Global Burden of Secondhand Smoke-Attributable ER-Negative Breast Cancer and Exploratory Immune-Metabolic Mediation Using Mendelian Randomization.

Abstract

Background: The global burden of breast cancer attributable to secondhand smoke (SHS) and its potential biological pathways, particularly for estrogen receptor (ER)-negative disease, remain incompletely characterized. This study aimed to evaluate temporal trends in SHS-attributable breast cancer burden and to explore whether immune and metabolic traits may represent candidate pathways linking SHS exposure with ER-negative breast cancer.

Methods: We analyzed the Global Burden of Disease 1990-2021 data to estimate temporal trends in SHS-attributable breast cancer burden using Joinpoint regression, Bayesian age-period-cohort modeling, decomposition analysis, and autoregressive integrated moving average forecasting to 2050. Two-sample Mendelian randomization (MR) was performed using 36-38 single-nucleotide polymorphism instruments for exposure to tobacco smoke outside the home. Exploratory two-step MR mediation analyses screened 637 serum metabolites and 731 immune phenotypes using the product-of-coefficients approach.

Results: Globally, absolute SHS-attributable breast cancer deaths increased from 4,949 in 1990 to 7,518 in 2021, and disability-adjusted life years (DALYs) increased from 1.74×10⁵ to 2.57×10⁵. In contrast, age-standardized death and DALY rates declined, with estimated annual percentage changes of -1.31% and -1.22% per year, respectively. Marked disparities were observed across Socio-Demographic Index regions. Projection analyses suggested a plateau around 2030, followed by widening uncertainty. Decomposition analysis indicated that population growth and population aging were major contributors to the increasing absolute burden. In MR analysis, genetically proxied SHS exposure outside the home showed a nominal positive association with ER-negative breast cancer in the inverse variance-weighted model (β = 0.9976; odds ratio = 2.71; 95% confidence interval = 1.08-6.82), although alternative MR estimators showed wider and less consistent estimates. Exploratory two-step MR identified lipid- and energy-related metabolites, including 10-nonadecenoate, eicosenoate, 3-hydroxybutyrate, and dihomo-linoleate 20:2n6, as candidate metabolic signals. A myeloid antigen-presentation phenotype, HLA-DR on CD33dim HLA-DR+ CD11b- cells, was also prioritized as a potential immune-related signal. However, the indirect-effect estimates were imprecise and should be interpreted as exploratory rather than confirmatory evidence of mediation.

Conclusion: The age-standardized burden of SHS-attributable breast cancer has declined globally, whereas absolute deaths and DALYs have increased due to demographic growth and aging, with persistent SDI-linked disparities. MR analyses provided suggestive genetic-instrument evidence linking SHS exposure outside the home with ER-negative breast cancer, and exploratory mediation analyses prioritized lipid metabolism, ketone-related energy pathways, and myeloid immune phenotypes for future validation. These findings support continued smoke-free policies and highlight the need for larger, diverse, and mechanistically informed studies to validate the potential immune-metabolic pathways implicated in SHS-related ER-negative breast cancer.