Available online at www.sciencedirect.com
ScienceDirect EJSO 41 (2015) 534e540
www.ejso.com
Impact of super-extended lymphadenectomy on relapse in advanced gastric cancer* G. de Manzoni a,*, G. Verlato b, M. Bencivenga a, D. Marrelli c, A. Di Leo d, S. Giacopuzzi a, C. Cipollari a, F. Roviello c a
Dept. of Surgery, General and Upper G.I. Surgery Division, University of Verona, Verona, Italy b Unit of Epidemiology and Medical Statistics, University of Verona, Verona, Italy c Dept. of Human Pathology and Oncology, Section of General Surgery and Surgical Oncology, Translational Research Laboratory, University of Siena, Siena, Italy d Unit of General Surgery, Rovereto Hospital, APSS of Trento, Trento, Italy Accepted 15 January 2015 Available online 4 February 2015
Abstract Background: In gastric cancer the incidence of loco-regional recurrences decreases when lymphadenectomy is expanded from D1 to D2. The present study aimed at evaluating whether the pattern of recurrence in advanced gastric cancer (AGC) is further modified when lymphadenectomy is expanded from D2 to D3. Methods: 568 patients undergoing curative gastrectomy for AGC (274 D2 and 294 D3) were considered; none of them received preoperative chemotherapy. MantelHaenszel test of homogeneity was used to verify whether the relation between extension of lymphadenectomy and recurrence varied as a function of each risk factor considered. The impact of D2 and D3 on relapse was further investigated by multivariable logistic regression model. Results: Cumulative incidence of recurrence did not significantly differ after D2 and after D3 in the whole series (45.3% vs 46.3%; p ¼ 0.866). However, the association between recurrence and extension of lymphadenectomy was significantly affected by histology (ManteleHaenszel test of homogeneity: p ¼ 0.007). The risk of recurrence was higher after D3 than after D2 (45.1% vs 35.3%) in the intestinal histotype while the pattern was reversed in the mixed/diffuse histotype (48.3% vs 61.5%). This pattern was confirmed in multivariable logistic regression: the interaction between histology and extension of lymphadenectomy was highly significant (p ¼ 0.004). In particular, cumulative incidence of locoregional recurrences was higher in the diffuse histotype after D2, while being higher in the intestinal histotype after D3. Conclusions: D3 reverses the negative impact of diffuse histotype on relapses, especially on locoregional recurrences. Therefore D3 could be considered a valid therapeutic option in histotype-oriented tailored treatment of AGC. Ó 2015 Elsevier Ltd. All rights reserved.
Keywords: Gastric cancer; Super-extended (D3) lymphadenectomy; Lauren’s diffuse; Loco-regional recurrences
Introduction Despite the declining incidence, gastric cancer is still the third cause of cancer related death worldwide.1 Radical * The present manuscript relates to the Best Proffered Paper session for ESSO 34 e BASO 2014. * Corresponding author. Dept. of Surgery, General and Upper G.I. Surgery Division, University of Verona, Piazzale Stefani 1, 37126 Verona, Italy. Tel.: þ39 (0) 45 8123063; fax: þ39 (0) 45 8122484. E-mail address:
[email protected] (G. de Manzoni).
http://dx.doi.org/10.1016/j.ejso.2015.01.023 0748-7983/Ó 2015 Elsevier Ltd. All rights reserved.
