Advertisement

Prognostic value of tumor deposits in lymph node-negative gastric cancer: A propensity score matching study

Open AccessPublished:December 13, 2022DOI:https://doi.org/10.1016/j.ejso.2022.12.004

      Abstract

      Background

      The purpose of this study was to assess the prognostic value of TD in lymph node-negative GC.

      Methods

      A retrospective study was conducted to collect the clinicopathological data from 1224 patients with lymph node-negative GC. According to their TD status, patients were categorized into TD-positive and TD-negative groups. Patients in both groups underwent a 1:1 propensity score matching analysis. Survival analysis was performed by the Kaplan-Meier method, and the differences between survival curves were measured by log-rank test. The cox proportional hazards model was used for univariate and multivariate analyses.

      Results

      The TD-negative group had higher 5-year overall survival(OS) rate than TD-positive group(69.4%VS.36.4%,P < 0.05). Further subgroup analysis indicated that patients in the TD-negative group had higher 5-year OS rates than those in the TD-positive group in the T1-2, T3, and T4 subgroups(all with P < 0.05).The OS rates were decreased with the increase of the number of TD.The univariate Cox regression analysis demonstrated that tumor location in antrum, distal gastrectomy, perineural invasion, T4-stage,lymphovascular invasion and the number of TD were all associated with prognosis in patients undergoing curative gastric resection (P < 0.05).The multivariable analysis revealed that the number of TD, perineural invasion, lymphovascular invasion and T4 stage were independently associated with OS.

      Conclusion

      In lymph node-negative GC, TD is an independent risk factor for prognosis, regardless of T-stage, and patients with ≥3 TD have a worse prognosis.

      Keywords

      1. Introduction

      Gastric cancer (GC) incidence and mortality have decreased in recent years because of improved medical techniques and treatments, but it is still one of the leading causes of cancer deaths [
      • Sung H.
      • Ferlay J.
      • Siegel R.L.
      • Laversanne M.
      • Soerjomataram I.
      • Jemal A.
      • et al.
      Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality.
      ]. In most parts of the world, surgery and postoperative adjuvant chemotherapy are considered the mainstay of treatment for advanced GC [
      • Cats A.
      • Jansen E.P.M.
      • van Grieken N.C.T.
      • Sikorska K.
      • Lind P.
      • Nordsmark M.
      • et al.
      Chemotherapy versus chemoradiotherapy after surgery and preoperative chemotherapy for resectable gastric cancer (CRITICS): an international, open-label, randomised phase 3 trial.
      ,
      Japanese Gastric Cancer Association
      Japanese gastric cancer treatment guidelines 2018.
      ]. The Tumor Node Metastasis(TNM)classification system is widely used to evaluate the outcome of patients with GC and to determine the best treatment options on the basis of tumor stage [
      • Hoda
      • Syed A.
      AJCC cancer staging manua.
      ,
      • Ajani J.A.
      • D'Amico T.A.
      • Bentrem D.J.
      • Chao J.
      • Cooke D.
      • Corvera C.
      • et al.
      Gastric cancer, version 2.2022, NCCN clinical practice guidelines in oncology.
      ].In our clinical experience, we found that the prognosis of patients with the same stage of GC differed significantly, even though they received the same standard treatment. Recent studies have also reported several clinicopathological factors with prognostic values for patients with GC, such as histological differentiation [
      • Pernot S.
      • Voron T.
      • Perkins G.
      • Lagorce-Pages C.
      • Berger A.
      • Taieb J.
      • et al.
      Signet-ring cell carcinoma of the stomach: impact on prognosis and specific therapeutic challenge.
      ], perineural invasion [
      • Chang K.
      • Song B.
      • Do I.G.
      • Koo D.H.
      • Lee H.W.
      • Son B.H.
      • et al.
      Venous invasion and perineural invasion as upstaging and poor prognostic factors in N0 gastric cancers.
      ], and TD [
      • Chen H.
      • Tang Z.
      • Chen L.
      • Li H.
      • Wang X.
      • Liu F.
      • et al.
      Evaluation of the impact of tumor deposits on prognosis in gastric cancer and a proposal for their incorporation into the AJCC staging system.
      ].
      TDs are irregular collections of discrete tumor cells detached from the original lesion in the soft tissue or fat [
      • Wang W.
      • Li Y.
      • Zhang Y.
      • Yuan X.
      • Xu D.
      • Guan Y.
      • et al.
      Incorporation of extranodal metastasis of gastric carcinoma into the 7th edition UICC TNM staging system.
      ].TD has been incorporated in the N stage in colorectal cancer since the 7th edition of American Joint Committee on Cancer (AJCC) TNM stage [
      • Edge S.B.
      • Compton C.C.
      The American Joint Committee on Cancer: the 7th edition of the AJCC cancer staging manual and the future of TNM.
      ].It was narrowly applied in the presense of negative lymph nodes. If TD is present, grading is attributed to N1C, regardless of the number of TD and T-stage. However, it is still intensely debated about the prognostic impact of TD on GC patients and its categorization in TMN staging [
      • Chen H.
      • Tang Z.
      • Chen L.
      • Li H.
      • Wang X.
      • Liu F.
      • et al.
      Evaluation of the impact of tumor deposits on prognosis in gastric cancer and a proposal for their incorporation into the AJCC staging system.
      ,
      • Etoh T.
      • Sasako M.
      • Ishikawa K.
      • Katai H.
      • Sano T.
      • Shimoda T.
      Extranodal metastasis is an indicator of poor prognosis in patients with gastric carcinoma.
      ,
      • Liang Y.
      • Wu L.
      • Liu L.
      • Ding X.
      • Wang X.
      • Liu H.
      • et al.
      Impact of extranodal tumor deposits on prognosis and N stage in gastric cancer.
      ,
      • Lee H.S.
      • Lee H.E.
      • Yang H.K.
      • Kim W.H.
      Perigastric tumor deposits in primary gastric cancer: implications for patient prognosis and staging.
      ,
      • Tan J.
      • Yang B.
      • Xu Z.
      • Zhou S.
      • Chen Z.
      • Huang J.
      • et al.
      Tumor deposit indicates worse prognosis than metastatic lymph node in gastric cancer: a propensity score matching study.
      ,
      • Sun Z.
      • Wang Z.N.
      • Xu Y.Y.
      • Zhu G.L.
      • Huang B.J.
      • Xu Y.
      • et al.
      Prognostic significance of tumor deposits in gastric cancer patients who underwent radical surgery.
      ,
      • Anup S.
      • Lu J.
      • Zheng C.H.
      • Li P.
      • Xie J.W.
      • Wang J.B.
      • et al.
      Prognostic significance of perigastric tumor deposits in patients with primary gastric cancer.
      ]. Patients with lymph node-negative GC were included in this study. Baseline confounding factors and selective bias in the TD-negative and TD-positive groups were overcome by propensity score matching method (PSM) [
      • Yao X.I.
      • Wang X.
      • Speicher P.J.
      • Hwang E.S.
      • Cheng P.
      • Harpole D.H.
      • et al.
      Reporting and guidelines in propensity score analysis: a systematic review of cancer and cancer surgical studies.
      ]. The aim of this study was to analyze more precisely the prognostic value of TD status and number on lymph node-negative GC for clinical diagnosis and management.

