Gastric subepithelial tumors (SET) are protruding mass-like lesions in the stomach that are covered by an intact mucosal surface. Although often asymptomatic, gastric SETs can present symptoms related to mass effect or bleeding [1]. However, in the absence of obvious clinical signs, these lesions are most commonly discovered incidentally during examinations for other clinical reasons. Owing to the widespread use of endoscopic examinations, particularly as part of Korea’s national health screening program, the detection of gastric SETs has become more common. A retrospective study reviewing the natural course of incidentally detected small SETs in the upper gastrointestinal (UGI) tract reported a prevalence of 1.9% among asymptomatic individuals. The authors concluded that, in Korea, the increasing use of endoscopy for cancer screening is likely to lead to the more frequent detection of SETs [2].

Gastric SETs can range from benign lesions to tumors with malignant potential. Those with malignant potential include neuroendocrine tumors, lymphomas, glomus tumors, and gastrointestinal stromal tumors (GISTs). Non-neoplastic lesions include leiomyomas, lipomas, ectopic pancreas, vascular anomalies such as gastric varices, submucosal cystic formations such as duplication cysts, and pseudolesions such as submucosal hematomas or extramural compressions. GISTs and leiomyomas are the most frequently encountered types of tumors found in the stomach [3,4]. Although identifying lesions with malignant potential is important, the differential diagnosis of these conditions, based solely on endoscopic findings, is difficult. Endoscopic ultrasound (EUS) can provide additional information, such as the layer of origin; more accurate estimation of size, echogenicity, and margin regularity; and characteristics of the lesion’s internal contents. EUS-guided fine-needle aspiration/biopsy (EUS-FNA/FNB) can aid in confirming the diagnosis and in evaluating the lesion’s malignant potential. However, studies have shown that EUS has a low diagnostic accuracy for gastric SETs (45.5%–48%), and EUS-FNA is both technically and economically demanding [2].

A noticeable increase in tumor size is considered a potential indicator of malignancy, and the risk factors for SET progression include large initial tumor size, changes in the overlying mucosa, and the presence of hypoechoic or anechoic areas during EUS [2]. Both the National Comprehensive Cancer Network (NCCN) and the European Society for Medical Oncology (ESMO) guidelines recommend resection for initially large SETs, while advising regular follow-up for asymptomatic small gastric SETs less than 20 mm in size [2]. However, the lack of sufficient studies on the natural history of small asymptomatic SETs limits a clinician’s ability to make confident recommendations regarding optimal surveillance and treatment strategies. This review aimed to enhance the understanding of the natural history of gastric SETs, using currently available studies, and to assist clinicians in making well-informed decisions for the optimal management of these tumor types.

The reported prevalence of SETs in the general population undergoing routine endoscopy ranges from 0.4% to 1.9% [5]. Owing to the increased accessibility of endoscopic examinations, the incidental detection of gastric SETs has become more frequent. A multicenter retrospective study conducted by the EUS Research Group of the Korean Gastrointestinal Endoscopy Society reported the endoscopic prevalence of UGI SETs to be 1.6%. However, this study showed that the prevalence differed significantly between institutions (0.7%–3.5%). The detection of SETs during endoscopy depends on the endoscopist’s visual assessment, which may result in variations in detection rates based on the endoscopist’s experience and the quality of endoscopic documentation [4].

Significant differences in the occurrence of gastric SETs have not been observed between men and women, and the detection rate is known to increase in individuals after 50 years of age [6]. A multicenter study conducted by the EUS Research Group of Korean Gastrointestinal Endoscopy Society showed that the most common location of gastric SETs is in the upper third of the stomach (39.9% of cases). In that study, leiomyomas (30.4%) were the most common UGI SET, followed by GISTs (24.6%), ectopic pancreas (16.7%), and lipomas (5.1%); in the stomach, specifically, GISTs (39.1%) were the most common, followed by ectopic pancreas (22.5%) and leiomyomas (18.3%) [4].

