International guidelines provide heterogeneous recommendations for the management of type 3 gNETs (Table 1). Endoscopic resection of tumors measuring <10 mm and of low grade (G1) is feasible, as per the recent European Neuroendocrine Tumor Society (ENETS) guidance paper, provided metastases are ruled out and the depth of gastric wall invasion is evaluated using endoscopic ultrasonography (4). Also, the European Society of Gastrointestinal Endoscopy (ESGE) suggests that endoscopic resection may be a viable option for type 3 gNETs that are less than 20 mm in diameter, show exclusive submucosal invasion, and have a negative gallium-68 DOTATOC scan beforehand (9). The American Society for Gastrointestinal Endoscopy (ASGE) states that, due to the high likelihood of lymph node metastases, type 3 gNETs should undergo surgical treatment. However, endoscopic excision might be considered for small, well-differentiated tumors (less than 10 mm) (10). Nonetheless, the North American Neuroendocrine Tumors Society (NANETS) guidelines issued in 2010 firmly maintain the indication for extensive surgery for type 3 gNETs (11). NCCN guidelines recommend surgical approaches, such as partial or total gastrectomy (with lymphadenectomy), as the “preferred” method for type 3 gNETs, considering endoscopic resection as an option when EUS or other imaging have ruled out regional lymphadenopathy without specifying indications regarding tumor size (12). Both the Nordic guidelines and the UK guidelines suggest to treat type 3 gNETs in the same way as adenocarcinomas, by performing surgical resection plus lymph node dissection (13, 14). The shift in recommendations toward a more conservative approach in the more recent guidelines (Table 1; Figure 1) is due to recent reports that, although based on small retrospective patient series, have highlighted the potential for safe and effective endoscopic therapy in these patients. The more conservative approach recommended by the ESGE (9) and ENETS (4) guidelines may be due to these guidelines being more recent than others. This underscores the importance of updating guidelines promptly as new scientific evidence becomes available, especially in type 3 gNETs where the scientific evidence is scant.

Even a few new publications can alter the existing recommendations. Excharcou et al. analyzed 229 individuals with type 3 gNETs in a recent systematic review that comprised 10 nonrandomized retrospective investigations on this subset of tumors. Overall, 51.5% (n = 118) of them were well-differentiated G1-G2 NETs, and 121 patients with small and confined lesions received endoscopic excision with Endoscopic Mucosal Resection (EMR) or Endoscopic Submucosal Dissection (ESD) (15). Only one patient, 68 months after the endoscopic resection of a 16 mm G1 NET, had a nodal recurrence during follow-up, implying that selected cases have a better prognosis than previously thought (15). Hirasawa and colleagues investigated similar findings, reporting data on 144 patients with well-differentiated G1-G2 type 3 gNET, 63 of whom had endoscopic resection (48–76.2% - without any other subsequent therapy) (16). Only one patient experienced disease progression during follow-up after receiving conservative endoscopic resection alone, and the 5-year overall survival in the endoscopic group was 100%, with a 5-year recurrence-free survival of 97.6%. It should be highlighted that all type 3 gNETs undergoing endoscopic resection were restricted to the mucosa and submucosa, with a median tumor size of 7 mm, indicating that the best prognostic results are reserved for carefully selected patients (16).

Since an endoscopic strategy to the care of type 3 gNETs had been proposed, several authors pondered what the optimal resection technique was to obtain a complete excision of these lesions. Excharcou et al. reported a rate of complete endoscopic resection (R0) for type 3 gNETs using EMR or ESD of 72–80%, with the highest values obtained using the ESD technique (8); whereas Min et al. obtained a rate of R0 resection in 86.4% of the cases, using primarily ESD but also EMR/modified-EMR (15). However, the optimum endoscopic resection approach in this situation is still debated, despite the fact that ESD appears to be the most successful (17).

The identification of prognostic factors capable of predicting disease progression and assisting clinicians in deciding which patient is eligible for endoscopic therapy in those with type 3 gNET is particularly challenging, given the lack of supporting scientific evidence.

Consistent with findings from Excharcou (15) and Hirasawa (16), numerous retrospective studies (8, 18–21) observed positive outcomes following endoscopic resection of well-differentiated, small, low-grade type 3 gNETs limited to the submucosal layers without lymphovascular invasion. Conversely, some studies reported poorer outcomes for type 3 gNETs, often involving cases with larger, higher-grade lesions (22–24). As a result, it is critical to stratify the population based on the presence of unfavorable risk factors, which should guide therapeutic decisions. The most important features that seem to affect the clinical outcomes of patients with a diagnosis of type 3 gNETs, according to a recent systematic review of the literature, are size, grading, and depth of gastric wall infiltration (25). In Hirasawa’s paper, larger G2 lesions with deeper invasion of the gastric wall had a statistically significant higher risk of lymph nodal involvement, with a consequent worse prognosis. The importance of tumor size is highlighted further by the fact that in some cases, extremely small (5 mm) type 3 gNETs were accidentally excised utilizing biopsies and did not recur during follow-up (16). In terms of grading, the Ki67 index is one of the most important risk factors for GEP-NENs in general (26), but its relevance in gNETs is yet unknown (21). In any case, it is prudent to exercise caution before planning an endoscopic resection for a gNET with a high Ki67, given the predictable risk of a more unfavorable biological behavior. Moreover, in this patient setting, there is no evidence to suggest that endoscopic resection is safe. Unsurprisingly, deeper infiltration of the stomach wall with involvement of the muscolaris propria (and beyond) is a risk factor for lymph nodal or distant metastases (18, 24, 27). Endoscopic ultrasound (EUS) is becoming increasingly important in the diagnostic path in this context, to better determine the depth of tumor invasion and lymph node status. Although the role of EUS is well-established in evaluating NETs originating from other parts of the digestive system, such as pancreatic primaries, its role in gastric-originating forms has not been thoroughly studied. The paucity of data on this subject is a major challenge when addressing the scientific literature related to the efficacy of endoscopic treatment of gNETs. Nevertheless, EUS is recommended in patients with type 3 gNETs by the major recent international guidelines, primarily for planning the appropriate mode of endoscopic resection and for conducting a local disease staging to exclude the presence of loco-regional lymph nodes (28).