Medical Resident Carilion Clinic/Virginia Tech University Roanoke, VA
Award: Presidential Poster Award
Vivian Ussui, MD, MS1, Meeta Desai, MD2, Airi Katoh, MD3, Danielle Guilliams, MS, CCRP4, Hunter Sharp, MS4, Jonathan Bern, MD5, Douglas J. Grider, MD6; 1Carilion Clinic/Virginia Tech University, Roanoke, VA; 2Penn State Health Gastroenterology, Hershey, PA; 3University of California San Francisco, Fresno, CA; 4Virginia Tech School of Medicine, Roanoke, VA; 5Henry Ford Allegiance Healthcare, Jackson, MI; 6Virginia Tech Carilion School of Medicine, Roanoke, VA
Background: Esophageal adenocarcinoma (EAC) can develop from Barrett esophagus (BE) via the dysplasia-carcinoma pathway. The American Gastroenterological Association (AGA) has defined BE as requiring the presence of goblet cells (GC). CDX2 is a homeobox transcription factor that regulates cellular proliferation and differentiation of intestinal mucosa. All intestinal mucosa is strongly positive for CDX2 by immunohistochemical (IHC) staining. Furthermore, CDX2 IHC staining is widely available in histopathology laboratories. Our past studies have shown CDX2 IHC positivity in the absence of GC in suspected BE patients.
Objective: The goals of this study are to evaluate the natural history of BE patients that were initially GC positive but had subsequent biopsies negative for GC, determine if CDX2 IHC more accurately defines the presence of BE (as compared to GC), and determine risk of dysplasia and EAC. Methods:
Methods: This was a pilot retrospective clinicopathologic study. Patients were eligible for inclusion based on review of the pathology database at the Carilion Clinic / Dominion Pathology Associates between 2010 and 2017. For inclusion, patients had at least two sets of biopsies, the first with the diagnosis of BE based on the presence of GC and the second pathology report read as not BE. All slides were re-read by a single expert gastrointestinal pathologist. Histology was based on hematoxylin and eosin (H&E) staining. CDX2 IHC staining was also performed on both sets of esophageal biopsies for all patients. In addition to pathology results, the electronic medical record of these patients was reviewed for clinical outcomes. Results:
Results: A total of 130 patients met inclusion criteria. Of these patients 47 gave informed consent to participate in the study. The initial set of biopsies by definition were all GC positive. Forty-four of the initial biopsies were positive for CDX2 IHC (94%) and three were negative (6%). On the second set of biopsies, one biopsy was both GC negative and CDX2 IHC negative, yet had high grade dysplasia (HGD). Thirty-four second esophageal biopsies were GC negative but CDX2 IHC positive. Two of these showed HGD (6%) and one had EAC (3%). Discussion: CDX2 IHC positivity maybe more accurate compared to GC (or at least has significant diagnostic value) for the diagnosis of BE. More importantly patients that are GC negative and CDX2 IHC positive on follow-up biopsies for BE in this study have significant risk for HGD and EAC.
Table comparing first and second set of biopsies in regards to presence of goblet cells and CDX2
Diagram of patients with negative goblet cells
Adenocarcinoma with overlying high grade dysplasia, no goblet cells seen (100 magnification)
Disclosures: Vivian Ussui indicated no relevant financial relationships. Meeta Desai indicated no relevant financial relationships. Airi Katoh indicated no relevant financial relationships. Danielle Guilliams indicated no relevant financial relationships. Hunter Sharp indicated no relevant financial relationships. Jonathan Bern indicated no relevant financial relationships. Douglas Grider indicated no relevant financial relationships.