How Many Biopsies Are Enough? Rethinking Biopsy and Surveillance Strategies for Barrett's Esophagus in Asia

The diagnosis and surveillance of Barrett's esophagus (BE) in Asia remain areas of uncertainty and heterogeneity, largely due to regional variations in clinical practice and the lack of established guidelines. In contrast to Western countries, where BE management is guided by standardized protocols, evidence from Asian populations is limited, particularly regarding the diagnostic yield of biopsies for intestinal metaplasia (IM) and the determination of surveillance strategies for BE. Most Western guidelines recommend systematic random biopsies when columnar-lined esophagus (CLE) is identified on endoscopy, applying the Seattle protocol to detect IM, given its well-established association with esophageal dysplasia and adenocarcinoma (EAC) [1, 2]. Accordingly, histologic confirmation of IM is considered essential for the diagnosis of BE and is considered the foundation of surveillance strategies. In contrast, many Asian guidelines, including those in Japan, do not require histological IM for the BE diagnosis. This contributes to uncertainty with regard to the prevalence of IM and the standardized surveillance protocols in Asian BE cohorts. A recent nationwide survey in Japan conducted in 2024, involving 804 endoscopists, evaluated real-world practices in the diagnosis and surveillance of BE. The results revealed substantial inconsistencies, particularly the very low adherence to the Seattle protocol, reported at only 0.6%, and the limited use of biopsies specifically intended to detect IM, which were performed in just 6.0% of cases [3]. Similar findings have been reported in previous surveys, indicating that Asian endoscopists generally perform fewer biopsies, largely due to practical and resource limitations [4, 5]. Together, these findings reveal a lack of uniform biopsy practices and highlight the need for careful consideration of their implementation. In this issue, Tsai et al. conducted a large retrospective study to evaluate the diagnostic yield of random biopsies for detecting IM in Asian patients, addressing the scarcity of data on both biopsy yield and the required number of biopsies [6]. A total of 117,526 consecutive esophagogastroduodenoscopies (EGDs) were retrospectively screened. Among 5963 EGDs, 3135 patients (5.1%) were diagnosed with CLE, endoscopically diagnosed BE. Analysis of 4675 EGDs with 8994 biopsies revealed an overall IM detection rate of 35.1% per endoscopy, with significantly higher yields in long-segment CLE (74.8%) compared with short-segment CLE (32.9%). A single biopsy detected IM in 26% of cases, whereas three biopsies increased detection rates comparable with the Seattle protocol. Among patients initially negative for IM, repeat endoscopy modestly increased detection, mainly at the second examination. These findings demonstrate the critical impact of biopsy number on IM detection and support the need for pragmatic biopsy strategies specifically for Asian populations. The authors' recommendation of obtaining three biopsies represents a practical and clinically relevant approach that bridges the gap between ideal Seattle protocol adherence and routine clinical practice. Given its comparable efficacy and greater simplicity, this approach may be a more acceptable alternative biopsy protocol for Asian endoscopists. Nevertheless, an unresolved question remains: Is a three-biopsy strategy sufficient for long-segment CLE in Asian patients? Despite its strengths, the study should not be interpreted as evidence that three biopsies are adequate in all patients, as IM is unlikely to be uniformly distributed within CLE. The observed plateau in yield after three biopsies may reflect the predominance of short-segment BE in Asians rather than a true plateau of sampling. Although long-segment BE (LSBE) is less prevalent in Asia than in Western countries, emerging evidence from Japan suggests that the risk of neoplastic progression in LSBE may be comparable. In a recent Japanese prospective registry study, a 10-year follow-up revealed an EAC incidence in LSBE cases, 1.01 (0.57–1.73)%/year, similar to that reported in Western countries [7]. Another nationwide, multi-institutional study showed that cancer risk steadily increases with length of BE, with EAC incidence in Japanese LSBE patients comparable to Western data [8]. In such patients, the risk of progression to neoplasia from IM is substantial, and a three-biopsy strategy potentially risks undersampling. These findings indicate that LSBE in Asia should not be considered low risk and may require more robust strategies. Importantly, it should be noted that this study was retrospective, and variation in endoscopists' experience has influenced biopsy numbers and potentially technical performance. The other concern is the study's follow-up methodology, which is based on chart review and telephone interviews rather than systematic surveillance or BE registry. Furthermore, no systematic use of narrow-band imaging, acetic acid, magnification endoscopy, or artificial intelligence was incorporated. Recent BE practice is rapidly shifting toward image-enhanced endoscopy (IEE) and targeted biopsy, and the role of random biopsies will continue to be debated. Multiple studies have already demonstrated that advanced imaging technologies improve IM and