Extended lymphadenectomy (D2) in patients with early gastric cancer

EJSO (2005) 31, 875–881

www.ejso.com

Extended lymphadenectomy (D2) in patients with early gastric cancer D. Nittia,*, A. Marcheta, E. Mammanoa, A. Ambrosia, C. Bellucoa, R. Mencarellia, M. Mainoa, G. Marconatoa, F. Farinatib, M. Lisea a

Department of Oncological and Surgical Sciences, University of Padova, Padova, Italy Department of Surgical and Gastroenterological Sciences, University of Padova, Padova, Italy

b

Accepted for publication 18 May 2005 Available online 26 July 2005

KEYWORDS Early gastric cancer; Lymphadenectomy; Prognosis; Survival

Abstract Aims: To investigate the survival benefit of extended lymphadenectomy (D2) in EGC patients in one European Institution. Methods: A review was made of our prospective gastric database from January 1980 to December 2001. Of 527 patients with primary gastric adenocarcinoma, 119 with EGC underwent potentially curative resection (R0) with D2 lymphadenectomy. Results: There were two post-operative deaths. Of the 117 evaluable cases, 96 were classified as N0 and 21 as NC, with metastases in the perigastric lymph nodes (level 1) in 13, and beyond this site (level 2) in eight. Five-year survival was 85.9 and 83.0% in N0 and NC patients, respectively. During a median follow-up of 90 months, five of the eight patients with level 2 metastases died of recurrent disease and three were alive. The estimated survival benefit for 119 patients with EGC was 2.5% (3/119 cases). Conclusions: In patients with EGC, metastases to level 2 are rare. Our results indicate that D2 lymphadenectomy has a limited survival benefit and that in these cases a less extensive lymphadenectomy (D1) could be performed. Q 2005 Elsevier Ltd. All rights reserved.

Introduction In 1962, the Japanese Society of Gastroenterological Endoscopy (JSGE) defined early gastric cancer (EGC) as adenocarcinoma involving the mucosa or submucosa of the gastric wall, regardless of lymph * Corresponding author. Address: Clinica Chirurgica Generale 2, Via Giustiniani, 2, 35128 Padova, Italy. Tel.: C39 49 8212073; fax: C39 49 651891. E-mail address: [email protected] (D. Nitti).

node metastasis.1 The need to define this category of gastric cancer is related to the difference between the prognosis for EGC and that for advanced gastric cancer (AGC): following radical surgery the survival rates of patients with EGC ranges from 80 to 90%2,3 whereas the survival rate of patients with AGC ranges from 15 to 60% depending on the subgroups considered.4–6 The frequency of EGC is growing steadily in patients who undergo resection for gastric cancer;3,7–9 currently it is 20% in Western countries, whereas

0748-7983/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.ejso.2005.05.013

876 in Eastern countries, it is in the range of 50%.3,7–9 The choice of optimal surgical resection for patients with EGC is a widely debated issue.3,8,10–16 Until the last decade, Japanese standard guidelines for the surgical treatment of gastric cancer did not take in account primary tumour stage, and extended lymphadenectomy (D2) was routinely employed also in patients with EGC.17,18 However, since EGC patients have a relatively good prognosis, even in the presence of lymph node involvement,3,8 and metastases to level 2 lymph nodes are significantly less frequent in patients with EGC than in those with AGC,3,10,17,18 the Japan Gastric Cancer Association (JGCA) has recently proposed a more conservative approach for some EGC subpopulations, such as endoscopic mucosal resection (EMR) or modified gastrectomy associated with less extended lymphadenectomy.19 Standard gastrectomy with traditional D2 lymphadenectomy is considered indicated in EGC patients at a high risk of lymph node involvement and in those with T2 to T4 stage gastric cancer.19–21 In Western countries, surgical guidelines for the treatment of EGC patients have not yet been standardized. In order to ascertain whether a less extended lymphadenectomy is also feasible in EGC patients in Western countries, in the present study we investigate the survival benefit following D2 lymphadenectomy in a group of patients treated in a single European Institution.

