Modified versus standard D2 lymphadenectomy in total gastrectomy for nonjunctional gastric carcinoma with lymph node metastasis

Stomach Modified versus standard D2 lymphadenectomy in total gastrectomy for nonjunctional gastric carcinoma with lymph node metastasis Gennaro Galizia, MD, PhD,a Eva Lieto, MD, PhD,a Ferdinando De Vita, MD, PhD,b Paolo Castellano, MD, PhD,a Francesca Ferraraccio, MD, PhD,c Anna Zamboli, MD,a Andrea Mabilia, MD,a Annamaria Auricchio, MD,a Gabriele De Sena, MD,a Lorenzo De Stefano, MD,a Francesca Cardella, MD,a Alfonso Barbarisi, MD, PhD,a and Michele Orditura, MD, PhD,b Naples, Italy

Background. Although D2 lymphadenectomy has been shown to improve outcomes in gastric cancer, it may increase postoperative morbidity, mainly owing to splenopancreatic complications. In addition, the effects of nodal dissection along the proper hepatic artery have not been extensively elucidated. We hypothesized that modified D2 (ie, D1+) lymphadenectomy may decrease surgical risks without impairing oncologic adequacy. Methods. Patients with node-positive gastric cancer undergoing curative total gastrectomy were intraoperatively randomized to D1+ (group 1, 36 patients) or standard D2 lymphadenectomy (group 2, 37 patients), the latter including splenectomy and nodal group 12a. The index of estimated benefit was used to assess the efficacy of dissection of each nodal station. The primary endpoint for oncologic adequacy was the disease-free survival (DFS) rate. Results. Surgical complications were significantly more common in group 2, which also included 2 postoperative deaths. Overall, 35 patients (49%) experienced tumor recurrence. The primary site of tumor relapse and the 5-year DFS rate were not different between the 2 groups. Involvement of the second nodal level was associated with a worse DFS rate; however, patients undergoing more extensive lymphadenectomy did not show a better DFS rate. The incidence of involvement of nodal stations 10, 11d, and 12a was 5%, and the 5-year DFS rate was zero. Consequently, the benefit to dissect such lymph nodes was null. Conclusion. These findings suggest that modified D2 lymphadenectomy confers the same oncologic adequacy as standard D2 lymphadenectomy, with a significant reduction of postoperative morbidity. (Surgery 2015;157:285-96.) From the Divisions of Surgical Oncologya and Medical Oncology,b and Unit of Pathology,c Department of Anesthesiological, Surgical and Emergency Sciences, Second University of Naples, School of Medicine, Naples, Italy

GASTRIC CANCER is the fourth most common malignant tumor and the second leading cause of cancer death worldwide, with almost 1 million new cases estimated to occur each year.1 Surgery plays a crucial role in the treatment of gastric cancer, although long-term outcome remains Gennaro Galizia and Eva Lieto contributed equally to this work. Accepted for publication September 10, 2014. Reprint requests: Gennaro Galizia, MD, PhD, Division of Surgical Oncology, Department of Anesthesiological, Surgical and Emergency Sciences, Second University of Naples School of Medicine, c/o II Policlinico, Edificio 17, Via Pansini, 5, 80131 Naples, Italy. E-mail: [email protected]. 0039-6060/$ - see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.surg.2014.09.012

disappointingly low in Western countries.2,3 In contrast, 5-year survival rates after potentially curative surgery have been reported to be >60% in Eastern Asia.4,5 These impressive differences have been attributed, other than to a more aggressive biology and advanced stages at presentation, to the fact that Western surgeons usually perform a more limited lymphadenectomy, which, in turn, may adversely affect oncologic adequacy.6,7 The extension of gastric lymphadenectomy is based on the Japanese classification of the nodal stations surrounding the stomach. Initially, the Japanese Gastric Cancer Association (JGCA), based on the analysis of the lymphatic flow and other important parameters, divided these stations into 4 levels according to the primary location of the SURGERY 285

286 Galizia et al

gastric tumor; consequently, the extension of lymphadenectomy depended on the site of the tumor.8-10 In an effort to meet Western requirements, the JGCA recently abandoned its complex system for a more practical classification of extension of lymphadenectomy that is now divided according to the type of gastric resection rather than the location of the tumor.11-14 In total gastrectomy, 3 different types of nodal excision have been considered. D1 lymphadenectomy includes the perigastric nodal stations from 1 to 7; it is indicated both for T1a tumors not suitable for endoscopic resection and for differentiated and #1.5 cm cT1bN0. D1+ lymphadenectomy (including D1 nodes plus the extraperigastric nodal stations 8a, 9, and 11p) is reserved to cT1N0 tumors other than the above or as a substitute for D2 in poor-risk patients. Finally, D2 lymphadenectomy (D1+ and nodal stations 10, 11d, and 12a) is indicated for potentially curable T2–T4 tumors, as well cT1N+ tumors.12,13 In Japan, D2 lymphadenectomy has been recommended as standard practice since the 1960s, and Eastern surgeons routinely perform this technique with excellent results.10,15 Although there is no clear evidence that extended lymphadenectomy can ensure a better long-term survival rate than more limited lymphadenectomy, reports from single centers, multicenter studies, and a recent metaanalysis have shown that D2 lymphadenectomy is the most effective procedure, at least in advanced stage and node-positive gastric cancers.3,7,16-19 However, D2 lymphadenectomy is a challenging procedure, requiring an appropriate learning curve20; moreover, it may be burdened with high postoperative morbidity and mortality rates,16,17 mainly owing to splenectomy and distal pancreasectomy, which are mandatory for complete clearance of nodal stations 10 and 11d.21,22 Although D2 lymphadenectomy can be performed with acceptable postoperative risks, particularly in high-volume centers,23,24 such an extensive nodal dissection is not a trivial task even for very skilled surgeons.25 The high risk associated with splenectomy or splenopancreasectomy, along with immunologic concerns, has led some authors to investigate whether spleen and pancreas preservation could reduce the complication rate with no detrimental effects on complete oncologic clearance.26,27 However, this question has yet to be clarified.28 Finally, it should be considered that excision of nodal station 12a along the proper hepatic artery, during standard D2 lymphadenectomy, is not frequently performed; therefore, its oncologic effect has not been completely elucidated.29,30