Because some gastric SETs may have malignant potential, establishing an appropriate management plan is important. Although even small SETs have the potential to become neoplastic, resection of all incidentally discovered gastric SETs is too invasive. Pathological confirmation can aid in decision-making; however, EUS and EUS-FNB are technically challenging and costly procedures that may not be available in all institutions. Moreover, the diagnostic yield of EUS-FNB for lesions smaller than 30 mm in size has been shown to be low, and obtaining tissue from anatomically difficult locations can be challenging [7]. Given its limitations, EUS-FNB should be reserved for lesions with high-risk features or for patients who are surgical candidates. Longitudinal follow-up imaging may be a more practical strategy for low-risk cases. Therefore, understanding the natural course of gastric SETs is essential for developing effective management guidelines. Several studies have assessed tumor progression based on tumor size and EUS features, offering valuable insights for lesion management.

Melzer and Fidder [8] evaluated the natural course of UGI SETs, using EUS, and found that most small SETs (<30 mm in size) showed no significant changes over a 19-month period, leading to the recommendation of conservative management with regular surveillance. The study included 25 consecutive patients with small UGI SETs (<40 mm in size). Follow-up EUS was performed at least six months after the initial examination. Gastric SETs were identified in 16 patients and esophageal SETs in 9. The initial maximal diameter measured 19.4±7.4 mm for gastric and 19.8±7.4 mm for esophageal lesions. Follow-up examinations were conducted 19.0±11.4 months after the initial EUS. At follow-up, no significant changes were observed in maximal diameter or echo patterns. Among the patients, four lesions (4/25; 16%) showed an increase of more than 25% in the maximal diameter during surveillance. One gastric lesion changed from hypoechoic to nonhomogeneous with irregular margins after 17 months of follow-up, and its size increased from 30 mm to 38 mm. The lesion was surgically resected and histological examination revealed a stromal tumor with high malignant potential. Three other patients showed no change in SET echogenic patterns, did not undergo surgical resection, and died of unrelated causes. The study concluded that because most small SETs (<40 mm in size) do not exhibit features suggestive of malignant potential, a conservative surveillance policy is generally sufficient [8]. This finding supports the current conservative approach, particularly for patients with asymptomatic SETs less than 30 mm in size and without high-risk EUS features.

As shown in the aforementioned studies, EUS is often considered a suitable method for evaluating SETs. Lok et al. [9] shared their experience of using EUS to monitor the progression of small SETs and concluded that the utility of EUS surveillance for small UGI SETs remains unclear. In their study, conducted from 2002 to 2007, 93 patients with SETs originating from the muscularis propria were included. Of these, 49 had small tumors without high-risk EUS features; 23 patients had a median tumor size of 12.9±6.9 mm. Endoscopic surveillance was performed over a period of 17.3±10.2 months, during which time only three patients (13.0%) showed an interval increase in tumor size, without changes in their EUS features. Surgical excisions were performed for these three patients, revealing schwannomas in two cases and a low-risk GIST in one. Given the low compliance and high cost, the usefulness of routine EUS surveillance for small SETs remains uncertain [9].

Another study evaluating the natural course of UGI SETs (<3 cm in size) using EUS was conducted by Gill et al. [7] They concluded that most SETs less than 30 mm in size did not show significant changes during follow-up. The study was a retrospective chart review of 51 SETs less than 30 mm in size originating from the second and fourth echolayers. Of these, 12 were located in the esophagus, one at the gastroesophageal junction, 36 in the stomach, and two in the duodenum. The mean duration was 29.7 months. Follow-up revealed size increases and/or echogenic feature changes in 7/51 (13.7%) patients, including one with an esophageal lesion and six with stomach lesions. Among the gastric SETs, 6/36 (17.7%) showed an increase in size, and 5/29 (17.2%) gastric SETs originating from the fourth echolayer demonstrated growth. Surgical follow-up was available for three of the seven patients with interval changes, and all were found to have GISTs. Most small SETs (<30 mm in size) remain stable over 23 months, but changes in size or features may suggest GISTs; thus, biennial follow-up is reasonable for asymptomatic cases [7]. Bruno et al. [10] also supported a surveillance approach for small benign-appearing SETs originating from the muscularis propria. A total of 49 asymptomatic patients with hypoechoic lesions arising from the muscularis propria were included in the follow-up EUS program. All lesions had a maximum diameter of less than 30 mm with regular margins and cystic spaces of less than 3 mm. After a mean follow-up of 31±20.8 months and a median of two EUS examinations per patient, no changes in size or echogenic features were observed in 44 subjects, whereas 5 (11.6%) showed increases of at least 25% in diameter. Surgical resection was performed in four of these patients. Histopathological examination revealed low- or very low-risk GISTs in three patients and a glomus tumor without malignant features in one. Given that the resected lesions demonstrated low malignant potential, the study concluded that although an isolated increase in size warrants attention, it is not always indicative of malignancy [10].

