Circulating C-Reactive Protein And Breast Cancer Risk ñ Meta-Analysis Of Prospective Studies
Doris CHAN, School of Public Health, Imperial College London, United Kingdom
BANDERA E. 2
,
GREENWOOD D. 3
,
NORAT T. 1
1 Department of Epidemiology and Biostatistics, School of Public Health, Imperial College London, London, United Kingdom
2 Rutgers Cancer Institute of New Jersey, Rutgers, The State University of New Jersey, New Jersey, United States
3 Leeds Institute of Cardiovascular and Metabolic Medicine, University of Leeds, Leeds, United Kingdom
Purpose
To explore the magnitude and the shape of the association between circulating C-reactive protein (CRP), a low-grade inflammation biomarker, and breast cancer risk in women through systematic literature review and meta-analysis.
Methods
Relevant publications from prospective studies on the association were identified in PubMed and Web of Science up to February 2015. Random-effects linear dose-response meta-analyses were conducted to calculate the summary relative risks (RRs) for breast cancer, overall and in postmenopausal women, and second order fractional polynomial models were used to examine any potential non-linear relationship.
Results
Overall twelve studies on any breast cancer (3 522 cases), and nine studies on postmenopausal breast cancer (2 516 cases) could be included in the meta-analyses. For each doubling of CRP concentration, the summary RRs were 1.07 (95% confidence interval (CI)=1.02–1.12; I2=47%, P heterogeneity=0.04) and 1.06 (95% CI=1.01–1.11; I2=32%, P heterogeneity=0.17), respectively. Positive associations remained in the studies that examined reverse causation by excluding the cases diagnosed in the early years of follow-up.
Subgroup analyses showed similar summary RRs in the studies adjusted or not adjusted for body mass index, but the associations attenuated in the studies adjusted for other lifestyle factors (smoking, physical activity, and alcohol use).
Although the test for departure from linearity was statistically significant (P non-linearity=0.01 overall and <0.001 in postmenopausal women), the associations appeared linear over most of the range of CRP concentrations. For postmenopausal women, the increase in risk was sharper and tailed off after 4 mg/L, possibly because of limited data points after this value.
Conclusions
Low-grade inflammation may be associated with breast cancer risk. More studies are needed to clarify the confounding factors in the association.
Funding source
This study was funded by the World Cancer Research Fund (WCRF) International (Grant number: 2007/SP01).