Surgery February 2015

Based these findings, we carried out a study in patients with locally advanced gastric cancer to determine whether a modified D2 lymphadenectomy, including nodal stations from 1 to 7, plus centrally sited nodes (nodal stations 8a, 9, and 11p), and excluding the groups 10, 11d, and 12a (ie, a formal D1+ lymphadenectomy), may adversely affect oncologic adequacy while warranting lower postoperative morbidity and mortality rates. MATERIALS AND METHODS Patients. From January 2006 to June 2012, all patients undergoing surgery for locally advanced nonmetastatic adenocarcinoma of the stomach (cT1N+, T2–T4, M0) were initially eligible for this study. Tumors of the gastroesophageal junction, which often require neoadjuvant therapy and have a particular lymph node spread, and distal tumors treated with subtotal gastrectomy and a more limited lymphadenectomy (particularly without inclusion of nodal groups 10 and 11d), were excluded.13 Tumors invading the greater curvature with 4sb-positive nodes and/or with apparent involvement of splenic nodal groups were also excluded, because splenectomy (with or without distal pancreasectomy) is mandatory to ensure complete nodal clearance in these cases.13 After exploratory laparotomy and decision to perform total gastrectomy, a sealed envelope, opened at the time of the operation, randomized the remaining patients to undergo modified D2 (D1+) lymphadenectomy (group 1) or standard D2 lymphadenectomy (group 2), including splenectomy and dissection of nodal group 12a. Based on postoperative pathologic assessment, patients undergoing nonradical resection, including positive cytology from the peritoneal washing, were excluded by further analyses. In addition, as the curative value of apparently complete resection of lymph node burden was examined, only positive node gastric cancers were retained into the study. Overall, 73 patients remained eligible; 36 patients underwent D1+ (group 1) and 37 patients underwent standard D2 lymphadenectomy (group 2). Follow-up and endpoints. For each patient, the following parameters were recorded: Age, gender, serum carcinoembryonic antigen (CEA) levels, performance status according to the Eastern Cooperative Oncology Group scale, tumor site (upper and middle third of the stomach), macroscopic type (mass, ulcerative, infiltrative ulcerative, and diffuse infiltrative),13 histologic type (differentiated type [well and moderately differentiated

Surgery Volume 157, Number 2

tubular and papillary adenocarcinoma], and undifferentiated type [poorly differentiated adenocarcinoma, signet-ring cell, and mucinous carcinoma]), Lauren’s classification (intestinal and diffuse), TNM stage,11 number of resected nodes, number of positive nodes, lymph node ratio (LNR; determined by dividing the total number of positive nodes by the total number of resected nodes), node position (first tier or perigastric lymph nodes, including nodal groups 1–7; second tier or extraperigastric lymph nodes, including nodal groups 8a–12a),13,14 and postoperative morbidity (complications were defined as grade II or higher of the Clavien-Dindo classification),31 and mortality rate. All but 2 patients were discharged from the hospital and underwent 5-fluorouracil plus oxaliplatin-based adjuvant chemotherapy.5 Patients were followed every 3 months for the first 2 years and every 6 months thereafter. No patient was lost to follow-up through December 31, 2013. All patients gave their informed consent and the study was approved by the Ethical Committee of the Department of Anesthesiological, Surgical, and Emergency Sciences of the Second University of Naples. Because a number of patients did not have a follow-up of >5 years, the disease-free survival (DFS) rate, which has recently been shown to be an acceptable surrogate for overall survival, was used for survival analysis.32 The primary endpoint for oncologic adequacy of the modified D2 lymphadenectomy was the DFS rate. Secondary endpoints were postoperative complication rate, second tier nodal involvement rate and its effect on outcome, and index of estimated benefit from lymph node dissection (IEBLD). IEBLD. The IEBLD, proposed by Sasako et al10 in 1995, was used to assess the efficacy of dissection of each lymph node station. In brief, each nodal station was carefully isolated immediately after the operation, classified according to the JGCA classification,13 and sent separately for pathologic examination for meticulously documented positive and negative nodes for each separate nodal station. Careful attention was paid to sorting out centrally sited nodal groups, namely, nodes 7, 8a, and 9. Quality control was checked by assessing noncompliance, defined as absence in the specimen of nodal groups that should be removed, and contamination, defined as pathologic proof of inclusion of nodal groups that should not have been excised in D1+ lymphadenectomy. The incidence of metastasis to each station was calculated by dividing the number of patients with

