In 1983, American pathologists Mazur and Clark, while studying gastrointestinal tumors, found that some tumors had neither smooth muscle nor Schwann cell characteristics, and were the first to introduce the concept of “gastric mesenchymal tumor” (GIST) in order to distinguish between different types of gastrointestinal tumors of non-epithelial tissue origin [1]. In 1998, Hirota et al [2] found c-kit activating mutations between the transmembrane and tyrosine kinase structural domains in GISTs, which were associated with the development of GIST; further immunohistochemical results showed that GISTs expressed CD34 and CD117, thus standardizing the diagnosis of GIST. The more accepted definition of gastrointestinal mesenchymal tumor (GIST) is that GIST originates from amorphous mesenchymal cells differentiated to Cajal mesenchymal cells and is a mesenchymal-derived tumor of the gastrointestinal tract enriched in CD117-expressing spindle-shaped, epithelioid or pleomorphic cells in addition to smooth muscle tumors, nerve sheath tumors and neurofibromas [3]. 1, GST epidemiology and clinical features The incidence of GST is not high, about 10 to 20 million per year worldwide, with slightly more men than women. It is the most common mesenchymal-derived tumor of the GI tract, probably originating from gastrointestinal mesenchymal stem cells, consisting of undifferentiated or pluripotent spindle or epithelioid cells, and accounting for <1% of all tumors of the GI tract [4], with approximately 60% occurring in the stomach, 25% in the small intestine, 10% in the ileum and rectum, and the rest in other parts of the GI tract. About 60% of the tumors are located in the submucosa and grow distensibly, 30% in the subplasma and the remaining 10% in the gastric wall. GST is characterized by non-directional differentiation, and there is no clear boundary between benign and malignant. To judge its biological behavior, we need to consider the indicators of tumor size, histomorphology and clinical condition, and there is no unified pathological grading standard. Currently, the more widely used is the National Institutes of Health (NIH) consensus scheme [6], which classifies GIST into four levels of invasive risk: very low, low, moderate, and high, based on tumor size, nuclear split image count, and tumor primary site as classification indicators. With the continuous research and application, it was found that abdominal cavity contamination due to tumor rupture is also an important indicator affecting the prognosis of patients, therefore, NIH added tumor rupture as another indicator to judge the prognosis on the basis of the original one after another discussion in 2008 [7]. 2, Diagnosis of GST CT is the main imaging method for the localization and diagnosis of gastrointestinal tract mesenchymal tumors. Wang Xiuhuan et al [8] reported that spiral CT had a high detection rate of 100% for gastric mesenchymal tumor, and its CT mainly showed a soft tissue mass that protruded into and out of the gastric lumen or both, and the density within the mass was mostly heterogeneous in the form of mixed density shadow, and the enhancement of the mass was more obvious after enhancement, and the irregular necrotic liquefied area within it was not enhanced, while the parenchymal density area had different degrees of enhancement. Spiral CT is also helpful for preoperative GST qualitative diagnosis. Kui Zhifeng reported [9] that the determination of benign and malignant degree of GST by spiral CT is in good agreement with pathological diagnosis, which provides a good reference for preoperative diagnosis. Endoscopy is another method for preoperative diagnosis, which is more valuable for diagnosing submucosal tumors and helps in the early detection of tumors; abdominal ultrasound is usually used as the preferred method for initial diagnosis of GST because of its advantages of being noninvasive, rapid, and inexpensive. Ultrasound endoscopy can accurately probe the level of origin, volume size, boundary, surrounding lymph nodes and echogenic characteristics of tumors, and its operation is simple and less invasive, therefore, it can improve the early diagnosis rate of submucosal lesions in the GI tract. Mu Hong et al [10] reported that the diagnostic accuracy of ultrasound endoscopy for gastric mesenchymal tumor was 63.4%, suggesting that the grading of invasive risk of gastric mesenchymal tumor under ultrasound endoscopy is in good agreement with pathological grading, which can provide a basis for the development of treatment plan. 3.GST benign and malignant grading GST has the characteristics of non-directional differentiation, and its biological behavior and benign and malignant properties are difficult to determine. There is no significant clinical difference in the incidence of benign and malignant GST, and recurrence is still possible in incomplete resection of benign GST. After excluding smooth muscle-derived and neurogenic tumors, the risk of GST was graded according to the National Institutes of Health (NIH) and WHO classification, and the tumors were classified into four levels, namely, very low, low, moderate, and high, based on tumor size, nuclear schizophrenia, necrosis, and cell density: ① tumor diameter < 2 cm, nuclear schizophrenia ≤ 5/50 high magnification field (HPF) for very low risk; ② tumor diameter 2-5 cm, nuclear schizophrenia ≤ 5/50 HPF for low risk; ③ tumor diameter 5-10 cm, nuclear schizophrenia < 5/50 HPF for low risk; ③ tumor diameter 5-10 cm, nuclear schizophrenia ≤ 5/50 HPF for low risk. (3) Tumor diameter of 5-10 cm, nuclear schizophrenia < 5/50 HPF or tumor diameter of 5-10 cm, nuclear schizophrenia 6-10/50 HPF is moderate risk; (4) Tumor diameter > 5 cm, nuclear schizophrenia > 5/50 HPF or tumor diameter > 10 cm, nuclear schizophrenia > 10/50 HPF is high risk. Regardless of the size of the tumor, once it ruptures, it is classified as high risk [11]. 4. Treatment of GST Before 2001, surgical resection was preferred for the treatment of GIST. In recent years, with the successive application of molecular targeting drugs such as imatinib, the surgical treatment pattern of GIST has changed. 