Comprehensive multi-omics analysis of tandem duplicator phenotypes in non-small cell lung cancer

Jie Li, Xinyan Pan, Yang Bi, Jian Huang, Jun Peng, Cx Bai, Jianhong Hou, Chen YangFrontiers in Medicine 2025 · DOI

This study, published in Frontiers in Medicine, provides a comprehensive look at how specific patterns of genomic instability—specifically the "Tandem Duplicator Phenotype" (TDP)—can serve as a roadmap for predicting patient outcomes and treatment responses in lung cancer.

TL;DR

Researchers have identified that the "span size" (length) of tandem duplications in the genomes of non-small cell lung cancer (NSCLC) patients is a critical biomarker. Patients with smaller duplication spans generally exhibit more active immune responses and better clinical outcomes, while those with larger spans face more aggressive disease and lower drug sensitivity.

What Problem Does This Paper Address?

Genomic instability is a hallmark of cancer, but it isn't a monolithic state. One specific type of instability is the Tandem Duplicator Phenotype (TDP), where segments of DNA are duplicated and placed head-to-tail. While TDP has been observed across various cancers, its specific clinical implications for NSCLC—the leading cause of cancer-related deaths—remained murky.

The authors sought to bridge the gap between raw genomic data and clinical reality. They asked: Can we classify patients based on the characteristics of these duplications to better predict who will survive longer, whose cancer is more aggressive, and who will respond best to chemotherapy?

Methods

The team employed a multi-omics approach, integrating DNA copy number variations (CNV), transcriptomics (RNA-seq), somatic mutations, and clinical data.

  1. TDP Scoring: They developed a "raw TDP score" to quantify the count and chromosomal distribution of tandem duplications.
  2. Gaussian Mixture Modeling: Using the mclust R package, they analyzed the distribution of duplication "span sizes" (the length of the duplicated DNA segments). This allowed them to statistically define distinct "modes" or groups.
  3. Subtyping: They categorized patients into six unique groups based on these span sizes, primarily distinguishing between Small Span Size Groups (SSG) and larger span groups (like TDP group 2/3 mix).
  4. Functional Analysis: They used tools like GISTIC 2.0 to identify target genes within duplicated regions and clusterProfiler to map out the biological pathways (like immune response or cell proliferation) that were most active in each group.

Datasets

The study relied on high-quality, large-scale public repositories:

  • TCGA (The Cancer Genome Atlas): Specifically the LUAD (Lung Adenocarcinoma) and LUSC (Lung Squamous Cell Carcinoma) cohorts, totaling over 1,000 samples.
  • CCLE (Cancer Cell Line Encyclopedia): Used to correlate genomic TDP profiles with sensitivity to 24 different anticancer drugs across 81 NSCLC cell lines.
  • UCSC Cancer Genomic Browser: For sourcing clinical timelines and relapse events.

Key Insights and Ideas

  • The "Long and Short" of It: The most striking insight is that the length of the duplication matters as much as the duplication itself. Smaller spans (SSG) correlate with a "moderate" pathology, whereas larger spans correlate with "severe" clinical stages and metastasis.
  • Immune System "Aha" Moment: The researchers found that SSG patients have significantly more active immune pathways (e.g., IL-17 and Chemokine signaling). This suggests that smaller genomic disruptions might "alert" the immune system more effectively or at least coexist with a more robust immune environment that curbs tumor growth.
  • Oncogene Overdrive: In larger span groups, the duplications frequently encompassed entire bodies of major oncogenes like EGFR and KRAS. This "DUP effect" essentially puts the cancer's growth signals into overdrive, leading to the observed worse prognosis.
  • Predicting Sensitivity: TDP classification proved to be a potent predictor of drug response. Cell lines with smaller span sizes were significantly more sensitive to targeted therapies like Lapatinib (targeting EGFR) and AZD6244 (targeting MEK).

Key Results

  • Survival Advantage: In the LUAD cohort, patients in the small span groups had significantly better recurrence-free survival compared to the TDP group 2/3 mix ($p = 0.0327$). Similar results were found in the LUSC cohort ($p = 0.0398$).
  • Clinical Correlation: SSG patients predominantly exhibited Stage I and II disease with no distal metastasis and showed a "complete" therapy response more frequently.
  • Drug Sensitivity: TDP Group 1 cell lines showed lower $IC_{50}$ values (meaning they are killed by lower doses of the drug) compared to "mix" groups for several key anticancer agents.

Limitations and Future Work

The authors are transparent about several hurdles:

  • Retrospective Nature: The study relies on existing datasets. While robust, prospective clinical trials are needed to validate these TDP groups as a standard diagnostic tool.
  • Sample Size Constraints: The drug susceptibility analysis for LUAD was limited by a small number of available cell lines in the CCLE (only six), which may affect the statistical power of those specific findings.
  • Mechanistic Gaps: While the paper establishes a strong correlation between span size and immune activity, the exact biological mechanism—why a small duplication triggers a better immune response than a large one—remains to be fully decoded.
  • Next Steps: The authors suggest that moving toward single-cell genome and transcript data could further refine our understanding of heterogeneity within these TDP subtypes.

Summary updated

Abstract

The integrated analysis underscores that an in-depth molecular assessment of TDP can offer invaluable insights into its role in cancers. Classification strategy of TDP could recognize the chemotherapy sensitivity of different NSCLC patients.
BibTeX
@inproceedings{li2025comprehensive,
  author = {Jie Li and Xinyan Pan and Yang Bi and Jian Huang and Jun Peng and CX Bai and Jianhong Hou and Chen Yang},
  title = {{Comprehensive multi-omics analysis of tandem duplicator phenotypes in non-small cell lung cancer}},
  booktitle = {Frontiers in Medicine},
  year = {2025},
  doi = {10.3389/fmed.2025.1556840},
}
Figure