Galizia et al 287

metastatic nodes at that station by the number in whom the station was dissected. The 5-year DFS rate was calculated for each nodal station irrespective of metastasis to other lymph node stations and of tumor site. The index was calculated by multiplying the incidence of metastasis at a node station by the 5-year DFS rate of patients with metastasis at that station. Statistical analysis. Statistical analysis was carried out using the SPSS statistical package (SPSS Inc, Chicago, IL) integrated by the Medcalc software version 9.4.2.0 (Mariakerke, Belgium). The equality of group means and comparisons between proportions were analyzed by unpaired Student t test and Chi-square test, respectively. Patients deceased for causes other than gastric cancer without evidence of disease were regarded as censored events for cancer-related relapse or DFS rate. Univariate statistical analysis was determined by log-rank test (Mantel-Cox); curves were plotted using the Kaplan-Meier method, and P values and hazard ratios (HR) with 95% CIs were obtained. The independent significance of each factor was determined by Cox’s proportional hazards model, after inclusion of prognostic variables showing a P < .10 on univariate analysis. Moreover, Cox’s models were used to identify possible interactions in treatment effect between subgroups, both with and without adjustment for prognostic factors. Subgroups were defined by factors showing significant value on univariate analysis and included gender, serum carcinoembryonic antigen levels, tumor site, pathologic T and N stage, node position, TNM stage, and LNR. RESULTS The clinicopathologic characteristics of the 2 groups matched well (Table I). Particularly, there were no differences both in the distribution of TNM stages and in the involvement rate of the first and second nodal level: 33% of patients in group 1 and 38% of patients in group 2 had positive extraperigastric lymph nodes. Patients in group 1 had significantly higher serum carcinoembryonic antigen levels. Pathologic examination ruled out events both of noncompliance and contamination. Overall, number of dissected nodes ranged from 16 to 60 (mean, 24.6 ± 9.6; median, 21). Interestingly, patients undergoing standard D2 lymphadenectomy had a significantly greater number of resected nodes than patients undergoing D1+ lymphadenectomy; however, both the number of positive nodes and LNR were the same in the 2 groups. Postoperative complications were

288 Galizia et al

Surgery February 2015

Table I. Clinicopathologic characteristics Characteristic Age (y) Range Mean ± SD Median Gender Male Female Serum CEA levels (ng/mL) #3.5 >3.5 Performance status 0 1 2 Tumor site Upper third Middle third Macroscopic type 1 2 3 4 Histologic type Differentiated Undifferentiated Lauren’s classification Intestinal Diffuse Tumor depth (T) 1b 2 3 4a 4b Node stage (N) 1 2 3a 3b TNM stage IB IIA IIB IIIA IIIB IIIC No of resected nodes Range Mean ± SD Median No of positive nodes Range Mean ± SD Median LNR

D1+ Lymphadenectomy (n = 36)

D2 Lymphadenectomy (n = 37)

38–78 62 ± 10 63

31–84 59 ± 14 63

20 16

23 14

25 11

34 3

13 18 5

17 15 5

15 21

19 18

2 12 16 6

1 13 17 6

22 14

17 20

20 16

16 21

3 5 7 21 0

1 4 12 19 1

17 13 3 3

12 13 9 3

1 5 3 15 9 3

1 3 1 14 13 5

P value* .2606y

.7371

.0325

.5259

.5521

.8494

.2873

.4134

.4758

.2782

.7387

<.0001y 16–41 19.6 ± 5.7 18

20–60 29.6 ± 10.2 26

1–20 4.4 ± 4.9 3

1–27 6.0 ± 6.2 4

.2315y

.9712y (continued)

Galizia et al 289

Surgery Volume 157, Number 2

Table I. (continued) Characteristic

D1+ Lymphadenectomy (n = 36)

D2 Lymphadenectomy (n = 37)

Range Mean ± SD Median Node positionz First tier Second tier PO complications No Yes

0.0476–0.8333 0.2028 ± 0.1697 0.1583

0.0212–0.7941 0.2044 ± 0.1964 0.1346

P value*

.8749 24 12

23 14

29 7

19 18

.0172

*Chi-square test unless indicated otherwise. yStudent t test. zFirst tier indicates perigastric nodes (nodal groups 1–7); second tier indicates extraperigastric nodes (nodal groups 8a–12a). Performance status according to the Eastern Cooperative Oncology Group scale. CEA, Carcinoembryonic antigen (normal value, 3.5 ng/mL); LNR, lymph node ratio (determined by dividing the total number of positive nodes by the total number of resected nodes); PO, postoperative; SD, standard deviation.

Table II. Postoperative complications

Complication Overall* Anastomotic leakage Pneumonia Pleural fluid Pancreas-related complications Intra-abdominal abscess Bleeding Postoperative death

D2 D1+ Lymphadenectomy, Lymphadenectomy, n (%) n (%) 7 (19.4) 2 (5.5)

18 (48.6) 2 (5.4)

1 (2.7) 4 (11.1) 0 (0.0)

4 (10.8) 8 (21.6) 6 (16.2)

1 (2.7)

4 (10.8)

1 (2.7) 0 (0.0)

2 (5.4) 2 (5.4)

*Grade II or higher of the Clavien-Dindo classification.31

markedly different because patients undergoing standard D2 lymphadenectomy experienced more frequently severe complications (Table II). In group 1, there were neither pancreas-related complications nor postoperative deaths; on the contrary, in group 2, 2 patients experienced inhospital death after intra-abdominal abscess and bleeding. Such patients were excluded by further survival analysis. DFS analysis related to 71 patients. At the end of the study, 35 patients (49.3%) experienced tumor recurrence. In terms of site of relapse, 10 recurrences (14.1%) were locoregional, 6 (8.4%) were distant metastases, and 19 (26.8%) were both locoregional and distant metastases. Recurrence rates were similar in both groups: 51.4% (18 cases) in patients undergoing D1+ lymphadenectomy and 47.2% in patients undergoing standard D2

