| Objective:To analyze the relationship between preoperative prognostic nutritional index(PNI),lymphocyte to monocyte ratio(LMR)and clinicopathological features of gastric cancer,and to further explore the preoperative prediction and prognostic value of PNI and LMR for gastric cancer.In the first part,we focused on the retrospective analysis of the collected data of 220 patients with gastric cancer in the northwest region,and assessed the correlation between PNI,LMR and clinicopathological features,and studied the predictive value of these two indicators on the invasion depth,lymph node metastasis and lymphovascular infiltration of gastric cancer and their effects on prognosis.In the second part,we used Meta-analysis to analyze the published data about PNI,LMR and prognosis,to explore the relationship between PNI,LMR and clinicopathological features and the prognostic value of gastric cancer patients through large sample data.Methods:In the first part,we retrospectively collected the clinical data of 220patients with gastric cancer who underwent surgical treatment in Lanzhou University Second Hospital from December 2013 to February 2015.We used T test,ANOVA test and nonparametric test to analyze the relationship between PNI,LMR and clinicopathological characteristics.Logistic regression was used to analyze the influencing factor of the invasion depth,lymph node metastasis and lymphovascular infiltration of gastric cancer.Receiver operating characteristic curve(ROC)was constructed to analyze the prediction value of PNI and LMR on the invasion depth,lymph node metastasis and lymphovascular infiltration of gastric cancer.The factors affecting prognosis of gastric cancer patients were investigated by Kaplan-Meier methods,and the COX proportional risk model was used for multivariate analysis.In the second part,we searched the articles which were published in Web of Science,Embase and PubMed databases until March 14th,2020.The articles that meet the conditions of our study were screened out according to the proposed exclusion criteria.Meta-analysis was used to explore the relationship between PNI,LMR and clinicopathological characteristics and the impact of two indicators on prognosis.STATA 14.0 was used to data analysis.Heterogeneity was evaluated by Q test and I~2test,whether there was publication bias was evaluated by Begg’s test,and the robustness of the included studies was evaluated by sensitivity analysis.Results:1.Preoperative PNI was correlated with patient’s gender,age,tumor maximum diameter,invasion depth,lymph node metastasis,TNM stage,lymphovascular infiltration and CA125 expression level(P<0.05).Female,age(≥55years),tumor maximum diameter(≥4.5cm),invasion depth(T3-T4),lymph node metastasis,TNM stage(stage III-IV),lymphovascular infiltration and CA125expression level(≥35U/mL)were significantly connected with reduced PNI.Preoperative LMR was correlated with patient’s tumor maximum diameter,lymph node metastasis,distant metastasis,TNM stage,lymphovascular infiltration,CEA expression level,CA125 expression level(P<0.05).Tumor maximum diameter(≥4.5cm),lymph node metastasis,distant metastasis,TNM stage(stage III-IV),lymphovascular infiltration,CEA expression level(≥3.4ng/mL)and CA125expression level(≥35U/mL)were significantly connected with reduced LMR.2.The PNI and LMR in the advanced gastric cancer group were lower than those in the early gastric cancer group(P<0.05).Logistic regression analysis showed that LMR is an influencing factor of invasion depth(T2-T4)in gastric cancer.The area under the curve where PNI predicts invasion depth(T2-T4)was 0.698(P=0.001),the specificity was 69.0%,and the sensitivity was 66.5%.The area under the curve where LMR predicts invasion depth(T2-T4)was 0.661(P=0.005),the specificity was 75.9%,and the sensitivity was 52.4%.The sensitivity is the highest when the two indicators are combined in parallel,which