# Neoplastic risk in hyperplastic esophagogastric junction lesions: Comprehensive multicenter study

**Authors:** Elena De Cristofaro, Federico Barbaro, Jérôme Rivory, Alexandru Lupu, Benedetto Neri, Dario Biasutto, Gianluca Andrisani, Rui Morais, Franscisco Mendes, João Santos-Antunes, Germana de Nucci, Sandro Sferrazza, Silvia Pecere, Yanis Dahel, Jean-Philippe Ratone, Laura Rovedatti, Cristiano Spada, Francesco Maria Di Matteo, Andrea Anderloni, Philippe Leclercq, Samanta Romeo, Guido Manfredi, Elisa Stasi, Jeremie Jacques, Edoardo Troncone, Lucile Héroin, Jerome Maitre, Arthur Berger, Giulio Antonelli, Simona Agazzi, Giovanna Del Vecchio Blanco, Giovanni Monteleone, Mathieu Pioche

PMC · DOI: 10.1055/a-2760-6753 · Endoscopy International Open · 2025-12-16

## TL;DR

This study finds that some esophagogastric junction lesions, previously thought benign, have a significant risk of becoming cancerous, emphasizing the need for careful endoscopic evaluation.

## Contribution

The study identifies specific endoscopic features that predict neoplastic transformation in hyperplastic esophagogastric junction lesions.

## Key findings

- Non-polypoid morphology, surface ulceration, and lesion size >12 mm are significant predictors of neoplastic transformation.
- 23% of hyperplastic esophagogastric junction lesions showed neoplastic transformation upon histological analysis.
- Accurate endoscopic assessment is crucial for identifying malignancy risk in these lesions.

## Abstract

Esophagogastric junction (EGJ) lesions are uncommon and histologically diverse. Among these, EGJ hyperplastic lesions are rare and generally considered benign. However, their nonspecific appearance makes accurate endoscopic identification challenging. Endoscopic resection is both a diagnostic and therapeutic approach, yet risk factors for neoplastic transformation in EGJ lesions remain unclear. This study aimed to identify predictive factors for neoplastic transformation in hyperplastic EGJ lesions.

This multicenter, retrospective study included patients with hyperplastic EGJ lesions endoscopically resected across 13 European hospitals. Data were collected from endoscopy and pathology reports. Neoplastic transformation was defined by presence of dysplasia or adenocarcinoma. A multivariable logistic regression model was conducted to assess predictive factors for neoplastic transformation in resected hyperplastic lesions.

From January 2015 to October 2024, 91 EGJ hyperplastic lesions were included. Polypectomy/endoscopic mucosal resection (EMR) was performed in 86% of cases, endoscopic submucosal dissection (ESD) in 19%. En bloc resection was successfully achieved in 93% of cases, whereas R0 resection rates were confirmed in 84% of cases. Twenty-one lesions (23%) showed neoplastic transformation on histology. Independent predictive factors for neoplastic transformation in hyperplastic lesions included non-polypoid morphology (odds ratio [OR] 5.48;
P
= 0.025), presence of surface ulceration (OR 11.5;
P
= 0.0005) and lesion size (OR 5.48;
P
= 0.021). Lesion size > 12 mm was identified as a significant predictor of neoplastic transformation in hyperplastic lesions.

EGJ hyperplastic lesions showed a non-negligible risk of neoplastic transformation. These findings highlight the need for careful endoscopic assessment to predict malignancy while promoting appropriate management strategies to ensure adequate R0 resection in case of undetected local malignancy.

## Linked entities

- **Diseases:** adenocarcinoma (MONDO:0004970)

## Full-text entities

- **Diseases:** lesions (MESH:D009059), hyperplastic (MESH:D000082242), adenocarcinoma (MESH:D000230), EGJ hyperplastic lesions (MESH:C537006), malignancy (MESH:D009369), dysplasia (MESH:D015792)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12817182/full.md

## Figures

3 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12817182/full.md

## References

32 references — full list in the complete paper: https://tomesphere.com/paper/PMC12817182/full.md

---
Source: https://tomesphere.com/paper/PMC12817182