lymphadenectomy (P = .9068). Site of recurrence was equivalent in both groups: Locoregional in 4 and 6 cases, distant metastases in 3 and 3 cases, both sites in 11 and 8 cases, respectively (P = .6551). Median follow-up was 55 months (mean, 46.7 ± 27; range, 6.5–96) for the 36 nonrelapsing patients and 14.5 months (mean, 16.8 ± 12; range, 7.1–59.7) for the 35 patients experiencing tumor recurrence; there was no difference between the 2 groups. The 1- to 5-year DFS rates for the whole cohort were 91.0, 71.0, 48.6, 46.7, and 42.3%, respectively, with a significant association with advanced node and TNM stage, greater number of positive nodes, elevated LNR, and presence of extraperigastric metastatic nodes (Table III). The 1- to 5-year DFS rates were 91.2, 69.9, 53.8, 50.4, and 42.9% in group 1, and 90.7, 71.9, 42.1, 42.1, and 42.1% in group 2, with no association (Fig 1, A). Interestingly, although involvement of the second nodal tier was associated with a worse outcome, a more extended lymphadenectomy did not yield a better prognosis. Indeed, among the 26 patients having extraperigastric metastatic nodes, the 5-year DFS rate was 18.7% in the 12 patients undergoing D1+ lymphadenectomy and 21.8% in the 14 patients undergoing standard D2 lymphadenectomy (HR, 0.80; 95% CI, 0.31–1.99; P = .6191), respectively (Fig 1, B). On multivariate analysis, advanced TNM and nodal stages (HR, 4.56; 95% CI, 1.60–13.00; P = .0046) and (HR, 2.18; 95% CI, 1.11–4.27; P = .0228), respectively and elevated LNR (HR, 2.10; 95% CI, 1.05–4.22; P = .0362) were shown to be independent prognostic factors related to poor recurrence rate. After adjusting for

290 Galizia et al

Surgery February 2015

Table III. Univariate analysis related to disease-free survival in 71 gastric cancer patients undergoing potentially curative surgery Characteristic Age (<63/>63 y)y Gender (M/F) Serum CEA levels (ng/mL) #3.5 >3.5 Performance status 0 1 2 Tumor site Upper third Middle third Macroscopic type 1 2 3 4 Histologic type Differentiated Undifferentiated Lauren’s classification Intestinal Diffuse Tumor depth (T) 1b 2 3 4a 4b Node stage (N) 1 2 3a 3b TNM stage IB IIA IIB IIIA IIIB IIIC No of resected nodes* #21 >21 No. of positive nodes* #3 >3 LNR* #0.1458 >0.1458 #0.0625y <0.1458 #0.2708 >0.2708

No. of patients

Recurrence

5-year disease-free survival (%)

Hazard ratio for recurrence

95% CI hazard ratio

P value

38/33 42/29

19/16 17/18

44.7/40.5 53.7/24.4

0.77 0.52 0.50

0.39–1.52 0.24–1.00 0.17–1.04

.4587 .0512 .0637

58 13

25 10

49.9 16.5 —

—

.2272

30 33 8

12 20 3

53.5 32.1 41.7

33 38

19 16

30.1 52.5

0.54

0.26–1.05

.0689

—

—

.8127

3 25 33 10

0 12 17 6

100 46.4 41.1 31.1 0.82

0.42–1.60

.5718

37 34

17 18

46.1 38.4 0.81

0.41–1.58

.5442

34 37

16 19

44.3 40.6 —

—

.1128

4 9 19 38 1

1 1 13 19 1

66.7 88.9 31.6 37.8 0.0

29 24 12 6

7 13 10 5

71.2 32.1 11.5 16.7

2 8 4 27 22 8

0 0 2 11 16 6

100 100 50.0 53.0 14.8 15.6

37 34

18 17

43.1 41.3

38 33

11 24

64 20.4

35 36 14 21 19 17

11 24 3 8 11 13

61.6 25.2 73.3 54.3 32.9 17.3

.0002

.0035

0.82

0.42–1.60

.5705

0.31

0.15–0.61

.0007

0.38

0.19–0.75

.0057

.0108

(continued)

Galizia et al 291

Surgery Volume 157, Number 2

Table III. (continued) Characteristic Node positionz First tier Second tier PO complications No Yes Group D1+ Lymphadenectomy D2 Lymphadenectomy

No. of patients

Recurrence

5-year disease-free survival (%)

45 26

16 19

57.4 17.9

48 23

22 13

45.6 36.2

36 35

18 17

42.9 42.1

Hazard ratio for recurrence

95% CI hazard ratio

P value

0.43

0.19–0.81

.0110

0.68

0.31–1.37

.2693

0.96

0.49–1.88

.9266

*Median value. yValues are obtained by using 25, 50, 75, and >75% percentile, respectively. zFirst tier indicates perigastric nodes (nodal groups 1–7); second tier indicates extraperigastric nodes (nodal groups 8a–12a). Performance status according to the Eastern Cooperative Oncology Group scale. CEA, Carcinoembryonic antigen (normal value, 3.5 ng/mL); LNR, lymph node ratio (determined by dividing the total number of positive nodes by the total number of resected nodes).

