Association between circulating white blood cell count and cancer mortality: a population-based cohort study.

BACKGROUND Inflammatory processes are implicated in the development and progression of cancer. However, it is not clear whether systemic markers of inflammation predict cancer. We examined the prospective relationship between circulating white blood cell (WBC) count and cancer mortality. METHODS Population-based cohort study of 3189 individuals, aged 49 to 84 years and free of cancer at the baseline examination (January 1, 1992, to December 31, 1994), in the Blue Mountains region, west of Sydney, Australia. The main outcome of interest was all cancer mortality ascertained from vital status as of December 31, 2001. RESULTS Higher WBC count was found to be associated with all cancer mortality. In proportional hazards models adjusting for age, sex, education, body mass index, hematocrit level, alcohol consumption, physical inactivity, smoking, weekly aspirin use, diabetes mellitus or fasting hyperglycemia status, and fasting glucose levels, the multivariable relative risk of all cancer mortality comparing quartile 4 of WBC count ( > or = 7400 cells/microL) with quartile 1 (< or = 5300 cells/microL) was 1.73 (95% confidence interval [CI], 1.18-2.55). In subgroup analyses, the relative risk of cancer mortality comparing quartile 4 of WBC count with quartile 1 was higher among those with diabetes or fasting hyperglycemia (3.03 [95% CI, 1.01-9.15]) than among those with normoglycemia (1.68 [95% CI, 1.12-2.52]). Also, the relative risk of cancer mortality associated with joint exposure to quartile 4 of WBC count and aspirin nonuse was 2.42 (95% CI, 1.46-4.01) compared with their absence. CONCLUSION These data provide new epidemiological evidence of an association between circulating WBC count, a widely available marker of inflammation, and subsequent cancer mortality.