Predictive Value Of Peripheral Blood Parameters For Prognosis Of Extensive Stage Small Cell Lung Cancer Receiving Immunotherapy And Establishment Of Model

Objective:To date,no validated predictive markers have been identified to predict the prognosis of patients with extensive stage small cell lung cancer(ES-SCLC)treated with immune checkpoint inhibitors(ICIs),and reliable predictive models are lacking to screen patients who benefit from ICIs.Therefore,the aim of this study was to investigate the effect of pretreatment peripheral blood parameters on the efficacy and their prognostic prediction in ES-SCLC patients treated with ICI.In addition,a model based on the results is constructed to predict the prognosis of patients and provide clinicians with a tool to screen patients who benefit.Methods:In this study,we retrospectively collected the data of 58 ES-SCLC patients who received ICI treatment from January 2020 to December 2020,and recorded the clinical characteristics of the patients,including gender,age,smoking history,ECOG score,height,weight,treatment history,survival status,and survival time after immunotherapy.Blood cell counts,albumin,cholesterol,and lactate dehydrogenase were collected from peripheral blood before immunotherapy,and the above indicators were analyzed to obtain pulmonary immune prognostic index(LIPI),prognostic nutritional index(PNI),controlled nutritional status(CONUT)score,systemic immune inflammatory index(SII),systemic inflammatory response index(SIRI),and advanced lung cancer inflammatory index(ALI).Patients were followed up with overall survival(OS)after ICI as the study endpoint.The optimal cut-off values of the above indicators were determined according to the receiver operating characteristic curve(ROC)and grouped based on the cut-off values.Survival curves were plotted by Kaplan-Meier method,log-rank test,and univariate analysis was used to determine prognostic variables.Significant variables in univariate analysis were subsequently included in multivariate Cox regression analysis using Forward: LR to identify independent prognostic factors for ICI in ES-SCLC.Finally,prognostic nomograms were created by R software using independent prognostic factors.The area under the curve(AUC)and C-index of the ROC curve were used to evaluate the predictive validity of the model.Bootstrap method was used to validate the prediction model and draw the calibration curve.Results:1.According to the peripheral blood parameters of patients before treatment,PNI,ALI,SII and SIRI were calculated,ROC curve was drawn,PNI = 44.48,ALI = 34.68,SII= 1092.76 and SIRI = 1.4 were calculated,which were the optimal cutoff values for predicting ES-SCLC in ICI treatment;in CONUT,recognized "2" was used as the cutoff value according to previous studies,which was divided into low and high groups according to the cut-off values.LIPI,on the other hand,divided patients into three groups(good,moderate,and poor)according to the number of high-risk factors.2.In Kaplan-Meier analysis,PNI(P < 0.001),SII(P = 0.005),ALI(P = 0.026),CONUT(P < 0.001),and LIPI(P = 0.001)were found to be significantly associated with OS in patients receiving immunotherapy,except SIRI(P = 0.861).3.PNI(P < 0.001),SII(P = 0.010),ALI(P = 0.039),CONUT(HR: P < 0.001)and LIPI(P = 0.001)were proved to be statistically significant in univariate analysis.Forward:LR was used to include the significant variables in univariate analysis in multivariate Cox regression analysis,and finally two significant prognostic factors were obtained.PNI(P <0.001)and LIPI(P = 0.025)were independent risk factors for OS.4.PNI and LIPI were used to make prediction models,nomograms for the prognosis of ES-SCLC patients were established,and the validity of the model was evaluated by plotting nomogram ROC curves;bootstrap method was used to internally validate the nomogram models and draw calibration curves.Conclusion:1.Pretreatment PNI and LIPI are independent prognostic factors for OS in patients receiving ICI for ES-SCLC and can better reflect the prognosis of patients.2.According to PNI and LIPI,the prediction model C-index is 0.763,which is reliable.The calibration curve shows that there is a good agreement between nomogram prediction and actual prediction.