interfering factors, no differences were found between D1+ and D2 lymphadenectomy in terms of DFS rate (HR, 0.84; 95% CI, 0.39–1.83; P = .6791). In addition, the equivalence of D1+ and D2 lymphadenectomy on recurrence rate was consistent across subgroups, without any interaction identified (Fig 2). IEBLD. The frequency of metastasis in each perigastric nodal station ranged from 15.4 to 67.6%, and the 5-year DFS rate ranged from 31.8 to 53.7%, respectively (Fig 3). Consequently, the IEBLD for these nodal stations ranged from 6.4 to 36.3% (mean, 18.8 ± 10; median, 19.3%; Table IV). Involvement of extraperigastric nodal stations dissected during D1+ lymphadenectomy (nodes 8a, 9, and 11p) occurred in almost 30% of the cases, with a 5-year DFS rate ranging from 9.6 to 30.2%, and an IEBLD ranging from 3.1 to 10.2% (mean, 6.9 ± 3; median, 7.4%); thus, the absolute benefit of such a nodal resection was up to 10%. In patients undergoing standard D2 lymphadenectomy, 2 patients (5.7%) had nodal station 10 involvement, 1 patient presented with metastasis in station 11d, and 1 patient had station 12a metastatic disease. The 5-year DFS rate was zero if nodal stations 10, 11d, or 12a were involved. Consequently, the IEBLD of stations that were located outside the D1+ lymphadenectomy was zero, and the benefit to dissect such nodal groups seemed to be null. DISCUSSION The present study shows that modified D2 lymphadenectomy (so-called D1+ lymphadenectomy) is a safe and effective operative approach

for locally advanced gastric cancer subjected to total gastrectomy. In fact, this procedure ensures the same oncologic adequacy as standard D2 lymphadenectomy; in addition, a significant reduction in postoperative morbidity and mortality rates is obtained. D2 lymphadenectomy is the standard operative procedure for treatment of locally advanced, albeit still resectable, gastric cancer in Eastern countries.13 On the contrary, in the Western world, such technique is still a matter of debate, with many surgeons thus continuing to perform D1 or even more limited lymphadenectomy.33,34 It seems intuitive that resecting as many nodes as possible would be the best option, but this is not always true.4,35 Basically, lymphadenectomy for cancer serves 3 purposes: Staging of disease, prevention of locoregional recurrence, and improvement in overall survival.6 The American Joint Committee on Cancer Staging Manual recommends a minimum of 16 lymph nodes be examined; harvesting <16 nodes prevents to categorize patients as N3b.11 Whereas standard D2 lymphadenectomy meets American Joint Committee on Cancer recommendations, D1 lymphadenectomy rarely achieves this goal.19,34,36 Rates of locoregional recurrence from centers performing limited lymphadenectomy may be disappointingly high (up to 80%).37 In contrast, lesser locoregional recurrence rates, ranging from 19 to 26%, have been reported by institutions performing extensive lymphadenectomy.38-40 Finally, impressive overall survival rates after extensive lymphadenectomy have been published by Eastern authors, although stage migration and possible differences in tumor

Surgery February 2015

100

100

80

80

Disease-Free Survival (%)

Disease-Free Survival (%)

292 Galizia et al

60 40

Group 1 20

Group 2

60 40

Group 1 20

Group 2

0

0 0

12

24

36

48

60

0

12

24

months Number at risk Group: 1 36 Group: 2 35

36

48

60

months

31

22

16

14

11

29

21

10

10

8

Number at risk Group: 1 12 Group: 2 14

12

7

5

4

2

11

8

2

2

1

Fig 1. (A), One- to 5-year disease-free survival rate in 36 patients undergoing modified D2 lymphadenectomy (group 1) and in 35 patients undergoing standard D2 lymphadenectomy (group 2). (B), One-to 5-year disease-free survival rate in 26 patients with extraperigastric metastatic nodes. Group 1 included 12 patients undergoing modified D2 lymphadenectomy. Group 2 included 14 patients undergoing standard D2 lymphadenectomy.

Fig 2. Forest plot showing adjusted hazard ratios (oblongs) and 95% CIs (bars) for 5-year disease-free survival rate in 71 gastric cancer patients undergoing potentially curative surgery, according to subgroup effects. Univariate hazard ratios, as well hazard ratios adjusted for baseline covariates and related P values, are listed on the right side. Subgroups were defined by factors showing significant value on univariate analysis. CEA, Carcinoembryonic antigen (normal value, 3.5 ng/mL); LNR, lymph node ratio (determined by dividing the total number of positive nodes by the total number of resected nodes). First tier indicates perigastric nodes (nodal groups 1–7); second tier indicates extraperigastric nodes (nodal groups 8a–12a).

biology have been advocated to explain such brilliant results.6,41 Thus, the effect of extensive lymphadenectomy on overall survival is still debated.35 For many years, Western surgeons

have not considered D2 lymphadenectomy because of the results of 2 large randomized, controlled trials carried out in the 1990s. The conclusions of these studies were not in favor of

Surgery Volume 157, Number 2

Fig 3. Incidence of lymph node metastasis determined for each nodal station and 5-year disease-free survival rates of patients who had nodal metastases at each nodal station.

routine use of this procedure because no survival advantage over D1 surgery was observed.16,17 However, these trials have been strongly questioned, mainly because of a high noncompliance rate and a significantly higher rate of complications and postoperative deaths in patients undergoing splenectomy.6,19 Subsequent analyses not including patients undergoing splenectomy and/ or pancreasectomy, which significantly affected postoperative morbidity rates, revealed that D2 lymphadenectomy could benefit patients with N2 status, reduce locoregional relapse rates, and improve disease-specific survival rates with respect to D1 surgery.21,22 Finally, D2 lymphadenectomy, performed with spleen and pancreas preservation, has been recently shown to offer the best survival benefit to gastric cancer patients with locally advanced disease in 2 recent meta-analyses.7,28 Consistent with what is reported in the Western world, we recorded a morbidity rate of nearly 50% as well as 2 postoperative deaths in patients undergoing standard D2 lymphadenectomy.30 In contrast, patients undergoing modified D2 lymphadenectomy displayed a significantly lower rate of operative morbidity and no occurrence of postoperative deaths. Modified D2 lymphadenectomy did not have any detrimental effect on the recurrence rate; the 2 groups showed the same percentage of tumor relapses. In addition, the primary site of recurrence was equivalent, with surgical locoregional control of 58.3% in group 1 and 60% in group 2. After adjusting for confounding factors, subgroup analysis failed to individuate any specific favorable condition for either procedure. Only in case of tumors sited in the upper third of the stomach, which are known to spread to the splenic

