Association between methylation quantitative trait loci and colorectal cancer risk, survival and cancer recurrence

Ines Mesa Eguiagaray; Andrii Iakovliev; Xue Li; Maria Timofeeva; Yazhou He; Xiaomeng Zhang; Farhat V N Din; Susan M Farrington; Athina Spiliopoulou; Malcolm G Dunlop; Evropi Theodoratou

doi:10.1038/s41416-025-03064-8

Association between methylation quantitative trait loci and colorectal cancer risk, survival and cancer recurrence

Ines Mesa Eguiagaray^*, Andrii Iakovliev, Xue Li, Maria Timofeeva, Yazhou He, Xiaomeng Zhang, Farhat V N Din, Susan M Farrington, Athina Spiliopoulou, Malcolm G Dunlop, Evropi Theodoratou

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

Abstract

Background: Epigenetic changes contribute to colorectal cancer (CRC) pathogenesis. We investigated whether methylation quantitative trait loci (mQTLs) are associated with CRC risk, survival and recurrence.

Methods: Using a well-characterised Scottish case-control study (6,821 CRC cases, 14,692 controls), we derived 118,982 mQTLs based on the Genetics of DNA Methylation Consortium (GoDMC). Association analysis between mQTLs and CRC risk, survival and recurrence was performed using logistic regression or Cox models respectively. Additionally, colocalisation analysis was performed.

Results: 19 mQTLs within 10 distinct genomic regions were associated with CRC risk. Two novel regions were mapped to MDGA2 (p value=2.97𝑥10−6 ) and STARD3 (p value=5.58𝑥10−6 ). Four regions mapped to POU5F1B, POU2AF2(c11orf53)/POU2AF3(COLCA2), GREM1 and CABLES2 were previously identified. Four regions mapped to PPA2, PANDAR/LAP3P2, POU6F1 and CTIF contained SNPs previously identified by CRC GWAS but with SNPs annotated to different genes. We found no evidence that any of the 19 mQTLs associated with CRC risk influenced survival or recurrence after FDR correction. Colocalisation analysis suggested that in three of the ten regions the causal variants were shared for methylation and CRC risk.

Conclusion: This study adds to the repertoire of CRC genes. However, we found no associations between methylation and CRC survival or recurrence.

Original language	English
Number of pages	10
Journal	British Journal of Cancer
Early online date	12 Jun 2025
DOIs	https://doi.org/10.1038/s41416-025-03064-8
Publication status	E-pub ahead of print - 12 Jun 2025

Access to Document

10.1038/s41416-025-03064-8Licence: Creative Commons: Attribution (CC-BY)

Association between methylation_EGUIAGARAY_DOA28042025_VOR_CC-BYFinal published version, 1.18 MBLicence: Creative Commons: Attribution (CC-BY)

https://www.nature.com/articles/s41416-025-03064-8Licence: Creative Commons: Attribution (CC-BY)

Colorectal cancer reduction through risk stratification of screening, follow-up and treatment
Theodoratou, E. (Principal Investigator)
Cancer Research UK
1/05/17 → 31/08/25
Project: Research
Integrative Genomics in Colorectal Cancer Susceptibility:Developing risk reducing interventions through understanding biology
Dunlop, M. (Principal Investigator), Campbell, H. (Co-investigator), Farrington, S. (Co-investigator) & Theodoratou, E. (Co-investigator)
UK-based charities
1/01/16 → 31/05/21
Project: Research

Cite this

Mesa Eguiagaray, I., Iakovliev, A., Li, X., Timofeeva, M., He, Y., Zhang, X., Din, F. V. N., Farrington, S. M., Spiliopoulou, A., Dunlop, M. G., & Theodoratou, E. (2025). Association between methylation quantitative trait loci and colorectal cancer risk, survival and cancer recurrence. British Journal of Cancer. Advance online publication. https://doi.org/10.1038/s41416-025-03064-8

@article{11a2e55320e04d3e8fdfc0318f25691a,

title = "Association between methylation quantitative trait loci and colorectal cancer risk, survival and cancer recurrence",

abstract = "Background: Epigenetic changes contribute to colorectal cancer (CRC) pathogenesis. We investigated whether methylation quantitative trait loci (mQTLs) are associated with CRC risk, survival and recurrence. Methods: Using a well-characterised Scottish case-control study (6,821 CRC cases, 14,692 controls), we derived 118,982 mQTLs based on the Genetics of DNA Methylation Consortium (GoDMC). Association analysis between mQTLs and CRC risk, survival and recurrence was performed using logistic regression or Cox models respectively. Additionally, colocalisation analysis was performed. Results: 19 mQTLs within 10 distinct genomic regions were associated with CRC risk. Two novel regions were mapped to MDGA2 (p value=2.97𝑥10−6 ) and STARD3 (p value=5.58𝑥10−6 ). Four regions mapped to POU5F1B, POU2AF2(c11orf53)/POU2AF3(COLCA2), GREM1 and CABLES2 were previously identified. Four regions mapped to PPA2, PANDAR/LAP3P2, POU6F1 and CTIF contained SNPs previously identified by CRC GWAS but with SNPs annotated to different genes. We found no evidence that any of the 19 mQTLs associated with CRC risk influenced survival or recurrence after FDR correction. Colocalisation analysis suggested that in three of the ten regions the causal variants were shared for methylation and CRC risk. Conclusion: This study adds to the repertoire of CRC genes. However, we found no associations between methylation and CRC survival or recurrence.",