surgery is the only chance of cure for gastric cancer patients, however the rate of relapse even after a curative resection remains high.2e4 Western authors have debated for a long time whether the extension of lymphadenectomy could affect the incidence of recurrences and survival in gastric cancer patients. Interestingly, the Dutch trial, after a median follow-up of 15 years, showed that D2 lymphadenectomy was associated with an improved disease-related survival and a lower locoregional recurrence rate as compared to D1.5 Hence it
G. de Manzoni et al. / EJSO 41 (2015) 534e540
could be speculated that extending lymphadenectomy beyond D2 dissection could further improve local control of advanced gastric tumours. The Japanese JCOG 9501 trial concluded that routinely adding para-aortic dissection to D2 lymphadenectomy does not affect the rate and the pattern of recurrences in advanced gastric cancer.6 Of note, in the Japanese trial D2 lymphadenectomy, for tumours located in the lower third of the stomach, included also dissection of retropancreatic nodes (No. 13) and nodes along the superior mesenteric vein (No. 14v),7 which are currently not comprised in the standard D2 dissection.8 According to the results of the JCOG 9501 trial, we believe that lymphadenectomy should not be routinely extended beyond D2 in patients with advanced gastric cancer. Nevertheless we hypothesized that D3 dissection could be useful in subgroups of patients9,10 regardless of clinical positivity of para-aortic nodes, as also suggested by the rather high 5-year survival (18.2%) reported in the Japanese trial8 in patients with pathologically positive para-aortic nodes after a prophylactic PAN dissection. Hence further investigations are necessary before definitely abandoning D3 lymphadenectomy. Moreover, it has been clearly shown that in high volumes Western centres, D3 lymphadenectomy can be performed without increasing postoperative morbidity and mortality.9,11e14 The present study aimed at evaluating the impact of super-extended D3 lymphadenectomy on overall and specific recurrences in a series of advanced gastric cancer patients. For this purpose, the GIRCG (Italian Research Group for Gastric Cancer) database was retrospectively reviewed. As the pattern of recurrence is affected by histotype, with peritoneal relapse being more common in diffuse tumours,15 both the whole series and subgroups of patients were examined, to evaluate whether D3 dissection could be useful in specific subsets when compared with D2. Patients and methods D3 lymphadenectomy has been routinely performed in two GIRCG centres, Verona and Siena, since the early Nineties. The series from Verona was operated on from January 1992 to May 2011, the series from Siena from January 1994 to June 2011, and the two series comprised 791 subjects altogether. Nineteen subjects with Bormann IV tumour were excluded as well as 2 subjects with neuroendocrine tumours and 75 subjects with early gastric cancer. Among the remaining 695 patients, 127 had non-curative resection: the proportion of R1 resection was similar in the D2 (39/ 350 ¼ 11.1%) and D3 groups (36/345 ¼ 10.4%), while the proportion of R2 resection was remarkably higher in the D2 group (37/350 ¼ 10.6% vs 15/345 ¼ 4.3%) (p ¼ 0.007). After excluding non-curative resections, 568 subjects (312 from Verona, 256 from Siena) were left for
535
the analysis. Of these 274 underwent D2 lymphadenectomy and 294 D3 lymphadenectomy. Postoperative deaths, defined as deaths occurred during the hospital stay, were included in our analysis. D2 and D3 lymphadenectomies were performed according to the Japanese Classification of Gastric Carcinoma e 2nd English Edition.16 Tumour invasion (pT) and lymph node status (pN) followed the UICC pathological tumour node metastasis (pTNM) criteria, 7th edition,17 while histological type was classified as intestinal or mixed/diffuse according to Lauren’s classification.18 Nearly all patients (539/565 ¼ 95.4%) could be adequately staged, as they had at least 15 lymph nodes retrieved. None of the patients received preoperative chemotherapy. Patients were generally followed-up regularly after hospital discharge. Blood tests including CEA and CA19-9 levels and chest/abdomen CT-scan were evaluated every six months for the first 5 years after the operation then once a year for the following 5 years. Upper GI endoscopy was performed once a year for 10 years after the operation. In this series patients were followed-up till December 2011. Median follow-up in surviving patients was 89 months (range 7e240). The pattern of recurrence was classified as locoregional, peritoneal or distant based on the findings of examinations or reoperation. When multiple recurrences had occurred, each of them was separately considered. Recurrences diagnosed in other hospitals, referred by other physicians or by the patient but lacking documentation or description, were defined as unknown. Statistical analysis Significance of differences in the proportion of relapses between different lymphadenectomy and different clinical and pathological groups was evaluated by Fisher’s exact test. ManteleHaenszel test of homogeneity was used to verify whether the relation between extension of lymphadenectomy and relapse varied as a function of each risk factor considered. The impact of D2 and D3 on relapse was further investigated by multivariable logistic regression model, controlling for centre, sex, age, site, histology, T and N status. Results were synthesized through the odds ratios, and p values were computed by the LR test. The interaction between extension of lymphadenectomy and each potential confounder was also tested. The statistical analysis was performed using STATA software, release 12 (StataCorp, College Station, TX, USA). Results Main demographic and clinico-pathological characteristics of the present series are shown in Table 1. Sixty-two percent of patients were male, and mean age SD was 66.4 11.7 years, ranging from 24 to 92 years. D2 was the procedure of choice among patients aged 75 years and over and in cancer arising from the antrum, while D3
536
G. de Manzoni et al. / EJSO 41 (2015) 534e540
Table 1 Main demographic and clinical characteristics at baseline of D2/D3 series from Verona and Siena.
