Pre-diagnostic derivatives of reactive oxygen metabolites and the occurrence of lung, colorectal, breast and prostate cancer: An individual participant data meta-analysis of two large population-based studies.

Abstract

Oxidative stress may be involved in carcinogenesis and biomarkers of oxidative stress like derivatives of reactive oxygen metabolites (d-ROM) may be useful for cancer prediction. However, no previous study assessed the association of pre-diagnostic d-ROM measurements with cancer incidence. We measured serum d-ROM levels in a cohort sample of n = 4,345 participants of the German ESTHER study and in a case-cohort sample of the Norwegian Tromsø study (cancer cases: n = 941; subcohort: n = 1,000). Moreover, d-ROM was repeatedly measured at follow-ups of both studies. Hazard ratios (HRs) and 95% confidence intervals (95% CIs) were derived by (weighted) multivariable-adjusted Cox regression with time-dependent modeling of d-ROM levels for incident lung, colorectal, breast and prostate cancer. Individual study results were pooled by random effects meta-analysis. The HRs (95% CI) for comparison of top and bottom d-ROM tertile were statistically significant for lung (1.90 [1.25-2.89]), colorectal (1.70 [1.15-2.51]) and breast cancer incidence (1.45 [1.01-2.09]) but not for prostate cancer incidence (1.20 [0.84-1.72]). In conclusion, this individual participant data meta-analysis of two large population-based cohort studies with repeated d-ROM measurements yielded evidence for an involvement of high oxidative stress in carcinogenesis. Given the observed associations of pre-diagnostic d-ROM measurements with lung, colorectal and breast cancer incidence, subjects with increased serum d-ROM levels should be recommended to reduce these levels by lifestyle changes including smoking cessation, a healthy diet and an increase in physical activity.