Galizia et al 293

nodal group,42 D2 lymphadenectomy seemed to be slightly more advantageous with respect to the standard procedure, although this difference was far from attaining significance. Interestingly, standard D2 lymphadenectomy allowed the harvest of a greater number of lymph nodes; however, this did not translate into a better DFS rate.35 This finding is intriguing and may suggest possible bias in the study. However, patients undergoing modified D2 lymphadenectomy consented to harvesting a greater number of lymph nodes than classical D1 lymphadenectomy; there was an absence of contamination and no difference both in the number of positive nodes and TNM stage distribution, thus avoiding confounding effects from stage migration, indicating that the 2 groups were optimally balanced. The involvement of the second nodal tier was associated with worsened long-term outcome in all patients, with no differences between those undergoing modified D1 lymphadenectomy and those subjected to standard D2 lymphadenectomy. Because the only difference between the 2 groups was the excision of nodal stations 10, 11d, and 12a, it is important to note that metastasis to these nodes comported a 5-year DFS rate equal to zero, as demonstrated by IEBLD. Previously, this analysis had provided a rational basis for the boundaries of gastric lymphadenectomy showing that the longterm survival rate after D2 was better than after D1 lymphadenectomy.6,10 Unfortunately, this index has been rarely reported and, to the best of our knowledge, this is the first time the IEBLD is calculated only in selected node-positive gastric cancer patients undergoing potentially curative surgery, without apparent spread to distal nodes.30,43 Our data confirm that removal of nodal stations beyond the boundaries of a D1 lymphadenectomy (namely, nodal stations 8a, 9, and 11p) can lead up to a 10% benefit, which is greater than previously reported.10,30 On the contrary, this benefit seemed to be null when nodal stations along the proper hepatic artery and the splenic vessels were metastatic. These data are only apparently in contrast with those reported by Sasako et al, who described very low survival rates for tumors of the upper third and the whole stomach, whereas they are in agreement with the general assumption that prognosis for patients with metastasis in such stations is very poor.6,10,21,27 Overall, the recurrence rate in our cohort of node-positive gastric cancer patients was disappointingly high, even after adequate lymphadenectomy. Curability after radical surgery for locally advanced gastric cancer may be overestimated

294 Galizia et al

Surgery February 2015

Table IV. Analyses related to the index of estimated benefit from lymph node dissection Lymph node station 1 2 3 4 5 6 7 8a 9 10 11p 11d 12a

Patients in whom the station was resected

Patients in whom the station was metastatic

Incidence of metastatic nodes (%)

5-year disease-free survival (%)

IEBLD (%)

71 71 71 71 71 71 71 71 71 35 71 35 35

25 17 48 35 11 29 33 24 20 2 23 1 1

35.2 23.9 67.6 49.3 15.4 40.8 46.4 33.8 28.1 5.7 32.3 2.8 2.8

45.1 31.8 53.7 51.4 41.6 47.3 44.5 30.2 26.4 0.0 9.6 0.0 0.0

15.8 7.6 36.3 25.3 6.4 19.3 20.6 10.2 7.4 0.0 3.1 0.0 0.0

IEBLD, Index of estimated benefit from lymph node dissection (calculated by multiplying the incidence of metastasis at a node station by the 5-year disease-free survival rate of patients with metastasis at that station).

because this disease is indeed a micrometastatic systemic disease14; therefore, surgery alone is no longer recommended for resectable gastric cancer.44 Ongoing trials, particularly ARTIST-2 (adjuvant chemotherapy alone to adjuvant chemoradiotherapy in node-positive patients) and CRITICS (chemotherapy alone versus chemoradiotherapy after induction chemotherapy) are expected to contribute to improvements in survival rates.15,45 A limitation of this study was the small number of patients analyzed. However, it has to be emphasized that a sample size of 36 patients per group would have been sufficient to identify a 15% difference in the 5-year DFS rate with a 80% power (type II error) for a 1-sided significance level of 0.10. These values were deemed acceptable, considering that type I error did not exceed 10%, no patient was lost for analyses, and both groups included only node-positive patients.46,47 In conclusion, although the small number of patients in this study could have adversely impacted its statistical power, thus requiring further validations by other trials, this study shows that modified D2 lymphadenectomy is effective for the operative treatment of locally advanced gastric cancer. While waiting for the final results, expected by mid-2014, from the Japan Clinical Oncology Group trial (JCOG 0110),25,48 these data seem to suggest that splenectomy (and distal pancreasectomy) as well resection of lymph nodes along the proper hepatic artery, which negatively affect postoperative morbidity rates with little oncologic significance, may be omitted without impairing curability and long-term outcome.