      2. Materials and methods

      2.1 Patients

      Case data of patients who had undergone radical GC surgery at Yi ji shan Hospital of Wannan Medical College were collected and screened between January 2012 and January 2018. Informed consent was given to each patient before their gastroscopy, surgery, or postoperative adjuvant chemotherapy. For GC patients with stage II–III disease, post-operative adjuvant chemotherapy was accompanied by regimens of CapeOX (capecitabine and oxaliplatin, CapeOX) or SOX(S-1 plus oxaliplatin, SOX) [
      Japanese Gastric Cancer Association
      Japanese gastric cancer treatment guidelines 2018.
      ].The patient inclusion criteria were as follows (a)postoperative pathology confirmed gastric adenocarcinoma and T1-4N0M0 stage, (b) tumors epicenter located >2 cm into the proximal stomach [
      • Hoda
      • Syed A.
      AJCC cancer staging manua.
      ],(c)completion of radical gastrectomy [
      Japanese Gastric Cancer Association
      Japanese gastric cancer treatment guidelines 2018.
      ], (d) complete follow-up (e) no history of neoadjuvant chemotherapy or radiotherapy. The exclusion criteria were as follows: (a) patients who had undergone a gastrectomy or who had another type of cancer; (b) patients who had undergone emergency surgery for GC; and (c) patients who died because of non-tumor related death. Clinicopathological data were recorded and analyzed according to the 8th edition of the AJCC-TNM staging system [
      • Hoda
      • Syed A.
      AJCC cancer staging manua.
      ]. The patient inclusion flow chart is presented in Fig. 1.All clinic-pathological data of enrolled patients were collected, including age, gender, histological types of the gastric adenocarcinomas, depth of tumor invasion, tumor location, tumor size, TD status, number of TD, presence or absence of lymphovascular invasion,
      perineural invasion, type of gastrectomy, and postoperative adjuvant therapy. This study complied with the standards of the Declaration of Helsinki and current ethical guidelines. The proposed investigation was examined and certified by the Ethics Committee of the Yi ji shan Hospital of Wannan Medical College.

      2.2 Definition of TD and group according TD status and number

      TDs are defined as discrete tumor nodules within the lymph drainage area of the original carcinoma, lacking lymph node tissue and vascular or neural structure, regardless of its shape, size, or contour [
      • Hoda
      • Syed A.
      AJCC cancer staging manua.
      ,
      • Wang W.
      • Li Y.
      • Zhang Y.
      • Yuan X.
      • Xu D.
      • Guan Y.
      • et al.
      Incorporation of extranodal metastasis of gastric carcinoma into the 7th edition UICC TNM staging system.
      ]. The pathological illustrations of TD is presented in Fig. 2. According to the status of TD, our included patients were grouped into either TD-negative or TD-positive groups. After matching, all paired patients were categorized into three groups according to the number of TD: L0, L1 and L2. Patients in the L0 group had no TD, those in the L1 group had one or two TDs, and those in the L2 group had three or more TDs.
      Fig. 2
      Fig. 2Cancer cells deposit in adipose tissue discontinuous with the primary lesion (A:H&E, × 20; B: H&E, × 100).

      2.3 Follow-up

      Patients were followed up using telephone, WeChat, or outpatient consultations every 3 months for 1 year, then every 6 months for 2 years, and finally every year until death. The OS time was calculated from the date of surgery to the last follow-up date or death. The median follow-up time was 41 months (range: 6–120 months). The final follow-up was scheduled on May 1, 2022.

      2.4 Statistical analysis

      Before PSM, baseline information was compared between the TD-negative and TD-positive groups. The χ2 test or Fisher test was used to compare the categorical variables. The Mann Whitney U test was used to compare the ordered categorical variables. 5-OS rates were calculated for each group using the Kaplan-Meier method. The log-rank test was used to compare survival differences between groups. COX regression models were used for univariate and multivariate analysis of prognostic factors. To reduce selection bias and differences in baseline information, a 1:1 PSM analysis was performed between the TD-positive and TD-negative groups. A dichotomous logistic regression model was used to assess the propensity score, and the matched variables were age, gender, tumor location, histological differentiation, tumor size, pT stage, lymphovascular invasion, perineural invasion, and type of gastrectomy.
      Based on this propensity score, patients in TD-positive group were matched with patients in TD-negative group by the nearest neighbor method, with the caliper set to 0.2. SPSS24.0(Chicago,IL)and R language version3.4.2 was used for all statistical analyses, and P < 0.05 was considered statistically significant.