dysplasia detection compared with random sampling. IEE has demonstrated utility in differentiating neoplastic from nonneoplastic lesions through detailed assessment of mucosal and vascular patterns, enabling real-time optical diagnosis and facilitating targeted biopsy. Targeted biopsy allows selective sampling of suspicious areas, thereby improving diagnostic yield while reducing the number of biopsies required. Nevertheless, this approach is highly operator-dependent and may be less reliable in less experienced hands, causing interobserver variability. Importantly, IEE and pathological examination serve as complementary roles, and histological diagnosis remains the gold standard for definitive diagnosis and for guiding clinical management. Therefore, rather than representing mutually exclusive strategies, a combined approach with careful mucosal inspection with IEE, targeted biopsy, and selective random sampling may currently represent the most pragmatic strategy. Looking ahead, the increasing adoption of artificial intelligence-assisted endoscopy may further enhance lesion detection, potentially shifting the paradigm away from extensive random sampling. From this perspective, the most relevant question for the future is not simply “how many biopsies are needed,” but rather “how many are required after careful, high-quality endoscopic assessment.” Another important discussion point raised is the need to develop BE surveillance protocols. There is currently no standardized accepted surveillance protocol in Asia, and surveillance intervals vary among institutions and countries. This lack of consensus hampers the development of evidence-based surveillance recommendations tailored to Asian populations. In this study, IM detection was independently associated with male sex, older age, longer CLE length, and a higher number of biopsies [6]. Addressing these gaps is essential for optimizing BE surveillance and improving early detection of neoplasia. Identifying high-risk subgroups within the Asian BE population, based on factors such as smoking, obesity, or other risks, will be important. Although chronic GERD is considered the primary causal risk factor for EAC based on Western data, previous studies have shown that up to 40% of patients with EAC have no history of chronic GERD [9]. In the aforementioned study in Japan, patients who developed EAC were significantly more likely to have a history of smoking [7]. Robust, population-specific data are needed to inform more individualized surveillance and biopsy strategies. Another particularly valuable aspect of this study is its discussion of follow-up strategies for patients with suspected BE but negative IM on initial endoscopy. Although most guidelines specify surveillance intervals for nondysplastic BE according to segment length, they often fail to address whether, and how, patients without IM on initial biopsy should be followed. In this regard, the study assessed whether repeat endoscopy improves the detection of IM in patients who were IM-negative at index endoscopy. Among 1772 IM-negative patients, IM was detected at the second endoscopy in 24.2%. If all IM-negative patients underwent a second endoscopy, an estimated 428 additional IM cases would be identified, increasing the additional yield of 16.9%. Based on these results, the authors recommend a repeat endoscopy for patients initially negative for IM. Their recommendation is both timely and clinically meaningful. This aligns with the 2022 American College of Gastroenterology guideline, which states that a repeat examination may be reasonable given that the yield for detecting IM on a second endoscopy can be 25% or higher, whereas further examinations beyond this are unlikely to be beneficial [1]. By incorporating this concept into an Asia-focused strategy, the present work fills an important gap in clinical guidance. What, then, should be the way forward? Management of BE in Asia requires pragmatic, data-driven approaches and establishment of refined guidelines. Most importantly, prospective, population-based registries with standardized biopsy and surveillance protocols are needed to define the true natural history of BE in Asia. Given the lower prevalence of LSBE, surveillance may be better to move beyond uniform models toward more refined, risk-adapted strategies. As the epidemiology of BE continues to evolve, standardized, evidence-based surveillance strategies will be essential to guide future studies and improve early neoplasia detection. In this context, the biopsy strategies proposed in this study provide an important reference point, enabling meaningful comparisons across studies. Equally important is the recognition of the need to optimize the quality of endoscopic examination. Key quality indicators, including adequate inspection time, measures to enhance neoplasia detection, and adherence to biopsy protocols, play a central role in improving diagnostic yield [10]. Ultimately, such practical, evidence-based approaches have the potential to shape the future of BE research and patient care in Asia. The authors have nothing to report. The authors have nothing to report. The authors have nothing to report. Y.S. serves as an Associate Editor for Digestive Endoscopy. Y.I. declares no conflicts of interest. This article is linked to Tsai et al. paper. To view this article, visit https://doi.org/10.1111/den.70077.