A Cohort Study of Systemic Markers of Inflammation and Oxidative Stress and Incidence of Esophageal Adenocarcinoma

Abstract

Esophageal adenocarcinoma (EA) incidence has increased dramatically in the Western world over several decades, while survival remains poor. Persons with Barrett's esophagus (BE) experience a higher risk for progression to EA; they are typically followed in long-term surveillance programs involving periodic endoscopy with biopsies so as to identify early-stage cancers. Currently, no medical, surgical or lifestyle interventions have been observed to safely lower EA risk. Recent studies have shown that elevated pre-diagnostic levels of serum inflammation markers may be predictive of cancers of the breast, colon, and lung, but their rolein predicting EA is unknown. In this dissertation project, we investigated whether elevated markers of inflammation, including C-reactive protein (CRP), interleukin-6 (IL6), and soluble tumor necrosis factor receptors I & II (sTNF-RI & sTNF-RII), and markers of oxidative stress, including F2-isoprostanes, could predict progression to EA in the Seattle Barrett's Esophagus Study (SBES), a prospective cohort of 397 BE patients 45 of whom developed EA. We also assessed the intra-individual variability and reliability of these inflammation markers. Additionally, we evaluated the correlates of telomere length in a subset of 234 persons from the SBES cohort. We observed that persons with CRP levels above the median value of 1.9 mg/L were at a two-fold increased risk for EA compared to those below (adjusted HR 1.77; 95% CI 0.93-3.37, p-trend for continuous CRP=0.04). Persons with IL6 levels above the median had a two-fold increased risk for EA (age- and gender-adjusted HR 1.95; 95%CI 1.03-3.72) but no significant trend was observed (p-trend = 0.94). No evidence of an association was found between EA risk and elevated levels of sTNF receptors or F2-isoprostanes. Analyses restricted to men revealed slightly stronger associations, but the overall conclusions remained the same. In a reliability study involving a subset of 360 participants from the SBES, we observed that the reliability over time, evaluated as intra-class correlations (ICCs), was excellent for sTNF receptors (ICCsTNF-RI=0.89, ICCsTNF-RII=0.85) and fair to good for CRP and IL-6 (ICCCRP=0.55, ICCIL-6=0.57). Moreover, the ICCs for CRP & IL-6 were lower among samples stored for over 13 years prior to laboratory analysis compared to samples stored for less than 13 years, but those for sTNF receptors were unaffected by storage time. In a cross-sectional analysis to assess the correlates of leukocyte telomere length (LTL), age, gender and cigarette pack-years of smoking were significantly associated with LTL. We observed that elevated sTNF-RI levels were associated with short telomeres (Adjusted OR comparing extreme tertiles = 2.19, 95% CI 1.00-4.85, p-trend for continuous sTNF-RI = 0.02). There were no significant associations observed between short LTL and higher levels of other inflammation markers, including CRP, IL-6, sTNF-RII, and F2-isoprostanes. We also did not find any association between short LTL and obesity or obesity-related biomarkers including leptin, adiponectin, glucose and insulin. Our findings suggest that systemic inflammation markers, including CRP and possibly IL-6, can predict progression to EA among persons with BE. Continued follow-up of this and other larger cohorts is needed to further understand the relationship between inflammation, telomeres and cancer, and possibly evaluate inflammation markers as tools for clinical risk stratification in persons with BE.