Endoscopic resection for gastric subepithelial lesions: overtreatment or the next step in optimization?

Subepithelial lesions (SELs) of the upper gastrointestinal (UGI) tract, particularly those located in the stomach, are increasingly detected because of the widespread adoption of routine endoscopic screening. Epidemiological studies have reported an overall prevalence ranging from 0.36% to 1.94%, with detection rates exceeding 9% in individuals over 70 years and up to 22% in autopsy studies of those older than 50 years.1-3 While many SELs are benign, a subset, including gastrointestinal stromal tumors (GISTs), possesses malignant potential, particularly when arising from the muscularis propria layer. The incidence of gastric GISTs, now classified as malignant soft tissue sarcomas under the 2020 World Health Organization classification, has been steadily increasing worldwide.4 Clinically, SELs are often asymptomatic and incidentally detected, yet their potential for malignant transformation mandates accurate diagnosis and risk stratification. Determining whether a lesion represents a benign leiomyoma or a GIST influences the treatment strategy and long-term prognosis. Therefore, refining the diagnostic accuracy and therapeutic safety of lesions in anatomically complex sites, such as the esophagogastric junction (EGJ) and gastric cardia, which are difficult surgical areas, remains an important frontier in gastrointestinal endoscopy.

In a study published in Clinical Endoscopy by Park et al.,5 they evaluated 26 consecutive patients who underwent endoscopic resection (ER) of 27 gastric SETs located in the EGJ (n=3) and gastric cardia (n=24) between 2012 and 2023. Most lesions originated from the muscularis propria layer, and pathology confirmed 23 leiomyomas and four GISTs. The mean tumor size was 21 mm, and ER achieved an en bloc rate of 100% and a complete resection rate of 88.9% with no reported bleeding, perforation, or infection. When comparing resection methods, endoscopic muscularis dissection (EMD) achieved complete resection in 82.4% of cases, whereas endoscopic submucosal dissection (ESD) and submucosal tunneling ER (STER) achieved 100% complete resection without complications. Importantly, the study highlighted a clear learning curve: complete resection improved from 76.9% before 2019 to 100% after 2019, reflecting the critical role of operator experience in achieving successful outcomes. Nevertheless, this study had several limitations. Its retrospective, single-center design, and small sample size limit its generalizability. Additionally, the follow-up duration was relatively short and no long-term histopathological outcomes were observed in the GIST subgroup. Furthermore, risk evaluation was not performed before resection using tissue acquisition or endoscopic ultrasound (EUS) findings. Despite these limitations, our results reaffirm that ER can be a safe and effective alternative to surgery for SELs at technically challenging locations, provided that the procedure is performed by experienced endoscopists.

Accurate diagnosis remains central to optimizing SEL management. A stepwise diagnostic algorithm should integrate the EUS, tissue acquisition, and risk stratification tools. EUS remains indispensable; however, a presumptive EUS diagnosis is correct to predict histology in only 48% of cases.6,7 Specific EUS features, such as heterogeneous echotexture, irregular borders, exophytic growth, and cystic changes or calcifications, are ominous indicators of malignancy. Emerging modalities—including contrast-enhanced harmonic EUS and EUS elastography (endoscopic ultrasound-strain elastography, endoscopic ultrasound-shear wave elastography)—further refine differentiation between GISTs and leiomyomas, achieving diagnostic accuracies of approximately 80% to 90%.8 Regarding tissue acquisition, mucosa-incision assisted biopsy and EUS-guided fine-needle biopsy currently yield the highest diagnostic accuracy (84%–88%) with minimal complications.9 For SELs smaller than 2 cm, mucosa-incision assisted biopsy or endoscopic ultrasound fine needle biopsy can be considered as first-line options when tissue acquisition is clinically indicated; however, routine sampling of all small lesions may lead to unnecessary procedures, as many are benign and do not require intervention. Therefore, tissue acquisition should be selectively performed for lesions with features suggestive of malignant potential, or when the results are expected to influence management. In addition, depending on institutional resources and operator expertise, ER of lesions approximately 2 cm in size may be a practical approach to achieve both diagnosis and treatment simultaneously. Artificial intelligence–assisted EUS has also emerged as a promising adjunct, achieving a pooled area under the curve of 0.95 for GIST identification in meta-analyses.10 Collectively, these advances are transforming SEL diagnosis from morphology-based to multimodal, data-driven precision imaging, aligning with the principles of personalized medicine.

ER has evolved from a diagnostic adjunct to a definitive therapeutic modality for SELs in the UGI tract. The current study adds to the growing body of evidence, particularly from experience in Asia, supporting endoscopic therapy as safe and effective. We conducted a large Taiwanese multicenter cohort study (317 lesions) that reported technical success, en bloc resection, R0 resection, and recurrence in 97%, 94.3%, 88.9%, and 0% of cases, respectively, with major complications in only 6.3% of cases.11 The available techniques tailored by EUS to identify the origins of tumors include EMD, ESD, STER, and endoscopic full-thickness resection. Additionally, advancements in endoscopic closure in addition to traction techniques and technologies allow en bloc removal and successful wound closure after resection using several methods. These are clip-based (with string, snare, loop, or rubber band), X-tack, and suturing systems, such as OverStitch. Thus, with similar outcomes of ER to surgical intervention, current guidelines from international societies recommend ER as an alternative to laparoscopic wedge resection for GISTs ≤35 mm or when biopsy is inconclusive.12,13 The current study results corroborate this evidence base, demonstrating that with experienced operators and appropriate technique selection, endoscopic therapy achieves curative outcomes equivalent to surgery while minimizing invasiveness and recovery time with better quality of life.

Although progress in the endoscopic management of SELs is undeniable, several frontiers warrant further exploration. First, prospective multicenter studies with long-term follow-ups are needed to clarify the risk of recurrence, particularly in endoscopically resected GISTs. The integration of artificial intelligence (AI)-based EUS interpretation and radiomic modeling may standardize lesion characterization and procedural decision-making. Second, risk stratification systems should evolve beyond tumor size to include patient age, lesion location, and EUS morphological indices. Such predictive scoring systems can personalize management and avoid unnecessary interventions or overtreatment of benign lesions. Third, as hybrid laparoscopic–endoscopic procedures and novel closure devices become more accessible, future research should compare their cost-effectiveness and learning curves relative to pure endoscopic methods. Moreover, training standardization and procedural competency thresholds should be incorporated into international guidelines. Finally, future directions should emphasize collaborative registries that integrate procedural, histological, and long-term oncological outcomes, enabling evidence synthesis across centers and populations.

The current study by Park et al.5 represents an important contribution to the growing body of literature supporting ER as a curative, minimally invasive therapy for SELs of the EGJ and gastric cardia. When coupled with modern diagnostic tools, including high-resolution EUS, tissue acquisition innovations, and emerging artificial intelligence applications, endoscopic management is poised to replace traditional surgery for most small-to-medium-sized SELs. The challenge lies not in demonstrating feasibility but in refining patient selection, ensuring operator competency, and unifying practice standards through multicenter, evidence-based collaboration.