| Objective:This study mainly investigated the prognostic value of hemoglobin,albumin,lymphocyte and platelet score(HALP),platelet to lymphocyte ratio(PLR),lymphocyte to monocyte ratio(LMR)on overall survival(OS)in patients with non-small cell lung cancer(NSCLC)before initial treatment and to establish a prediction model.Methods:In this study,432 patients diagnosed with NSCLC in Qingdao Municipal Hospital between January 1,2015 and December 31,2018 were retrospectively analyzed.Clinical,pathological and hematological data of enrolled patients were collected,and survival data of patients were obtained through follow-up.The values of HALP,PLR and LMR were calculated according to the formula,and X-tile software was used to determine the optimal cut-off values of HALP,PLR and LMR.Patients were divided into low HALP group,high HALP group,low PLR group,high PLR group,low LMR group and high LMR group according to the optimal cut-off values.X~2 test was used to compare clinical features among HALP,PLR,and LMR groups.Kaplan-Meier method was used to draw the survival curve,and Log-Rank test was used to compare the survival of NSCLC patients in different HALP,PLR and LMR groups.At the same time,subgroup survival analysis was conducted according to TNM stage and pathological type.Receiver operating characteristic(ROC)curves was plotted to compare the predictive value of HALP,PLR and LMR for OS.Univariate and multivariate Cox regression analyses were used to investigate the independent prognostic factors of NSCLC.Finally,a nomogram was established to predict the survival rate of NSCLC patients,and the predictive efficacy of the nomogram was verified by drawing calibration curves,ROC curves and calculating the consistency index(C index).Results:The optimal cut-off values of HALP,PLR and LMR were 42.2,129.6 and 3.9,respectively.Kaplan-Meier survival analysis showed that patients in the low HALP,high PLR and low LMR groups had shorter OS than those in the high HALP,low PLR and high LMR groups,respectively(P<0.001).Subgroup analysis based on TNM staging showed that low HALP was associated with poor OS in stage Ⅰ A～ Ⅱ B patients(P=0.001),while PLR and LMR levels were not significantly associated with OS(P=0.060,P=0.068).Low HALP,high PLR,and low LMR were associated with poor OS in stage ⅢA～ⅣB patients(P<0.001).Survival analysis based on pathological type showed that low HALP,high PLR,and low LMR were associated with poor OS in both adenocarcinoma patients and non-adenocarcinoma patients(P < 0.05).ROC curves showed that the area under the curve(AUC)of HALP,PLR and LMR were 0.729(95%CI:0.683 ～ 0.776),0.694(95%CI:0.645 ～ 0.743),0.656(95%CI:0.604 ～ 0.707),respectively.The AUC of HALP was statistically significant compared with that of PLR and LMR(P=0.002,P=0.005).Multivariate Cox regression analysis showed that HALP,age,gender and TNM stage were independent prognostic factors for OS in NSCLC patients(P < 0.05),while PLR(P=0.523)and LMR(P=0.423)were not independent prognostic factors.The C index of the nomogram based on HALP was 0.792(95%CI:0.766～0.820),the AUC of 1-year,3-year and 5-year ROC curves were 0.855,0.855 and0.854,respectively.And the calibration curve showed that the predicted survival rates were in good agreement with the actual survival rates.Conclusion:1.For NSCLC patients with stageⅠA～ⅡB,HALP level was correlated with OS,while PLR and LMR level were not significantly correlated with OS.HALP,PLR and LMR levels were all correlated with OS in stage ⅢA～ⅣB patients.Lower HALP level,higher PLR level and lower LMR level all suggested poor prognosis in NSCLC patients.2.HALP,PLR and LMR all have certain clinical value for the prognosis of NSCLC patients,but among the three,HALP has the best prognostic value.3.HALP,gender,age,and TNM stage were independent prognostic factors for OS in NSCLC patients,while PLR and LMR were not.4.The prediction model based on HALP has good calibration and differentiation,which can provide reference for prognosis assessment of NSCLC patients. |
PDF Full Text Request