      3. Result

      3.1 Patient's clinicopathologic characteristics

      A total of 1224 patients were enrolled in the study. There were 91 patients in the TD-positive group and 1133 patients in the TD-negative group. The number of TD ranged from 1 to 9. Of the 91 patients, 30.8% (28) had 1 TD, 35.1% (32) had 2, 34.1% (31) had 3 or more. There were statistically significant differences in age, tumor size, location of tumor, T stage, lymphovascular invasion, perineural invasion, and type of gastrectomy between the two groups (P < 0.05). Compared with patients in the TD-negative group, patients in the TD-positive group were older,and had a larger tumor size, a higher proportion of T3-4 stage, and a higher proportion of perineural invasion and lymphovascular invasion. The incidence of tumor location in the antrum was lower in the TD-positive group, and patients in the TD-positive group were less likely to undergo distal partial gastrectomy (P < 0.05).There were no statistically significant differences between the two groups in terms of gender and histological differentiation of patients (P > 0.05).The above observations show that TD is significantly associated with the following 6 clinicopathological factors, namely older age, large tumor diameter, tumor location in the cardia fundus or gastric body, T3-4 stage, perineural invasion and lymphovascular invasion(Table 1).
      Table 1Clinicopathologic characteristics of included patients before and after Propensity Score Matching.
      VariablesBefore PSM(n = 1224)After PSM(n = 160)
      TD(+)group

      (n = 91)
      TD(−)group

      (n = 1133)
      χ2/Zp

      value
      TD(+)group(n = 80)TD(−)group (n = 80)χ2/ZP value
      Age(years)5.4630.0190.0350.851
      <60193711819
      ≥60727626261
      Gender0.0020.9670.8480.357
      Male668245863
      Female253092217
      Tumor size60.2200.0000.2380.625
      ≤4 cm338433229
      >4 cm582904851
      Histologic grade0.7060.4011.5600.693
      Differentiated738656365
      undifferentiated182681715
      Tumor location7.7250.0210.4200.811
      Cardia fundus353132826
      Gastric body151411215
      Antrum416794039
      Depth of invasion
      According to the 8th edition of the AJCC-TNM staging system.
      −5.8050.000−0.6520.514
      T1350932
      T2912288
      T3551934643
      T4243092327
      Neural invasion33.9120.0000.0001.000
      (+)603995050
      (−)317343030
      Vascular invasion90.7510.0000.0260.872
      (+)592334847
      (−)328993233
      Type of gastrectomy32.8680.0001.1430.565
      Pro171971712
      Dis367333536
      Total382032832
      Differentiated, papillary or well/moderately differentiated tubular adenocarcinoma.
      Undifferentiated, poorly differentiated,or mucinous adenocarcinoma or signet-ring cell carcinoma.
      a According to the 8th edition of the AJCC-TNM staging system.

      3.2 Propensity score-matching analysis

      After 1:1 PSM, sensitivity analysis was performed on 160 GC patients who were successfully matched. All baseline variables were well balanced between the two groups of patients(Table 1). The scores of TD-positive and TD-negative groups before and after matching are shown by histograms. Fig. 3A and B shows that the TD-negative and TD-positive groups are significantly different before matching, while Fig. 3C,and 3D shows the two groups with similar scores after matching. The visual data suggest that the matching was successful(Fig. 3).
      Fig. 3
      Fig. 3Histograms of propensity scores before and after matching.(A,B before matching,C,D after matching).

      3.3 Survival analysis in the matched cohort(n = 160)

      All patients in the matched cohort received clinical follow-up (range, 6–120 months; median follow-up, 41 months). The 5-year OS rates of TD-positive groups and TD-negative groups were 69.4%, 36.4%, respectively. Patients in the TD-negative group had a significantly better prognosis than those in the TD-positive group. The OS curves of these two groups differed significantly(P < 0.05). (Fig. 4A).
      Fig. 4
      Fig. 4(A) Kaplan Meier survival curve of patients according to the status of tumor deposits(positive/negative) (B)Kaplan Meier survival curve of patients according to the number of tumor deposits (L0/L1/L2).

      3.4 Survival analysis of TD status in T1-2, T3, T4 subgroups

      In the T1-2, T3, and T4 staging subgroups, the 5-year OS rates was higher in the TD-negative group than in the positive group(P < 0.05).This result suggests that the prognosis of patients with GC without lymph node metastasis is poor if TD is present, which is not affected by T stage.Further study found that the prognosis of patients in the T1-2 stage TD-positive group was similar to that of patients in the T3 stage TD-negative group and the prognosis of patients in the T3 stage TD-positive group was similar to that of patients in the T4 stage TD-negative group. This result suggests that TD may elevate the T stage grade in lymph node-negative GC.

      3.5 Prognostic impact of the TD number

      The 5-year OS rates of groups L0, L1,and L2 were 70.3%,44.9%, and 21.5%, respectively. The OS curves of these three groups differed significantly(P < 0.05). Further studies found that patients in the L1 group had a better prognosis than patients in the L2 group (P < 0.05), while patients in the L0 group had a better prognosis than patients in the L1 group (P < 0.05) (Fig. 4B).The results suggest that in lymph node-negative GC, the number of TD is a risk factor for prognosis, and the prognosis worsens as the number of TD increases.

