Mitochondrial Epigenetic Alterations Induced by Biomass Smoke Exposure and Their Role in Age-Related Breast Cancer Risk: A Machine Learning-Based Predictive Study

Document Type : Research Articles

Authors

1 Division of Environmental Biotechnology, Genetics & Molecular Biology (EBGMB), ICMR-National Institute for Research in Environmental Health (NIREH), Bhopal, India.

2 Faculty of Medical Research, Academy of Scientific and Innovative Research (AcSIR), Ghaziabad, India.

3 Faculty of Science, Ram Krishna Dharmarth Foundation (RKDF) University, Bhopal, India.

4 Department of Biotechnology, All India Institute of Medical Sciences (AIIMS), New Delhi, India.

Abstract

Objective: To investigate the relationship between age, mitochondrial epigenetics, and BC risk among women exposed to biomass smoke, and the development of a predictive model for BC detection. Methods: A cross-sectional study was conducted among a total of 205 women exposed to biomass smoke and were divided into two age groups (18-25 and >25 years). mtDNA methylation, inflammatory cytokines (IL-6, TNF-α, IL-10), and carcinoembryonic antigen (CEA) levels were assessed. Machine learning models were developed using clinical and molecular data to predict BC risk. Results: Prolonged HAP exposure was assoc to increased mitochondrial dysfunction, particularly in older women. mtDNA methylation changes were significantly correlated with elevated CEA levels, signifies a role in BC risk. Multivariate analysis revealed strong positive correlations between age and inflammatory cytokines: IL-6 (R = 0.95, p < 0.001), TNF-α (R = 0.99, p < 0.000), and IL-10 (R = 0.88, p < 0.005), indicating heightened inflammation with age. Logistic Regression outperform predictive performance with accuracy: 90.18% and AUC: 1.00. Conclusion: Age and mitochondrial epigenetic changes such as mtDNA methylation and inflammatory cytokine levels are strongly linked to BC risk in women exposed to biomass smoke. These results highlight the role of mitochondrial epigenetics in BC and the potential of AI-based tools for early detection in high-risk populations. However, the study’s cross-sectional design limits causal inference, emphasizing the need for longitudinal studies to clarify timing and causality.

Keywords

Main Subjects