A Ruler Needs a Zero Point: Defining the Gastroesophageal Junction for Length‐Based Risk Stratification in Barrett's Esophagus

Adenocarcinoma of the gastroesophageal junction (GEJ) is increasing not only in Western countries but also in Japan, where the proportion of adenocarcinoma among esophageal cancers is gradually rising. Against this background, the review by Mukaisho et al. [1] is a timely and important contribution. Defining the GEJ may look like a concern for anatomists and pathologists, removed from daily endoscopic practice. For the endoscopist, however, the GEJ is the starting point for measuring the length of Barrett's esophagus (BE), and that length lies at the center of risk stratification for esophageal adenocarcinoma. In addition, the definition of BE itself differs between Japan and the West, and the definition of the GEJ is closely tied to this East–West difference. Where we set the GEJ therefore has a direct effect on diagnosis, surveillance, and treatment decisions. This commentary considers its clinical value.

The greatest strength of this review is that it brings together knowledge about the GEJ across anatomy, pathology, endoscopy, oncology, and surgery within one framework. The authors first organize the three landmarks used to identify the GEJ: the squamocolumnar junction (SCJ), the proximal end of the gastric folds (PEGF), and the distal end of the palisade vessels (DEPV). The DEPV marks the termination of the lower esophageal sphincter and, compared with the other landmarks, is relatively robust to conditions such as atrophy of the gastric folds. In routine practice, however, the DEPV is not always easy to identify, as factors such as inflammation or insufficient insufflation can obscure it. Based on the Kyoto international consensus [2], they present the area extending 1 cm above and below the DEPV as the gastroesophageal junctional zone (GEJZ), supported by immunohistochemical validation. The review then goes further, into the origin and expansion of cardiac‐type mucosa, the role of CDX2 in intestinal metaplasia, and the gut regenerative cell lineage. On this basis, it argues that GEJ cancer has two distinct etiologies, a Helicobacter pylori ‐negative reflux‐related type that resembles esophageal adenocarcinoma and an atrophy‐related type that resembles gastric cancer. The discussion further extends to treatment, from endoscopic submucosal dissection (ESD) to chemotherapy and surgery. Few reviews carry the argument this consistently, from embryology and pathology through to treatment. For daily practice, however, its most useful contribution is more specific: The review sets out why the DEPV is the anatomically appropriate reference for the GEJ. Because we measure the length of BE from the GEJ, an anatomically sound landmark is also a reliable starting point for that measurement, and this is the foundation on which length‐based risk stratification rests.

The definitions of the GEJ and of BE, however, differ clearly between the West and Japan. The major Western guidelines emphasize the link with cancer risk: They require intestinal metaplasia for diagnosis and do not define columnar epithelium shorter than 1 cm as BE. Indeed, diagnostic guidelines for BE vary substantially across countries, as summarized by Kusano et al. [3], reflecting these divergent priorities. Behind this lies a difference in what a diagnosis of BE means. In the West, it places the patient on a strict surveillance protocol, so overdiagnosing low‐risk cases is seen as undesirable. For this reason, the West builds criteria directly tied to cancer risk, intestinal metaplasia and a length of at least 1 cm, into the definition itself. The Rio de Janeiro consensus [4] from the World Endoscopy Organization, which involved experts from both the West and Japan, likewise defines BE as columnar‐lined esophagus with intestinal metaplasia on biopsy extending at least 1 cm above the GEJ, citing the limited evidence for cancer risk in columnar epithelium without intestinal metaplasia. It also recommends the PEGF as the most suitable landmark for the GEJ, since the palisade vessels can be obscured in BE. In contrast, the Kyoto international consensus [2] removed both the length criterion and the intestinal metaplasia requirement and adopted the DEPV as the anatomically appropriate landmark. This position is almost identical to Japanese practice, which has defined BE by anatomical continuity regardless of length or intestinal metaplasia. Rather than asking which is correct, the difference is one of emphasis: a West that values cancer risk and a Japan that values anatomical accuracy. How to identify the GEJ thus remains under international discussion, and even the adoption of the DEPV in the Kyoto consensus was not unanimous. This lack of agreement has a practical side. Because the DEPV and the PEGF do not always mark the same point, a length measured from one landmark cannot be directly compared with a length measured from the other, which matters whenever length is used to compare cancer risk across studies and countries. It should also be acknowledged that direct evidence showing that DEPV‐based measurement improves cancer risk stratification or patient outcomes, compared with PEGF‐based measurement, remains limited. Harmonizing Eastern and Western approaches will therefore require prospective, and preferably international, cohorts that record BE length from a defined landmark under a common protocol and link it to hard outcomes such as adenocarcinoma incidence.

