TL;DR
This study investigates how different molecular layers (genomic, epigenomic, transcriptomic) contribute to intratumoral heterogeneity (ITH) in esophageal squamous cell carcinoma (ESCC). By analyzing multi-omic data from 36 patients, they reveal that chromosomal instability fuels ITH across these layers and that the tumor's immune microenvironment actively shapes this evolution, leading to immune evasion mechanisms.
What Problem Does This Paper Address?
Intratumoral heterogeneity (ITH), the variation of molecular features within a single tumor, is a major hurdle in cancer treatment, often leading to drug resistance and treatment failure. While ITH has been studied at genomic, transcriptomic, and epigenomic levels in various cancers, a comprehensive, multi-omic understanding of ITH in esophageal squamous cell carcinoma (ESCC) has been lacking. Furthermore, the precise role of tumor-infiltrated T lymphocytes (TILs) as a selective force driving this heterogeneity and shaping cancer evolution in ESCC remained unclear. This paper aims to bridge this gap by integrating multi-omic data to dissect the complex interplay of factors contributing to ESCC ITH and its relationship with the immune microenvironment.
Methods
The researchers employed a multi-pronged approach, integrating whole-exome sequencing (WES), reduced representation bisulfite sequencing (RRBS) for DNA methylation analysis, and RNA sequencing (RNA-seq) on 186 samples from 36 primary ESCC patients.
- Multi-omic Sequencing: WES provided detailed genomic information, RRBS captured genome-wide methylation patterns, and RNA-seq revealed gene expression profiles.
- Genomic ITH Analysis: Somatic mutations and copy-number alterations (SCNA) were classified as clonal or subclonal. Tumor phylogenies were constructed to infer evolutionary patterns (branched vs. linear). Chromosomal instability was assessed using the Weighted Genome Instability Index (wGII) and by identifying genome doubling (GD) events and mirrored subclonal allelic imbalance (MSAI).
- Epigenomic ITH Analysis: DNA methylation levels were analyzed across the genome, with a focus on differentially methylated regions (DMRs). An average pairwise ITH (APITH) metric was developed to quantify methylation heterogeneity.
- Transcriptomic ITH Analysis: Genes were classified into four classes based on their intra- and inter-tumor heterogeneity scores. RNA heterogeneity was correlated with genomic and epigenomic features.
- Immune Infiltration Analysis: RNA-seq data was used to estimate immune cell populations using the Danaher method and validated with ConsensusTME. Immune infiltration levels were classified as high, low, or heterogeneous.
- Integration and Correlation: The study extensively correlated findings across different omics layers, investigating how genomic and epigenomic alterations influence gene expression and how immune infiltration impacts these processes. Mechanisms of neoantigen evasion, including HLA class I allele loss (HLA LOH), were also investigated.
Datasets
The study utilized a prospectively collected cohort of 186 multi-regional samples from 36 primary ESCC patients. Adjacent normal esophageal tissue was also collected. The data generated includes:
- Whole-Exome Sequencing (WES) data
- Reduced Representation Bisulfite Sequencing (RRBS) data
- RNA Sequencing (RNA-seq) data
The raw sequence data has been deposited in the European Genome-phenome Archive (EGA) under accession code EGAS00001003832.
Key Insights and Ideas
- Chromosomal Instability as a Universal Driver: The study strongly suggests that ongoing chromosomal instability (CIN) is a fundamental driver of ITH across genomic, epigenomic, and transcriptomic levels in ESCC. This is a unifying concept that connects diverse molecular alterations.
- Genome Doubling Fuels SCNA: The presence of genome doubling (GD) events was significantly associated with subclonal copy-number alterations, providing a direct link between a major genomic event and the generation of subclonal genomic diversity.
- Epigenetic Instability Mirrors Genomic Instability: Regions with higher chromosomal instability showed increased entropy and PDR (proportion of discordant reads), indicating that CIN can promote epigenetic instability and contribute to epigenetic ITH.
- Immune Microenvironment as a Selective Pressure: The study provides compelling evidence that the immune microenvironment is not just a bystander but an active selective force. Diverse immune infiltration levels within and between tumors correlate with specific molecular evolutionary patterns, suggesting immune cells are shaping tumor evolution.
- Neoantigen Evasion is Multi-faceted: The research highlights that cancer cells employ multiple strategies to evade immune recognition, including neoantigen heterogeneity and HLA class I allele loss (HLA LOH), particularly in the context of high immune infiltration.
- Intertumoral Heterogeneity Dominates: Across all omics levels, the study found that intertumoral heterogeneity (differences between tumors from different patients) significantly exceeded intratumoral heterogeneity (differences within a single tumor). This implies that each tumor has a distinct molecular signature.
Key Results
- High Subclonal Proportion: A median of 33.3% of somatic mutations and 55.1% of SCNA were found to be subclonal, underscoring the significant intra-tumor diversity.
- Branched Evolution is Predominant: 92% of ESCC tumors exhibited a branched evolutionary pattern, characterized by coexisting subclones, rather than a linear progression.
- GD and SCNA Link: A significant association was found between genome doubling and subclonal copy-number alterations (P = 4.33×10⁻⁸).
- Epigenetic ITH Correlates with SCNA: Pairwise DNA methylation distances were positively correlated with pairwise SCNA distances (Spearman's rho = 0.63, P = 4.46×10⁻⁵²).
- RNA ITH and SCNA Link: A positive correlation was observed between median RNA-ITH score and SCNA-ITH (Spearman's rho = 0.38, P = 0.027).
- Immune Infiltration and HLA Expression: High HLA gene expression showed a significant positive association with total immune infiltration, CD8 T cells, and cytotoxic cells.
- HLA LOH and Immune Evasion: Nine out of 36 patients showed clonal HLA LOH, and two had subclonal HLA LOH, suggesting it's a prevalent immune evasion mechanism. HLA LOH was also associated with higher expression of TIL markers.
- Intertumoral Heterogeneity Outperforms Intratumoral: Significant differences were observed in Euclidean distances between intra-patient and inter-patient tumor pairs across SCNA, gene expression, and DNA methylation (P < 2.2×10⁻¹⁶ for all).
Limitations and Future Work
- Sample Size for Linear Evolution: The study notes that only 3 out of 36 patients showed a linear evolutionary pattern, limiting definitive conclusions about its clinical implications, though it suggests monotherapies might be more effective in these cases.
- Immune Cell Heterogeneity and RNA-ITH: While a trend was observed, the correlation between immune-cell ITH and RNA-ITH was not statistically significant (R² = -0.27, P = 0.17), suggesting that other factors, like tumor purity, might also play a role in RNA-ITH.
- Future Directions: The authors suggest that their findings provide an avenue for dissecting the complex contributions of multi-omics to ITH in ESCC, which can enhance the development of clinical therapies. The detailed multi-omic characterization and the insights into immune selection pressures could inform more targeted and effective treatment strategies. The study also highlights the insufficiency of single-site sampling for clinical diagnosis in ESCC.
Summary updated