      3.6 Prognostic impact of postoperative chemotherapy

      In this study, a total of 139 patients were required to receive chemotherapy for postoperative pathological stage II-III after matching. However, only 97 patients received postoperative chemotherapy(including some stage I patients). Among the 97 patients who received chemotherapy, 50 were from the TD-positive group, and 47 were from the TD-negative group. The 5-year OS rate was higher in the TD-negative group than that in the TD-positive group(P < 0.05).Of the 63 patients who did not receive chemotherapy, 30 were from the TD-positive group,and 33 were from the TD-negative group. The 5-year OS rate was lower in the TD-positive group than that in the TD-negative group(P < 0.05).The above results suggest that TD is a prognostic factor for pathological stage II-III patients regardless of postoperative chemotherapy. According to the 8th edition of the AJCC staging system, 21 patients had postoperative pathological staging of stage I, including 11 patients with TD. Among them, 5 received chemotherapy and 6 did not receive chemotherapy. In terms of 5-year survival rate, patients in the group receiving postoperative adjuvant chemotherapy were higher than those in the group not receiving postoperative adjuvant chemotherapy (53.3% vs. 50.0%), however, the difference between the two groups was not statistically significant (P > 0.05).

      3.7 Prognostic factors of patients after PSM in univariate and multivariable analysis

      Table 2 shows the results of univariate and multivariable Cox proportional hazard analysis of patient prognosis after PSM(n = 160). The univariate Cox regression analysis demonstrated that tumor location in antrum, distal gastrectomy, T4-stage, lymphovascular invasion, perineural invasion and the number of TD were all associated with prognosis in patients undergoing curative gastric resection (P < 0.05).The multivariable analysis identified that the number of TD[L1 group (HR, 3.518; P < 0.001),L2 group(HR, 6.404; P < 0.001)], perineural invasion (HR, 3.344; P < 0.001), lymphovascular invasion(HR, 2.360; P = 0.003) and T4 stage(HR, 1.813; P = 0.026) were independently associated with OS. Distal gastrectomy (HR, 0.329; P = 0.033) is an independent protective factor affecting prognosis.
      Table 2Univariate and Multivariate survival analyses of 160 patients with gastric cancer after matching.
      VariablesNumber (n = 160)5-OS(%)Univariate analysisMultivariate analysis
      HR (95%CI)P valueHR (95%CI)P value
      Age(years)0.946(0.549–1.627)0.840
      <603752.3%
      ≥6012353.6%
      Gender0.833(0.478–1.450)0.518
      Male12152.3%
      Female3957.0%
      Tumor size1.453(0.891–2.369)0.134
      ≤4 cm6161.7%
      >4 cm9948.2%
      Histologic grade0.816(0.440–1.516)0.521
      Differentiated12851.6%
      undifferentiated3261.3%
      Tumor location0.032
      Cardia fundus5437.7%RefRef
      Gastric body2759.0%0.712(0.359–1.408)0.3290.820(0.388–1.733)0.602
      Antrum7962.8%0.522(0.316–860)0.0111.044(0.422–2.581)0.926
      Depth of invasion ∗2.106(1.322–3.354)0.0021.813(1.075–3.057)0.026
      T1-311059.6%
      T45039.1%
      Neural invasion2.928(1.697–5.055)0.0003.344(1.839–6.080)0.000
      (−)6073.1%
      (+)10041.2%
      Vascular invasion2.330(1.401–3.874)0.0012.360(1.348–4.132)0.003
      (−)6566.4%
      (+)9544.3%
      Type of gastrectomy0.002
      Pro2927.7%RefRef
      Dis7166.2%0.358(0.199–0.644)0.0010.329(0.118–0.916)0.033
      Total6050.4%0.578(0.328–1.019)0.0580.580(0.303–1.112)0.101
      Chemotherapy0.758(0.478–1.202)0.239
      NO6345.1%
      YES9758.4%
      Number of TD0.000
      L0 group8069.4%RefRef
      L1 group5742.8%2.534(1.474–4.354)0.0013.518(1.994–6.205)0.000
      L2 group2321.7%4.948(2.678–9.142)0.0006.404(3.375–12.151)0.000
      ∗ according to the 8th edition of the AJCC-TNM staging system.
      Differentiated, papillary or well/moderately differentiated tubular adenocarcinoma.
      Undifferentiated, poorly differentiated,or mucinous adenocarcinoma or signet-ring cell carcinoma.