Centre Verona Siena Sex Men Women Age <65 years 65e74.9 years 75 years Sitea Fundus Body Antrum Histologya Intestinal Mixed/Diffuse Depth of tumour invasiona T2 T3 T4a T4b Node metastases N0 N1 N2 N3
Extended (D2) (n ¼ 274)
Superextended (D3) (n ¼ 294)
111 (35.6%) 163 (63.7%)
201 (64.4%) 93 (36.3%)
162 (46.2%) 112 (51.6%)
189 (53.8%) 105 (48.4%)
75 (32.5%) 96 (49%) 103 (73%)
156 (67.5%) 100 (51%) 38 (27%)
49 (36.8%) 56 (42.1%) 159 (54.8%)
84 (63.2%) 77 (57.9%) 131 (45.2%)
170 (49.6%) 104 (46.4%)
173 (50.4%) 120 (53.6%)
69 57 121 18
(61.6%) (37.3%) (48.4%) (46.2%)
43 96 129 21
(38.4%) (62.7%) (51.6%) (53.8%)
102 31 47 94
(59%) (43.7%) (43.5%) (43.5%)
71 40 61 122
(41%) (56.3%) (56.5%) (56.5%)
P value <0.001
0.227
<0.001
0.001
0.492
0.001
0.010
Significant P values (<0.05) were highlighted in bold. a Information on site, histology and depth of tumour invasion was missing respectively in 12, 1 and 14 patients.
in patients younger than 65 years and in cancer arising from the body/fundus. As regards stage, less advanced cancer (muscularis propria-limited tumours and N0) were more frequently treated with D2 lymphadenectomy, while more advanced cancer (T3 and Nþ) more frequently underwent D3. No difference in lymphadenectomy extension emerged as regards Lauren histotype (p ¼ 0.492). As expected, para-aortic nodes were retrieved in nearly all patients undergoing D3 (279/294 ¼ 94.9%) and in just one patient undergoing D2 (0.4%). Para-aortic node invasion was detected in 34 patients undergoing D3, which corresponds to 11.6% of the whole series (n ¼ 294) and to 12.2% of patients with retrieved para-aortic nodes (n ¼ 279). In-hospital mortality was slightly higher after D2 (11 patients, 4%) than after D3 (7 patients, 2.4%) (p ¼ 0.340). Cumulative incidence of any relapse During the follow-up 260 patients were diagnosed with cancer recurrence or died from recurrence. Cumulative incidence of relapse did not significantly differ after D2 and after D3 both in the whole series (45.3% vs 46.3%; p ¼ 0.866) (Table 2) and when patients were stratified by centre, sex, age, site, pN stage according to the ManteleHaenszel test of homogeneity. The association between relapse and extension of lymphadenectomy was significantly affected by histology (test of homogeneity: p ¼ 0.007) while the influence
of depth of tumour invasion approached statistical significance (p ¼ 0.061). The risk of relapse was higher after D3 than after D2 (45.1% vs 35.3%; p ¼ 0.078) in the intestinal histotype while the pattern was reversed in the mixed/diffuse histotype (48.3% vs 61.5%; p ¼ 0.060). This pattern was substantially confirmed in multivariable logistic regression (Table 3): the interaction between histology and extension of lymphadenectomy was highly significant (p ¼ 0.004). Of note, none of the other risk factors presented a significant interaction with the extension of lymphadenectomy. As expected, depth of tumour invasion, nodal status and site were significant prognostic factors. Cumulative incidence of specific relapse D2 and D3 groups did not significantly differ not only as regards overall relapse, but also when separately considering each type of relapse (Table 2). Table 2 Pattern of relapse as a function of extension of lymphadenectomy.