author = "\{Mesa Eguiagaray\}, Ines and Andrii Iakovliev and Xue Li and Maria Timofeeva and Yazhou He and Xiaomeng Zhang and Din, \{Farhat V N\} and Farrington, \{Susan M\} and Athina Spiliopoulou and Dunlop, \{Malcolm G\} and Evropi Theodoratou",

year = "2025",

month = jun,

day = "12",

doi = "10.1038/s41416-025-03064-8",

language = "English",

journal = "British Journal of Cancer",

issn = "0007-0920",

publisher = "Springer Nature",

}

TY - JOUR

T1 - Association between methylation quantitative trait loci and colorectal cancer risk, survival and cancer recurrence

AU - Mesa Eguiagaray, Ines

AU - Iakovliev, Andrii

AU - Li, Xue

AU - Timofeeva, Maria

AU - He, Yazhou

AU - Zhang, Xiaomeng

AU - Din, Farhat V N

AU - Farrington, Susan M

AU - Spiliopoulou, Athina

AU - Dunlop, Malcolm G

AU - Theodoratou, Evropi

PY - 2025/6/12

Y1 - 2025/6/12

N2 - Background: Epigenetic changes contribute to colorectal cancer (CRC) pathogenesis. We investigated whether methylation quantitative trait loci (mQTLs) are associated with CRC risk, survival and recurrence. Methods: Using a well-characterised Scottish case-control study (6,821 CRC cases, 14,692 controls), we derived 118,982 mQTLs based on the Genetics of DNA Methylation Consortium (GoDMC). Association analysis between mQTLs and CRC risk, survival and recurrence was performed using logistic regression or Cox models respectively. Additionally, colocalisation analysis was performed. Results: 19 mQTLs within 10 distinct genomic regions were associated with CRC risk. Two novel regions were mapped to MDGA2 (p value=2.97𝑥10−6 ) and STARD3 (p value=5.58𝑥10−6 ). Four regions mapped to POU5F1B, POU2AF2(c11orf53)/POU2AF3(COLCA2), GREM1 and CABLES2 were previously identified. Four regions mapped to PPA2, PANDAR/LAP3P2, POU6F1 and CTIF contained SNPs previously identified by CRC GWAS but with SNPs annotated to different genes. We found no evidence that any of the 19 mQTLs associated with CRC risk influenced survival or recurrence after FDR correction. Colocalisation analysis suggested that in three of the ten regions the causal variants were shared for methylation and CRC risk. Conclusion: This study adds to the repertoire of CRC genes. However, we found no associations between methylation and CRC survival or recurrence.

AB - Background: Epigenetic changes contribute to colorectal cancer (CRC) pathogenesis. We investigated whether methylation quantitative trait loci (mQTLs) are associated with CRC risk, survival and recurrence. Methods: Using a well-characterised Scottish case-control study (6,821 CRC cases, 14,692 controls), we derived 118,982 mQTLs based on the Genetics of DNA Methylation Consortium (GoDMC). Association analysis between mQTLs and CRC risk, survival and recurrence was performed using logistic regression or Cox models respectively. Additionally, colocalisation analysis was performed. Results: 19 mQTLs within 10 distinct genomic regions were associated with CRC risk. Two novel regions were mapped to MDGA2 (p value=2.97𝑥10−6 ) and STARD3 (p value=5.58𝑥10−6 ). Four regions mapped to POU5F1B, POU2AF2(c11orf53)/POU2AF3(COLCA2), GREM1 and CABLES2 were previously identified. Four regions mapped to PPA2, PANDAR/LAP3P2, POU6F1 and CTIF contained SNPs previously identified by CRC GWAS but with SNPs annotated to different genes. We found no evidence that any of the 19 mQTLs associated with CRC risk influenced survival or recurrence after FDR correction. Colocalisation analysis suggested that in three of the ten regions the causal variants were shared for methylation and CRC risk. Conclusion: This study adds to the repertoire of CRC genes. However, we found no associations between methylation and CRC survival or recurrence.

U2 - 10.1038/s41416-025-03064-8

DO - 10.1038/s41416-025-03064-8

M3 - Article

SN - 0007-0920

JO - British Journal of Cancer

JF - British Journal of Cancer

ER -

Association between methylation quantitative trait loci and colorectal cancer risk, survival and cancer recurrence

Abstract

Access to Document

Fingerprint

Projects

Colorectal cancer reduction through risk stratification of screening, follow-up and treatment

Integrative Genomics in Colorectal Cancer Susceptibility:Developing risk reducing interventions through understanding biology

Cite this