APSR2026-258

 
Late-Breaking
Submitted
Poster Presentation
 
AP14-009
P28
 
Format: PPTX Only
https://storage.unitedwebnetwork.com/files/1285/1212517-46904-APSR2026-258_432890.pdf
Format: PDF Only
08. Lung Cancer
3

No more than 15 authors can be listed.
Please ensure the authors are listed in the right order.
If there are multiple affiliations, please use the "+" button next to "Select Country/Region" to add additional affiliation fields.

Lu Xu * Tianjin Medical University General Hospital Medical Oncology Department Tianjin China
Diansheng Zhong - Tianjin Medical University General Hospital Medical Oncology Department Tianjin China
Linlin Zhang - Tianjin Medical University General Hospital Medical Oncology Department Tianjin China
 
 
 
 
 
 
 
 
 
 
 
 
Presenting Author Details
Lu
Xu
X17647550138@163.com
+8617647550138
Under 30
China
Original Research
CD8+ T Cell-Derived 13-Gene Signature Predicts Prognosis and Tumor Microenvironment Phenotypes in Non-Small Cell Lung Cancer
Maximum 25 words
Abstract Content
CD8+ T cells are key mediators of anti-tumor immunity. However, their phenotypic and functional heterogeneity in the tumor microenvironment (TME) of non-small cell lung cancer (NSCLC) remains incompletely characterized. This study employed integrative multi-omics approaches to develop and validate a CD8+ T cell-derived gene signature for prognostic prediction and TME characterization in NSCLC.
Differentially expressed genes in CD8+ T cells from single-cell RNA sequencing data of NSCLC were intersected with immune-related gene sets from InnateDB and IMMPORT. Univariate Cox regression analysis was performed on these immune core genes using The Cancer Genome Atlas (TCGA) cohort to screen for candidates significantly associated with overall survival (OS). An optimal prognostic model was selected from 101 algorithm combinations using TCGA and Gene Expression Omnibus (GEO) cohorts. Model performance was evaluated via Kaplan-Meier analysis and time-dependent receiver operating characteristic (ROC) curves, and further validated in independent datasets. Differences in TME profiles between risk groups were also assessed.
Univariate Cox analysis of 171 immune core genes in the TCGA cohort identified 16 prognostic candidates. The optimal prognostic model was constructed using CoxBoost + Elastic Net (α = 0.2), identifying 13 core genes: CX3CR1, CISH, PELI1, CXCR6, SLAMF1, OASL, HSPA1A, LGALS1, OAS3, F2R, ITGB1, BIRC3, and CTSL. In the TCGA training cohort, high-risk patients had significantly worse OS (P < 0.001), with 1-, 3-, and 5-year area under the ROC curve (AUC) of 0.653, 0.664, and 0.648, respectively. External validation in GSE37745 (P = 0.010, AUC = 0.605–0.633), GSE42127 (P < 0.001, AUC = 0.580–0.772), and the IMvigor210 immunotherapy cohort (P = 0.006, AUC = 0.527–0.682) demonstrated consistent prognostic stratification. Mechanistically, the low-risk group showed an immune-active "hot" TME with enhanced infiltration of plasma cells, CD8+ T cells, and activated natural killer cells, whereas the high-risk group exhibited an immune-suppressive "cold" TME with predominant M2 macrophages and neutrophils.
This CD8+ T cell-derived 13-gene signature effectively predicts prognosis and TME phenotypes in NSCLC, providing a robust tool for risk stratification and potential guidance for immunotherapy strategies.
 
 
 
 
 
No
No
 
JPG or JPEG or PNG format ONLY
 
maximum 3 keywords and no abbreviations, separated by commas
 
336