625 Interobserver Agreement in the Use of p53 Immunohistochemistry for Barrett Esophagus with Dysplasia

Document Type

Conference Proceeding

Publication Date

3-23-2026

Publication Title

Lab Invest

Keywords

biological marker, protein p53, aged, Barrett esophagus, cauterization, conference abstract, controlled study, diagnosis, diagnostic accuracy, dysplasia, electronic medical record, endoscopic mucosal resection, female, histology, human, human tissue, immunofluorescence, immunohistochemistry, interrater reliability, major clinical study, male, surgery

Abstract

Disclosures: Pegah Dejban: None; Qing Chang: None; Zongshan Lai: None; Sandeep Kumar: None; Imeelda Altone: None; Beena Ahsan: None Background: Grading of dysplasia in Barrett esophagus (BE) remains a well-recognized diagnostic challenge with important clinical and therapeutic consequences. Immunohistochemical staining for p53 has been increasingly used as an adjunct to improve diagnostic accuracy and confidence in this setting. We therefore conducted a concordance study among subspecialty-trained gastrointestinal pathologists (GI-Ps) to assess both the role of p53 in diagnostic interpretation and its impact on interobserver agreement in the evaluation of BE with dysplasia. Design: Fifty-three consecutive cases of BE on which p53 immunostaining was performed as part of the initial work-up between 2023 and 2025 were retrieved from our institutional archives. All slides, including H&E and immunostains, were independently reviewed by 4 (GI-Ps) with subspecialty expertise. For each case, pathologists rendered a categorical diagnosis (negative, indefinite for dysplasia [IND], low-grade dysplasia [LGD], high-grade dysplasia [HGD], intramucosal carcinoma [IMC], or adenocarcinoma) and responded to structured questions regarding the use of p53 immunostaining, its influence on their diagnosis, its prognostic role, and its application in assessing endoscopic mucosal resection (EMR) margins. Inter-rater agreement across questions was analyzed using Fleiss’ kappa statistics and interpreted according to the Landis and Koch scale. Results: The highest concordance was observed for histologic diagnosis (κ = 0.54; moderate–substantial agreement). In contrast, agreement was poor to slight regarding whether p53 should be performed (κ = 0.05), whether it altered the rendered diagnosis (κ = –0.05), and its role in assessing EMR margin status (κ = –0.09). Pathologists most frequently reported ordering p53 in cases showing small or focal areas of atypia, in the setting of active inflammation or ulceration, in biopsies with extensive cautery artifact, and when crypt dysplasia was suspected. Despite variability in its application, all pathologists consistently reported that p53 carried both diagnostic and prognostic significance in BE with dysplasia. [Formula presented] Conclusions: Although p53 immunostaining was uniformly regarded as having diagnostic and prognostic value, its use did not improve interobserver diagnostic concordance among gastrointestinal pathologists evaluating BE with dysplasia. These findings suggest that while p53 may provide interpretive support in selected histologically challenging scenarios, it should be applied judiciously in routine practice.

Volume

106

Issue

3

Find It @ Sladen

Share

COinS