Association between Levels of C-Reactive Protein and Leukocytes and Cancer: Three Repeated Measurements in the Swedish AMORIS Study

作者：Van Hemelrijck Mieke^*; Holmberg Lars; Garmo Hans; Hammar Niklas; Walldius Goran; Binda Elisa; Lambe Mats; Jungner Ingmar

来源：Cancer Epidemiology Biomarkers & Prevention, 2011, 20(3): 428-437.

DOI：10.1158/1055-9965.EPI-10-1190

Objective: To study levels of C-reactive protein (CRP) and leukocytes, as inflammatory markers, in the context of cancer risk.
Methods: From the Apolipoprotein MOrtality RISk (AMORIS) study, we selected 102,749 persons with one measurement and 9,273 persons with three repeated measurements of CRP and leukocytes. Multivariate Cox proportional hazards regression was applied to categories of CRP (< 10, 10-15, 15-25, 25-50, > 50 g/L) and quartiles of leukocytes. An inflammation-based predictive score (IPS) indicated whether someone had CRP levels of more than 10 mg/L combined with leukocytes of more than 10 x 10(9) /L. Reverse causality was assessed by excluding those with less than 3, 5, or 7 years of follow-up. To analyze repeated measurements of CRP and leukocytes, the repeated IPS (IPS(r)) was calculated by adding the IPS of each measurement.
Results: In the cohort with one measurement, there was a positive trend between CRP and risk of developing cancer, with the lowest category being the 0.99 (0.92-1.06), 1.28 (1.11-1.47), 1.27 (1.09-1.49), and 1.22 (1.01-1.48) for the second to fifth categories, respectively. This association disappeared when excluding those with follow-up of less than 3, 5, or 7 years. The association between leukocytes and cancer was slightly stronger. In the cohort with repeated measurements, the IPSr was strongly associated with cancer risk: 1.87 (1.33-2.63), 1.51 (0.56-4.06), and 4.46 (1.43-13.87) for IPSr 1, 2, and 3 compared with IPS(r) = 0. The association remained after excluding those with follow-up of less than 1 year.
Conclusions and Impact: Our large, prospective cohort study adds evidence for a link between inflammatory markers and cancer risk by using repeated measurements and ascertaining reverse causality. Cancer Epidemiol Biomarkers Prev; 20(3); 428-37.

• 出版日期2011-3