Patients and methods Patients characteristics A review was made of our prospective gastric cancer database of cases collected from January 1980 to December 2001. Among 527 patients who underwent surgery for primary gastric adenocarcinoma, 157 had palliative surgery (R1–R2) for unresectable loco-regional tumour, distant metastases and microscopic residual disease in the resection margins or in the gastric bed. Of 370 who underwent potentially curative resection (R0), 251 had AGC, whereas 119 had diagnosis of EGC, 61 with mucosal (T1a), and 58 with submucosal (T1b) involvement (Fig. 1). All 119 patients underwent D2 lymphadenectomy, which included dissection of lymph node stations 1–12 according to the JGCA guidelines.22 Two patients died in the post-operative period (one of cardiac failure and the other of multiorgan failure secondary to sepsis). The present analysis is, therefore, based on data from 117 EGC patients. A single pathologist (RM) classified nodal

D. Nitti et al.

Figure 1 Among 527 patients treated for gastric cancer at our institution, 119 who underwent radical resection for early gastric cancer formed the series for the present analysis.

status and reviewed all the pathology data. Lymph nodes were classified according to the JGCA (n0, n1, n2, n3)22 and to the International Union Against Cancer (UICC) (N0–N1–N2–N3)23 staging systems. The post-operative outcomes of patients were obtained by making follow-up at regular intervals, the final data being collected on June 30, 2003. The median follow-up was 90 months (range 12–239 months).

Statistical analysis The statistical analysis was made using the Statistical Product and Service Solutions, SPSS 11.5 for Windows, 2003, SPSS Inc., Chicago, and R environment (R Foundation for Statistical Computing). A robust regression analysis24 was used to consider the number of metastatic lymph nodes as the function of the number of lymph nodes examined, related to the different T categories, taking into account the possible presence of outliers. Clinico-pathological variables analysed were: age (! or O 70 years), sex (male or female), tumour site (distal or proximal), type of surgical resection (subtotal gastric resection or total gastrectomy), depth of tumour penetration into the gastric wall (T1a or T1b), venous and lymphatic vessels involvement (yes or no), grading (G1, G2, G3 or G4),25 Lauren’s histological type (intestinal or diffuse), anatomical location of lymph-node metastases (n0Zno lymph-node metastases; n1Zmetastasis to node stations 1–6, level 1; n2Zmetastases to node stations 7–12, level 2), and number of lymph-node metastases (N0Zno lymph-node metastases; N1Z1–6 lymph-node metastases; and N2Z7–15 lymph-node metastases). Survival was determined using the Kaplan–Meier method, and the log-rank test was used to assess statistical differences between groups.26 P values of !.05 were considered significant.

Lymphadenectomy in early gastric cancer

Results Clinico-pathological findings The clinico-pathological characteristics of the 117 EGC evaluated patients, according to lymph node involvement, are reported in Table 1. In 105 patients, a distal subtotal gastrectomy was performed, whereas 12 patients underwent total gastrectomy. The total number of lymph nodes examined was 2899 (median, 24.78; range, 10–52). At pathology, 82 of 2899 lymph nodes were involved. Ninety-six patients were classified as n0, according to the JGCA. Metastases to level 1 (n1) were found in 13 patients, and to level 2 (n2) in eight. Depending on the number of lymph nodes involved (N), patients were classified as N0 (96 cases), N1 (15 cases), and N2 (six cases). When lymph node involvement was considered according to depth of tumour penetration into the gastric wall, lymph-node metastases were found in seven of the 60 patients with mucosal, and in 14 of the 57 patients with submucosal tumours (Table 2). Findings at robust regression analysis, considering the number of lymph nodes involved as a function of the number of lymph nodes examined in relation to the different T categories, are reported in Fig. 2. While in neither of the groups of patients with T1a

877 and T1b tumours did the number of metastatic lymph nodes increase with the number of nodes examined in patients with T2-4 tumours, there was a monotonic increase in the pattern for metastatic lymph nodes and the number of lymph nodes examined. During the follow-up, seven of the 117 patients with EGC died of distant or peritoneal metastasis: all had lymph node involvement. Disease recurrence occurred in two of the 13 patients with n1 and in five of the eight patients with n2 metastasis. Only three of the eight patients with n2 lymph node involvement were alive and disease-free.