A study conducted by Lim et al. [11] reviewed the medical records of 104159 patients who underwent UGI endoscopy (EGD) and detected SETs in 795 patients (0.8%). Among these, 252 patients were followed up using EGD for an average of 82.5±29.2 months, with a median interval between follow-up endoscopies of 12 months; the mean lesion size at the initial examination was 8.9 mm. Of the 252 lesions, 244 (96.8%) remained unchanged, whereas 8 (3.2%) showed a significant increase in size after a mean follow-up period of 59.1±27.5 months. Surgical resection was performed for lesions equal to or greater than 30 mm in diameter. Among these, two were diagnosed as GISTs with intermediate or high malignant potential, and one was classified as a schwannoma. The remaining five patients, who did not undergo surgery, were monitored using EUS or endoscopy and showed no further changes over follow-up periods of 1–5 years. Although the recommended follow-up duration varies across studies, Lim et al. [11] concluded that regular follow-up with endoscopy or EUS is sufficient for small, asymptomatic SETs, particularly those less than 10 mm in size.

In agreement with a conservative surveillance strategy for small gastric SETs, Kim et al. [12] suggested that a more stratified management plan should be developed based on a tumor’s initial findings. In their study, Kim et al. [12] retrospectively reviewed the EGD and/or EUS findings of 989 gastric SETs measuring 30 mm or less in size, with a minimal follow-up duration of three months. Among these lesions, 84 (8.5%) demonstrated significant changes in size, echogenicity, and/or morphology over a median follow-up period of 24 months. Notably, the estimated growth rates differed significantly according to the initial tumor size: <10 mm, 0.14 mm/month; 10–20 mm, 0.22 mm/month; 20–30 mm, 0.31 mm/month (p=0.003), underscoring the need for size-specific management. Of the patients with SETs that showed interval changes, 25 underwent surgical or endoscopic resection. Nineteen of these were diagnosed with GISTs, including three high-risk, four intermediate-risk, ten low-risk, and two very low-risk cases. A univariate analysis revealed that GISTs demonstrated significantly greater size changes during follow-up than did other benign tumors (p=0.037). However, the multivariate analysis did not confirm a statistically significant difference (p=0.326). The study concluded that SETs less than 10 mm in size had a significantly slower growth rate than those measuring 10–30 mm, highlighting the importance of individualized follow-up protocols. Kim et al. [12] recommended annual endoscopy for gastric SETs less than 10 mm in size, whereas SETs measuring 10–30 mm in size and located in the muscularis propria should undergo EUS 6 months after the initial diagnosis, with subsequent follow-up intervals determined by EUS findings.

As mentioned above, surgical intervention is generally recommended for gastric GISTs greater than 20 mm in diameter, whereas conservative management is often recommended for lesions less than 20 mm in size. However, Fang et al. [13] proposed a more refined cutoff size to guide treatment decisions that was based on their analysis of the clinical course of small gastric SETs. In a retrospective study, Fang et al. [13] evaluated 50 patients with EUS-suspected gastric GISTs measuring less than 3 cm in size, each patient was followed, using EUS, at least twice over a period of more than 24 months. During follow-up, significant tumor growth, defined as at least a 20% increase in maximal diameter, was observed in 14 patients (28.0%). The average tumor growth rate in the progressive disease group was 17.1% per year, which was significantly higher than that in the stationary disease group (1.0% per year). Because a substantial proportion of tumors in the progression group measured less than 20 mm in size at baseline, a receiver operating characteristic (ROC) curve analysis was performed. The analysis identified 14 mm as the optimal cutoff size for predicting tumor progression, with a sensitivity of 85.7%, specificity of 86.1%, and overall accuracy of 86.0%. Additionally, a larger tumor size (p=0.005) and the presence of irregular tumor margins during EUS (p=0.004) were significantly associated with tumor progression. Fang et al. [13] concluded that EUS-suspected gastric GISTs greater than 14 mm in size and those with irregular margins were more likely to exhibit tumor progression, warranting closer surveillance through more intensive follow-up protocols.

