Introduction

Acetic acid chromoendoscopy (AAC) is frequently used in Barrett esophagus (BE) surveillance endoscopies. AAC is believed to be useful as a ‘red flag’ technique, thereby improving the overall neoplasia detection and the diagnostic yield of targeted biopsies. Accurate targeted neoplasia detection with AAC therefore holds the promise of abandoning random four-quadrant biopsies at 2cm intervals of the BE segment. Although some data exist on the use of AAC in expert centers, the added value in a non-dysplastic surveillance population in routine clinical practice is currently unclear.

Aims & Methods

A prospective stepped wedge cluster randomized trial was conducted in 18 Dutch community hospitals. BE patients undergoing surveillance endoscopies were eligible, while those with known or previously treated dysplasia were excluded. Participating hospitals shifted from usual care (no AAC; control group) to the incorporation of AAC (AAC; intervention group) in randomized order. Endoscopists were trained in the use and assessment of AAC immediately prior to its implementation. In both study groups, targeted biopsies were obtained in the presence of visible lesions before the acquisition of random four-quadrant biopsies. Primary endpoint was the overall neoplasia detection rate, defined as the percentage of patients with low-grade dysplasia (LGD), high-grade dysplasia (HGD) and esophageal adenocarcinoma (EAC) in both targeted and random biopsies, confirmed by an experienced GI pathologist. Secondary endpoint was the neoplasia detection rate of targeted biopsies. Data were analyzed using an as-treated approach, employing mixed effects regression analyses with a random intercept per hospital and correction for calendar time and BE length.

Results

We included 2,267 patients (1570 males; mean age 66) between June 2020 and November 2023. A total of 1,438/2,267 (63%) patients were allocated to usual care and 829/2,267 (37%) to AAC. In patients allocated to AAC, AAC was applied in 679/829 patients (82%). Overall, 163/2,267 patients (7.2%) were diagnosed with neoplasia, which encompassed LGD in 111/2,267 patients (4.9%), HGD in 30/2,267 patients (1.3%), and EAC in 22/2,267 patients (1.0%). The use of AAC did not result in a significantly higher overall neoplasia detection rate (7.4% with AAC [50/679 patients] versus 7.1% without AAC [113/1,588 patients]; OR 0.97; 95% CI 0.65-1.44). Visible lesions were detected in 95/679 (14.0%) patients receiving AAC versus 101/1,588 (6.4%) patients receiving usual care (p<0.001), yet the targeted neoplasia detection rate was not significantly higher (2.1% [14/679 patients] versus 1.8% [29/1,588 patients]; OR 1.05; 95% CI 0.49-2.24). Neoplasia was detected by random four-quadrant biopsies in the majority of patients: 39/50 patients (78%) receiving AAC and 81/113 patients (72%) receiving usual care.

Conclusion

The use of AAC does not improve the overall or targeted neoplasia detection rate in BE patients undergoing endoscopic surveillance in routine clinical practice. Endoscopic surveillance with AAC-guided targeted biopsies cannot replace the random biopsy protocol.

Disclosure

Pentax Medical, C2 Therapeutics