Mitoses, Genomic Instability and Neoplastic Progression in Barrett's Esophagus.
Dominique Coco, Amitabh Srivastava, Carissa Sanchez, Pui Y Fong, Xiaohong Li, David Cowan, Carlo Maley, Patricia Blount, Brian Reid, Robert D Odze. Brigham and Women's Hospital, Boston; Dartmouth Hitchcock Medical Center, Lebanon; Fred Hutchinson Cancer Research Center, Seattle; UCSF, San Francisco, CA
Background: Barrett's esophagus (BE)-associated adenocarcinoma (AdCa) develops as a consequence of chromosomal instability which is, in part, related to defects in cell division. A recently proposed theory suggests that Barrett's metaplasia is a successful adaptation to acidic reflux. Intestinal crypt stem cells produce transient amplifying cells that ultimately lead to differentiated cells that function in mucosal defense. The aim of this study was to evaluate epithelial mitoses as a biomarker for AdCa in BE in a large prospective study that included 32 patients who developed cancer.
Design: 3970 mucosal biopsies from 214 BE patients (M/F ratio: 170/44, mean age: 63 years, mean BE length: 5.7 cm), all of whom had a baseline endoscopy and at least one follow up endoscopy (mean follow up: 90.4 months), were evaluated in a blinded fashion for the presence, number (#), location (surface versus crypts) and type (typical versus atypical) of mitoses in both dysplastic and non-dysplastic epithelium. Weighted Cox regression model was used to assess mitoses and cancer outcome after adjustment for age, gender, and BE segment length. Wilcoxon rank sum test was used to evaluate mitoses and aneuploidy (Yes/No) and ploidy (≤2.7N and >2.7N), and tetraploidy (4N ≤6% and >6%).
Results: Both total # of surface mitoses and mean # of surface mitoses per crypt, in baseline biopsies, were both significantly associated with progression to AdCa (p<0.0001 for both comparisons). However, neither total, nor mean, # of crypt mitoses were associated with progression (p=0.8 for both comparisons). Regarding genomic instability measured by flow cytometry, a significant association was noted between surface mitoses and aneuploidy (p<0.0001), DNA ploidy (> 2.7N versus ≤2.7N, p=0.0001), and tetraploidy (4N ≤6% versus >6%, p<0.0001). However, no relationship was observed between crypt mitoses and aneuploidy.
Conclusions: Surface epithelial mitoses are a morphologic biomarker of increased risk of progression to AdCa in patients with BE, and are significantly associated with genomic instability. The lack of association of crypt mitoses with these parameters supports the hypothesis that crypt mitoses are a component of the epithelial adaptation to acidic reflux, and may have a protective effect from progression to AdCa.
Monday, February 28, 2011 1:00 PM
Poster Session II # 88, Monday Afternoon