Analysis Of The Impact Of EGFR Co-mutations On The Efficacy Of Immune Checkpoint Inhibitors And Immune-Related Features In Advanced Non-small Cell Lung Cancer

Part Ⅰ Impact of EGFR co-mutations on the efficacy of immune checkpoint inhibitors in advanced non-small cell lung cancer patientsBackground:Lung cancer is the leading malignancy in China in terms of both morbidity and mortality.According to the pathological type,lung cancer is divided into small cell lung cancer and nonsmall cell lung cancer(NSCLC),of which NSCLC accounts for more than 80%.Most NSCLC patients are already at an advanced stage when they are diagnosed.Thus,Platinumbased chemotherapy has been the standard first-line regimen for patients with advanced NSCLC in the past.Nevertheless,patients have had limited benefit from it.For advanced NSCLC with sensitive driver mutations,targeted therapy is their first choice.Epithelial growth factor receptor(EGFR)is the most common driver gene mutation.Multiple clinical trials have shown that tyrosine kinase inhibitors(TKIs)targeting EGFR(first-,second-or third-generation TKIs)significantly prolong the survival of patients with EGFR-mutated advanced NSCLC compared to traditional double platinum-containing chemotherapy.Therefore,lung cancer guidelines recommend "TKIs single-agent targeted therapy " and "TKIs combined with chemotherapy" as the new first-line standard therapy for advanced NSCLC patients with EGFR mutation.Nevertheless,all NSCLC patients inevitably develop resistance to TKIs,and there is a lack of efficient alternative treatment options.Immune checkpoint inhibitors(ICIs)represented by PD-1/PD-L1 antibodies is another major milestone following targeted therapy in the history of non-small cell lung cancer therapy,which significantly prolonged patient survival.KEYNOTE-024 study showing a 5year survival rate of 31.9%for advanced NSCLC patients treated with Pembrolizumab.At present,treatment strategies such as "ICIs monotherapy" and "ICIs combined with chemotherapy" have been approved as first-line treatment for advanced driver-negative NSCLC.In contrast,regarding EGFR mutation-positive NSCLC patients,the antitumor efficacy of ICIs is significantly diminished,thereby clinical guidelines do not recommend ICIs as first-line therapy for these patients.Studies have shown that the immunomodulatory effects mediated by EGFR mutation are the main reason for weakening the efficacy of ICIs.Indeed,a growing body of clinical evidence suggests the critical role of gene mutations in modulating the efficacy of immunotherapy.In addition to EGFR mutation,mutations such as POLE,KEAP1,and STK11 have been revealed to contribute to sensitization or resistance to ICIs.With the large-scale application of testing technologies such as NGS,most NSCLC patients have been found to share mutations in multiple genes.Notably,co-mutations with multi-genes may exert a more important function in immunotherapy than single mutation.Currently,extensive studies have reported that NSCLC patients always carrying multi-gene co-mutations.For example,mutations in TP53 and KRAS are mostly accompanied by mutations in other genes,and such co-mutation patterns serve as important modulators of immunotherapy.For example,KRAS&TP53 co-mutated NSCLC are more likely to benefit from immunotherapy in terms of survival,while patients with KRAS&STK11 co-mutations benefit less from ICIs.For EGFR mutated NSCLC patients,they are always accompanied by other genes mutations,such as TP53,LRP1B,PIK3CA,etc.There is strong evidence that EGFR mutation diminish the efficacy of ICIs,while the effect of varying EGFFR co-mutation patterns on the efficacy of ICIs is unclear.Objective:(1)To investigate the impact of distinct EGFR co-mutation patterns on the efficacy of ICIs in advanced NSCLC patients.(2)To screen potential EGFR-mutated NSCLC patients who could benefit from ICIs,thereby providing evidence for clinician decision-making.Methods:(1)Data collection:Clinical data(e.g.,gender,age,smoking history,ethnicity,treatment medication,etc.)and mutation profile data were downloaded from the original studies,the OAK study(Phase Ⅲ randomized controlled study)(NCT01903993,N=1225)and the POPLAR study(Phase Ⅱ randomized controlled study)(NCT02008227,N=287).(2)Data integration:Consolidation of data from the OAK study and POPLAR study into two groups,the EGFR-mutated&atezolizumab-treated group and the EGFR wild-type and atelizumab-treated group,based on EGFR mutation status and therapeutic modality.