Development And Assessment Of A Novel Nomogram Based On NLR And PLR For Predicting Tumor Response Rate And Survival Rate Of Neoadjuvant Chemotherapy For Locally Advanced Gastric Cancer

Objective: Locally advanced gastric cancer(LAGC)remains high incidence and mortality worldwide and is still one of the top tumor-related burden for global health system.This study aimed to investigate the correlation between systemic inflammatory indicators(neutrophil-lymphocyte ratio(NLR),platelet-lymphocyte ratio(PLR))and short and long-term efficacy in patients with locally advanced gastric cancer after neoadjuvant chemotherapy(NACT).Two novel nomograms were developed and assessed as clinical prediction model to explore the significance of comprehensive treatment options for locally advanced gastric cancer.Methods: The clinical data of 141 patients with locally advanced gastric cancer who received surgery and postoperative chemotherapy after neoadjuvant chemotherapy in Lanzhou University Second Hospital from January 2013 to December 2017 were collected,including complete blood count before neoadjuvant chemotherapy,gender,age,neoadjuvant chemotherapy regimen,Eastern Cooperative Oncology Group(ECOG)physical score,Borrmann classification,Lauren classification,tumor location,clinical TNM stage before neoadjuvant chemotherapy(cTNM),TNM stage after neoadjuvant chemotherapy(ypTNM),surgical methods and survival time.15 patients whose follow-up date were incomplete,nonspecific survival time and died due to other reasons were screened out.Finally,a total of 126 patients were included in the study.All patients received neoadjuvant chemotherapy and underwent surgery and postoperative chemotherapy after evaluating the efficacy of chemotherapy.The optimal cutoff values of NLR and PLR were calculated using X-tile software;the independent predicting factors of overall response rate(ORR)of neoadjuvant chemotherapy were analyzed using Lasso-logistic regression model,the independent risk factors for overall survival(OS)were analyzed using Cox proportional hazards regression model,the survival analysis was performed using Kaplan-Meier method and log-rank test;the nomograms of overall response rate(ORR)and 3-year and 5-year survival rates of neoadjuvant chemotherapy were developed using R software,the Bootstrap concordance prediction model was used for internal validation,and the index of concordance(C-index),receiver operating characteristic curve(ROC),calibration plot and decision curve analysis(DCA)were used to assess the discrimination,calibration and clinical practicability of the prediction model.Results:1.The optimal cutoff values of NLR and PLR were 3.7 and 185.1,respectively.2.Among the factors influencing the overall response rate of neoadjuvant chemotherapy analyzed by Lasso-logistic regression model,age,NLR,PLR,cN,Borrmann classification and tumor location were independent predicting factors for overall response rate,respectively(P < 0.05).A nomogram prediction model was developed according to the above indicators.The index of concordance(C-index)was 0.823(95% CI: 0.638-0.832),the internal validation model C-index was 0.792(95% CI: 0.579-0.812),the area under the receiver operating characteristic curve(AUC)was 0.802,the calibration plot showed good conformity,and the decision curve(DCA)showed that the net benefit rate of patients ranged from 19% to 84%.The model had moderate discrimination,calibration and clinical practicability;3.Cox regression showed the independent risk factors affecting the overall survival as follows: age,NLR,PLR,neoadjuvant chemotherapy regimen,ECOG performance score and ypN(P < 0.05).The nomogram prediction model was developed according to the above indicators,the concordance index(C-index)of the model was 0.868(95% CI: 0.587 – 0.794),the calibration plot of 3-year and 5-year survival rates showed good concordance,the C-index was 0.836(95% CI: 0.626 – 0.742)in internal validation,the decision curve showed that the net benefit rate range was < 87%,which all indicating that the model had high discrimination,calibration,and clinical utility.Conclusions:1.High NLR(> 3.7)and high PLR(> 185.1)before neoadjuvant chemotherapy suggest low overall response rate and low 3-year and 5-year survival rates;2.Age,NLR,PLR,cN,Borrmann classification and tumor location have correlation with the overall response rate of neoadjuvant chemotherapy;3.Age,NLR,PLR,neoadjuvant chemotherapy regimen,ECOG performance score and ypN are independent risk factors for overall survival;4.The prediction model presented by nomograms in this study have high discrimination,calibration and clinical practicability.