Abstract:
http://jco.ascopubs.org/cgi/content/abstract/27/21/3437
Editorial:
http://jco.ascopubs.org/cgi/content/full/27/21/3418
Supplementary editorial provided by OncologySTAT
TAKE-HOME MESSAGE
Biomarkers associated with inflammation, C-reactive protein and serum amyloid A, are independent prognostic markers for breast cancer survival.
STUDY IN CONTEXT
Persistent cytokine production by liver cells induces chronic inflammation, which is a known mediator of tumor development and progression. Two hepatic proteins, C-reactive protein (CRP) and serum amyloid A (SAA), are secreted in response to cytokine secretion. Both CRP and SAA are biomarkers for low-grade chronic inflammation and presumably also of cancer risk. Both CRP and SAA have been associated with poor survival in several types of cancers. In particular, chronic inflammation appears to be involved in mammary tumor development. However, the relationship between CRP and SAA and breast cancer survival has not been investigated. CRP has been studied in relationship to tumor burden and breast cancer progression, but SAA has never been evaluated in terms of breast cancer outcomes. The aim of this study by Pierce et al was to determine the prognostic significance of circulating CRP and SAA levels in breast cancer survivors at 31 months after diagnosis. The analysis involved 734 women enrolled in the Health, Eating, Activity, and Lifestyle (HEAL) Study. The HEAL study is a prospective cohort study of breast cancer survivors who were recruited through Surveillance, Epidemiology, and End Results (SEER) registries. Among the study cohort (mean age at 24 months, 57.5 years), 23.3%, 54.5%, and 22.3% of women had in situ, localized, and regional stage disease, respectively. As of September 30, 2004, after a median follow-up period of 6.9 years for overall survival (OS) and 4.1 years for disease-free survival (DFS), 88 deaths and 91 DFS events had occurred.
Higher levels of SAA were significantly associated with shorter OS (P for trend < .0001). Hazard ratios (HRs) among the SAA tertiles indicated a threshold effect rather than a dose-dependant response. In a model adjusted for age, disease stage, race/study site, and body mass index, no significant difference in SAA was observed between the lowest and middle tertiles (hazard ratio [HR], 0.98; 95% CI, 0.50-1.90). However, between the lowest and highest SAA tertiles (SAA <4.2 mg/L vs >8.1 mg/L), a significant difference in OS was observed (HR, 3.08; 95% CI, 1.73-5.47). Further adjustment for hormone receptor status resulted in no significant change in the interaction. SAA levels had borderline associations with reduced DFS (P for trend = .07).
Elevated CRP was significantly associated with shorter OS only for the highest tertile (>3.9 mg/L), and to a lesser extent than with SAA. CRP was associated with DFS in a dose-dependant manner. In a model adjusted for age, disease stage, race/study site, body-mass index, estrogen receptor/progesterone receptor status, and cardiovascular events, the HR for recurrence was 1.91 (95% CI, 1.04-3.51; P for trend = .04).
Further adjustment for cardiovascular comorbidity (which is correlated with inflammation) slightly diminished the effect for all interactions between survival and SAA or CRP, suggesting that this is a potentially confounding factor.
In conclusion, significant associations between shortened OS and elevated concentrations of SAA and CRP were observed when these inflammation biomarkers were measured at approximately 31 months after diagnosis. Elevated CRP was also associated with shortened DFS. The associations between elevated CRP and SAA levels and survival were independent of other factors. Therefore, chronic systemic inflammation may be an important prognostic indicator of long-term survival in women with breast cancer.
Hopeful
Linear Mode
