Radiomics Based Approach To Predict The Risk Of Cachexia On Immunotherapy In Extensive-stage Small Cell Lung Cancer

Part ⅠBackground and objectives:Cachexia is closely related to the efficacy of chemoradiotherapy and immunotherapy for non-small cell lung cancer.At present,programmed death-ligand(PD-L1)inhibitor combined with chemotherapy has become the first-line standard treatment in Extensive-stage small cell lung cancer(ES-SCLC).However,the impact of cachexia on its efficacy is not clear.This study aims to explore the association between cachexia and the efficacy of first-line immunotherapy in ES-SCLC.Methods:ES-SCLC patients receiving first line immunotherapy were enrolled in analysis.The development of cachexia after the initiation of ICI was defined as patients with more than 5%weight loss over past 6 months,or more than 2%weight loss with body mass index(BMI)less than 20 kg/m2.The endpoints were objective response rate(ORR),disease control rate(DCR),progression free survival(PFS),overall survival(OS)and cachexia.Kaplan-Meier survival curve was used to evaluate the efficacy of cachexia on survival.Results:Data of 123 ES-SCLC patients receiving first-line immunotherapy were analyzed.There were statistical differences in ORR and DCR between cachexia group and non-cachexia group(ORR,43.1%vs.66.7%,P=0.009;DCR,43.1%vs.66.7%,P=0.009,respectively).PFS and OS in cachexia group were significantly worse than non-cachexia group(median PFS,6.5 months vs.8.3 months,P=0.013;median OS,9.5 months vs.not reached,P=0.015,respectively).Conclusion:Cachexia affects the efficacy of first-line immunotherapy in ES-SCLC.Part ⅡBackground and objectives:Cachexia affects the efficacy of first-line immunotherapy in ES-SCLC.Accurate prediction of cachexia can better guide the intervention of early nutritional support treatment and maximize the clinical benefits of patients.However,there are no available biomarkers to predict the risk of cachexia.Radiomic uses the high-throughput extraction of advanced quantitative features to objectively and quantitatively describe tissue heterogeneity.This study aims to construct and validate radiomics signature to predict the risk of cachexia of ES-SCLC patients with first line immunotherapy.Methods:ES-SCLC patients receiving first line immunotherapy were enrolled in analysis,then radiomics features were extracted.LASSO-Logistic was performed to identify radiomics features,and generate radiomics signature(RS).Results:Total of 123 ES-SCLC patients were enrolled in analysis.874 radiomics features were extracted,and RS consisted of 5 selected features.The AUC value of RS for predicting the risk of cachexia were 0.7419 in the training cohort,and an AUC of 0.6568 in the validation cohort.Then the clinical model and clinical-radiomics model were constructed to predict the risk of cachexia respectively.The AUC of the training cohort were 0.74 and 0.806 respectively and the results of the validation cohort also confirmed that the clinical-radiomics model had higher prediction efficiency,with AUC of 0.658 and 0.753 respectively.The risk score of clinical-radiomics model was significantly with PFS both in training cohort and validation cohort(median PFS:8.4months vs.5.6 months,p=0.0022;median PFS:9.5months vs.6.5 months,p=0.039,respectively).There was difference but no statistical significance in OS.Conclusion:Radiomics exhibited a favorable effect to predict the risk of cachexia in ES-SCLC patients receiving first line immunotherapy.