Hata et al. [14] conducted a retrospective review of 228 gastric SETs evaluated using EUS and identified the factors associated with tumor progression. Among the evaluated SETs, 118 were diagnosed as gastrointestinal mesenchymal tumors (GIMTs), and 50 underwent serial EUS with a mean follow-up of 24.3±19.3 months. Fourteen GIMTs (28%) showed significant growth and were classified into the “increased size group,” whereas the remainder formed the “no change group.” Patients in the “increased size group” were significantly older (66.4±8.3 years vs. 58.0±11.4 years, p=0.017) and their tumors were larger at the end of the follow-up (21.7±7.3 mm vs. 12.1±5.4 mm, p<0.01). The mean tumor doubling time in the increased size group was 19.2±18.2 months. The pathologic diagnoses included 13 GISTs and one schwannoma. A univariate analysis identified patient age, tumor location, and the presence of anechoic lesions during EUS as significant factors associated with tumor growth. Multivariate analysis confirmed that older age, tumor location in the middle or lower stomach, and anechoic features during EUS were predictors of progression. Based on these findings, Hata et al. [14] recommended closer follow-up of gastric SETs in older patients, particularly if those lesions were located in the middle or lower stomach and were anechoic, even if the initial size was small. They also noted that although doubling time does not necessarily reflect GIST aggressiveness, intermediate-risk GISTs tend to have short doubling times.

A prospective cohort study by Kushnir et al. [15] enrolled 187 consecutive adult patients who were referred for EUS evaluation of foregut SETs. EUS surveillance was recommended for 65 patients, 29 of whom underwent follow-up for a median duration of 30.3 months. During surveillance, 16 SETs (24.6%) increased in size, with a mean growth of 3.4±3.9 mm; no changes in echotexture were observed. Three patients were referred for surgery during follow-up: two due to tumor growth (both with pathology-confirmed GISTs) and one based on patient preference. The mean initial size of the SETs, measured using EUS, was 13.0 mm. When comparing size measurements between EGD and EUS, a clinically significant discrepancy was observed in only three cases; overall, the correlation between the two modalities was strong. This finding suggests that endoscopy may be a suitable alternative to EUS for surveillance purposes. Notably, fewer than 50% of the patients completed at least one follow-up EUS. This low compliance rate raises concerns about the effectiveness of surveillance as a management strategy for small asymptomatic SETs (<30 mm in size). The authors concluded that, when feasible, tissue acquisition should be pursued to help reassure patients and better guide surveillance recommendations [15].

Gao et al. [16] also proposed a tumor size cutoff for treatment guidelines and recommended an appropriate EUS surveillance duration for patients with suspected GISTs. In a single-institution retrospective study of 69 patients with EUS-suspected GISTs, those with minimal GISTs (≤20 mm in size) were followed for a mean period of 29 months. During the follow-up, 16 GISTs (23.2%) showed significant changes in size. Specifically, 11 GISTs (11/69, 15.9%) showed size changes at one year, 6 (6/43, 14.0%) at two years, and 7 (7/30, 23.3%) after more than 3 years. An ROC curve analysis identified a tumor size cutoff of 9.5 mm, which was smaller than the 14 mm cutoff previously suggested by Fang et al. [13] Only 4.7% of GISTs less than 9.5 mm in size showed significant changes in size at one year and 3.7% showed changes at two years; however, 9.5% showed growth after three years. In contrast, 34.6%, 31.3%, and 55.6% of GISTs that were at least 9.5 mm in size showed significant growth at one, two, and more than years, respectively. Based on these findings, the study recommended that patients with GISTs at least 9.5 mm in size should undergo EUS surveillance every 6–12 months, whereas those with GISTs less than 9.5 mm in size may extend the interval to every 2–3 years [16].

On the other hand, Hu et al. [17], like Fang et al. [13], proposed that an initial tumor size greater than 14 mm should be considered the cutoff for predicting tumor progression. By reviewing electronic medical records over a 14-year period, they enrolled patients with 10–30 mm gastric SETs originating from the muscularis propria who had undergone at least two EUS examinations. A total of 88 patients were included in the study; 25 (28.4%) were classified into the progression group. The mean duration of EUS surveillance was 24.6 months in the stationary subgroup and 30.7 months in the progression subgroup. A comparison between the two groups revealed that initially larger tumor sizes and irregular tumor borders were predictive of tumor progression. An ROC curve analysis identified a tumor diameter of 14 mm as the optimal cutoff size for predicting progression, with a sensitivity of 68.0%, specificity of 66.7%, and accuracy of 67.0%. A total of 17 patients underwent surgery during the study period. Of those, 13 were from the progression subgroup and were pathologically confirmed with GISTs; four were from the stationary subgroup and were diagnosed with leiomyomas. Most gastric SETs (71.6%) were indolent and tumor progression was observed only in GISTs, suggesting that progression may serve as a useful marker for distinguishing GISTs from leiomyomas [17].