Survival The overall 5-year survival was 85.4%. Survival at 5 years was 90.2% for T1a patients and 80.1% for T1b patients (Fig. 3). When 5-year survival was considered in relation to lymph node involvement, it was 85.9% in the 96 patients classified as N0 and 83.0% in the 21 patients with lymph node metastasis. Five-year survival was 91 and 75% for n1 and n2 patients, respectively, and 80 and 66% for N1 and N2 patients, respectively. At univariate analysis, the only factor of statistical significance for survival was age: patients younger than 70 years had a higher 5-year survival rate than older patients: 88 vs 74% (PZ.03).

Figure 2 Robust regression analysis of metastatic and examined lymph nodes related to T categories in 331 patients (117 with early gastric cancer and 214 with advanced gastric cancer) who underwent radical resection for gastric cancer.

878

D. Nitti et al.

Table 1 Clinico-pathologic characteristics of 117 patients with early gastric cancer according to lymph nodes status Factor

Number of patients NK

All 96 Sex Male 56 Female 40 Age !70 years 67 R70 years 29 Site Proximal 6 Distal 90 Surgery Gastric resection 85 Total gastrectomy 11 Grading Well differentiated (G1) 44 12 Moderately differentiated (G2) Poorly differentiated (G3) 33 Undifferentiated (G4) 7 Lauren’s histotype Intestinal 50 Diffuse 17 Unknown 29 Lymphatic invasion No 89 Yes 7 Venous invasion No 92 Yes 4 Depth of invasion Mucosal (T1a) 53 Submucosal (T1b) 43 Level of metastatic lymph nodes n0 96 n1 – n2 – Number of metastatic lymph nodes N0 96 N1 – N2 –

NC

Total

21

117

13 8

69 48

18 3

85 32

1 20

7 110

20 1

105 12

10 4

54 16

6 1

39 8

10 5 6

60 22 35

18 3

107 10

20 1

112 5

7 14

60 57

– 13 8

96 13 8

– 15 6

96 15 6

EGC and lymph nodes involvement Since 25% of our patients with involvement of the submucosa (T1b) presented lymph node metastasis, as also reported by other authors,3,10,11 it is important to identify EGC patients at a higher risk of lymph node metastasis, who could benefit from a more extended lymphadenectomy. At pre-operative staging for EGC patients, endoscopic ultrasonography can differentiate between mucosal and submucosal involvement. Various authors have reported that the accuracy of endosonography for T stage ranges from 65 to 85%.8,27 The size of the tumour8 and its macroscopic characteristics at endoscopy28 are further factors predictive of node metastasis in patients with EGC. In their analysis of 165 patients who underwent radical surgery for EGC, Hochwald et al.8 observed a rate of positive nodes of 4% in tumours limited to the mucosa and measuring less than 4.5 cm. On the other hand, lymph node metastasis was found in 56% of tumours measuring more than 4.5 cm and penetrating the submucosa. Sano et al.28 found that variant protruding-type mucosal EGC (type I and IIA according to the JGCA)22 with a diameter of less than 25 mm, and excavated-type mucosal EGC (type IIC) with a diameter of less than 20 mm, were free of lymph node metastasis. However, in their retrospective study on 1051 EGC patients, Shimada et al.11 found lymph node metastasis in 4.8% of cases of mucosal tumours with ulceration or ulceration scar, but no lymph node involvement in patients with tumours without ulceration. Lymphatic mapping and sentinel node biopsy are emerging as technically feasible and accurate tools in the assessment of nodal status in patients with gastric cancer. Moreover, this approach appears promising in patients with EGC, as described by Kitagawa et al.,29 who reported a detection rate of 99% and a sensitivity rate of 100% in their experience on 140 consecutive EGC patients. However, several problems need to be taken into account before this technique can be routinely used. In fact, in patients with gastric cancer the frequency of skip metastases has been reported to be quite high (15–20%), and the number of patients

Discussion The optimal surgical treatment of patients with EGC is widely debated by surgical oncologists. In order to investigate the value of extended lymphadenectomy in patients with radically resected EGC, we focused on the incidence and characteristics of lymph node metastasis and their relation to longterm outcome in 117 patients.