4.1 The extent of surgical resection for primary resectable GST should depend on the tumor site, and overall due to following the principle of non-expansion surgery, the 2004 ESMO symposium recommended: expanded resection of GST cannot benefit patients; it was emphasized that preoperative biopsy is not recommended to avoid tumor rupture, bleeding, abdominal implantation, and needle tract metastasis. Complete resection is the goal pursued for surgery, and the 2007 re-revised NCCN guidelines re-established that the goal of surgery should include obtaining negative margins, but whether it is negative microscopic margins remains controversial. The tumor-free principle is another principle that must be followed during GST surgery, which should be carefully performed to avoid rupture of the tumor and to ensure the integrity of the pseudoperitoneum, i.e., not touching the tumor as much as possible during surgery, and avoiding excessive turning of the gastric wall, intestinal canal and mesentery, and performing non-contact surgical resection. The decision of local excision, wedge resection or whole excision can be made based on the size of the tumor, the absence of necrosis and invasive adhesions. Lin Guole [12] reported 24 cases, including 19 cases of partial gastrectomy, 4 cases of major gastrectomy, and 1 case of total gastrectomy with regional lymph node dissection. The operation went smoothly and there were no fatal cases. The average follow-up time was 33 (4-108) months. None of the patients experienced local recurrence of tumor or distant metastasis during the follow-up period. Tumor enucleation is not suitable for the treatment of GST because of tumor residue and tumor rupture, and it is not suitable even for the protection of important organs. With the popularization and development of laparoscopic techniques, reports of laparoscopic surgical resection of GST have gradually increased in recent years [13 ]. Prior to 2004, laparoscopic or hand-assisted laparoscopic resection of GIST was not recommended for reasons of caution, and a subsequent retrospective analysis found that for GIST of about 4.0 cm in diameter, the success rate and prognosis of laparoscopic resection did not differ significantly from the open group, but there is a lack of The 2007 revision of the NCCN guidelines recommended laparoscopic resection for tumors ≤5.0 cm in diameter and hand-assisted laparoscopic resection for those >5.0 cm in diameter. According to Mingliang Wang, Wei-Lin Tan, et al, laparoscopic resection as a surgical approach for GST should be done by a laparoscopically experienced physician, operated strictly according to the principles of tumor-free and non-contact, and incisional implantation should be avoided as much as possible. If the tumor is locally infiltrated and combined organ resection is possible, then timely intermediate open surgery is recommended to be considered [14]. Endoscopic mucosal resection with ligator device (EMR-L) is suitable for submucosal gastric mesenchymal tumor. Li Jiao [15] reported that for the endoscopic minimally invasive treatment of gastric mesenchymal tumors in 30 cases, the results showed that EMR-L is a simple, rapid, safe, effective and feasible treatment for limited upper gastrointestinal mesenchymal tumors ≤2.5 cm. 4.2 Primary unresectable or recurrent metastases Prior to 2002, there were almost no effective treatments for unresectable or recurrent or metastatic GST. With the advent of molecularly targeted therapeutic agents such as imatinib mesylate, the treatment paradigm of GST has undergone a revolutionary change. Combination therapy based on imatinib treatment has become the treatment of choice for primary unresectable or recurrent metastatic GST. Imatinib mesylate, a tyrosine kinase inhibitor, is a new class of drugs that inhibits cell proliferation with high inhibition and high selectivity only for ab1 tyrosine kinase, c-Kit, and platelet-derived growth factor receptor (PDGFR). Imatinib mesylate binds to the ATP-binding site in the functional region of the intracytoplasmic tyrosine kinase of c-Kit, blocking the transfer of phosphate groups from ATP to the tyrosine residues of the protein substrate and inhibiting c-Kit tyrosine phosphorylation, leading to inhibition of cell proliferation and recovery of apoptosis. The median survival time for unresectable or recurrent metastatic GST was only about 1 year before the introduction of imatinib, while the median survival time for such patients was about 5 years after imatinib-based combination therapy. Patients had significantly longer disease stabilization (SD) and significantly better progression-free survival (PFS) and overall survival (OS) than the placebo group. In a retrospective analysis of 74 patients with advanced GST treated with imatinib, Zheng L.P. [17] showed a treatment efficiency of 87.8% (65/74) and a disease control rate of 94.6% (70/74). Foreign data suggest that it is possible for unresectable GST to shrink or even meet the criteria for surgical resection after treatment with imatinib. If the tumor can still be resected, reoperation should be performed; if it cannot be resected, then imatinib should be increased to 800mg/d or sunitinib 37.5mg/d. Imatinib has different degrees of toxic effects. The most common side effects are gastrointestinal reactions, hepatic and renal impairment, bone marrow hematopoietic suppression, eyelid edema and rash, but most of them are mild to moderate and can be relieved or disappeared with symptomatic treatment. The problem of drug dependence occurs with long-term use. Sunitinib is the only tyrosine kinase inhibitor currently recommended by the NCCN and ESMO guidelines as a second-line treatment for patients with GIST who have failed imatinib therapy. Du Yu [18] summarized that imatinib mesylate plus dosing, switching to sunitinib malate, and symptomatic management, three follow-up treatments for GIST, and the efficiency (28.6% ) and clinical benefit rate (92.9% ) of sunitinib malate group were significantly higher than the other two groups. 5. In conclusion, GST is clinically rare and lacks specificity of symptoms. Therefore, the difficulty of early diagnosis, lack of uniform grading criteria, high tumor recurrence rate and drug resistance are still difficult problems that we must face and solve. It is believed that as the understanding of gastrointestinal mesenchymal tumors continues to deepen and the diagnosis and treatment continues to improve, the prognosis of GST will further improve.