Cadmium exposure and cancer mortality in a prospective cohort: the strong heart study

Abstract

Background: Cadmium (Cd) is a toxic metal classified as a human carcinogen by the International Agency for Research on Cancer.

Objective: We evaluated the association of long-term Cd exposure, as measured in urine, with cancer mortality in American Indians from Arizona, Oklahoma, and North and South Dakota who participated in the Strong Heart Study during 1989-1991.

Methods: The Strong Heart Study was a prospective cohort study of 3,792 men and women 45-74 years of age who were followed for up to 20 years. Baseline urinary Cd (U-Cd) was measured using inductively coupled plasma mass spectrometry. We assessed cancer events by annual mortality surveillance.

Results: The median (interquintile range) U-Cd concentration was 0.93 (0.55, 1.63) μg/g creatinine. After adjusting for sex, age, smoking status, cigarette pack-years, and body mass index, the adjusted hazard ratios (HRs) comparing the 80th versus the 20th percentiles of U-Cd were 1.30 (95% CI: 1.09, 1.55) for total cancer, 2.27 (95% CI: 1.58, 3.27) for lung cancer, and 2.40 (95% CI: 1.39, 4.17) for pancreatic cancer mortality. For all smoking-related cancers combined, the corresponding HR was 1.56 (95% CI: 1.24, 1.96). Cd was not significantly associated with liver, esophagus and stomach, colon and rectum, breast, prostate, kidney, or lymphatic and hematopoietic cancer mortality. On the basis of mediation analysis, we estimated that the percentage of lung cancer deaths due to tobacco smoking that could be attributed to Cd exposure was 9.0% (95% CI: 2.8, 21.8).

Conclusions: Low-to-moderate Cd exposure was prospectively associated with total cancer mortality and with mortality from cancers of the lung and pancreas. The implementation of population-based preventive measures to decrease Cd exposure could contribute to reducing the burden of cancer.

Figure 1

HRs (95% CIs) for overall, smoking-related, pancreatic, and lung cancer mortality based on restricted cubic splines for ln-transformed U-Cd concentrations with knots at the 10th (0.4 μg/g creatinine), 50th (0.93 μg/g creatinine), and 90th (2.15 μg/g creatinine) percentiles. The reference value is set at the 10th percentile of the Cd distribution. HRs were adjusted for sex, age, smoking status, cigarette pack-years, and BMI. Lines represent the HR (thick line) and 95% CIs (dotted lines). The p-value for the linear and nonlinear components of the dose–response relationship were, respectively, 0.03 and 0.26 for all cancers, 0.02 and 0.25 for smoking-related cancers, 0.02 and 0.09 for pancreatic cancer, and 0.01 and 0.10 for lung cancer. The p-value for the nonlinear component was estimated using the Wald test.

Figure 2

Fully adjusted HRs (95% CIs) for overall (A) and smoking-related (B) cancer mortality comparing the 80th versus the 20th percentile of Cd (μg/g creatinine) by participant characteristics at baseline.