Table 2 Lymph node involvement (n) related to T categories in 117 patients with early gastric cancer Number of patients T1a T1b Total

n0

n1

n2

Total

53 43 96

5 8 13

2 6 8

60 57 117

Lymphadenectomy in early gastric cancer

Figure 3

879

Survival in relation to the depth of tumour invasion in 117 patients with early gastric cancer.

with sentinel lymph node as the only node with metastatic involvement is lower than that in patients with other types of tumours, such as breast carcinoma and melanoma.30 Moreover, the frequency of false-negative sentinel nodes ranges from 0 to 39%.30

D2 lymphadenectomy and outcome in patients with EGC Extended lymphadenectomy (D2) is advocated in patients with gastric cancer because disease down-staging can occur if a limited lymphadenectomy (D1) is performed. Moreover, since the incidence of immunohistochemistry-detected micrometastases in the lymph nodes of patients with EGC has been reported to be about 20%,31 it can be argued that micrometasases in N2 nodes may negatively impact on the survival benefit of D1 dissection. On the other hand, as recently reported by Morgagni et al. in their study on 300 cases,32 the presence of lymph node micrometastases in EGC does not seem to significantly influence patient prognosis. On considering our data from regression analysis of lymph nodes involved and examined, a different pathway was found between EGC and AGC. The curve fitting the relationship between these variables had a constant pattern for early gastric cancer: the number of metastatic lymph nodes did not change with the number of lymph nodes examined, even when mucosal and submucosal involvement were considered separately. In AGC

patients (T2–T4), the number of metastatic lymph nodes increased in relation to the depth of tumour penetration and to the number of lymph nodes examined. This finding confirms that the significance of D2 lymphadenectomy in the staging, and probably in the outcomes of EGC patients is different from that in AGC patients.33–35 In our series, the estimated survival benefit following D2 lymphadenectomy for EGC was, therefore, 2.5% (3/119). Our data are similar to those reported by Yoshikawa et al.3 in their retrospective study on 1041 EGC patients who underwent D2 lymphadenectomy: in this study, the estimated survival benefit of D2 lymphadenectomy was 1.7% (18/1041 patients). Our findings confirm that, although rare, most recurrences in EGC patients who undergo radical resection, are distant and/or peritoneal.3,8,36,37 In their study on 1041 patients who underwent curative gastrectomy with D2 lymphadenectomy for primary EGC, Yoshikawa et al.3 found that only 4 of the 15 patients (1.4%) who died of recurrence developed loco-regional relapse. In their study on 461 patients, Kunisaki et al.36 reported that eight patients (1.7%) died of recurrence: of these deaths, three were due to distant metastases, three to peritoneal dissemination and two to lymph node recurrence. In their studies, Isozaki et al.,9 and Sano et al.,37 confirm the prevalence of distant recurrence, indicating that extended lymphadenectomy cannot prevent distant metastases: treatment failure is not due to local or lymph-nodes metastases but to systemic tumour spread.

880 In conclusion, in patients with EGC, D2 lymphadenectomy provides a limited survival benefit. This treatment should be reserved for patients at high risk of lymph node metastasis (i.e. with tumours involving the submucosa, or limited to the mucosa but larger than 4.5 cm) or with a positive sentinel node; a less extended lymphadenectomy can be pursued in most other cases.

Acknowledgements This work was supported in part by a grant (Progetti di Ricerca di Rilevante Interesse Nazionale 2002) from the M.I.U.R. (Ministero dell’Istruzione, dell’Universita ` e della Ricerca) and by the A.I.R.C. (Associazione Italiana per la Ricerca sul Cancro) Regional Grant 2005 Veneto. The authors thank Ms Sara Pearcey for her editorial assistance.