We refine these definitions not just to classify disease, but to tailor management to each patient's risk. The length of BE is one of the most reproducible endoscopic risk factors for neoplastic progression, and we measure this length from the GEJ. The Kyoto international consensus [2] also states that we cannot determine the exact length of BE without first deciding the point that separates esophagus from stomach. In Japan, Fukuda et al. [5] showed that cancer risk differs by length: The annual incidence was only 0.0032% for BE shorter than 1 cm, but 0.58% for BE of 3 cm or longer. In a 10‐year prospective cohort study, Ishimura et al. [6] reported an annual incidence of esophageal adenocarcinoma of 1.01% for long‐segment BE (LSBE), similar to Western reports. LSBE in Japan is therefore not a low‐risk condition and deserves the same attention as in the West. Accurate measurement of length is therefore central to risk stratification, and it depends on correctly identifying the GEJ. This is why the precise definition the review provides matters in daily practice: It determines whether the length we record can be trusted.

Risk stratification has another purpose, which is to avoid overdiagnosis of low‐risk cases. Ultra‐short‐segment BE (USSBE) shorter than 1 cm carries a very low cancer risk and its endoscopic diagnosis is poorly reproducible, so such a diagnosis can bring patients unnecessary anxiety and excessive surveillance. The Kyoto international consensus [2] also does not recommend uniform surveillance for such short segments. We must reliably detect high‐risk LSBE while avoiding overtreatment of low‐risk cases, but in Japan the basis for this stratification is not yet fully in place. In our nationwide survey [7], only about one‐third of endoscopists used the Prague classification, the standard scale for recording the length of BE, and even when LSBE was known in advance, only half used magnifying endoscopy. Establishing a practice of recording length accurately and moving to detailed examination according to risk is a task for the future.

The importance of length also extends to follow‐up after treatment in both Japan and the West. Studying EMR and radiofrequency ablation in the West, Shimamura et al. [8] found that the longer the BE segment, the more likely complete eradication was to fail. Similarly in Japan, Ikenoyama et al. [9] showed that metachronous and synchronous multifocal cancers after ESD for esophageal adenocarcinoma are clearly stratified by BE length, with a markedly higher cumulative incidence in LSBE. Length is thus an important marker for post‐ESD surveillance and for the management of residual BE, which in the West is regarded as an area at high risk of cancer. Endoscopic eradication therapy, which eradicates this residual BE mainly by radiofrequency ablation, has become established there. In Japan, however, where radiofrequency ablation is not widely available for this indication, a treatment strategy for residual BE after ESD for LSBE has not been established. To remove the high‐risk residual BE, the available option is circumferential ESD, which removes the cancer together with the surrounding BE in one piece. However, the longer the resected area, the higher the risk of stricture. Stepwise ESD [10] has been reported as an attempt to solve this problem, although fibrosis from repeated resections makes later sessions challenging. In Japan, building a management strategy for LSBE after ESD is an urgent issue.

A precise definition of the GEJ is not an end in itself, but the zero point from which clinically meaningful measurement begins. The length of BE matters across the entire clinical pathway, from diagnosis and surveillance to risk stratification after endoscopic treatment, but it can guide practice only when the starting point is reliable. In this regard, the review by Mukaisho et al. is highly valuable because it defines the GEJ not from a single viewpoint but through an integrated discussion of embryology, histology, pathology, endoscopy, carcinogenesis, and treatment, providing a common language for the endoscopists, pathologists, and surgeons who manage BE and GEJ cancer. On this basis, we must now address several unresolved issues: reconciling Western risk‐based definitions with Japanese anatomy‐based definitions, standardizing the recording of BE length in daily practice, avoiding unnecessary surveillance for very low‐risk short segments, and developing management strategies for residual BE after ESD, especially in LSBE. Defining the GEJ is only the beginning; the ultimate goal is to use that definition to deliver reproducible, risk‐adapted, and patient‐centered care.

Author Contributions

Yugo Iwaya conceived, drafted, revised, and approved the final manuscript.

Funding

The author has nothing to report.

Conflicts of Interest

The author declares no conflicts of interest.

Linked Article

This article is linked with Mukaisho et al. http://doi.org/10.1111/den.70193.

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.