A study examined the association between World Trade Center site exposure and DNA methylation changes potentially related to cancer risk.
Researchers analyzed DNA methylation patterns in peripheral blood samples from 64 World Trade Center (WTC)-exposed individuals and 32 unexposed controls using the Infinium MethylationEPIC array.
Published in Environmental Epidemiology, the researchers observed global hypermethylation in WTC-exposed participants. Among the top 5,000 differentially methylated CpG sites, hypermethylated sites (β > 0.8) were more prevalent in WTC-exposed cancer-free participants compared to unexposed controls (14.3% vs 4.5%). A similar pattern was seen in breast cancer cases (24.3% vs 6.6%).
Several tumor suppressor genes showed altered methylation in WTC-exposed individuals. In cancer-free women, these included NF1 (β = 0.023, P < 0.001), NOTCH1 (β = 0.048, P < 0.001), PTEN (β = -0.061, P < 0.001), and RUNX1 (β = -0.054, P < 0.001). In breast cancer cases, differentially methylated genes included NOTCH1 (β = 0.08, P < 0.001), PML (β = -0.03, P < 0.001), WRN (β = -0.07, P < 0.001), ATM (β = -0.04, P < 0.001), and TP53 (β = 0.05, P < 0.001).
Gene set overrepresentation analysis revealed enrichment of cancer-related pathways among WTC-exposed participants. In cancer-free women, top enriched pathways included endocytosis (adjusted P = 1.58e-05) and axon guidance (adjusted P = 3.16e-05). For breast cancer cases, enriched pathways included human papillomavirus infection (adjusted P = 5.01e-08) and cGMP-PKG signaling (adjusted P = 1.58e-06).
Network analysis identified a cluster of 47 genes involved in immune regulation and carcinogenesis. This included components of the Wnt/β-catenin signaling pathway (WNT4, TCF7L2) and immune-related genes (JAK3, HDAC1, and IGF2R).
Methodology and Study Population
The study employed various methodologies to control for potential confounders:
- DNA extraction used the PicoPure DNA extraction kit, with bisulfite conversion via the EZ-96 DNA methylation kit.
- Data processing and analysis utilized the R package "minfi," with cell type composition adjustments using "FlowSorted.Blood.EPIC."
- Linear regression models for the top 5,000 differentially methylated cytosine–phosphate–guanine sites were adjusted for age, race/ethnicity, smoking history, body mass index (BMI), education, and white blood cell type composition.
- Functional analyses employed R packages "missMethyl" and "clusterProfiler," with pathway visualization using the KEGG database.
- Gene networks were explored using the STRING 2023 database and visualized with Cytoscape software.
A combined analysis with previous pilot data increased the sample size to 50 WTC-exposed and 40 unexposed cancer-free women, as well as 60 WTC-exposed and 40 unexposed breast cancer cases.
Demographic data showed that among cancer-free participants, WTC-exposed women had a mean age at blood donation of 58.7 years compared to 55.5 years for unexposed women. For breast cancer cases, the mean age was 56.9 years for WTC-exposed and 54.7 years for unexposed participants.
The study population was diverse, with 30.2% non-Hispanic Black and 41.6% non-Hispanic White participants across all groups. Most participants (62.5%) were never-smokers, and the majority fell into overweight (33.3%) or obese (31.2%) BMI categories.
Study limitations included the relatively small sample size, potential differences in sample storage time between WTC-exposed and unexposed blood samples, and the use of postdiagnostic samples for WTC-exposed breast cancer cases compared to prediagnostic samples for unexposed cases.
The authors declared no conflicts of interest.