Multivariate Prognostic Analysis And Prognostic Model Establishment Of Small Cell Lung Cancer

Background and objective: Small cell lung cancer(SCLC)is a fast multiplying,early metastatic malignant tumor with short overall survival(OS).The purpose of this study was to determine the prognostic value of peripheral blood inflammation and immune indicators on SCLC by analyzing the factors affecting the prognosis of SCLC,and to establish a scoring system of SCLC based on systemic inflammatory indicators.Methods: The clinical data of 438 patients with SCLC diagnosed and treated in the First Hospital of Jilin University between January 1st,2012 and December 31,2016 were retrospectively analyzed.All the enrolled patients were pathologically diagnosed with SCLC and had received first-line chemotherapy.Records in patients with clinical features include age,sex,smoking index,KPS score,with or without pleural effusion or superior vena cava syndrome,peripheral blood cell count,albumin level and tumor markers test indicators before treatment,therapeutic options(chemotherapy alone/radiation and chemotherapy combined),cycle of first-line chemotherapy;Patients were followed up and progression free survival(PFS)and OS were recorded monthly.SPSS 24.0 software was used for statistical calculation,ROC curve analysis was used to determine the cut-off value of continuous variable,and Kaplan-Meier method was used to conduct single-factor survival analysis and draw the survival curve to determine the factors influencing the prognosis.Multivariate analysis was performed by COX regression,and independent prognostic factors affecting PFS or OS were determined.P<0.05 was considered statistically significant.R3.6.0 software was used to establish the predictive nomogram model for the selected variables,so as to make the results of the predictive model more readable and facilitate the evaluation of patients.The Harrell concordance index(c-index)is used to evaluate the predictive value of the nomogram model.Research results: 1.The pre-treatment NLR,CAR,PLR and LMR were calculated by blood routine,serum CRP and serum albumin,respectively,and the ROC curve was drawn.The critical point for predicting the prognosis of SCLC patients was taken as NLR=2.68,CAR=0.11,PLR=165.20,LMR=4.97.2.The results of single-factor Kaplan-Meier analysis showed that higher KPS score,limited disease SCLC(LD-SCLC),low CAR,NSE,FBG,without hyponatremia,radiation and chemotherapy combined for first-line treatment,and first-line chemotherapy more than 4 cycles were associated with extend OS.Higher KPS score,LD-SCLC,low CAR,NSE,without hyponatremia,radiation and chemotherapy combined for first-line treatment,first-line chemotherapy more than 4 cycles were associated with extend PFS.3.Multivariate COX regression analysis showed that: KPS score less than 80,ED-SCLC,high CAR,NSE,hyponatremia,first-line treatment was chemotherapy alone,and first-line chemotherapy cycle less than 4 were independent prognostic factors of death risk of SCLC patients.KPS score less than 80,ED-SCLC,high NSE,first-line treatment with chemotherapy alone and first-line chemotherapy cycle less than 4 were independent prognostic factors for the risk of recurrence or progression of SCLC patients.4.All independent prognostic factors were included in the nomogram based on COX multi-factor regression.The nomogram showed significant accuracy in predicting OS,with a consistency index of 0.725.Conclusions: 1.Multivariate analysis removed confounding factors and predicted the prognosis more accurately.The ED-SCLC,KPS score less than 80,first-line treatment was chemotherapy alone,and the first-line chemotherapy cycle <4,diagonosis with hyponatremia,increased serum NSE and CAR levels were independent prognostic factors for the mortality risk of SCLC patients.2.In this study,the model c-index of PFS and OS nomograms obtained were 0.66 and 0.725,respectively,which improved the predictive power,made it more intuitive and convenient for clinical application.