REFERENCES 1. Jemal A, Center MM, DeSantis C, Ward EM. Global patterns of cancer incidence and mortality rates and trends. Cancer Epidemiol Biomarkers Prev 2010;19:1893-907. 2. Marrelli D, De Stefano A, de Manzoni G, Morgagni P, Di Leo A, Roviello F. Prediction of recurrence after radical surgery for gastric cancer: a scoring system obtained from a prospective multicenter study. Ann Surg 2005;241: 247-55. 3. Ferri L. Extended lymphadenectomy in gastric cancer is crucial. World J Surg 2013;37:1768-72. 4. Sasako M, Sano T, Yamamoto S, Kurokawa Y, Nashimoto A, Kurita A, et al. D2 lymphadenectomy alone or with paraaortic nodal dissection for gastric cancer. N Engl J Med 2008;359:453-62. 5. Bang YJ, Kim YW, Yang HK, Chung HC, Park YK, Lee KH, et al. Adjuvant capecitabine and oxaliplatin for gastric cancer after D2 gastrectomy (CLASSIC): a phase 3 open-label, randomised controlled trial. Lancet 2012;379:315-21. 6. Schmidt B, Yoon SS. D1 versus D2 lymphadenectomy for gastric cancer. J Surg Oncol 2013;107:259-64. 7. Jiang L, Yang KH, Guan QL, Zhao P, Chen Y, Tian JH. Survival and recurrence free benefits with different lymphadenectomy for resectable gastric cancer: a meta-analysis. J Surg Oncol 2013;107:807-14. 8. Kampsch€ oer GH, Maruyama K, van de Velde CJ, Sasako M, Kinoshita T, Okabayashi K. Computer analysis in making preoperative decisions: a rational approach to lymph node dissection in gastric cancer patients. Br J Surg 1989; 76:905-8. 9. Nishi M, Omori Y, Miwa K. Japanese classification of gastric carcinoma. Tokyo: Kanehara & Co., Ltd; 1995. 10. Sasako M, McCulloch P, Kinoshita T, Maruyama K. New method to evaluate the therapeutic value of lymph node dissection for gastric cancer. Br J Surg 1995;82:346-51. 11. Sobin LH, Gospodarowicz MK, Wittekind C. International Union Against Cancer (UICC) TNM classification of malignant tumors. 7th ed. New York: Wiley-Liss; 2010. 12. Sano T, Aiko T. Japanese classifications and treatment guidelines for gastric cancer: revision concepts and major revised points. Gastric Cancer 2011;14:97-100.

Surgery Volume 157, Number 2

13. Japanese Gastric Cancer Association. Japanese gastric cancer treatment guidelines 2010 (ver. 3). Gastric Cancer 2011;14:101-23. 14. Rausei S, Dionigi G, Sano T, Sasako M, Biondi A, Morgagni P, et al. Updates on surgical management of advanced gastric cancer: new evidence and trends. Insights from the First International Course on Upper Gastrointestinal Surgery- Varese (Italy), December 2, 2011. Ann Surg Oncol 2013;20:3942-7. 15. Lee J, Lim do H, Kim S, Park SH, Park JO, Park YS, et al. Phase III trial comparing capecitabine plus cisplatin versus capecitabine plus cisplatin with concurrent capecitabine radiotherapy in completely resected gastric cancer with D2 lymph node dissection: the ARTIST trial. J Clin Oncol 2012;30:268-73. 16. Bonenkamp JJ, Hermans J, Sasako M, van de Velde CJ, Welvaart K, Songun I, et al. Extended lymph-node dissection for gastric cancer. N Engl J Med 1999;340:908-14. 17. Cuschieri A, Weeden S, Fielding J, Bancewicz J, Craven J, Joypaul V, et al. Patient survival after D1 and D2 resections for gastric cancer: long-term results of the MRC randomized surgical trial. Surgical Co-operative Group. Br J Cancer 1999;79:1522-30. 18. Strong VE, Yoon SS. Extended lymphadenectomy in gastric cancer is debatable. World J Surg 2013;37:1773-7. 19. Degiuli M, Sasako M, Ponti A, Vendrame A, Tomatis M, Mazza C, et al. Randomized clinical trial comparing survival after D1 or D2 gastrectomy for gastric cancer. Br J Surg 2014;101:23-31. 20. Lee JH, Ryu KW, Lee JH, Park SR, Kim CG, Kook MC, et al. Learning curve for total gastrectomy with D2 lymph node dissection: cumulative sum analysis for qualified surgery. Ann Surg Oncol 2006;13:1175-81. 21. Hartgrink HH, van de Velde CJ, Putter H, Bonenkamp JJ, Klein Kranenbarg E, Songun I, et al. Extended lymph node dissection for gastric cancer: who may benefit? Final results of the randomized Dutch gastric cancer group trial. J Clin Oncol 2004;22:2069-77. 22. Songun I, Putter H, Kranenbarg EM, Sasako M, van de Velde CJ. Surgical treatment of gastric cancer: 15-year follow-up results of the randomised nationwide Dutch D1D2 trial. Lancet Oncol 2010;11:439-49. 23. Sano T, Sasako M, Yamamoto S, Nashimoto A, Kurita A, Hiratsuka M, et al. Gastric cancer surgery: morbidity and mortality results from a prospective randomized controlled trial comparing D2 and extended para-aortic lymphadenectomy–Japan Clinical Oncology Group study 9501. J Clin Oncol 2004;22:2767-73. 24. Degiuli M, Sasako M, Ponti A, Italian Gastric Cancer Study Group. Morbidity and mortality in the Italian Gastric Cancer Study Group randomized clinical trial of D1 versus D2 resection for gastric cancer. Br J Surg 2010;97:643-9. 25. Sano T, Sasako M, Shibata T, Yamamoto S, Tsuburaya A, Nashimoto A, et al. Randomized controlled trial to evaluate splenectomy in total gastrectomy for proximal gastric carcinoma (JCOG0110): analyses of operative morbidity, operation time, and blood loss. J Clin Oncol 2010;28(15 Suppl):4020; (ASCO meeting abstract). 26. Erturk S, Ersan Y, Cicek Y, Dogusoy G, Senocak M. Effect of simultaneous splenectomy on the survival of patients undergoing curative gastrectomy for proximal gastric carcinoma. Surg Today 2003;33:254-8. 27. Yu W, Choi GS, Chung HY. Randomized clinical trial of splenectomy versus splenic preservation in patients with proximal gastric cancer. Br J Surg 2006;93:559-63.

