Role of Inflammation in Barrett's Esophagus-Related Cancer Development.
Maria McIntire, Carissa A Sanchez, David S Cowan, Brian Reid, Pui Y Fong, Patricia L Blount, Robert D Odze. Brigham and Women's Hospital, Boston; Fred Hutchinson Cancer Center, Seattle
Background: Cancer in Barrett's esophagus (BE) develops via a metaplasia-dysplasia-carcinoma sequence, which is initiated by reflux-induced inflammation and associated epithelial damage. However, the role of inflammation in the development of cancer in BE has never been evaluated. The aim of this case-control study was to determine whether the degree, and type, of inflammation in the esophagus is related to cancer risk in BE.
Design: All esophageal mucosal biopsies (N=3,240) obtained from endoscopies at baseline until the patient's final outcome (cancer versus no cancer) were evaluated from 38 BE patients enrolled in a long term prospective surveillance program (mean follow up: cases; 24.4 months, controls; 56.7 months). Cases consisted of 19 patients who developed cancer (mean age: 68 years, M/F ratio:18/1, mean BE length 5.8 cm, mean # biopsies/patient:92) and 19 without cancer (mean age: 68 years, M/F ratio:18/1, mean BE length 5.0 cm, mean # biopsies/patient:83). The biopsies were evaluated in a blinded fashion for the mean # of eosinophils and the mean # of mononuclear cells per high power field (HPF) in the lamina propria, and the degree of active inflammation, measured on a four-point scale (0=none, 1=<50% of mucosa involved, 2=>50% of mucosa, 3=ulceration). The results were compared between cases and controls.
Results: Overall, the mean # of mononuclear cells/HPF and mean # of eosinophils/HPF were statistically similar in the cases (33.2±24.8, 4.7±4.7, respectively) and controls (30.4±24.6, 5.4±5.7, respectively) (p>0.05 for both comparisons). No differences were noted in the presence, or degree, of active inflammation between the cases (mean score: 0.83±0.77) and controls (mean score 0.86±0.75, p>0.05). The degree of eosinophilic and mononuclear inflammation, and the activity scores, did not differ significantly in individual biopsies that contained BE without dysplasia compared to those with either low or high-grade dysplasia, in either of the two patient groups.
Conclusions: In this case-control study, the degree of mononuclear, eosinophilic, and active inflammation were not risk factors for cancer development in BE. Once inflammation-induced epithelial damage and columnar metaplasia has occurred, further progression to cancer may be unrelated to the degree of inflammation.
Monday, February 28, 2011 1:00 PM
Poster Session II # 85, Monday Afternoon