A retrospective study conducted by Kim et al. [5] evaluated the natural history and risk factors for progression of gastric SETs in a large number of patients. Of the 3237 enrolled patients, 1859 underwent serial EGDs over a period of more than six months. EUS was performed in 733 (39.4%) patients and 73 (3.9%) patients underwent resection. Tumor progression was observed in 138 (7.4%) patients over a mean follow-up period of 59.4 months. In these cases, the mean initial tumor size was 15.0 mm, with a mean size increase of 12.86 mm. A univariate analysis identified larger initial tumor size, mid-third gastric location, overlying mucosal hyperemia, ulceration or erosion, and lobulated shape as significant risk factors for progression. In the multivariate analysis, large initial size, surface ulceration or erosion, lobulated shape, and mid-third gastric location remained independent predictors of progression. Large SETs were associated with faster growth and higher progression rates (p<0.001). Kim et al. [5] also failed to find significant differences in SET size measurements between EUS and endoscopy. Interestingly, variable tumor cell proliferation rates within a lesion may contribute to its lobulated shape. Additionally, rapid growth and larger tumor size may cause focal mucosal ischemia, leading to surface ulceration or erosion. Given that the majority of gastric SETs did not increase in size during the long-term follow up, the authors suggested that serial endoscopy may be sufficient for monitoring small SETs with intact mucosa and no high-risk features.

Doubling times can provide valuable insights into the malignant potential of tumors and help determine appropriate surveillance intervals. Several studies have explored the significance of tumor doubling times, particularly for GISTs. As noted by Hata et al. [14], tumors with shorter doubling times are more likely to be GISTs and demonstrate significant growth. However, Kim et al. [12] observed that although doubling time does not always correlate directly with malignancy risk, intermediate-risk GISTs tend to have longer doubling times than do low-risk GISTs, suggesting some predictive value. Sawaki et al. [18] also addressed the importance of tumor growth rates and doubling times for determining the appropriate monitoring intervals following an initial SET diagnosis. The study included 16 patients immunohistochemically diagnosed with GISTs. Over a median follow-up period of 4.9 years, 14 tumors remained stable in terms of size and shape. However, two showed notable growth: one doubled in diameter over eight years, whereas the other increased from 18 mm at diagnosis to 100 mm within just two years; the latter had a calculated doubling time of 3.1 months. Based on these findings, the authors recommended follow-up imaging every three months after an initial SET diagnosis. Koizumi et al. [19] further evaluated the efficacy doubling times for distinguishing GISTs from other gastric SETs. Out of 323 patients with final histopathological diagnoses of gastric SETs following surgical resection or EUS-FNA, 53 had tumors that could be measured on at least two consecutive EUS examinations; the estimated doubling times were retrospectively analyzed. Among the 53 cases, the histopathological diagnoses were as follows: GIST (34 patients), leiomyoma (5), schwannoma (3), ectopic pancreas (1), hamartoma (1), cyst (1), Brunner’s adenoma (1), and spindle-cell tumor (7). The median doubling time for GISTs was 17.2 months, significantly shorter than those for leiomyomas (232.2 months), schwannomas (104.7 months), ectopic pancreas (274.9 months), hamartomas (61.2 months), cysts (49.0 months), and Brunner’s adenomas (134.7 months). The difference in doubling times between GISTs and the combined leiomyoma and schwannoma group was statistically significant (p=0.005). Furthermore, the doubling times of GISTs varied by risk classification: 24.0 months for extremely low-risk and low-risk GISTs, 17.1 months for intermediate-risk GISTs, and 3.9 months for high-risk GISTs. These findings suggest that high-risk GISTs are associated with faster growth. Notably, the doubling times for both intermediate- and high-risk GISTs were less than 6 months, leading the authors to recommend that even patients with small SETs (<20 mm in size) should undergo follow-up imaging within six months of diagnosis [19]. Overall, tumor doubling time is a useful indicator of growth rate and may guide individualized risk-based surveillance strategies. Doubling time can serve as an indicator of the tumor growth rate, helping to guide surveillance frequency by balancing the need for early detection of aggressive tumors against the risk of unnecessary monitoring of indolent ones.

