The Effect Of Tumor-associated Neutrophils And Granulocyte Colony-stimulating Agents On The Efficacy Of Immunotherapy In Non-small Cell Lung Cancer

Part I The predictive value of tumor-associated neutrophil gene signature for immunotherapy in non-small cell lung cancerObjectives:Tumor-associated neutrophils(TANs)are main components of tumor microenvironment(TME)and play an important role in cancer development,metastasis and immunotherapy resistance.The correlation between TANs and the efficacy of immune checkpoint inhibitors(ICIs)has been confirmed in liver cancer,melanoma,breast cancer.In this study,we aim to explore the effect of TANs on ICIs efficacy in NSCLC by means of bioinformatics methods.Materials and methods:The TCGA-LUAD dataset and three GEO cohorts were downloaded for further analysis.From single cell sequencing data of NSCLC,twenty genes specifically expressed in TANs were screened out and defined as TAN characteristic genes.Based on TAN characteristic genes,a prognostic model of NSCLC was constructed by LASSO regression.According to the risk score calculated by prognostic model,patients were divided into highrisk group and low-risk group.Kaplan-Meier curves and Log-rank test were used to validate the predictive value of the prognostic model.Gene expression profiles of high-risk group and low-risk group were compared by differential analysis.GO,KEGG and GSEA enrichment analysis were applied to elucidate the function of differentially expressed genes.The composition of TME and immune cell infiltration were analyzed by ESTIMATE algorithm and ssGSEA algorithm.Results:In TCGA-LUAD dataset,a prognostic model containing 14 TAN characteristic genes was established by LASSO regression.The 14 genes were IL1RN CD44,RHOH,RIPK2,HLA-DMB,HLA-DRB1,SNRPG,ASAH1,LGALS3,GNPDA1,PLEKHB2,RPL23,RPN2 and RPS12.The prognostic model evaluated the mortality risk of each individual and effectively predicted the overall survival of NSCLC patients.A total of 198 genes were differentially expressed between high-risk group and low-risk group.The differentially expressed genes were enriched in antigen processing and presentation,immune cell function,immune cell differentiation and other signaling pathways.Compared with low-risk group,the TME of high-risk group exhibited decreased infiltration level of T cell,B cell,NK cell and other immune cells,presenting an immune deserted phenotype.Finally,in three GEO cohorts,the prognostic model based on TAN characteristic genes was verified to have a good performance in predicting ICIs efficacy in NSCLC.Conclusions:The prognostic model based on TAN characteristic genes effectively predicted ICIs efficacy of NSCLC patients.TANs were important influencing factors and potential prognostic markers for ICIs efficacy in NSCLC.PART II The effect of granulocyte colony-stimulating agents on the efGcacy of immunotherapy in advanced non-small cell lung cancerObjectives: The results of our study and other research have suggested that TANs were main factors affecting the efficacy of ICIs treatment in NSCLC.Granulocyte colony stimulating factor(G-CSF)is the major cytokine that regulates TAN development and recruitment,and GCSF agents are widely used in clinical setting to prevent and treat neutropenia caused by myelosuppressive chemotiierapy.However,the correlation between G-CSF agents and ICIs efficacy of NSCLC stays unclear.In the current study,we aim to explore the effect of G-CSF agents on the prognosis of advanced NSCLC patients receiving chemotherapy combined with ICIs.Materials and methods: Locally advanced and advanced NSCLC patients receiving chemotherapy combined with ICIs in Qilu Hospital of Shandong University from January 2018 to December 2021 were enrolled.The clinicopathologic information,computed tomography images,hematological examination results,and G-CSF drug usage of enrolled patients were retrieved from electronic medical records.The survival data was obtained through telephone follow-up.The endpoints included progression free survival(PFS)and overall survival(OS).Kaplan-Meier survival analysis and COX hazards regression analysis were used to explore the influence of G-CSF agents on PFS and OS of NSCLC patients receiving chemotherapy combined with ICIs.The confounding factors were controlled by propensity score and inverse probability weighting methods to increase baseline comparability.Results: A total of 120 patients were eligible for analysis.57 patients(47.5%)received G-CSF drugs during treatment while 63(52.5%)did not.The median PFS was 12.9 months in G-CSF group,and 9.6 months in the group without G-CSF usage.There was no statistical difference(P = 0.92,HR = 0.97,95%CI = 0.60-1.58).The median OS of G-CSF group tended to be longer than the untreated group(17.6 months vs.15.3 months),but no statistical significance was observed(P=0.99,HR=0.99,95%CI=0.50-1.98).According to multivariate analysis,clinical stage(?<0.01,HR=2.58? 95%CI=1.21-5.51)and baseline peripheral neutrophil counts(P<0.01?HR=3.065 95% CI=1,29-7.21)were significantly correlated with PFS,while age(P<0.01,HR=6.35,95%CI=1.81-22.30)and baseline peripheral neutrophil counts(P < 0.01,HR=11.45,95%CI=2.83-46.33)were independent risk factors for OS.The propensity score and inverse probability weighting further validated that G-CSF agents had no significant effect on the ICIs efficacy ofNSCLC patients(PFS: P=0.92,HR=1.01,95%CI=0.61-1.66;OS: P=0.91,HR-1.04,95%CI=0.50-2.17).Conclusions: In this study,no correlation was observed between G-CSF agents and the prognosis of advanced NSCLC patients receiving chemotherapy plus ICIs,suggesting that GCSF agents can be used rationally in clinic condition.