ABBA the new face of ACID

If you change your mind, I'm the first in line. Honey I'm still free. Take a chance on me.

Take A Chance On Me, ABBA, 1977

In recent years we have seen impressive developments in advanced endoscopic imaging modalities that help us to better diagnose luminal gastrointestinal diseases and guide subsequent (endoscopic) therapy [1]. With the evolution of these new techniques, the responsibility for pathological diagnosis seems to be constantly shifting from the pathologist to the endoscopist.

However, although the approach of optical biopsies has become well established for colorectal polyps, nontargeting mapping biopsies (Seattle protocol) are still the standard method for surveillance of patients with Barrett’s esophagus in all available guidelines [2]. In contrast to the generally well recognizable colon polyps, the majority of neoplasia in Barrett’s esophagus is flat and often not visible with standard white-light endoscopy alone. Only about 13 % of early neoplasia in Barrett’s appears as macroscopically visible nodules. As lesions in the presence of dysplasia are already often missed, detection of subtle lesions in a surveillance setting remains the greatest challenge [3].

Although the need for surveillance gastroscopy in Barrett’s esophagus is obvious, mapping biopsies according to the Seattle protocol have major limitations. Similarly to every nonsystematic and nontargeted approach in endoscopy, mapping biopsies are time consuming, expensive, and less acceptable to both patients and endoscopists. Indeed, endoscopists only follow the recommendation of the Seattle protocol in up to 50 % of cases [4]. Therefore, different techniques for surveillance of patients with Barrett’s esophagus are highly advocated, and the path of abandoning random biopsies that was followed in studies of surveillance for long-standing colitis should also be followed for Barrett’s surveillance [5].

“The ABBA trial provides strong data that advanced pathology is not overlooked with the targeted biopsy approach, and now allows for powering of a definitive noninferiority study.”

Chromoendoscopy with acetic acid has been shown to be effective for neoplasia detection in high risk Barrett’s populations [6]. Acetic acid results in a reversible acetowhitening of the Barrett’s mucosa by causing acetylation of cytoplasmic proteins. Neoplastic areas have reduced cytoplasmic protein, causing an early loss of acetowhitening. Accordingly, after acetic acid application to the Barrett’s mucosa, neoplastic areas appear red. The advantages of acetic acid include its universe availability (e. g. vinegar from the grocery store) and the short learning curve in interpreting the images (just evaluate for the “red-in-white-sign”) [6]. However, despite its advantages, no randomized controlled trials comparing acetic acid chromoendoscopy with standard mapping biopsies are currently available.

In this issue of Endoscopy, Longcroft-Wheaton et al. present the results of their multicenter randomized ABBA trial (Acetic acid-targeted Biopsies versus nontargeted quadrantic biopsies during BArrett’s surveillance), comparing neoplasia detection rates of nontargeted biopsies (Seattle protocol) and acetic acid targeted biopsies [7]. With this trial, the authors have introduced a new tune in the study of advanced imaging in Barrett’s esophagus, as the study was conducted in a standard surveillance setting and accurately assessed the power and acceptability of a back-to-back study with two separate endoscopies.

The study involved six centers, from university hospitals to smaller district hospitals, and included 200 patients under Barrett’s surveillance with no history of neoplasia. In a crossover trial, patients were randomized to either the Seattle protocol or acetic acid targeted biopsies first. At 6 – 8 weeks after the first gastroscopy, 87 % of patients agreed to participate in the second gastroscopy, thereby serving as their own control. To avoid bias, the pathologist did not release the histology report (unless cancer was detected) until both endoscopies had been performed.

The authors showed a zero miss rate for high risk neoplasia. Notably, low grade intraepithelial neoplasia was missed by both approaches. Interestingly, none of these cases have been found on follow-up gastroscopy and biopsy, thereby potentially highlighting the low interobserver agreement of pathologists for diagnosing low grade Barrett’s dysplasia. Up to 12 biopsies per patient were required for the Seattle protocol to diagnose neoplasia, whereas targeted biopsies after acetic acid spraying only required 1.2 biopsies per patient. Based on these data, the authors also provided a cost-effectiveness analysis for the UK, showing a 9.5-fold difference between the costs of nontargeted and targeted biopsies.

This study contributes major knowledge to the current literature and highlights the importance of a well-structured study design and an exploratory trial to accurately power a subsequent randomized trial. As highlighted by the authors, some clinicians included in the study were not comfortable with the approach of not taking biopsies during the acetic acid evaluation when all Barrett’s tissue appeared homogenously “white.” However, they felt reassured by the tandem endoscopy design. Given the reduction in number of biopsies with the targeted approach, one can fully appreciate their concerns that important pathology might be overlooked. However, the ABBA trial provides strong data that advanced pathology is not overlooked with the targeted biopsy approach, and now allows for powering of a definitive noninferiority study. The authors estimated that 2828 patients would need to be recruited for such a trial, and we strongly advocate conducting this study as soon as possible in order to finally prove the effectiveness of acetic acid chromoendoscopy with targeted biopsies for surveillance of Barrett’s patients.

Coming back to one of the most popular songs by ABBA, indeed, we can change our mind and maybe we should increasingly use targeted biopsy approaches instead of nonsystematic nontargeted protocols. We should take a chance on acetic acid, as the new sound of the future for surveillance in Barrett’s. Hopefully, the subsequent trial will provide sufficient evidence to follow the more effective path of targeted biopsies, as in chronic colitis. However, based on the current ABBA trial, one cannot yet recommend hasty discontinuation of the Seattle mapping biopsies.

• References

• 1 Neumann H, Bisschops R. Artificial intelligence and the future of endoscopy. Dig Endosc 2019; 31: 389-390
• 2 Weusten B, Bisschops R, Coron E. et al. Endoscopic management of Barrett’s esophagus: European Society of Gastrointestinal Endoscopy (ESGE) Position Statement. Endoscopy 2017; 49: 191-198
• 3 Schölvinck DW, van der Meulen K, Bergman JJGHM. et al. Detection of lesions in dysplastic Barrett’s esophagus by community and expert endoscopists. Endoscopy 2017; 49: 113-120
• 4 Shaheen NJ, Falk GW, Iyer PG. et al. ACG Clinical Guideline: Diagnosis and management of Barrett’s esophagus. Am J Gastroenterol 2016; 111: 30-50
• 5 Bisschops R, Bessissow T, Joseph JA. et al. Chromoendoscopy versus narrow band imaging in UC: a prospective randomised controlled trial. Gut 2018; 67: 1087-1094
• 6 Kandiah K, Chedgy FJQ, Subramaniam S. et al. International development and validation of a classification system for the identification of Barrett’s neoplasia using acetic acid chromoendoscopy: the Portsmouth acetic acid classification (PREDICT). Gut 2018; 67: 2085-2091
• 7 Longcroft-Wheaton G, Fogg C, Chedgy F. et al. A feasibility trial of Acetic acid-targeted Biopsies versus nontargeted quadrantic biopsies during BArrett’s surveillance: the ABBA trial. Endoscopy 2020; 52: 29-36