Cancer
Urinary Heavy Metal Mixtures and Colorectal Cancer Incidence: Pilot Results from the Louisville Environmental Assessments Pilot Study (LEAPS) in Colorectal Cancer Olufunmilayo Babarinde* Olufunmilayo Babarinde Sandy Kavalukas Lu Cai Jianxiang Xu Allie Jin Adhikari Jeevan Daniel Riggs Jeremy Gaskins Kira Taylor Kathy Baumgartner Natalie DuPre
Introduction
Epidemiologic evidence on the association between carcinogenic metal exposures, individual or mixtures, and the risk of colorectal cancer (CRC) is scarce.
Objective
To examine the individual and joint associations between urinary arsenic (As), cadmium (Cd), antimony (Sb), and thallium (Tl) exposures and the odds of CRC incidence in an area with a high CRC incidence and metal-emitting industries.
Methods
In an ongoing case-control study, eligible consented participants included CRC cases diagnosed and receiving surgical treatment and controls screening negative for CRC. Participants provided biospecimens and completed a detailed questionnaire. Urinary metals were measured using inductively coupled plasma mass spectrometry (ICP-MS) in 41 CRC cases and 84 controls and creatinine normalized. The incidence odds ratio (OR) and 95% confidence interval (CI) of CRC were estimated using multivariable logistic regression modeling metal exposures as continuous with a log transformation and binary variables based on the median value of controls. Bayesian Kernel machine regression (BKMR) was used to assess the association between metal mixtures and the odds of CRC. Models were adjusted for age at diagnosis, gender, family history of CRC, and urinary cotinine indicating smoking status.
Results
Median concentrations of metals were higher among cases than controls for As (3.62±9.4 vs. 2.8±3.2ng/mg creatinine), Cd (0.43±0.6 vs. 0.29±0.6ng/mg creatinine), Sb (0.13±0.2 vs. 0.10±0.09ng/mg creatinine), and not Tl (0.11±0.2 vs. 0.15±0.5ng/mg creatinine). A one-unit increase in the natural log of urinary Cd was associated with 90% higher odds of CRC (Adjusted OR: 1.90, 95%CI: 1.12, 3.21); and those with Cd levels above the median had 3.0 times higher odds of CRC (95%CI: 1.2, 7.50) compared to those below the median. In the BKMR modeling, the association between the metal mixture and CRC risk was not statistically significant. However, the posterior inclusion probabilities (PIP) show that Cd is the most commonly included in the mixture for predicting CRC risk (PIP=0.802).
Conclusion
Higher urinary Cd was associated with higher incidence odds of CRC independent of urinary cotinine. Our study findings suggest that interventions mitigating Cd exposure may be beneficial in reducing CRC risk.