Abstract:Background: The robustness and reproducibility of transcriptomics-derived models were limited by the various microenvironmental conditions like stromal context and tumor purity. We aimed to explore a robust prognostic and immunotherapeutic predictive model by differentiating cell origins in lung adenocarcinoma (LUAD).Methods: A total of 2,451 patients in 12 independent cohorts with transcriptomic data, including 10 public datasets and two in-house cohorts, were analyzed. The tumor immune cell score (TICS) and tumor stromal score (TSCS) was developed depending on the candidate immune or stromal cells in the training meta-cohorts (n=6), and then verified in four datasets. Therapeutic responses including immune checkpoint inhibitors (ICIs), chemotherapy and tyrosine kinase inhibitor (TKI) were compared between high- and low- TICS groups in the advanced-stage LUAD.Finding: The TICS, instead of TSCS, contributed majorly for predicting clinical outcomes in LUAD. The high TICS was associated with a prolonged overall survival (OS) in the training meta-cohorts (HR 0·55, 95%CI 0·45-0·66) and was validated in three public cohorts and one in-house cohort. The prognostic efficacy of TICS was more significant in the early-stage LUAD (HR=0·49, 95%CI: 0·38-0·63). For the advanced-stage LUAD, high TICS predicted a superior OS in patients who treated with ICIs (HR 0·09, 95% CI 0·02-0·32) instead of chemotherapy (HR 0·84, 95% CI 0·30-2·30) or TKI (HR 1·20, 95% CI 0·08-20·00).Interpretation: TICS is associated with survival in patients with LUAD, and may serve as a predictor for the benefit of ICIs in advanced LUAD.Funding Information: National Natural Sciences Foundation (82102886 for JX); Beijing Nova Program (Z211100002121055 for ZY); the Special Research Fund for Central Universities, Peking Union Medical College (3332021029 for JX); National Natural Sciences Foundation (82203827 for ZY); the Special Research Fund for Central Universities, Peking Union Medical College (3332021028 for ZY); Beijing Natural Science Foundation (7222146 for ZY); Guangdong Association of Clinical Trials (GACT) /Chinese Thoracic Oncology Group (CTONG) and Guangdong Provincial Key Lab of Translational Medicine in Lung Cancer (Grant No. 2017B030314120); Beijing Hope Run Special Fund of Cancer Foundation of China (LC2021B22, LC2020B09, LC2020A33); the National Natural Sciences Foundation (81871889 and 82072586 for ZW) and Beijing Natural Science Foundation (7212084 for ZW).Declaration of Interests: The authors declare that they have no competing interests.Ethics Approval Statement: All patients have informed consent and this study was approved by the ethics committees of the National Cancer Center (NCC-22/250-3454, NCC-22/429-3631). Patients without follow-up information were removed from further evaluation.

Deciphering key roles of B cells in prognostication and tailored therapeutic strategies for lung adenocarcinoma: a multi-omics and machine learning approach towards predictive, preventive, and personalized treatment strategies

Unraveling the immune landscape of lung adenocarcinoma: insights for tailoring therapeutic approaches

Advancing predictive markers in lung adenocarcinoma: A machine learning‐based immunotherapy prognostic prediction signature

Multi-omics characterization and machine learning of lung adenocarcinoma molecular subtypes to guide precise chemotherapy and immunotherapy

Immune, metabolic landscapes of prognostic signatures for lung adenocarcinoma based on a novel deep learning framework

Machine-learning and scRNA-Seq-based diagnostic and prognostic models illustrating survival and therapy response of lung adenocarcinoma

Lung Adenocarcinoma Systems Biomarker and Drug Candidates Identified by Machine Learning, Gene Expression Data, and Integrative Bioinformatics Pipeline

A&nbsp;Novel Prognostic and Immunotherapeutic Predictive Model Based On&nbsp;cell-Originated Characterization of Tumor Microenvironment in Lung Adenocarcinoma

Machine learning-based integration of CD8 T cell-related gene signatures for comprehensive prognostic assessment in lung adenocarcinoma

Integrated multi-omics analysis and machine learning to refine molecular subtypes, prognosis, and immunotherapy in lung adenocarcinoma

A novel gene expression signature-based on B-cell proportion to predict prognosis of patients with lung adenocarcinoma

Multi-omics identification of GPCR gene features in lung adenocarcinoma based on multiple machine learning combinations

The Predictive Role of Immune Related Subgroup Classification in Immune Checkpoint Blockade Therapy for Lung Adenocarcinoma

Subclassification of lung adenocarcinoma through comprehensive multi-omics data to benefit survival outcomes

A large cohort study identifying a novel prognosis prediction model for lung adenocarcinoma through machine learning strategies

Comprehensive Analysis and Reinforcement Learning of Hypoxic Genes Based on Four Machine Learning Algorithms for Estimating the Immune Landscape, Clinical Outcomes, and Therapeutic Implications in Patients With Lung Adenocarcinoma

Identification of a Costimulatory Molecule-Based Signature for Predicting Prognosis Risk and Immunotherapy Response in Patients with Lung Adenocarcinoma

Machine Learning-Based Immune Prognostic Model and Cerna Network Construction for Lung Adenocarcinoma

Comprehensive analysis of PPP4C's impact on prognosis, immune microenvironment, and immunotherapy response in lung adenocarcinoma using single-cell sequencing and multi-omics

Integrated machine learning survival framework to decipher diverse cell death patterns for predicting prognosis in lung adenocarcinoma

Multi-omics profiling and experimental verification of tertiary lymphoid structure-related genes: molecular subgroups, immune infiltration, and prognostic implications in lung adenocarcinoma