Analysis of clinical-imaging-pathological features and preoperative misdiagnosis status of 47 cases of gastric schwannomas: a single-center retrospective study over 13 years.

OBJECTIVE: To identify the diagnostic difficulties of gastric schwannomas (GS) and provide evidence for improving the accuracy of clinical diagnosis. METHODS: A retrospective analysis was conducted on the data of 47 patients with pathologically confirmed GS in Northern Jiangsu People’s Hospital of Jiangsu Province from September 2012 to March 2025. Demographic characteristics, tumor markers (CA125, CA199, CEA), preoperative diagnosis, surgical methods, and postoperative immunohistochemical results (S-100, CD117, Ki67, etc.) of the patients were statistically analyzed. Immunohistochemical detection was performed using the streptavidin-peroxidase (SP) method, with a positive criterion defined as the presence of brown-yellow granules in the cytoplasm or nucleus and a positive cell proportion ≥ 5%. RESULTS: Among the 47 patients, the male-to-female ratio was 1:2.1, with a mean age of (59.83 ± 12.02) years, and 76.6% of the patients were aged ≥ 50 years. Most preoperative tumor marker levels were within the normal range (only 1 case had slightly elevated CA125, and 5 cases had abnormal CEA, including 2 cases complicated with other malignant tumors). No patients received a definitive preoperative histological diagnosis of gastric schwannoma, and the preoperative clinical diagnostic impressions were mainly classified into two most common alternative categories: non-specific gastric submucosal tumors (66.0%) and gastrointestinal stromal tumors (GISTs, 27.7%), with only a small proportion diagnosed as gastric cancer and gastric cancer combined with other tumors (6.4%), which is attributed to the intrinsic limitations of current imaging and endoscopic modalities in distinguishing GS preoperatively. Laparoscopic gastric lesion resection was the main surgical method (55.3%), followed by endoscopic resection (31.9%) and open surgery (12.8%). Postoperative immunohistochemical results showed that all patients were positive for S-100 and negative for CD117; the positive rate of other immunohistochemical markers ranged from 2.1% to 27.7%, and 89.36% of the patients had a Ki67 index ≤ 5%. CONCLUSION: GS lacks specific clinical manifestations and is prone to preoperative misdiagnosis. Postoperative pathological examination combined with immunohistochemistry (defined by S-100 positivity and CD117 negativity) is the gold standard for GS diagnosis, and patients have a good prognosis after complete surgical resection of the tumor. This study provides a reference for the clinical diagnosis and treatment of GS; however, due to the single-center design, small sample size, and lack of long-term follow-up data, the conclusions need further verification in subsequent studies.