Abstract:Phenotypic plasticity is thought to assist immune evasion, metastasis, and therapeutic resistance in tumor evolution. Epigenetic mechanisms such as DNA methylation provide for a more easily acquired and flexible, yet semi-stable adaptation by the tumor cell than hardwired genetic changes. Cancer-associated gain of DNA methylation at promoter CpG islands or distal regulatory elements is associated with transcriptional silencing, whereas loss of DNA methylation is observed at late-replicating, lamina-attached regions of the genome, tracking with cumulative cell divisions. One of the most well-documented genomic features predisposing to DNA hypermethylation is Polycomb Repressive Complex (PRC) occupancy in precursor cells, consistent with an epigenetic block to cellular differentiation as an early or predisposing event in carcinogenesis. A central question is whether this epigenetic switch from reversible PRC repression to more stable DNA hypermethylation occurs before onset of clonal expansion, or whether it represents an ongoing process during tumor evolution. Clonal expansion is a pivotal characteristic of cancer, and is thought to be initiated by a genetic alteration in a key gatekeeper driver gene. However, not all normal cells appear to be susceptible to malignant transformation following such an event. Multiple lines of evidence suggest that epigenetic heterogeneity among normal cells may affect their cancer-initiating potential. We have developed a tumor processing pipeline to deeply and comprehensively map DNA methylation patterns in primary human colorectal cancer (CRC) and adjacent tissues, while preserving spatial information, enabling inference of the temporal order of molecular alterations. We employ a suite of technologies to map the spatial heterogeneity of DNA methylation patterns in primary human CRC. These include a deep single-cell whole-genome bisulfite sequencing technique that improves genomic coverage several-fold over existing published protocols, providing DNA methylation measurements at the majority of CpGs in the genome in individual tumor cells. We have also calibrated and validated an epigenetic clock that allows us to accurately measure the relative replicative history of individual cells. We infer phylogenetic relationships among subclones, and relate these to their spatial distribution in the primary tumor. We find that some tumors have high proportions of PRC-associated clonal hypermethylation, indicating that these hypermethylation events were already present prior to clonal expansion, possibly contributing to susceptibility to malignant transformation in the cell-of-origin, whereas other tumors appear to have active PRC-associated CpG island hypermethylation ongoing during tumor expansion, generating subclonal heterogeneity, which may influence post-therapeutic recurrence potential. Citation Format: Peter W. Laird, Nathan J. Spix, Hsiao-yun Milliron, Manpreet Kalkat, Emily Eugster, David Sokol, Emily Jung, Paula Nolte, Kelly K. Krzyzanowski, Toshinori Hinoue, Hui Shen. Spatial reconstruction and temporal inference of DNA methylation alterations in primary human colorectal cancer [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Translating Cancer Evolution and Data Science: The Next Frontier; 2023 Dec 3-6; Boston, Massachusetts. Philadelphia (PA): AACR; Cancer Res 2024;84(3 Suppl_2) nr IA023.

Predicting DNA methylation from genetic data lacking racial diversity using shared classified random effects

Abstract 4288: Predicting DNA methylation in cervical cancer using somatic mutations in a classified mixed model prediction

Racial variation in breast tumor promoter methylation in the Carolina Breast Cancer Study

Advancing epigenetic profiling in cervical cancer: machine learning techniques for classifying DNA methylation patterns

A new parsimonious method for classifying Cancer Tissue-of-Origin Based on DNA Methylation 450K data

Accurate Prediction of Pan-Cancer Types Using Machine Learning with Minimal Number of DNA Methylation Sites.

A Sparse and Low-Rank Regression Model for Identifying the Relationships Between DNA Methylation and Gene Expression Levels in Gastric Cancer and the Prediction of Prognosis

A Bayesian hierarchical model to estimate DNA methylation conservation in colorectal tumors

Robust prediction of gene regulation in colorectal cancer tissues from DNA methylation profiles

Abstract IA023: Spatial reconstruction and temporal inference of DNA methylation alterations in primary human colorectal cancer

DNA methylation data–based molecular subtype classification related to the prognosis of patients with cervical cancer

Abstract A002: In-silico screening of race-specific endometrial cancer methylation markers using machine learning techniques

Predicting genome-wide DNA methylation using methylation marks, genomic position, and DNA regulatory elements

Prognostic evaluation of the novel blueprint of DNA methylation sites by integrating bulk RNA-sequencing and methylation modification data in endometrial cancer

Risk scoring based on DNA methylation-driven related DEGs for colorectal cancer prognosis with systematic insights

PretiMeth: Precise Prediction Models for DNA Methylation Based on Single Methylation Mark

Predicting DNA methylation level across human tissues

Abstract C114: DNA methylation signatures linked to different grades of prostate tumors from African American and European American patients

Predict disease-related RNA methylation sites from the methylation-expression association by using hypergeometric test

DNA methylation arrays as surrogate measures of cell mixture distribution

A novel method for detecting association between DNA methylation and diseases using spatial information