Galizia et al 295

28. Brar SS, Seevaratnam R, Cardoso R, Law C, Helyer L, Coburn N. A systematic review of spleen and pancreas preservation in extended lymphadenectomy for gastric cancer. Gastric Cancer 2012;15(Suppl 1):S89-99. 29. Cuschieri A, Fayers P, Fielding J, Craven J, Bancewicz J, Joypaul V, et al. Postoperative morbidity and mortality after D1 and D2 resections for gastric cancer: preliminary results of the MRC randomised controlled surgical trial. The Surgical Cooperative Group. Lancet 1996;347:995-9. 30. Goto H, Tokunaga M, Sugisawa N, Tanizawa Y, Bando E, Kawamura T, et al. Value of splenectomy in patients with Siewert type II adenocarcinoma of the esophagogastric junction. Gastric Cancer 2013;16:590-5. 31. Clavien PA, Barkun J, de Oliveira ML, Vauthey JN, Dindo D, Schulick RD, et al. The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg 2009;250: 187-96. 32. Oba K, Paoletti X, Alberts S, Bang YJ, Benedetti J, Bleiberg H, et al. Disease-free survival as a surrogate for overall survival in adjuvant trials of gastric cancer: a meta-analysis. J Natl Cancer Inst 2013;105:1600-7. 33. Reid-Lombardo KM, Gay G, Patel-Parekh L, Ajani JA, Donohue JH, Gastric Patient Care Evaluation Group from the Commission on Cancer. Treatment of gastric adenocarcinoma may differ among hospital types in the United States, a report from the National Cancer Data Base. J Gastrointest Surg 2007;11:410-9. 34. Wang J, Dang P, Raut CP, Pandalai PK, Maduekwe UN, Rattner DW, et al. Comparison of a lymph node ratio-based staging system with the 7th AJCC system for gastric cancer: analysis of 18,043 patients from the SEER database. Ann Surg 2012;255:478-85. 35. Memon MA, Subramanya MS, Khan S, Hossain MB, Osland E, Memon B. Meta-analysis of D1 versus D2 gastrectomy for gastric adenocarcinoma. Ann Surg 2011;253:900-11. 36. Schmidt B, Chang KK, Maduekwe UN, Look-Hong N, Rattner DW, Lauwers GY, et al. D2 lymphadenectomy with surgical ex vivo dissection into node stations for gastric adenocarcinoma can be performed safely in Western patients and ensures optimal staging. Ann Surg Oncol 2013; 20:2991-9. 37. Macdonald JS, Smalley SR, Benedetti J, Hundahl SA, Estes NC, Stemmermann GN, et al. Chemoradiotherapy after surgery compared with surgery alone for adenocarcinoma of the stomach or gastroesophageal junction. N Engl J Med 2001;345:725-30. 38. D’Angelica M, Gonen M, Brennan MF, Turnbull AD, Bains M, Karpeh MS. Patterns of initial recurrence in completely resected gastric adenocarcinoma. Ann Surg 2004;240:808-16. 39. Sakuramoto S, Sasako M, Yamaguchi T, Kinoshita T, Fujii M, Nashimoto A, et al. Adjuvant chemotherapy for gastric cancer with S-1, an oral fluoropyrimidine. N Engl J Med 2007; 357:1810-20. 40. Ahn HS, Lee HJ, Yoo MW, Jeong SH, Park DJ, Kim HH, et al. Changes in clinicopathological features and survival after gastrectomy for gastric cancer over a 20-year period. Br J Surg 2011;98:255-60. 41. Wu CW, Hsiung CA, Lo SS, Hsieh MC, Chen JH, Li AF, et al. Nodal dissection for patients with gastric cancer: a randomised controlled trial. Lancet Oncol 2006;7:309-15. 42. Ito H, Inoue H, Odaka N, Satodate H, Mukai S, Omoto T, et al. Prognostic impact of prophylactic splenectomy for upper-third gastric cancer: a cohort study. Anticancer Res 2013;33:277-82.

296 Galizia et al

43. Nunobe S, Ohyama S, Sonoo H, Hiki N, Fukunaga T, Seto Y, et al. Benefit of mediastinal and para-aortic lymph-node dissection for advanced gastric cancer with esophageal invasion. J Surg Oncol 2008;97:392-5. 44. Oh DY, Bang YJ. Adjuvant and neoadjuvant therapy for gastric cancer. Curr Treat Options Oncol 2013;14:311-20. 45. Dikken JL, van Sandick JW, Maurits Swellengrebel HA, Lind PA, Putter H, Jansen EP, et al. Neo-adjuvant chemotherapy followed by surgery and chemotherapy or by surgery and chemoradiotherapy for patients with resectable gastric cancer (CRITICS). BMC Cancer 2011;11:329.

Surgery February 2015

46. Donner A. Approaches to sample size estimation in the design of clinical trials - a review. Stat Med 1984;3: 199-214. 47. Cannistra SA. Phase II trial in Journal of Clinical Oncology. J Clin Oncol 2009;27:3073-6. 48. Sano T, Yamamoto S, Sasako M, Japan Clinical Oncology Group Study LCOG 0110-MF. Randomized controlled trial to evaluate splenectomy in total gastrectomy for proximal gastric carcinoma: Japan clinical oncology group study JCOG 0110-MF. Jpn J Clin Oncol 2002;32: 363-4.