References 1. Murakami T. Pathomorphological diagnosis. Definition and gross classification of early gastric cancer. Gann Monogr Cancer Res 1971;11:53–5. 2. Itoh H, Oohata Y, Nakamura K, Nagata T, Mibu R, Nakayama F. Complete ten-year postgastrectomy follow-up of early gastric cancer. Am J Surg 1989;158:14–16. 3. Yoshikawa T, Tsuburaya A, Kobayashi O, Sairenji M, Motohashi H, Noguchi Y. Is D2 lymph node dissection necessary for early gastric cancer? Ann Surg Oncol 2002;9: 401–5. 4. Hundahl SA, Phillips JL, Menck HR. The National Cancer Data Base Report on poor survival of US gastric carcinoma patients treated with gastrectomy. Fifth Edition American Joint Committee on Cancer Staging, proximal disease, and the ‘different disease’ hypothesis. Cancer 2000;88:921–32. 5. Siewert JR, Bottcher K, Roder JC, Busch R, Hermanek P, Meyer HJ. Prognostic relevance of lymph node dissection in gastric carcinoma. German Gastric Carcinoma Study Group. Br J Surg 1993;80:1015–8. 6. McCulloch P, Niita ME, Kazi H, Gama-Rodrigues JJ. Gastrectomy with extended lymphadenectomy for primary treatment of gastric cancer. Br J Surg 2005;92:5–13. 7. Hohemberger P, Gretschel S. Gastric cancer. Seminar. Lancet 2003;362:305–15. 8. Hochwald SN, Brennan MF, Klimstra DS, Kim S, Karpeh MS. Is lymphadenectomy necessary for early gastric cancer? Ann Surg Oncol 1999;6:664–70. 9. Isozaki H, Tanaca N, Okajima K. General and specific prognostic factors of early gastric carcinoma treated with curative surgery. Hepatogastroenterology 1999;46: 1800–8. 10. Kunisaki C, Shimada H, Nomura M, Akiyama H. Appropriate lymph node dissection for early gastric cancer based on lymph node metastases. Surgery 2001;129:153–7. 11. Shimada S, Yagi Y, Shiomori K, Honmyo U, Hayashi N, Matsuo A, et al. Characterization of early gastric cancer and proposal of the optimal therapeutic strategy. Surgery 2001; 129:714–9.

