| The non-small cell lung cancer(NSCLC)clinical routine treatment mode entered the era of immunotherapy from the age of traditional chemotherapy,radiotherapy,and targeted therapy.However,quite a subset of patients failed to respond to immune checkpoint blockades(ICBs).Correctly identifying and screening patients who can benefit from ICBs is a foremost challenge faced by precision medicine.Previous studies of immunotherapy biomarkers mainly focused on PD-L1 immunohistochemistry(IHC),tumor mutation burden(TMB),microsatellite instability(MSI),but they still cannot meet the demand of precisely screening responders.Recently,multiple transcriptomic studies established numerous biomarkers calculated based on machine learning algorithm and high-throughput data,which reflect multifaceted characterization of tumor microenvironment and predict immunotherapy efficacy across multiple cancer types.Therefore,we aim to generate a reliable biomarker based on different pre-treatment NSCLC immunotherapy datasets by integrating tumor microenvironment,tumor intrinsic pathways,the biological signatures associated with immune response,and further explore the underlying immune and genomic mechanisms.Results:1.The Establishment of Immune-Keratin-Checkpoint score(IKCscore).Transcriptome sequencing was performed in tumor specimens from 65 patients with advanced NSCLC who received anti-PD-1 therapy in NFH cohort.We identified 60 genes associated with immunotherapy response,derived from differentially expressed gene analysis and signatures comprising tumor microenvironment,metabolic pathways,tumor intrinsic pathways.Based on the 60 response-relevant genes,we generated 5 transcriptome patterns including immune pattern and keratin pattern which were positively and negatively associated with therapeutic response to immune checkpoint blockade respectively.Integrating the genes from above two patterns and immune-checkpoint genes,we constructed the IKCscore for predicting immunotherapeutic response and precision medicine.2.The predictive value of IKCscore for immunotherapeutic sensitivity.The immune score,immune-checkpoint score and IKCscore which established in this research,were correlated to favourable immunotherapy response and improved progress free survival(PFS),whereas the KRTscore was a negative indicator for immunotherapy efficacy,suggesting IKCscore as a promising biomarker for immunotherapy sensitivity and patient prognosis.3.The Comparison of predictive capacity between IKCscore and exiting biomarkers.IKCscore exhibited superior predictive power than PD-L1 expression level and TMB in three immunotherapy cohorts.Significant correlation was observed between IKCscore and PD-L1 expression,but not between IKCscore and TMB,demonstrating that IKCscore and TMB mediated ICB responses via two distinct mechanisms.4.The correlation between IKCscore and tumor-microenvironment subtypes.An increase of CD8+T cell and DC cell infiltration was observed in high-IKCscore patients.Moreover,the IKCscore was remarkedly elevated in the inflamed phenotype compared with desert phenotype,unveiling a close correlation between IKCscore and immune activation.5.The genome atlas and intrinsic mutations associated with IKCscore.We revealed genome alterations associated with high-and low-IKCscore subgroups in lung cancer cohorts.STK11 and KEAP1 mutations,which recognized as an indicator for decreased efficacy of ICB treatment in NSCLC,were significantly associated with a decrease of IKCscore.6.The single-cell landscape and cell-cell crosstalk associated with immunotherapy response.Single-cell analysis implied that cancer cells potentially provide a permissive microenvironment for tumor progression and resistance to immunotherapy via inducing endothelial cell proliferation,angiogenesis,and vascular permeability.ConclusionThe Immune-Keratin-Checkpoint signature is a promising and capable biomarker for prediction of immunotherapy efficacy and precision medicine of non-small cell lung cancer. |
PDF Full Text Request