      4. Discussion

      TD was initially described by Dr. Gabriel in the rectum in 1935 [
      • Gabriel W.B.
      • Dukes C.
      • Bussey H.J.R.
      Lymphatic spread in cancer of the rectum.
      ]. Since then, several studies have confirmed that patients with TD in colorectal cancer have a poor prognosis [
      • Belt E.J.
      • van Stijn M.F.
      • Bril H.
      • de Lange-de Klerk E.S.
      • Meijer G.A.
      • Meijer S.
      • et al.
      Lymph node negative colorectal cancers with isolated tumor deposits should be classified and treated as stage III.
      ,
      • Tong L.L.
      • Gao P.
      • Wang Z.N.
      • Song Y.X.
      • Xu Y.Y.
      • Sun Z.
      • et al.
      Is the seventh edition of the UICC/AJCC TNM staging system reasonable for patients with tumor deposits in colorectal cancer?.
      ]. TD was formally included in the 7th edition of the AJCC on colorectal cancer N staging [
      • Edge S.B.
      • Compton C.C.
      The American Joint Committee on Cancer: the 7th edition of the AJCC cancer staging manual and the future of TNM.
      ]. It is frequently observed in GC and lately has become a non-negligible element in determining the prognosis of GC patients [
      • Wang W.
      • Li Y.
      • Zhang Y.
      • Yuan X.
      • Xu D.
      • Guan Y.
      • et al.
      Incorporation of extranodal metastasis of gastric carcinoma into the 7th edition UICC TNM staging system.
      ,
      • Lee H.S.
      • Lee H.E.
      • Yang H.K.
      • Kim W.H.
      Perigastric tumor deposits in primary gastric cancer: implications for patient prognosis and staging.
      ,
      • Sun Z.
      • Wang Z.N.
      • Xu Y.Y.
      • Zhu G.L.
      • Huang B.J.
      • Xu Y.
      • et al.
      Prognostic significance of tumor deposits in gastric cancer patients who underwent radical surgery.
      ].However, the issue of TD categorization in TNM staging and its number on the prognosis of gastric cancer has been controversial [
      • Chen H.
      • Tang Z.
      • Chen L.
      • Li H.
      • Wang X.
      • Liu F.
      • et al.
      Evaluation of the impact of tumor deposits on prognosis in gastric cancer and a proposal for their incorporation into the AJCC staging system.
      ,
      • Liang Y.
      • Wu L.
      • Liu L.
      • Ding X.
      • Wang X.
      • Liu H.
      • et al.
      Impact of extranodal tumor deposits on prognosis and N stage in gastric cancer.
      ,
      • Sun Z.
      • Wang Z.N.
      • Xu Y.Y.
      • Zhu G.L.
      • Huang B.J.
      • Xu Y.
      • et al.
      Prognostic significance of tumor deposits in gastric cancer patients who underwent radical surgery.
      ,
      • Anup S.
      • Lu J.
      • Zheng C.H.
      • Li P.
      • Xie J.W.
      • Wang J.B.
      • et al.
      Prognostic significance of perigastric tumor deposits in patients with primary gastric cancer.
      ].Therefore, patients with lymph node-negative GC were included in this study and the PSM method was used to overcome the selective bias and control for differences in baseline data. The aim of this study was to explore the prognostic value of TD status and number on lymph node-negative GC.
      In this study, TD-related factors include tumor size >4 cm, T3-4 stage, perineural invasion and lymphovascular invasion, all associated with tumor progression. In all T-stage subgroups, the prognosis of patients in the TD-positive group was worse than that in the TD-negative group, and the prognostic impact of TD-positive status on such patients was not limited by T-stage. As the number of TD increased, the 5-OS rate of patients decreased,and the prognosis worsened.
      A total of 1224 patients with lymph node negative GC were included in this study, and the incidence of TD was 7.4%.A recent report showed that the incidence of TD in GC was 10.6%–36.7% [
      • Graham Martínez C.
      • Knijn N.
      • Verheij M.
      • Nagtegaal I.D.
      • van der Post R.S.
      Tumour deposits are a significant prognostic factor in gastric cancer - a systematic review and meta-analysis.
      ]. The low incidence of TD in our study, compared with previous studies, may be attributed to the inclusion of lymph node-negative GC. TD may be associated with lymph node metastasis, which has been confirmed in other studies [
      • Etoh T.
      • Sasako M.
      • Ishikawa K.
      • Katai H.
      • Sano T.
      • Shimoda T.
      Extranodal metastasis is an indicator of poor prognosis in patients with gastric carcinoma.
      ,
      • Lee H.S.
      • Lee H.E.
      • Yang H.K.
      • Kim W.H.
      Perigastric tumor deposits in primary gastric cancer: implications for patient prognosis and staging.
      ,
      • Graham Martínez C.
      • Knijn N.
      • Verheij M.
      • Nagtegaal I.D.
      • van der Post R.S.
      Tumour deposits are a significant prognostic factor in gastric cancer - a systematic review and meta-analysis.
      ].Currently, the mechanism of TD formation is unclear. The mainstream view is that TD originates from the lymph node pathway, where the metastatic lymph node expands outside the capsule and the normal lymph node tissue is completely replaced by tumor cells. Secondly, TD is an intermediate condition of tumor metastasis. In the lymphatic network, tumor cells grow and form TD [
      • Balch C.M.
      • Buzaid A.C.
      • Soong S.J.
      • Atkins M.B.
      • Cascinelli N.
      • Coit D.G.
      • et al.
      Final version of the American Joint Committee on Cancer staging system for cutaneous melanoma.
      ]. However, this view cannot explain the phenomenon that GC has TD without lymph node metastasis.Others have proposed that TD may be a result of cancer cell dissemination along blood vessels.Alternatively, the primary lesion may invade the plasma membrane and the tumor cells detach in the perifocal area and implant metastasis [
      • Ueno H.
      • Mochizuki H.
      • Tamakuma S.
      Prognostic significance of extranodal microscopic foci discontinuous with primary lesion in rectal cancer.
      ].
      In the present study, pre-match data showed that TD-positive patients had a higher rate of perineural invasion and lymphovascular invasion compared to TD-negative patients, which is consistent with previous findings [
      • Chen H.
      • Tang Z.
      • Chen L.
      • Li H.
      • Wang X.
      • Liu F.
      • et al.
      Evaluation of the impact of tumor deposits on prognosis in gastric cancer and a proposal for their incorporation into the AJCC staging system.
      ,
      • Sun Z.
      • Wang Z.N.
      • Xu Y.Y.
      • Zhu G.L.
      • Huang B.J.
      • Xu Y.
      • et al.
      Prognostic significance of tumor deposits in gastric cancer patients who underwent radical surgery.
      ].We have identified four potential factors associated with TD, including tumor size>4 cm, T3-4 stage, lymphovascular invasion, and perineural invasion; all four factors are involved in tumor progression. Although TD is clearly associated with T3-4 stage, it may also be present in T1-2 stage. Sun et al. reported that TD occurred in 4.3% (21/491) of pT1 gastric cancer and 10%(53/529) of pT2 gastric cancer [
      • Sun Z.
      • Wang Z.N.
      • Xu Y.Y.
      • Zhu G.L.
      • Huang B.J.
      • Xu Y.
      • et al.
      Prognostic significance of tumor deposits in gastric cancer patients who underwent radical surgery.
      ].In our study, the incidence of TD occurring at pT1stage was 0.6% (3/512)and 6.9%(9/131) at pT2stage, which is evidently lower than prior results. Therefore, for T1-2 stage GC, en bloc resection of perigastric lymphatic adipose tissue should be performed to avoid TD leakage.