Overall relapse Locoregional relapse Peritoneal relapse Distant relapse
Extended (D2) (n ¼ 274)
Superextended (D3) (n ¼ 294)
P value
124 45 40 29
136 50 34 38
0.866 0.911 0.319 0.436
(45.3%) (16.4%) (14.6%) (10.6%)
P values were computed by Fisher’s exact test.
(46.3%) (17.0%) (11.6%) (12.9%)
G. de Manzoni et al. / EJSO 41 (2015) 534e540 Table 3 Odds ratio (OR) of relapse, evaluated by a multivariable logistic regression, controlling for all other variables included in the model. The joint effect of histology and lymphadenectomy is presented as the interaction between the two variables was highly significant. OR (95% CI) Siena vs Verona Sex (women vs men) Age 65e74 vs <65 years 75 vs <65 years Site Body vs fundus Antrum vs fundus T stage T3 vs T2 T4a vs T2 T4b vs T2 N stage N1 vs N0 N2 vs N0 N3 vs N0 Interaction histologya lymphadenectomy D2 intestinal D2 diffuse D3 intestinal D3 diffuse a
0.88 (0.56e1.36) 0.74 (0.49e1.14)
a
P value 0.555 0.171 0.166
1.49 (0.93e2.40) 0.99 (0.56e1.74) <0.001 0.32 (0.18e0.57) 0.41 (0.24e0.69) <0.001 1.76 (0.91e3.40) 4.04 (2.19e7.44) 5.23 (1.98e13.83) <0.001 1.07 (0.52e2.23) 3.71 (2.05e6.71) 4.68 (2.72e8.05) 0.004 1 2.65 (1.40e5.04) 1.27 (0.72e2.22) 1.02 (0.54e1.93)
P values were computed by the LR test.
When analysing type of recurrences as a function of both histology and extension of lymphadenectomy, cumulative incidence of locoregional recurrences was higher in the diffuse histotype after D2, while being higher in the intestinal histotype after D3 (Fig. 1). According to the ManteleHaenszel test of homogeneity, this discrepancy was significant (p ¼ 0.017). The relation between histology and specific relapse was not modified by extension of lymphadenectomy for the other types of relapse (p ¼ 0.271 for peritoneal relapse, p ¼ 0.368 for distant relapse).
Figure 1. Type of recurrence as a function of histology and extension of lymphadenectomy. When multiple recurrences had occurred, each of them was separately considered. P values were computed by ManteleHaenszel test of homogeneity.
537
Distribution of T and N categories according to tumour histotype and extension of lymphadenectomy As shown in Table 4, the relation between T status, N status and the extension of lymphadenectomy was similar when comparing intestinal and mixed/diffuse tumours (Table 4). Indeed for both histotypes a D2 dissection was more frequently performed for pT2 tumours while a D3 lymphadenectomy was more frequently chosen for pT3 and pT4b (Table 4). Also the pattern of N status was rather similar when comparing intestinal and mixed-diffuse histotypes, with N0 patients preferentially undergoing D2 and Nþ patients preferentially undergoing D3, however these differences were more pronounced and significant in the diffuse histotype (Table 4). Discussion The main results of the present study were: 1) The extension of lymphadenectomy did not significantly affect the cumulative incidence of relapse. 2) The risk of relapse was higher in Lauren diffuse tumours with respect to intestinal tumours after D2 lymphadenectomy. However this pattern was reversed after D3 lymphadenectomy. 3) The qualititative interaction between extension of lymphadenectomy and histotype, observed when Table 4 Baseline differences in tumour stage between the D2 and D3 groups, as a function of Lauren histotype.