(3)Co-mutation spectrum analysis:The mutation profiles of EGFR mutated NSCLC patients treated with atezolizumab were counted,with the top 20 genes(TOP20)co-mutated with EGFR screened.(4)Survival analysis:Comparison of overall survival and progression-free survival between patients with EGFR co-mutations and those with EGFR single mutations by KaplanMeier and Log-rank tests,respectively.(5)Objective response rate(ORR)analysis:Comparison of the difference in ORR between patients with EGFR co-mutations and EGFR single mutations by Fisher’s exact test.(6)Multivariable Cox regression analyses:The independent impact of co-mutation patterns on the efficacy of ICIs was examined and the survival prediction model was constructed by multivariable Cox regression analyses.Results:(1)A total of 56 EGFR-mutated NSCLC patients receiving atezolizumab monotherapy were identified.(2)TOP20 genes co-mutated with EGFR were identified as as TP53,LRP1B,DNMT3A,SPTA1,CDK12,ERBB4,MLL3,FAT3,SLIT2,ASXL1,EPHB1,PRKDC,BARD1,NOTCH2,RB1,TSC1,STAG2,ATM,MLL2,and FAT1.(3)Survival analyses demonstrated that the addition of LRP1B mutation to EGFRmutated patients yield superior survival benefit over those with EGFR single mutation(median overall survival(OS):EGFR MT and LRP1B MT vs EGFR MT and LRP1B WT:not reached(NR)vs 7.82 months,HR(hazard ration)0.43,95%confidence interval(CI)0.220.86,P=0.0377).There was a tendency for prolonged survival in patients with EGFR and FAT3 co-mutation patterns as well(EGFR MT and FAT3 MT vs EGFR MT and FAT3 WT:NR vs 10.45 months,HR 0.30,95%CI 0.12-0.71,P=0.0723).Compared to EGFR WT patients,EGFR&LRP1B co-mutated(EGFR MT&LRP1B MT vs EGFR WT&LRP1B MT:NR vs 13.24 months,P=0.0588)and EGFR&FAT3 co-mutated(EGFR MT&FAT3 MT vs EGFR WT&FAT3 MT:NR vs 13.50 months,P=0.0746)patients both showed a trend towards longer survival.Besides,there was a mild tendency for some other co-mutation patterns to affect the efficacy of ICIs,with SLIT2 and BPHB1 mutations tending to strengthen ICIs efficacy,while TP53,CDK12,MLL2,MLL3,NOTCH2 and RBI tending to weaken ICIs efficacy,but none of these results reached statistical difference(all P>0.05).(4)ORR showed that patients with EGFR and LRP1B co-mutations had significantly higher ORR than those with EGFR single mutation(28.57%vs 2.5%,P=0.013);patients with EGFR and FAT3 co-mutations also had higher ORR than those with EGFR single mutation(33.33%vs 6.25%,P=0.0894),but did not reach a statistical difference.(5)Multivariable Cox regression analyses showed that EGFR&LRP1B co-mutation or EGFR&FAT3 co-mutation was independent predictor for benefit from atezolizumab.A nomogram with great predictability was constructed by incorporating gender,age,race,ECOG score,smoking history,and EGFR and LRP1B/FAT3 co-mutation pattern,with area under curve(AUC)of 0.88 for predicting 1-year survival and 0.85 for 2-year survival.Conclusions:(1)There was considerable impact of EGFR co-mutation patterns on the efficacy of ICs,which differed across different co-mutation patterns.(2)Patients with EGFR&LRP1B co-mutation or EGFR&FAT3 co-mutation patterns could benefit more from atezolizumab monotherapy,indicating that the atezolizumab monotherapy regimen could be a potential treatment option for those patients.Part Ⅱ Analysis of immune features in patients with EGFR&LRP1B and EGFR&FAT3 co-mutated non-small cell lung cancerBackground:In part Ⅰ study,EGFR&LRP1B co-mutation and EGFR&FAT3 co-mutation pattern were found to synergistically facilitate the anti-tumor effect of ICIs,which was associated with high survival benefit of patients from ICIs.However,the underlying mechanisms is unclear.Currently,numerous studies have been dedicated to the regulation of sensitivity and resistance to immune checkpoint inhibitors,with programmed cell death ligand 1(PD-L1)expression levels,tumor mutation burden(TMB),and tumor immune microenvironment revealed to be strongly related to the anti-tumor efficacy of immunotherapy.PD-L1 is an important biomarker widely used clinically to predict whether NSCLC patients will benefit from ICIs,as it has been found that patients with high PD-L1 expression in NSCLC are more likely to benefit from immunotherapy.In addition,clinical studies have shown that treatment with ICIs monotherapy can significantly prolong the survival of advanced NSCLC patients with PD-L1 expression levels≥50%.TMB is another key biomarker widely used in clinical practice to predict the efficacy of ICIs.TMB is regarded to be closely related to the production of tumor neoantigens,with patients with high TMB