Given the potential for the malignant transformation of gastric SETs, particularly GISTs, a careful evidence-based surveillance strategy is essential. Although even small asymptomatic SETs can progress to become high-risk tumors, numerous studies have shown that most small SETs remain stable over time. Based on the studies discussed above, the reported growth rates of SETs range from 3.2% to 28.4% for all UGI SETs and from 5.4% to 28.4% for gastric SETs (Table 1). Most studies have identified initial tumor size as a key risk factor for progression, with some also highlighting additional factors such as older patient age, tumor location, and certain EUS features, including irregular borders and the presence of anechoic areas. Endoscopic findings, such as surface ulceration, erosion, and lobulated shape, have also been proposed as indicators of high-risk tumors. Interestingly, three studies conducted in Korea reported lower tumor growth rates compared with others [5,11,12]. Unlike studies that relied exclusively on EUS for follow-up, these Korean studies included endoscopy to measure both the initial and follow-up tumor sizes. The initial tumor sizes reported in these studies (mean diameter 8.9 mm [11], 11.0 mm [12], and 11.7±7.3 mm[5]) were smaller than those reported in other studies, where the largest reported tumors had a mean size of 19.4±7.3 mm [8]. Since tumor size is a significant factor in progression risk, the smaller baseline sizes in the Korean studies may have contributed to the lower reported growth rates. Several studies have supported the adoption of individualized surveillance strategies based on tumor characteristics. Fang et al. [13] and Hu et al. [17] proposed a tumor cutoff size of 14 mm for predicting an elevated risk of progression, whereas Gao et al. [16] suggested an even lower cutoff size of 9.5 mm. Therefore, surveillance intervals should be individualized based on the initial tumor size and presence of high-risk features, with more intensive EUS follow-up recommended for larger or suspicious lesions. An individualized, risk-adapted management strategy is essential to ensure the early detection of aggressive tumors and avoidance of unnecessary interventions for indolent lesions. In Table 2, we suggest a general surveillance strategy based on the literature descriptions of tumor natural histories and guidelines from Korea [20] and Europe [21].

GISTs are the most common type of SET in the stomach. Due to their malignant potential, they have particular clinical significance and underscore the need for appropriate surveillance and management. GISTs are diagnosed pathologically and are typically characterized using positive immunohistochemical staining for KIT (CD117) or CD34 [18]. These lesions are usually located in the fourth layer (muscularis propria) of the stomach and can be differentiated from other gastric SETs using EUS findings and pathological confirmation via EUS-guided FNA/FNB [6]. The malignant potential of GISTs is primarily determined by their mitotic index, tumor size, and anatomical location; their risk is commonly stratified into very low, low, moderate, or high categories [22]. Miettinen and Lasota [23] proposed a widely accepted risk stratification system that was based on tumor size and mitotic count. This stratification system suggests that gastric SETs less than or equal to 2 cm in size, with 0–5 per 50 high-power fields (HPFs), have very low malignancy potential, whereas tumors greater than 5 cm in size, with mitotic counts of more than 5 per 50 HPFs, are associated with high risk (Table 3).

Although GISTs are relatively rare, with an estimated annual incidence of 10–20 cases per million, their prevalence increases to 22.5%–35% if micro-GISTs (those only detected microscopically) are included. However, as noted by Kawanowa et al. [24], only a small fraction of micro-GISTs progress to clinically significant tumors with malignant potential, suggesting that most remain indolent [6]. Among the aforementioned studies, few have specifically focused on the natural course of GISTs. Sawaki et al. [18] reported that 12.5% (2/16) of patients experienced tumor growth, with the fastest growing tumor demonstrating a doubling time of 3.1 months. Fang et al. [13] found that 28.0% (14/50) exhibited significant tumor growth (defined as ≥20% increase in size), and an ROC analysis identified 1.4 cm as the optimal cutoff size for predicting progression. Koizumi et al. [19] reported a median doubling time of 17.2 months for GISTs, overall, with faster doubling times for high-risk tumors (3.9 months) than for very low- and low-risk tumors (24.0 months). Based on these findings, they recommended follow-up imaging within six months of diagnosis [15]. In another study, Gao et al. [16] analyzed 69 patients with EUS-suspected GISTs at least 20 mm in size and found that 23.2% (16/69) demonstrated significant changes during the follow-up period. They identified 9.5 mm as a critical threshold for predicting growth and recommended EUS surveillance every 6–12 months for GISTs at least 9.5 mm in size. Although most small gastric GISTs are indolent and exhibit slow growth rates over time, a subset, particularly those that are larger at the time of diagnosis, carries a higher risk of progression; this highlights the need for more frequent surveillance based on tumor size and individual risk.