Intestinal histotype Depth of tumour invasion* T2 T3 T4a T4b Node metastases N0 N1 N2 N3 Mixed/diffuse histotype Depth of tumour invasion** T2 T3 T4a T4b Node metastases N0 N1 N2 N3
Extended (D2) (n ¼ 274)
Superextended (D3) (n ¼ 294)
49 46 62 10
(64%) (40%) (50%) (45%)
28 69 61 12
(36%) (60%) (50%) (55%)
74 23 30 43
(56%) (50%) (43%) (44%)
57 23 39 54
(44%) (50%) (57%) (56%)
20 11 59 8
(57%) (29%) (47%) (47%)
15 27 67 9
(43%) (71%) (53%) (53%)
28 8 17 51
(68%) (32%) (44%) (43%)
13 17 22 68
(32%) (68%) (56%) (57%)
P value
0.014
0.204
0.099
0.014
Information on histology was missing in one subject. Information on depth of tumour invasion was missing in 6 and 8 patients, respectively in intestinal (*) and mixed/diffuse histotype (**). Significant P values (<0.05) were highlighted in bold.
538
G. de Manzoni et al. / EJSO 41 (2015) 534e540
considering overall relapse, mainly reflected a similar pattern in locoregional relapse. Up to 50% of patients with gastric cancer develop a recurrence after radical resection. The main patterns of relapse are loco-regional, peritoneal and hematogenous.2e4,19e22 Loco-regional relapses comprise recurrences that occur at gastric stump, at anastomosis, in the gastric bed or in lymph nodes. There are many observations suggesting that the rate of loco-regional relapse is affected by the surgical approach, mainly by the extension of lymphadenectomy. Indeed in Eastern series,6,22,23 where at least a D2 dissection is performed, the rate of loco-regional recurrence is lower compared to Western series in which several patients undergo a limited (D1) lymphadenectomy.20,21 Interestingly a recent GIRCG study2 reported that in Western high volumes centres the increase in the proportion of extended (D2/D3) lymphadenectomies was associated with a simultaneous decrease in the incidence of locoregional relapses, which approached values similar to those observed in Eastern series. Nowadays D2 lymphadenectomy has become the standard of surgical management in advanced gastric cancer also in Western countries24e29; nevertheless it is still controversial whether a super-extended lymphadenectomy could provide a better local control of gastric tumours compared to D2 dissection. In the JCOG 9501 trial no noticeable difference in the pattern of relapse between D2 and D2 þ PAND was observed.6 Of note in the Japanese trial, for tumours of the lower third of the stomach, D2 dissection comprised also the removal of retropancreatic nodes (No. 13) and nodes along the superior mesenteric vein (No. 14v)7: it should be reminded that removal of No. 13 station was not part of a standard D2 lymphadenectomy,16 while a routine dissection of No. 14v station is currently no more indicated for distal tumours.8 An observational study from Japan30 reported no differences in the incidence of recurrences nor modification in relapse pattern between D2 and D3 patients. Anyway the incidence of lymphatic recurrence was lower after D3 compared with D2 in tumours of 50e100 mm of diameter suggesting the utility of D3 dissection in a subgroup of patients with advanced gastric cancer. In the present study D3 lymphadenectomy was associated with a lower risk of recurrence in tumours with mixed-diffuse histology compared to D2. In detail, tumours with mixed-diffuse histology showed a lower rate of locoregional relapse after D3 dissection, while for intestinal tumours the pattern was reversed. Our results suggest that D3 lymphadenectomy, which includes removal of posterior (No. 12p, No. 13, No. 14v) and para-aortic (No. 16) nodes, provides a better local control of disease in advanced gastric tumours with mixed-diffuse histotype. It should be underlined that Lauren diffuse
