Identification and validation of blood leukocyte DNA methylation biomarkers for early detection of colorectal neoplasm.

Abstract

Background: Identifying high-risk populations for colorectal cancer (CRC) is critical for precise screening. This study aimed to develop a novel risk prediction model using blood DNA methylation biomarkers to identify individuals at high risk for colorectal neoplasms.

Methods: The biomarker discovery phase involved 106 samples (56 advanced adenomas and 50 healthy controls) collected from the TARGET-C screening cohort between May 2018 and May 2021, which were analyzed using the Illumina Infinium MethylationEPIC v2.0 BeadChip, and 72 samples (22 CRC, 20 advanced adenomas, and 30 healthy controls) collected from clinical cohorts between July 2023 and July 2024, which were analyzed using reduced representation bisulfite sequencing (RRBS). Differentially methylated positions (DMPs) and regions (DMRs) were identified and independently validated in 147 samples (48 CRC, 50 advanced adenomas, and 49 healthy controls) collected from an independent clinical cohort between June 2022 and May 2024 using targeted bisulfite sequencing (TBS). A multi-marker prediction model was constructed using logistic regression, and its diagnostic performance was evaluated through receiver operating characteristic (ROC) curve analysis.

Results: In the discovery set, 48 DMPs and 74 DMRs were identified, exhibiting significant differences between CRC/advanced adenomas and healthy controls. Of these, three DMPs and 11 DMRs were successfully validated in the independent set using TBS. Through machine learning approaches, five stable methylation markers were identified and incorporated into a multi-marker prediction model. This model demonstrated excellent diagnostic performance for detecting colorectal neoplasms, with an area under the curve (AUC) of 0.85 (95% confidence interval [CI]: 0.74-0.94), outperforming the traditional lifestyle score (AUC = 0.55, 95% CI: 0.46-0.68). Combining methylation markers with lifestyle scores further improved diagnostic accuracy, achieving an AUC of 0.89. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses of the significant markers indicated their involvement in tumorigenesis through pathways regulating developmental processes, transcriptional activation, and cancer-related signaling.

Conclusions: Blood leukocyte DNA methylation markers show significant potential for identifying high-risk populations for CRC. The identified markers could contribute to the development of novel, effective tools for CRC screening, facilitating precision screening strategies.