 120
a
18 24
11 8
11 5
30
60
90
120
Time after surgery (months) 9 3
HDLN-positive stage I–III versus HDLN-negative stage IV
No. at risk HDLN-negative, stage I 737 HDLN-negative, stage II 379 HDLN-negative, stage III 568 HDLN-positive, stage I–III 41
b
663 306 321 18
583 263 224 11
441 212 167 11
347 154 136 9
HDLN-positive stage I–III versus HDLN-negative stages I, II and III
Overall survival of patients with stage I–III cancer and hepatoduodenal ligament lymph node (HDLN) metastasis versus a patients with stage IV disease without HDLN metastasis and b patients with stage I, II or III disease without HDLN metastasis. a P < 0·001; b P = 0·184 ( HDLN-positive versus HDLN-negative stage III) (log rank test)
Fig. 2
analysis. Multivariable analysis for survival was carried out using a Cox proportional hazards model with the backward logistic regression method. Statistical analyses were performed with SPSS version 13.0 software (IBM, Armonk, New York, USA). P < 0·050 was considered to indicate statistical significance. Table 2
Results
The mean(s.d.) follow-up time for the 1872 enrolled patients was 93·1(72·7) (range 0·3–267·4) months. Of these, 68 (3·6 per cent) had a HDLN metastasis. The characteristics of patients with and without HDLN
Multivariable logistic regression analysis of prognostic factors for survival
Age (≥ 60 versus < 60 years) Sex (M versus F) Extent of resection (total versus subtotal) Histology (undifferentiated versus differentiated) Depth of invasion T2 versus T1 T3 versus T1 T4 versus T1 Lymph node metastasis N1 versus N0 N2 versus N0 N3 versus N0 Distant metastasis (M1 versus M0) HDLN (positive versus negative) Station 8a lymph node (positive versus negative) Station 9 lymph node (positive versus negative) Station 13 lymph node (positive versus negative) Station 14v lymph node (positive versus negative) Station 16 lymph node (positive versus negative)
Coefficient
s.e.
Hazard ratio
P
0·475 0·224 0·065 0·048
0·077 0·084 0·088 0·081
1·61 (1·38, 1·87) 1·25 (1·06, 1·47) 1·07 (0·90, 1·27) 1·05 (0·90, 1·23)
< 0·001 0·008 0·461 0·552
0·358 0·820 1·320
0·166 0·149 0·148
1·43 (1·03, 1·98) 2·27 (1·70, 3·05) 3·74 (2·80, 5·00)
0·031 < 0·001 < 0·001
0·120 0·435 1·032 0·987 0·029 0·615 0·138 0·639 0·208 0·824
0·145 0·137 0·125 0·112 0·164 0·346 0·120 0·277 0·107 0·205
1·13 (0·84, 1·50) 1·55 (1·18, 2·02) 2·81 (2·20, 3·59) 2·66 (2·14, 3·31) 1·03 (0·75, 1·42) 1·85 (0·94, 3·64) 1·15 (0·91, 1·45) 1·89 (1·10, 3·26) 1·23 (1·00, 1·52) 2·28 (1·52, 3·41)
0·409 0·002 < 0·001 < 0·001 0·860 0·075 0·248 0·021 0·051 < 0·001
Values in parentheses are 95 per cent confidence intervals. HDLN, hepatoduodenal lymph node.