tumours have a more pronounced lymphotropism and a greater propensity to metastatize to third level lymph nodes31 as compared to intestinal tumours, and these observations could at least partly explain the lower loco-regional recurrence rates recorded in this tumour type after a super-extended lymphadenectomy. The present study was based on one of the largest series undergoing D3 lymphadenectomy in the Western world. The two centres involved, although presenting differences in tumour characteristics, could rely on surgeons sharing the same surgical training and approach. Moreover, clinical data, although retrospectively analysed, were prospectively collected, and all patients with advanced adenocarcinoma were consecutively enrolled and regularly followed-up. Of course, the main limitation of the present study is its observational nature. Indeed surgeons were more prone to adopt super-extended lymphadenectomy in younger patients with more advanced tumours arising from the upper third of the stomach, in the effort to increase patients’ chance to survive. In this respect, the comparison between D2 and D3 is surely biased and randomized trials are necessary to fully elucidate the matter. However, it is unlikely to attribute the different impact of lymphadenectomy on relapse in different histotypes to surgeon’s preference. Indeed the choice of lymphadenectomy extension was not affected by histotype in the present series and similar distribution of T and N categories in the D2 and the D3 groups was observed when comparing intestinal and mixed-diffuse tumours. So if D3 reverses the negative impact of diffuse histotype on relapses, this finding is less likely to be biased. None of the patients of the present series had received preoperative chemotherapy. A recent Japanese study32 showed that D2 þ PAND is feasible after neo-adjuvant chemotherapy. Although further evaluations are needed to confirm our findings also in a neo-adjuvant setting, our results could help in designing new trials on super-extended lymphadenectomy in the era of multimodal therapy, where patients are stratified according to tumour histotype. In conclusion, D3 lymphadenectomy reverses the negative impact of diffuse histotype on relapses, especially on locoregional recurrences. It is not conceivable that this result can be the consequence of surgeon’s preferences. Therefore D3 could be considered a valid therapeutic option in histotype-oriented tailored treatment of AGC. The role of D3 in advanced gastric cancer, although restricted to a minority of patients, is further increased by current epidemiological trends: indeed there is a higher proportion of diffuse-type gastric cancer, especially in Western countries, due to the declining incidence of distal, intestinaltype gastric tumours, at least partly related to the decreasing prevalence of Helicobacter Pylori infection.2,33,34 Conflict of interest statement All the authors declare that there is no financial interest or any other potential conflict of interest.
G. de Manzoni et al. / EJSO 41 (2015) 534e540
References 1. Ferlay J, Soerjomataram I, Ervik M, et al. GLOBOCAN 2012 v1.0, Cancer Incidence and Mortality Worldwide: IARC CancerBase No. 11 [Internet]. Lyon, France: International Agency for Research on Cancer; 2013. Available from: http://globocan.iarc.fr. [accessed on 3.1.14]. 2. Marrelli D, Pedrazzani C, Morgagni D, et al. On behalf of the Italian Research Group for Gastric Cancer (IRGGC). Changing clinical and pathological features of gastric cancer over time. Br J Surg 2011;98: 1273–83. http://dx.doi.org/10.1002/bjs.7528 PMID: 21560122. 3. Deng J, Liang H, Wang D, Sun D, Pan Y, Liu Y. Investigation of the recurrence patterns of gastric cancer following a curative resection. Surg Today 2011;41:210–5. http://dx.doi.org/10.1007/s00595-0094251-y PMID: 21264756. 