      The present study showed that the 5-year OS rate was significantly lower in the TD-positive group than that in the TD-negative group (36.4% vs. 69.4%), which is similar to results from previous studies [
      • Liang Y.
      • Wu L.
      • Liu L.
      • Ding X.
      • Wang X.
      • Liu H.
      • et al.
      Impact of extranodal tumor deposits on prognosis and N stage in gastric cancer.
      ,
      • Lee H.S.
      • Lee H.E.
      • Yang H.K.
      • Kim W.H.
      Perigastric tumor deposits in primary gastric cancer: implications for patient prognosis and staging.
      ,
      • Anup S.
      • Lu J.
      • Zheng C.H.
      • Li P.
      • Xie J.W.
      • Wang J.B.
      • et al.
      Prognostic significance of perigastric tumor deposits in patients with primary gastric cancer.
      ,
      • Graham Martínez C.
      • Knijn N.
      • Verheij M.
      • Nagtegaal I.D.
      • van der Post R.S.
      Tumour deposits are a significant prognostic factor in gastric cancer - a systematic review and meta-analysis.
      ]. The prognosis of patients in the TD-positive group was worse than that of patients in the TD-negative group in each T1-2,T3,and T4 subgroup. This result suggests that in lymph node-negative GC, regardless of T-stage, TD-positive patients have a poor prognosis and that the prognostic impact of TD on this group of patients is not limited by T-stage. We additionally found that the prognosis of patients in the T1-2 stage TD-positive group was similar to that of patients in the T3 stage TD-negative group and that the prognosis of patients in the T3 stage TD-positive group was similar to that of patients in the T4 stage TD-negative group. Although the above results suggest that TD may upgrade the T stage in lymph node-negative GC, they differ from the study conducted by Sun et al. [
      • Sun Z.
      • Wang Z.N.
      • Xu Y.Y.
      • Zhu G.L.
      • Huang B.J.
      • Xu Y.
      • et al.
      Prognostic significance of tumor deposits in gastric cancer patients who underwent radical surgery.
      ], in which there were 1134 patients with GC in the N0 subgroup and 90 TD-positive patients. In the T1,T2,and T3 subgroups, TD-positive patients had a worse prognosis than TD-negative patients. However, in the T4a,T4b subgroups, TD-positive patients had a similar prognosis to TD-negative patients. It was considered that TD should be regarded as T4a and included in T staging. Based on our results, TD may not be classified as T4a. Therefore, we cannot ignore the prognostic impact of TD in GC patients without lymph node metastasis.
      The prognostic value of TD number for GC patients was controversial. Anup et al. reported that 1250 patients with potentially curative GC surgery were grouped according to the number of TD (1, 2, ≥3), and there was no significant association with the number of TD and patient prognosis [
      • Anup S.
      • Lu J.
      • Zheng C.H.
      • Li P.
      • Xie J.W.
      • Wang J.B.
      • et al.
      Prognostic significance of perigastric tumor deposits in patients with primary gastric cancer.
      ].However, in three other studies with various sample sizes and subtle variations in TD number classification, a difference in prognosis amongst the TD groups was observed, along with a decrease in 5-year OS rate when the number of TD increased [
      • Wang W.
      • Li Y.
      • Zhang Y.
      • Yuan X.
      • Xu D.
      • Guan Y.
      • et al.
      Incorporation of extranodal metastasis of gastric carcinoma into the 7th edition UICC TNM staging system.
      ,
      • Etoh T.
      • Sasako M.
      • Ishikawa K.
      • Katai H.
      • Sano T.
      • Shimoda T.
      Extranodal metastasis is an indicator of poor prognosis in patients with gastric carcinoma.
      ,
      • Mayo E.
      • Llanos A.A.
      • Yi X.
      • Duan S.Z.
      • Zhang L.
      Prognostic value of tumour deposit and perineural invasion status in colorectal cancer patients: a SEER-based population study.
      ]. Our data, including 160 patients with a TD number classification of 0, 1–2, and ≥3, strongly agreed with the results of the above three studies. The prognostic significance of the number of TD was confirmed by multivariate Cox regression analysis. The number of TD has a negative impact on the prognosis of GC patients without lymph node metastasis. Therefore, clinicians cannot ignore TD and its number in the prognosis assessment and treatment strategy selection of lymph node-negative GC patients.We believe that the possible explanations for the discrepancies across the TD number impact studies are as follows: 1. possible differences in tumor stage among the included study subjects; and 2. inconsistent criteria for grouping according to the number of TD. We also believe that the risk of TD and its number is greater in early-stage GC but less so in advanced GC, which needs to be further confirmed by follow-up studies. In colorectal cancer, Mayo et al. showed that TD had a greater prognostic impact on patients with low N stage [
      • Mayo E.
      • Llanos A.A.
      • Yi X.
      • Duan S.Z.
      • Zhang L.
      Prognostic value of tumour deposit and perineural invasion status in colorectal cancer patients: a SEER-based population study.
      ]. In GC, Eoth et al. showed that in the N0,N1,N2 subgroups, TD-positive patients have a worse prognosis than TD-negative patients, whereas in the N3 subgroup, TD-positive patients have a similar prognosis to TD-negative patients [
      • Etoh T.
      • Sasako M.
      • Ishikawa K.
      • Katai H.
      • Sano T.
      • Shimoda T.
      Extranodal metastasis is an indicator of poor prognosis in patients with gastric carcinoma.
      ].
      At present, according to the 5th edition of the Japanese Gastric Cancer Treatment Guidelines, adjuvant chemotherapy is not required after radical surgery for T1-2 lymph node negative GC. There is no consensus on whether adjuvant chemotherapy is needed after surgery for these patients if they contain TD.Lee et al. found that TD is highly associated with early GC prognosis and therefore suggested that postoperative adjuvant chemotherapy be considered for early GC combined with TD [
      • Lee I.S.
      • Park Y.S.
      • Ryu M.H.
      • Song M.J.
      • Yook J.H.
      • Oh S.T.
      • et al.
      Impact of extranodal extension on prognosis in lymph node-positive gastric cancer.
      ]. A retrospective study from China showed that patients with lymph node-negative GC combined with TD in the T1-2 subgroup had a significantly better prognosis with postoperative chemotherapy than those without chemotherapy (63.6%vs.16.7%). It suggests that for these patients, postoperative adjuvant chemotherapy may be beneficial and can prolong survival time [
      • Zhi C.
      • Yang W.
      • Li N.
      • Zhang Z.
      • Hua Y.
      • Liu H.
      Prognostic value of the tumor deposit in N0 gastric cancer by propensity score matching analysis.
      ]. In this study, there were 11 TD-positive patients in the T1-2 staging subgroup, and we did not identify any statistical difference in prognosis between chemotherapy and non-chemotherapy groups(53.3% vs. 50.0%). Clinicians should pay more attention to TD, even in patients with T1-2 lymph node negative GC, and postoperative adjuvant chemotherapy and close follow up should be considered. Nonetheless, this conclusion needs to be further substantiated by follow-up studies.
      Our study has some limitations. First, it was a single-center small sample retrospective study. Second, the issue of TD belonging in TNM staging was not explored.