is 84.1%.The specificity is the highest when the two indicators are combined in series,which is 92.5%.3.The PNI and LMR in the lymph node metastasis group are lower than those in the no lymph node metastasis group(P<0.05).Logistic regression analysis shows that PNI is an influencing factor of lymph node metastasis in gastric cancer.The area under the curve where PNI predicts lymph node metastasis is 0.693(P<0.001),the specificity was 73.5%,and the sensitivity was 61.8%.The area under the curve where LMR predicts lymph node metastasis is 0.620(P=0.004),the specificity was 57.4%,and the sensitivity was 67.1%.The sensitivity is the highest when the two indicators are combined in parallel,which is 87.4%.The specificity is the highest when the two indicators are combined in series,which is 88.7%.4.The PNI and LMR in the lymphovascular infiltration group are lower than those in the no lymphovascular infiltration(P<0.05).Logistic regression analysis shows that PNI is an influencing factor of lymphovascular infiltration in gastric cancer.The area under the curve where PNI predicts lymphovascular infiltration is 0.630(P=0.005),the specificity is 61.5%,and the sensitivity is 60.7%;The area under the curve where LMR predicts lymphovascular infiltration is 0.630(P=0.005),the specificity is 90.4%,and the sensitivity is 33.3%.The sensitivity is the highest when the two indicators are combined in parallel,which is 73.8%.The specificity is the highest when the two indicators are combined in series,which is 96.3%.5.Univariate analysis showed that PNI and LMR were correlated with overall survival(OS)(P<0.05).The results revealed that PNI(<46.825)and LMR(<5.36)were significantly connected with reduced OS.The joint analysis showed that patients with both low PNI and low LMR had lowest OS rate than others(P<0.001).In the multivariate COX regression model,PNI[Hazard Ratio(HR)=0.441;95%confidence interval(CI):0.233-0.837;P=0.012]and LMR(HR=0.445;95%CI:0.224-0.884;P=0.021)were independent prognostic factors for gastric cancer patients.The 5-year survival rates of the PNI<46.825 group and the PNI≥46.825group were 70.0%and 90.0%,respectively.The 5-year survival rates of the LMR<5.36 group and the LMR≥5.36 group were 73.9%and 91.7%,respectively.6.A total of 20,756 patients from 24 studies and 11,471 patients from 11 studies are included to explore the correlation between PNI,LMR and clinicopathological features and prognosis of gastric cancer.The results showed that low PNI value is correlated with invasion depth(T2-T4)[Odds Ratio(OR)=2.40,95%CI:1.63-3.54;P<0.001],lymph node metastasis(OR=1.59,95%CI:1.16-2.18;P=0.004),TNM stage(stage III-IV)(OR=2.13,95%CI:1.69-2.68;P<0.001),vascular invasion(OR=3.22,95%CI:2.10-4.96;P<0.001),lymphatic invasion(OR=2.16,95%CI:1.54-3.03;P<0.001)and poor prognosis[OS(pooled HR=1.74,95%CI:1.50-2.03;P<0.001)and disease free survival(DFS)/relapse free survival(RFS)(pooled HR=2.06,95%CI:1.33-3.19;P<0.001)].Low LMR value is correlated with invasion depth(T2-T4)(OR=2.16,95%CI:1.80-2.60;P<0.001),lymph node metastasis(OR=1.89,95%CI:1.36-2.64;P<0.001),TNM stage(stage III-IV)(OR=1.77,95%CI:1.28-2.44;P=0.001),vascular invasion(OR=1.69,95%CI:1.10-2.58;P=0.016)and poor prognosis[OS(pooled HR=1.44,95%CI:1.26-1.66;P<0.001)and DFS/RFS(pooled HR=1.58,95%CI:1.05-2.38;P=0.027)].Conclusions:1.The reduction of PNI and LMR is related to the progress of gastric cancer.2.Low LMR is an influencing factor for the invasion depth of gastric cancer(T2-T4),and low PNI is an influencing factor for lymph node metastasis and lymphovascular infiltration of gastric cancer.3.PNI and LMR combined detection can improve the predictive value of gastric cancer invasion depth,lymph node metastasis and lymphovascular infiltration.4.Low PNI and low LMR are independent predictors of poor prognosis in patients with gastric cancer.5.Patients with both low PNI and low LMR had lowest OS. |