4. Spolverato G, Ejaz A, Kim Y, et al. Rates and patterns of recurrence after curative intent resection for gastric cancer: a United States multi-institutional analysis. J Am Coll Surg 2014;219(4):664–75. http://dx.doi.org/10.1016/j.jamcollsurg.2014.03.062 PMID: 25154671. 5. 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. http://dx.doi.org/10.1016/S1470-2045(10)70070-X PMID: 20409751. 6. Sasako M, Sano T, Yamamoto S, et al. D2 lymphadenectomy alone or with para-aortic nodal dissection for gastric cancer. N Engl J Med 2008; 359:453–62. http://dx.doi.org/10.1056/NEJ- Moa0707035 PMID: 18669424. 7. Sano T, Sasako M, Yamamoto S, et al. Gastric cancer surgery: morbidity and mortality results from a prospective randomized controlled trial comparing D2 and extended para-aortic lymphadenectomy e Japan Clinical Oncology Group Study 9501. J Clin Oncol 2004;22:2767–73. http://dx.doi.org/10.1200/JCO.2004.10.184 PMID: 15199090. 8. Japanese Gastric Cancer Association. Japanese gastric cancer treatment guidelines 2010 (ver. 3). Gastric Cancer 2011;14:113–23. http://dx.doi.org/10.1007/s10120-011-0042-4 PMID: 1573742. 9. Roviello F, Pedrazzani C, Marrelli D, et al. Super-extended (D3) lymphadenectomy in advanced gastric cancer. Eur J Surg Oncol 2010; 36(5):439–46. http://dx.doi.org/10.1016/j.ejso.2010.03.008 PMID: 20392590. 10. de Manzoni G, Di Leo A, Roviello F, et al. Tumor site and perigastric nodal status are the most important predictors of para-aortic nodal involvement in advanced gastric Cancer. Ann Surg Oncol 2011;18: 2273–80. http://dx.doi.org/10.1245/s10434-010-1547-5 PMID: 21286941. 11. Marrelli D, Pedrazzani C, Neri A, et al. Complications after extended (D2) and superextended (D3) lymphadenectomy for gastric cancer: analysis of potential risk factors. Ann Surg Oncol 2007;14:25–33. http://dx.doi.org/10.1245/s10434-006-9063-3 PMID: 17024558. 12. G€ unther K, Horbach T, Merkel S, et al. D3 lymph node dissection in gastric cancer: evaluation of postoperative mortality and complications. Surg Today 2000;30:700–5. http://dx.doi.org/10.1007/ s005950070080 PMID: 10955732. 13. Bostanci EB, Kayaalp C, Ozogul Y, Aydin C, Atalay F, Akoglu M. Comparison of complications after D2 and D3 dissection for gastric cancer. Eur J Surg Oncol 2004;30:20–5. http://dx.doi.org/10.1016/ j.ejso.2003.10.008 PMID: 14736518. 14. Verlato G, Roviello F, Marchet A, et al. Indexes of surgical quality in gastric cancer surgery: experience of an Italian network. Ann Surg Oncol 2009;16:594–602. http://dx.doi.org/10.1245/s10434-008-0271-x PMID: 19118437. 15. Marrelli D, Roviello F, de Manzoni G, et al. Different patterns of recurrence in gastric cancer depending on Lauren’s histological type: longitudinal study. World J Surg 2002;26(9):1160–5. http:// dx.doi.org/10.1007/s00268-002-6344-2 PMID: 12209247.
539
16. Japanese Gastric Cancer Association. Japanese classification of gastric carcinomad2nd English editiond. Gastric Cancer 1998;1:10–24. PMID: 11957040. 17. Sobin LH, Gospodarowicz MK, Wittekind CH, editors. International Union Against Cancer (UICC) TNM classification of malignant tumors. 7th ed. Oxford: Wiley-Blackwell; 2009. 18. Lauren PA, Nevalainen TJ. Epidemiology of intestinal and diffuse types of gastric carcinoma. A time-trend study in Finland with comparison between studies from high- and low-risk areas. Cancer 1993;71:2926–33. PMID: 8490820. 19. Marrelli D, De Stefano A, de Manzoni G, et al. Prediction of recurrence after radical surgery for gastric cancer: a scoring system obtained from a prospective multicenter study. Ann Surg 2005;241: 247–55. http://dx.doi.org/10.1097/01.sla.0000152019.14741.97 PMID: 15650634. 