      5. Conclusion

      In lymph node-negative GC, the presence of TD suggests a poor prognosis, which is not influenced by T stage. The number of TD is an independent risk factor for the prognosis of patients with lymph node-negative GC. Clinicians should pay more attention to this risk factor to benefit more patients through early multidisciplinary intervention.

      Funding

      This study is supported by The Natural Science Foundation of Anhui Province (grant to.2008085MH294).

      Ethics statement

      The study was examined and certified by the Ethics Committee of the Yijishan Hospital of Wannan Medical College. All patients gave informed consent prior to gastroscopy, surgery, or chemotherapy, and every procedure was performed according to the rules of clinical practice. This study complied with the standards of the Declaration of Helsinki and current ethical guidelines.

      Data availability statement

      All the data used to support the findings of this study are included in the article.

      CRediT authorship contribution statement

      Ran Xu: Conceptualization, Methodology, Formal analysis, Data curation, Resources, Writing – original draft, Writing – review & editing, Supervision, Project administration, Funding acquisition. Yisheng Zhang: Investigation, Validation, Formal analysis, Data curation, Writing – original draft. Jun Zhao: Formal analysis, Investigation, Data curation. Ke Chen: Resources, Writing – review & editing, Validation, Funding acquisition. Zhengguang Wang: Resources, Writing – review & editing, Supervision, Project administration, Funding acquisition.

      Declaration of competing interest

      The authors declare that they have no competing interests.

