AUTHOR=Bao Guangyao , Li Tian , Guan Xiaojiao , Yao Yao , Liang Jie , Xiang Yifan , Zhong Xinwen 

TITLE=Development of a Prognostic Alternative Splicing Signature Associated With Tumor Microenvironment Immune Profiles in Lung Adenocarcinoma

JOURNAL=Frontiers in Oncology

VOLUME=Volume 12 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2022.880478

DOI=10.3389/fonc.2022.880478

ISSN=2234-943X

ABSTRACT=Background: Alternative splicing (AS), a pivotal post-transcriptional process across over 95% human transcripts, involved in transcript structural variations and protein complexity. Clinical implication of AS events and their interaction with tumor immunity were systematically analyzed in lung adenocarcinoma (LUAD).
Methods: Transcriptome profiling as well as AS data of LUAD were retrospectively curated. Then, the network of the overall survival (OS)-relevant AS events with splicing factors was established. After screening OS-relevant AS events, a LASSO prognostic model was conducted and evaluated with ROC curves. A nomogram that integrated independent prognostic indicators was created. Immune response and immune cell infiltration were estimated with ESTIMATE, CIBERSORT and ssGSEA algorithms. Drug sensitivity was inferred with pRRophetic package.
Results: In total, 2415 OS-relevant AS events were identified across LUAD patients. The interaction network of splicing factors with OS-relevant AS events uncovered the underlying regulatory mechanisms of AS events in LUAD. Thereafter, a prognostic model containing 12 AS events was developed, which acted as a reliable and independent prognostic indicator following verification. A nomogram that constituted stage and risk score displayed the great effectiveness in evaluating the survival likelihood. Moreover, the AS-based prognostic model was in relation to immune response and immune cell infiltration. Patients with high risk score displayed therapeutic superiority to cisplatin, erlotinib, gefitinib, and gemcitabine. Finally, three AS-relevant genes (CDKN2A, TTC39C, and PKIB) were identified as prognostic markers.
Conclusion: Collectively, our findings developed an AS event signature with powerful prognostic predictive efficacy in LUAD.