2014 BJS Society Ltd Published by John Wiley & Sons Ltd
www.bjs.co.uk
BJS 2014; 101: 518–522
Hepatoduodenal lymph node metastasis in gastric cancer
metastasis are shown in Table 1. HDLN metastasis was found more frequently in patients who underwent total gastrectomy. The proportion of patients with HDLN metastasis was higher with more advanced tumours. Sixtyone (90 per cent) of 68 patients with metastasis in the HDLN had N3 disease. The 5-year survival rate for the 68 patients with HDLN metastasis was 30 per cent, which was intermediate between and significantly different from 5-year survival rates for patients with stage III (47·7 per cent; P < 0·001) and stage IV (9·8 per cent, P = 0·007) gastric cancer without HDLN metastasis (Fig. 1). The 5-year survival rate of 41 patients with HDLN metastasis, after excluding 27 patients with definite distant metastasis, was analysed to clarify the intrinsic effect of HDLN metastasis on survival. These patients had a significantly better 5-year survival rate than the 120 patients with stage IV disease without HDLN metastasis (42 versus 9·8 per cent respectively; P < 0·001) (Fig. 2a). The 5-year survival rate did not differ significantly between the 41 patients with stage I–III disease with HDLN metastasis and 568 patients who had stage III tumours without HDLN metastasis (42 versus 47·7 per cent respectively; P = 0·184) (Fig. 2b). Although overall 5-year survival differed significantly according to HDLN involvement (P < 0·001), HDLN metastasis was not an independent prognostic factor in the multivariable analysis (Table 2). Discussion
HDLN metastasis was associated with advanced gastric cancer. The 5-year survival rate of 41 patients with HDLN metastasis and no evidence of distant metastasis was similar to that of patients with stage III tumours without HDLN metastasis. Their survival was significantly better than that of patients with stage IV disease without HDLN metastasis. These results indicate that HDLN metastasis does not carry the same prognosis as established distant metastasis, as implied by the seventh AJCC classification system. The inclusion of patients with a positive HDLN in AJCC stage IV creates apparently improved survival for this group compared with that for other classification systems, where the 5-year survival rate is usually below 10 per cent4,17 . The removal of these patients with HDLN-positive disease from those otherwise classified as having AJCC stage III disease has little effect on stage-specific survival as patients who are HDLN-positive have a 5-year survival rate similar to that of patients with AJCC stage III disease, although clearly the new staging system must reduce the proportion 2014 BJS Society Ltd Published by John Wiley & Sons Ltd
521
of patients ascribed to stage III disease at the expense of an increase in stage IV disease. A Cox proportional hazards regression model demonstrated that HDLN metastasis was not an independent prognostic factor, as most patients with HDLN metastasis were in the N3 group. Although lymph node metastasis was a significant prognostic factor, the N1 category of lymph node metastasis had no significant survival value. Controversy exists regarding the seventh AJCC classification system for lymph node metastasis6,18 . Many studies7,8,19 have commented on the changes in node status criteria compared with the sixth AJCC classification criteria. Previous studies have demonstrated that the seventh AJCC classification is poor at discriminating between N1 and N2 categories3,20 , and this is confirmed by the present results. The HDLN has three components based on lymph node location: lymph nodes along the proper hepatic artery (12a), along the bile duct (12b) and along the portal vein (12p)15 . Even though station 12 is classified as a regional lymph node in the JGCA classification system, D2 dissection during a distal and total gastrectomy includes only station 12a according to the JGCA treatment guidelines16 . Although the three components may have differing degrees of importance in terms of prognosis, the present study could not evaluate these nodes individually. HDLN metastasis in gastric cancer is not associated with the same survival as other distant metastases included in the seventh AJCC classification. HDLN should be appropriately recognized as a regional lymph node. Survival comparisons between centres in the East and West will continue to be hampered by variations in staging systems. Acknowledgements