Although EUS plays an important role in the diagnosis of GISTs, its diagnostic accuracy for small gastric SETs is limited [25,26]. Several studies have revealed that pathological diagnoses involving FNA or FNB can improve the diagnostic yield compared with EUS alone. Akahoshi et al. [27] reported that the diagnostic rate of EUS-FNA for hypoechoic, solid, gastric SETs measuring less than 20 mm is 73%. Joo et al. [28] demonstrated a diagnostic yield of 92.1% for EUS-FNB in gastric SETs measuring 20–50 mm in size, along with a 100% concordance rate between EUS-FNB and surgically resected specimens. In a comparative study of EUS-FNA and EUS-FNB for the histopathological diagnosis of gastric SETs, Kim et al. [29] found that EUS-FNB was superior for obtaining histological core samples with a higher diagnostic sufficiency rate than EUS-FNA. EUS features, combined with histopathological information from biopsies, provide important information about tumor progression and support informed management decisions.

As supported by most of the aforementioned studies, GISTs smaller than 20 mm in size, without any high-risk features, can be managed with regular surveillance. The American Gastroenterological Association recommends surgical resection of gastric SETs greater than 30 mm in size that originate from the muscularis propria and exhibit malignant echo features. However, the NCCN advises resection of all GISTs measuring at least 2 cm in size due to their malignant potential. [12] Although laparoscopic resection is the standard treatment for gastric GISTs, recent advancements in endoscopic techniques have made it a potential alternative for relatively small gastric GISTs. A retrospective study by Joo et al. [30] compared the baseline characteristics and clinical outcomes of endoscopic and surgical resections of GISTs. The study found that the procedure time and hospital stay were significantly shorter for the endoscopic group, whereas the R0 resection rate was higher in the surgical group. Notably, there was no significant difference in recurrence rates between the two groups (p=0.586). Huang et al. [31] also evaluated the effectiveness of endoscopic full-thickness resection compared with laparoscopic surgery for gastric SETs originating from the muscularis propria. They concluded that operation times, complete resection rates, and lengths of hospital stay were comparable between the two groups. Importantly, no complications were reported in the endoscopic group, whereas two patients in the surgical group experienced postoperative problems. Although endoscopic resection may be an acceptable alternative to surgery for select patients, it requires a high level of technical expertise and carries a notable risk of perforation. Therefore, endoscopic resection should only be performed in facilities with immediate surgical backup.

Gastric SETs are increasingly being identified owing to the expanded use of endoscopy, particularly in countries, such as Korea, with national screening programs. Although most SETs are asymptomatic and benign, a subset of tumors, particularly GISTs, can be malignant. Distinguishing these tumor types based solely on their endoscopic appearance is difficult.

This review demonstrates that most small SETs (<30 mm in size), and especially those less than 10 mm in size, remain stable over time, supporting a conservative surveillance approach. The risk of progression is strongly associated with larger initial tumor sizes and irregular tumor margins observed during EUS. Features such as a mid-to lower-gastric location, older patient age, mucosal surface changes, and a lobulated or heterogeneous appearance are also suggested to be associated with an increased risk. Doubling times, particularly those under six months, may indicate more aggressive behavior, especially for intermediate- to high-risk GISTs. An individualized risk-based approach is essential for effective management of gastric SETs. Small low-risk lesions can be monitored with periodic endoscopy, whereas larger lesions or those with particular risk factors warrant closer EUS follow-ups every 6–12 months. Tissue sampling should be performed when feasible, particularly for growing lesions and those that exhibit atypical features.

Although current evidence supports a conservative approach for most small gastric SETs, individualized follow-up protocols based on tumor size, location, EUS characteristics, and patient factors are essential for optimizing patient outcomes. Further prospective studies are needed to optimize the surveillance intervals and establish appropriate clinical management strategies.