Global Profiling of Alternative Splicing in Non-Small Cell Lung Cancer Reveals Novel Histological and Population Differences

Academic Background

Lung cancer is one of the most frequently diagnosed cancers in the United States, with non-small cell lung cancer (NSCLC) accounting for the majority of cases. Among NSCLC, lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) are the most common subtypes. Despite significant progress in understanding the molecular mechanisms of lung cancer, minority populations, such as African Americans (AA), remain underrepresented in lung cancer research. African American men are more likely to develop lung cancer compared to European American (EA) men, with higher incidence and mortality rates. These disparities are partly attributed to differences in access to healthcare and treatment decisions.

Recent genomic studies have shown that alternative splicing (AS) plays a crucial role in cancer development and progression. AS refers to the process by which precursor mRNA is spliced to produce different mRNA isoforms, thereby influencing protein function. Aberrant splicing events are closely associated with tumorigenesis in various cancers, particularly in lung cancer, where AS may drive tumor progression by affecting the expression and function of oncogenes. However, research on AS differences between racial and histological subtypes remains limited.

Source of the Paper

This paper was co-authored by Saman Zeeshan, Bhavik Dalal, Rony F. Arauz, Adriana Zingone, Curtis C. Harris, Hossein Khiabanian, Sharon R. Pine, and Bríd M. Ryan. The authors are affiliated with multiple research institutions, including the Rutgers Cancer Institute of New Jersey and the National Cancer Institute. The paper was published in 2025 in the journal Oncogene (DOI: 10.1038/s41388-024-03267-y).

Research Process and Results

Research Process

1. Sample Collection and Processing
   The study utilized two independent patient cohorts for discovery and validation. The discovery cohort included 75 evaluable tumor samples (35 AA, 40 EA) and 77 non-tumor adjacent tissue (NAT) samples. The validation cohort consisted of 191 tissue samples (94 tumor samples, 23 AA, 71 EA, 95 NAT). All samples were obtained from NSCLC patients, including LUAD and LUSC subtypes.

2. RNA Sequencing and Data Analysis
   High-throughput, high-depth total RNA sequencing (total RNA-seq) was performed on tumor and NAT samples. Sequences were aligned to the human reference genome (hg38) using the HISAT2 software. Differential AS events were analyzed using the rMATS software, calculating the Percent Spliced In (PSI) for each splicing event. High-confidence splicing events were identified with a PSI difference greater than 10% and a false discovery rate (FDR) less than 0.1.

3. Validation Experiments
   High-throughput reverse transcription PCR (RT-PCR) was used to validate the identified splicing events, ensuring their reproducibility across different cohorts.

Key Results

1. Histology-Specific Splicing Events
   A total of 990 and 668 high-confidence AS events were identified in LUAD and LUSC tumors, respectively. These events predominantly occurred in protein-coding genes, with LUAD exhibiting more splicing events than LUSC. Validation confirmed 40% of LUAD and 50% of LUSC splicing events. These events were associated with multiple signaling pathways, including IL-15, STAT3, and VEGF signaling.

2. Population-Specific Splicing Events
   Over 1,000 and 300 high-confidence splicing events were identified in AA and EA populations, respectively. Approximately 50% of these events were shared between the two populations, while the remainder were population-specific. For example, TPM1 and VEGFA genes exhibited multiple splicing events in both AA and EA, whereas ADAM15 and FGFR2 showed splicing events only in AA.

3. Splicing Events in Driver Genes and Surface Proteins
   The study also identified splicing events associated with cancer driver genes and cell surface proteins. For instance, the AFDN gene exhibited significant PSI differences in both AA and EA, correlating with tumor invasiveness and migration. Additionally, splicing events in the CD44 gene showed significant PSI differences in LUAD, suggesting its potential role in tumor development.

4. Impact of Environmental and Host Factors
   Linear regression analysis revealed that smoking and medication use (e.g., aspirin) significantly influenced splicing events. Smoking intensity and duration were associated with multiple splicing events, while aspirin use was linked to specific splicing events.

Conclusions and Significance

This study provides the first comprehensive analysis of global AS features in NSCLC, revealing splicing differences between histological subtypes and racial groups. These findings offer new insights into the molecular mechanisms of lung cancer and present potential targets for personalized therapies tailored to specific populations. Importantly, the study highlights the role of AS in tumorigenesis and immunotherapy, suggesting its potential as a novel direction for cancer treatment.

Research Highlights

1. High-Confidence Splicing Events: Through high-throughput RNA sequencing and validation experiments, the study identified numerous high-confidence AS events with significant differences between tumors and NATs.
2. Population-Specific Splicing Events: The study revealed AS differences between AA and EA populations, providing new evidence for the biological basis of racial disparities in lung cancer.
3. Splicing Events in Driver Genes and Surface Proteins: The study identified multiple splicing events associated with cancer driver genes and cell surface proteins, suggesting their potential roles in tumor development.
4. Impact of Environmental and Host Factors: The study demonstrated the significant influence of smoking and medication use on splicing events, offering new directions for research on environmental factors in lung cancer.

Summary

Through comprehensive genomic analysis, this study elucidates the complexity and diversity of AS in NSCLC, particularly the differences between racial and histological subtypes. These findings not only provide new perspectives on the molecular mechanisms of lung cancer but also offer potential targets for personalized therapies tailored to specific populations. Future research should further explore the specific roles of these splicing events in tumorigenesis, progression, and treatment, advancing the precision of lung cancer therapy.