This study was supported by a grant from the National Research Foundation of Korea (no. 2012R1A1A1043576). Disclosure: The authors declare no conflict of interest. References 1 Edge SB, Byrd DR, Compton CC, Fritz AG, Greene FL, Trotti A. American Joint Committee on Cancer (AJCC) Cancer Staging Manual (7th edn). Springer: New York, 2010. 2 Edge SB, Compton CC. The American Joint Committee on Cancer: the 7th edition of the AJCC cancer staging manual and the future of TNM. Ann Surg Oncol 2010; 17: 1471–1474. 3 Jung H, Lee HH, Song KY, Jeon HM, Park CH. Validation of the seventh edition of the American Joint Committee on Cancer TNM staging system for gastric cancer. Cancer 2011; 117: 2371–2378. 4 Marrelli D, Morgagni P, de Manzoni G, Coniglio A, Marchet A, Saragoni L et al.; Italian Research Group for
www.bjs.co.uk
BJS 2014; 101: 518–522
522
5
6
7
8
9
10
11
12
S. L. Lee, H. H. Lee, Y. H. Ko, K. Y. Song, C. H. Park, H. M. Jeon and S. S. Kim
Gastric Cancer (IRGGC). Prognostic value of the 7th AJCC/UICC TNM classification of noncardia gastric cancer: analysis of a large series from specialized Western centers. Ann Surg 2012; 255: 486–491. Dikken JL, van de Velde CJ, Gonen M, Verheij M, Brennan MF, Coit DG. The New American Joint Committee on Cancer/International Union Against Cancer staging system for adenocarcinoma of the stomach: increased complexity without clear improvement in predictive accuracy. Ann Surg Oncol 2012; 19: 2443–2451. Kim SH, Ha TK, Kwon SJ. Evaluation of the 7th AJCC TNM staging system in point of lymph node classification. J Gastric Cancer 2011; 11: 94–100. Fang WL, Huang KH, Chen JH, Lo SS, Hsieh MC, Shen KH et al. Comparison of the survival difference between AJCC 6th and 7th editions for gastric cancer patients. World J Surg 2011; 35: 2723–2729. Chae S, Lee A, Lee JH. The effectiveness of the new (7th) UICC N classification in the prognosis evaluation of gastric cancer patients: a comparative study between the 5th/6th and 7th UICC N classification. Gastric Cancer 2011; 14: 166–171. Mezhir JJ, Shah MA, Jacks LM, Brennan MF, Coit DG, Strong VE. Positive peritoneal cytology in patients with gastric cancer: natural history and outcome of 291 patients. Ann Surg Oncol 2010; 17: 3173–3180. Tokunaga M, Ohyama S, Hiki N, Fukunaga T, Aikou S, Yamaguchi T. Can superextended lymph node dissection be justified for gastric cancer with pathologically positive para-aortic lymph nodes? Ann Surg Oncol 2010; 17: 2031–2036. An JY, Pak KH, Inaba K, Cheong JH, Hyung WJ, Noh SH. Relevance of lymph node metastasis along the superior mesenteric vein in gastric cancer. Br J Surg 2011; 98: 667–672. Masuda TA, Sakaguchi Y, Toh Y, Aoki Y, Harimoto N,
2014 BJS Society Ltd Published by John Wiley & Sons Ltd
13
14
15
16
17
18
19
20
Taomoto J et al. Clinical characteristics of gastric cancer with metastasis to the lymph node along the superior mesenteric vein (14v). Dig Surg 2008; 25: 351–358. Tokunaga M, Ohyama S, Hiki N, Fukunaga T, Inoue H, Yamada K et al. Therapeutic value of lymph node dissection in advanced gastric cancer with macroscopic duodenum invasion: is the posterior pancreatic head lymph node dissection beneficial? Ann Surg Oncol 2009; 16: 1241–1246. Sobin LH, Gospodarowicz MK, Wittekind C. International Union Against Cancer (UICC) TNM Classification of Malignant Tumors (7th edn). Wiley-Blackwell: Oxford, 2010. Japanese Gastric Cancer Association. Japanese classification of gastric carcinoma: 3rd English edition. Gastric Cancer 2011; 14: 101–112. Japanese Gastric Cancer Association. Japanese gastric cancer treatment guidelines 2010 (ver. 3). Gastric Cancer 2011; 14: 113–123. Kim DH, Oh CA, Oh SJ, Choi MG, Noh JH, Sohn TS et al. Validation of seventh edition AJCC gastric cancer staging modifications. J Surg Oncol 2012; 105: 26–30. Sun Z, Wang ZN, Zhu Z, Xu YY, Xu Y, Huang BJ et al. Evaluation of the seventh edition of American Joint Committee on Cancer TNM staging system for gastric cancer: results from a Chinese monoinstitutional study. Ann Surg Oncol 2012; 19: 1918–1927. Ahn HS, Lee HJ, Hahn S, Kim WH, Lee KU, Sano T et al. Evaluation of the seventh American Joint Committee on Cancer/International Union Against Cancer Classification of gastric adenocarcinoma in comparison with the sixth classification. Cancer 2010; 116: 5592–5598. Lee HH, Song KY, Park CH, Jeon HM. Undifferentiated-type gastric adenocarcinoma: prognostic impact of three histological types. World J Surg Oncol 2012; 10: 254.
www.bjs.co.uk
BJS 2014; 101: 518–522