D. Nitti et al. 12. Yamada H, Nihei Z, Yamashita T, Shirota Y, Ichikawa W, Sugihara K. Is lymphadenectomy needed for all submucosal gastric cancers? Eur J Surg 2001;167:199–203. 13. Adachi Y, Shiraishi N, Kitano S. Modern treatment of early gastric cancer: Review of the Japanese experience. Dig Surg 2002;19:333–9. 14. Yokota T, Saito T, Teshima S, Yamada Y, Iwamoto K, Takahashi M, et al. Probability of lymph node metastasis in small gastric cancer tumor: Is it an indication for limited surgery? Int Surg 2001;86:206–9. 15. Isozaki H, Okajima K, Ichinona T, Fujii K, Nomura E, Izumi N, et al. Distant lymph node metastasis of early gastric cancer. Surg Today 1997;27:600–5. 16. Inoue K, Tobe T, Kan N, Nio Y, Sakai M, Takeuchi E, et al. Problems in the definition and treatment of early gastric cancer. Br J Surg 1991;78:818–21. 17. Okamura T, Tsujitani S, Korenaga D, Haraguchi M, Baba H, Hiramoto Y, Bazzocchi F, et al. Lymphadenectomy for cure in patients with early gastric cancer and lymph node metastasis. Am J Surg 1988;155:476–80. 18. Noguchi Y, Imada T, Matsumoto A, Coit DG, Brennan MF. Radical surgery for gastric cancer: A review of the Japanese experience. Cancer 1989;64:2053–62. 19. Nakajima T. Gastric cancer treatment guidelines in Japan. Gastric Cancer 2002;5:1–5. 20. Maehara Y, Kakeji Y, Koga T, Emi Y, Baba H, Akazawa K, Sugimachi K, et al. Therapeutic value of lymph node dissection and the clinical outcome for patients with gastric cancer. Surgery 2002;131:S85–S91. 21. Kasakura Y, Mochizuki F, Wakabayashi K, Kochi M, Fuji M, Takayama T. An evaluation of the effectiveness of extended lymph node dissection in patients with gastric cancer: A retrospective study of 1403 cases at a single institution. J Surg Res 2002;103:252–9. 22. Japanese Gastric Cancer Association. Japanese classification of gastric carcinoma—2nd English edition. Gastric Cancer 1998;1:10–24. 23. Sobin LH, Wittekind CH. International Union Against Cancer (UICC). TNM classification of malignant tumors. 5th ed. New York: Wiley; 1997. 24. Hampel R, Ronchetti EM, Rousseew PJ, Atahel WA. Robust statistic: The approach based on influence functions. New York: Wiley; 1986. 25. Compton C, Sobin LH, for the Cancer Committee, College of American Pathologists, and the Task Force for Protocols on the Examination of Specimens from Patients with Gastric Cancer. Protocol for the examination of specimens removed from patients with gastric carcinoma: A basis for checklists. Arch Pathol Lab Med 1998;122:9–14. 26. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457–81. 27. Yanai H, Noguchi T, Mizumachi S, Tokiyama H, Nakamura H, Tada M, et al. A blind comparison of the effectiveness of endoscopic ultrasonography and endoscopy in staging early gastric cancer. Gastrointest Endosc 1999;44:361–5. 28. Sano T, Kobori O, Muto T. Lymph node metastasis from early gastric cancer: Endoscopic resection of tumor. Br J Surg 1992;79:241–4. 29. Kitagawa Y, Kitajima M. Lymphatic mapping for upper gastrointestinal malignancies. Semin Oncol 2004;31:409–14. 30. Schlag PM, Bembenek A, Schulze T. Sentinel node biopsy in gastrointestinal-tract cancer. Eur J Cancer 2004;40: 2022–32. 31. Cai J, Ikeguchi M, Maeta M, Kaibara N, et al. Micrometastases in lymph nodes and microinvasion of the muscolaris propria in primary lesions of submucosal gastric cancer. Surgery 2000;127:32–9.

Lymphadenectomy in early gastric cancer 32. Morgagni P, Saragoni L, Scarpi E, Zattini PS, Zaccaroni A, Morgagni D, et al. Lymph node micrometastases in early gastric cancer and their impact on prognosis. World J Surg 2003;27:558–61. 33. Lise M, Nitti D, Marchet A, Sahmoud T, Duez N, Fornasiero A, et al. Prognostic factors in resectable gastric cancer: Results of EORTC study no. 40813 on FAM adjuvant chemotherapy. Ann Surg Oncol 1995;2:495–501. 34. Nitti D, Marchet A, Olivieri M, Ambrosi A, Mencarelli R, Farinati F, et al. Lymphadenectomy in patients with gastric cancer: A critical review. Tumori 2003;2:S35–S8.

881 35. Nitti D, Marchet A, Belluco C, Oliviera M, Ambrosi A, Mammano E, et al. Ratio between metastatic and examined lymph nodes is an independent prognostic factor after D2 resection for gastric cancer: Analysis of a large European monoinstitutionalexperience.AnnSurgOncol2003;10:1077–85. 36. Kunisaki C, Shimada H, Takahashi M, Ookubo K, Moriwaki Y, Akiyama H, et al. Prognostic factors in early gastric cancer. Hepatogastroenterology 2001;48:294–8. 37. Sano T, Sasako M, Kinoshita T, Maruyama K, et al. Recurrence of early gastric cancer. Follow-up of 1475 patients and review of the Japanese literature. Cancer 1993;72:3174–8.