20. Roviello F, Marrelli D, de Manzoni G, et al. Prospective study of peritoneal recurrence after curative surgery for gastric cancer. Br J Surg 2003;90:1113–9. http://dx.doi.org/10.1002/bjs.4164 PMID: 12945079. 21. D’Angelica M, Gonen M, Brennan MF, et al. Patterns of initial recurrence in completely resected gastric adenocarcinoma. Ann Surg 2004; 240:808–16. http://dx.doi.org/10.1097/01.sla.0000143245.28656.15 PMID: 15492562. 22. Yoo CH, Noh SH, Shin DW, et al. Recurrence following curative resection for gastric carcinoma. Br J Surg 2000;87:236–42. http:// dx.doi.org/10.1046/j.1365-2168.2000.01360.x PMID: 10671934. 23. Maehara Y, Hasuda S, Koga T, Tokunaga E, Kakeji Y, Sugimachi K. Postoperative outcome and sites of recurrence in patients following curative resection of gastric cancer. Br J Surg 2000;87(3):353–7. http://dx.doi.org/10.1046/j.1365-2168.2000.01358.x PMID: 10718807. 24. Meyer HJ, H€olscher AH, Lordick F, et al. Current S3 guidelines on surgical treatment of gastric carcinoma. Chirurg 2012;83:31–7. http://dx.doi.org/10.1007/s00104-011-2149-x PMID: 22127381. 25. Allum WH, Blazeby JM, Griffin SM, Cunningham D, Jankowski JA, Wong R. Association of Upper Gastrointestinal Surgeons of Great Britain and Ireland, the British Society of Gastroenterology and the British Association of Surgical Oncology. Guidelines for the management of oesophageal and gastric cancer. Gut 2011;60:1449–72. http:// dx.doi.org/10.1136/gut.2010.228254 PMID: 2170545654. 26. Okines A, Verheij M, Allum W, Cunningham D, Cervantes A. Gastric cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol 2010;21(Suppl. 5):v50–4. http:// dx.doi.org/10.1093/annonc/mdq164 PMID: 20555102. 27. Waddell T, Verheij M, Allum W, Cunningham D, Cervantes A, Arnold D. Gastric cancer: ESMO-ESSO-ESTRO Clinical 55 Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol 2013; 24(Suppl. 6):vi57–63. http://dx.doi.org/10.1093/annonc/mdt344 PMID: 24078663. 28. De Manzoni G, Baiocchi GL, Framarini M, et al. The SIC-GIRCG 2013 Consensus Conference on gastric cancer. Updates Surg 2014;66(1):1–6. http://dx.doi.org/10.1007/s13304-014-0248-1 PMID: 24523031. 29. Verlato G, Giacopuzzi S, Bencivenga M, Morgagni P, De Manzoni G. Problems faced by evidence-based medicine in evaluating lymphadenectomy for gastric cancer. World J Gastroenterol 2014;20(36):12883–91. http://dx.doi.org/10.3748/wjg.v20.i36.12883. Review. PMID: 25278685. 30. Kunisaki C, Akiyama H, Nomura M, et al. Comparison of surgical results of D2 versus D3 gastrectomy (para-aortic lymph node dissection) for advanced gastric carcinoma: a multi-institutional study. Ann Surg Oncol 2006;13(5):659–67. http://dx.doi.org/10.1245/ASO.2006.07.015 PMID: 16538414. 31. Esaki Y, Hirayama R, Hirokawa K. Comparison of patterns of metastasis in gastric cancer by histologic type and age. Cancer 1990;65(9): 2086–90. PMID: 2372774. 32. Tsuburaya A, Mizusawa J, Tanaka Y, Fukushima N, Nashimoto A, Sasako M. Stomach Cancer Study Group of the Japan Clinical Oncology Group. Neoadjuvant chemotherapy with S-1 and cisplatin followed by D2 gastrectomy with para-aortic lymph node dissection for gastric
540
G. de Manzoni et al. / EJSO 41 (2015) 534e540
cancer with extensive lymph node metastasis. Br J Surg 2014;101(6): 653–60. http://dx.doi.org/10.1002/bjs.9484 PMID: 24668391. 33. de Martel C, Forman D, Plummer M. Gastric cancer: epidemiology and risk factors. Gastroenterol Clin North Am 2013;42(2):219–40. http://dx.doi.org/10.1016/j.gtc.2013.01.003 PMID: 23639638.
34. Henson DE, Dittus C, Younes M, et al. Differential trends in the intestinal and diffuse types of gastric carcinoma in the United States, 1973e2000: increase in the signet ring cell type. Arch Pathol Lab Med 2004;128:765–70. PMID: 15214826.