being more likely to produce tumor neoantigens,thus facilitating ICIs to stimulate anti-tumor immune responses and mobilize anti-tumor immune cells to recognize and eradicate tumor cells.Besides,the tumor immune microenvironment(TIME)also plays crucial roles in ICIs therapy.TIME is the environment for tumorigenesis and development,consisting of a complex network system of immune cells,fibroblasts,extracellular matrix and various signaling molecules.Based on the composition of TIME,TIME can be classified into "hot" TIME and "cold" TIME."Hot"TIME tends to show high levels of anti-tumor immune cell infiltration and activation of antitumor-related immune-related signaling pathways,whereas "cold" TIME tends to show low levels of anti-tumor immune cell infiltration and suppression of anti-tumor-related immunerelated signaling pathways.In this part,we systematically analyzed the influence of the co-mutation pattern on PDL1 expression level,TMB level and TIME,which are closely related to the efficacy of ICIs,with the aim of providing theoretical evidence that patients bearing EGFR&LRP1B comutation or EGFR&FAT3 co-mutation pattern are more likely to benefit from ICIs.Objective:To investigate the potential mechanism by which EGFR&LRP1B co-mutation pattern or EGFR&FAT3 co-mutation pattern enhance the anti-tumor effects of ICIs from three aspects:PD-L1 expression level,TMB level and TIME type.Methods:(1)PD-L1 expression analysis:Advanced NSCLC patients treated with atezolizumab monotherapy in the OAK and POPLAR study were divided into 4 groups based on the mutational status of EGFR and LRP1B:EGFR MT&LRP1B MT、EGFR MT&LRP1B WT、EGFR WT&LRP1B MT、EGFR WT&LRP1B WT.The same methodology divided patients into four groups based on the mutational status of EGFR and FAT3:EGFR MT&FAT3 MT、EGFR MT&FAT3 WT、EGFR WT&FAT3 MT、EGFR WT&FAT3 WT.Comparison of the difference in PD-L1 levels among different groups by Kruskal-Wallisjia test.(2)b-TMB analysis:grouping methods were the same as in PD-L1 expression analysis,and the differences in b-TMB levels among different groups were compared by KruskalWallisjia test.(3)t-TMB analysis:The Cancer Genome Atlas(TCGA)(PanCancer Atlas)lung adenocarcinoma and lung squamous carcinoma gene mutation profiles(EGFR,LRP1B,FAT3)and t-TMB data were downloaded from cBioPortal.Based on the co-mutation status of EGFR with LRP1B or FAT3,the groups were divided into 4 groups(EGFR MT&LRP1B MT、EGFR MT&LRP1B WT、EGFR WT&LRP1B MT、EGFR WT&LRP1B WT;EGFR MT&FAT3 MT、EGFR MT&FAT3 WT、EGFR WT&FAT3 MT、EGFR WT&FAT3 WT),and then comparing the difference in t-TMB levels among groups by Kruskal-Wallisjia test.(4)TIME analysis:NSCLC patients were divided into 2 groups based on EGFR with LRP1B or FAT3 mutation status:EGFR MT&LRP1B MT and EGFR MT&LRP1B WT,EGFR MT&FAT3 MT and EGFR MT&FAT3 WT,respectively.Statistical analysis was performed using the Mann-Whitney U test.Results:(1)PD-L1 expression results:no significant difference in PD-L1 expression levels among the 4 groups of EGFR MT&LRP1B MT,EGFR MT&LRP1B WT,EGFR WT&LRP1B MT,EGFR WT&LRP1B WT;no significant difference in PD-L1 expression levels among the 4 groups of EGFR MT&FAT3 MT,EGFR MT&FAT3 WT,EGFR WT&FAT3 MT,EGFR WT&FAT3 WT.(2)b-TMB levels:patients in the EGFR MT&LRP1B MT group exhibited the highest bTMB levels,while patients in the EGFR MT&LRP1B WT group exhibited the lowest b-TMB levels;patients in the EGFR MT&FAT3 MT group exhibited the highest b-TMB levels,while patients in the EGFR MT&FAT3 WT group exhibited the lowest b-TMB levels.(3)t-TMB levels:patients in the EGFR MT&LRP1B MT group showed the highest tTMB levels,while patients in the EGFR MT&LRP1B WT group showed the lowest t-TMB levels;patients in the EGFR MT&FAT3 MT group showed the highest t-TMB levels,while patients in the EGFR MT&FAT3 WT group showed the lowest t-TMB levels.(4)TIME results:patients with EGFR MT&LRP1B MT presented higher proportions of anti-tumor immune cell infiltration(activated CD4+T cells,effector CD4+ T cells and memory B cells);patients with EGFR MT&FAT3 MT presented higher proportions of antitumor immune cell infiltration(activated CD4+T cells).Conclusions:(1)EGFR&LRP1B/FAT3 co-mutations have no effect on PD-L1 expression.(2)EGFR&LRP1B/FAT3 co-mutations are associated with high tumor immunogenicity in NSCLC.(3)EGFR&LRP1B/FAT3 co-mutations are associated with "hot" TIME formation.(4)The modulation of TMB and TIME by EGFR&LRP1B/FAT3 co-mutations could be the potential mechanism by which patients in this group are more likely to benefit from ICIs.