      References

        • Sung H.
        • Ferlay J.
        • Siegel R.L.
        • Laversanne M.
        • Soerjomataram I.
        • Jemal A.
        • et al.
        Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality.
        CA A Cancer J Clin. 2021; 71: 209-249https://doi.org/10.3322/caac.21660
        • Cats A.
        • Jansen E.P.M.
        • van Grieken N.C.T.
        • Sikorska K.
        • Lind P.
        • Nordsmark M.
        • et al.
        Chemotherapy versus chemoradiotherapy after surgery and preoperative chemotherapy for resectable gastric cancer (CRITICS): an international, open-label, randomised phase 3 trial.
        Lancet Oncol. 2018; 19: 616-628https://doi.org/10.1016/S1470-2045(18)30132-3
        • Japanese Gastric Cancer Association
        Japanese gastric cancer treatment guidelines 2018.
        Gastric Cancer. 2021; 24 (5th edition): 1-21https://doi.org/10.1007/s10120-020-01042-y
        • Hoda
        • Syed A.
        AJCC cancer staging manua.
        Adv Anat Pathol. 2017; 24 (8th edition)
        • Ajani J.A.
        • D'Amico T.A.
        • Bentrem D.J.
        • Chao J.
        • Cooke D.
        • Corvera C.
        • et al.
        Gastric cancer, version 2.2022, NCCN clinical practice guidelines in oncology.
        J Natl Compr Cancer Netw. 2022; 20: 167-192https://doi.org/10.6004/jnccn.2022.0008
        • Pernot S.
        • Voron T.
        • Perkins G.
        • Lagorce-Pages C.
        • Berger A.
        • Taieb J.
        • et al.
        Signet-ring cell carcinoma of the stomach: impact on prognosis and specific therapeutic challenge.
        World J Gastroenterol. 2015; 21: 11428-11438https://doi.org/10.3748/wjg.v21.i40.11428
        • Chang K.
        • Song B.
        • Do I.G.
        • Koo D.H.
        • Lee H.W.
        • Son B.H.
        • et al.
        Venous invasion and perineural invasion as upstaging and poor prognostic factors in N0 gastric cancers.
        Anticancer Res. 2021; 41: 5803-5810https://doi.org/10.21873/anticanres.15397
        • Chen H.
        • Tang Z.
        • Chen L.
        • Li H.
        • Wang X.
        • Liu F.
        • et al.
        Evaluation of the impact of tumor deposits on prognosis in gastric cancer and a proposal for their incorporation into the AJCC staging system.
        Eur J Surg Oncol. 2018; 44: 1990-1996https://doi.org/10.1016/j.ejso.2018.10.062
        • Wang W.
        • Li Y.
        • Zhang Y.
        • Yuan X.
        • Xu D.
        • Guan Y.
        • et al.
        Incorporation of extranodal metastasis of gastric carcinoma into the 7th edition UICC TNM staging system.
        PLoS One. 2011; 6e19557https://doi.org/10.1371/journal.pone.0019557
        • Edge S.B.
        • Compton C.C.
        The American Joint Committee on Cancer: the 7th edition of the AJCC cancer staging manual and the future of TNM.
        Ann Surg Oncol. 2010; 17: 1471-1474https://doi.org/10.1245/s10434-010-0985-4
        • Etoh T.
        • Sasako M.
        • Ishikawa K.
        • Katai H.
        • Sano T.
        • Shimoda T.
        Extranodal metastasis is an indicator of poor prognosis in patients with gastric carcinoma.
        Br J Surg. 2006; 93: 369-373https://doi.org/10.1002/bjs.5240
        • Liang Y.
        • Wu L.
        • Liu L.
        • Ding X.
        • Wang X.
        • Liu H.
        • et al.
        Impact of extranodal tumor deposits on prognosis and N stage in gastric cancer.
        Surgery. 2019; 166: 305-313https://doi.org/10.1016/j.surg.2019.04.027
        • Lee H.S.
        • Lee H.E.
        • Yang H.K.
        • Kim W.H.
        Perigastric tumor deposits in primary gastric cancer: implications for patient prognosis and staging.
        Ann Surg Oncol. 2013; 20: 1604-1613https://doi.org/10.1245/s10434-012-2692-9
        • Tan J.
        • Yang B.
        • Xu Z.
        • Zhou S.
        • Chen Z.
        • Huang J.
        • et al.
        Tumor deposit indicates worse prognosis than metastatic lymph node in gastric cancer: a propensity score matching study.
        Ann Transl Med. 2019; 7: 671https://doi.org/10.21037/atm.2019.10.33
        • Sun Z.
        • Wang Z.N.
        • Xu Y.Y.
        • Zhu G.L.
        • Huang B.J.
        • Xu Y.
        • et al.
        Prognostic significance of tumor deposits in gastric cancer patients who underwent radical surgery.
        Surgery. 2012; 151: 871-881https://doi.org/10.1016/j.surg.2011.12.027
        • Anup S.
        • Lu J.
        • Zheng C.H.
        • Li P.
        • Xie J.W.
        • Wang J.B.
        • et al.
        Prognostic significance of perigastric tumor deposits in patients with primary gastric cancer.
        BMC Surg. 2017; 17 (19): 84https://doi.org/10.1186/s12893-017-0280-4
        • Yao X.I.
        • Wang X.
        • Speicher P.J.
        • Hwang E.S.
        • Cheng P.
        • Harpole D.H.
        • et al.
        Reporting and guidelines in propensity score analysis: a systematic review of cancer and cancer surgical studies.
        J Natl Cancer Inst. 2017; 109 (1)https://doi.org/10.1093/jnci/djw323
        • Gabriel W.B.
        • Dukes C.
        • Bussey H.J.R.
        Lymphatic spread in cancer of the rectum.
        Br J Surg. 2010; 23: 395-413https://doi.org/10.1002/bjs.1800239017
        • Belt E.J.
        • van Stijn M.F.
        • Bril H.
        • de Lange-de Klerk E.S.
        • Meijer G.A.
        • Meijer S.
        • et al.
        Lymph node negative colorectal cancers with isolated tumor deposits should be classified and treated as stage III.
        Ann Surg Oncol. 2010; 17: 3203-3211https://doi.org/10.1245/s10434-010-1152-7
        • Tong L.L.
        • Gao P.
        • Wang Z.N.
        • Song Y.X.
        • Xu Y.Y.
        • Sun Z.
        • et al.
        Is the seventh edition of the UICC/AJCC TNM staging system reasonable for patients with tumor deposits in colorectal cancer?.
        Ann Surg. 2012; 255: 208-213https://doi.org/10.1097/SLA.0b013e31821ad8a2
        • Graham Martínez C.
        • Knijn N.
        • Verheij M.
        • Nagtegaal I.D.
        • van der Post R.S.
        Tumour deposits are a significant prognostic factor in gastric cancer - a systematic review and meta-analysis.
        Histopathology. 2019; 74: 809-816https://doi.org/10.1111/his.13781
        • Balch C.M.
        • Buzaid A.C.
        • Soong S.J.
        • Atkins M.B.
        • Cascinelli N.
        • Coit D.G.
        • et al.
        Final version of the American Joint Committee on Cancer staging system for cutaneous melanoma.
        J Clin Oncol. 2001; 19 (15): 3635-3648https://doi.org/10.1200/JCO.2001.19.16.3635
        • Ueno H.
        • Mochizuki H.
        • Tamakuma S.
        Prognostic significance of extranodal microscopic foci discontinuous with primary lesion in rectal cancer.
        Dis Colon Rectum. 1998; 41: 55-61https://doi.org/10.1007/BF02236896
        • Mayo E.
        • Llanos A.A.
        • Yi X.
        • Duan S.Z.
        • Zhang L.
        Prognostic value of tumour deposit and perineural invasion status in colorectal cancer patients: a SEER-based population study.
        Histopathology. 2016; 69: 230-238https://doi.org/10.1111/his.12936
        • Lee I.S.
        • Park Y.S.
        • Ryu M.H.
        • Song M.J.
        • Yook J.H.
        • Oh S.T.
        • et al.
        Impact of extranodal extension on prognosis in lymph node-positive gastric cancer.
        Br J Surg. 2014; 101: 1576-1584https://doi.org/10.1002/bjs.9640
        • Zhi C.
        • Yang W.
        • Li N.
        • Zhang Z.
        • Hua Y.
        • Liu H.
        Prognostic value of the tumor deposit in N0 gastric cancer by propensity score matching analysis.
        Zhonghua Wei Chang Wai Ke Za Zhi. 2019 Feb 25; 22: 172-179https://doi.org/10.3760/cma.j.issn.1671-0274