Search

Background: Till present, there are still controversies over the epidemiology, pathological features, types of surgical treatment, and prognoses of primary small gastric GISTs (gGISTs)., Methods: From January 1998 to January 2015, patients with primary small gGIST admitted from four high-volume medical centers of the Southern China were enrolled and their data were analyzed to evaluate their clinicopathological features, treatment and prognostic factors to provide evidence-based medical experience for clinical practice., Results: A total of 276 primary small gGIST cases over a period of 18 years were investigated and had a median age of 60 years (range 27-91 years old). Regarding the tumor sites, 24 (8.7%) cases were in the cardia of the stomach, 107 (38.8%) in the fundus, 117 (42.4%) in the gastric body, and 28 (10.1%) in the gastric antrum. Eleven patients (4.0%) underwent a preoperative biopsy. A total of 137 (49.6%), 75 (27.2%), and 64 (23.2%) patients underwent laparoscopic, open resection, and endoscopic resection, respectively. Sixty-four patients (23.2%) had local endoscopic resection, 172 (62.3%) had wedge resection, 7 (2.5%) had proximal gastrectomy, 19 (6.9%) had distal gastrectomy, and 14 (5.1%) had total gastrectomy. Mitotic counts were ≤ 5/50, (5-10)/50, and > 10/50 per HPF in 259 (93.8%), 7 (2.5%), and 10 (3.6%) cases, respectively. There were 259 cases (97.1%) of spindle cell type, 7 (2.5%) epithelial cell types and one case (0.4%) of mixed type. Immunohistochemistry showed 74.6% (206/276), 98.2% (271/276), and 97.4% (269/276) of the patients had co-expression of CD34+, CD117+, and DOG-1+, respectively. Thirty-nine patients underwent genetic testing (39/276, 14.1%). Three patients (1.1%) had positive resection margin. Five high-risk patients received follow-up treatment with imatinib with a median follow-up time of 38 months (range 3-156 months). The overall 1-, 3-, and 5-year overall survival rates were 100%, 99.6%, and 99.1%, respectively., Conclusion: Though the incidence of primary small gGISTs increased per annum, the overall survival prognoses were high. Surgery or endoscopic resection was the primary mode of treatment. Pathological features of primary small gGISTs were similar to large gGISTs, and to achieve a timely surgical intervention, the identification of intermediate- and high-risk cases should be a future focus of study.

Background: The most common site of gastrointestinal stromal tumors (GISTs) is the stomach, and gastric GISTs (gGISTs) occur most often in elderly patients. However, the clinicopathological features, treatment patterns, and prognosis of elderly patients with gGISTs remain unclear., Methods: We retrospectively collected clinicopathological and prognostic data for patients with primary gGISTs who underwent curative-intent resection at 10 medical centers in China from 1998 to 2015., Results: Over the 18 years, 10 medical centers treated 1846 patients with primary gGISTs by curative-intent resection. The median age was 59 (range 18-91) years. The patients were classified into two groups according to age, namely an elderly group (≥ 65 years of age) and a nonelderly group (< 65 years of age). The elderly group had more comorbidities (40.7% vs 23.5%, p = 0.011), a higher rate of postoperative complications (14.4% vs 8.7%, p = 0.031), and a lower proportion of intermediate/high-risk patients who received adjuvant therapy (30.0% vs 66.8%, p = 0.001) than did the nonelderly group. Regarding pathological outcomes, a significant difference in tumor necrosis was observed between the two groups (p = 0.002), and more cases of tumor necrosis occurred in the elderly group than in the nonelderly group. Regarding postoperative recovery outcomes, no significant difference was observed between the two groups. Univariate analysis showed that age, postoperative complications, adjuvant therapy, tumor size, mitotic count, modified National Institutes of Health (NIH) risk category, and tumor necrosis were factors that affected disease-free survival (DFS). Multivariate analysis showed that modified NIH risk category was the only independent factor affecting DFS. The 5-year DFS rates in the nonelderly and elderly groups were 88.1% and 81.4%, respectively (p = 0.034), and the 5-year overall survival (OS) rates were 90.4% and 85.5% (p = 0.038), respectively., Conclusions: Currently, the treatment patterns for elderly patients with gGISTs remain the same as those for young patients with gGISTs. Elderly gGIST patients had more comorbidities and postoperative complications than did nonelderly gGIST patients, and fewer elderly gGIST patients received postoperative adjuvant therapy. Elderly gGIST patients also had a higher rate of tumor necrosis and worse DFS and OS than did young gGIST patients. Further exploration into the diagnosis and treatment patterns of elderly patients is therefore essential.

Objective: To develop and evaluate a radiomics nomogram for differentiating the malignant risk of gastrointestinal stromal tumours (GISTs)., Methods: A total of 222 patients (primary cohort: n = 130, our centre; external validation cohort: n = 92, two other centres) with pathologically diagnosed GISTs were enrolled. A Relief algorithm was used to select the feature subset with the best distinguishing characteristics and to establish a radiomics model with a support vector machine (SVM) classifier for malignant risk differentiation. Determinant clinical characteristics and subjective CT features were assessed to separately construct a corresponding model. The models showing statistical significance in a multivariable logistic regression analysis were used to develop a nomogram. The diagnostic performance of these models was evaluated using ROC curves. Further calibration of the nomogram was evaluated by calibration curves., Results: The generated radiomics model had an AUC value of 0.867 (95% CI 0.803-0.932) in the primary cohort and 0.847 (95% CI 0.765-0.930) in the external cohort. In the entire cohort, the AUCs for the radiomics model, subjective CT findings model, clinical index model and radiomics nomogram were 0.858 (95% CI 0.807-0.908), 0.774 (95% CI 0.713-0.835), 0.759 (95% CI 0.697-0.821) and 0.867 (95% CI 0.818-0.915), respectively. The nomogram showed good calibration., Conclusions: This radiomics nomogram predicted the malignant potential of GISTs with excellent accuracy and may be used as an effective tool to guide preoperative clinical decision-making., Key Points: • CT-based radiomics model can differentiate low- and high-malignant-potential GISTs with satisfactory accuracy compared with subjective CT findings and clinical indexes. • Radiomics nomogram integrated with the radiomics signature, subjective CT findings and clinical indexes can achieve individualised risk prediction with improved diagnostic performance. • This study might provide significant and valuable background information for further studies such as response evaluation of neoadjuvant imatinib and recurrence risk prediction.

This study aimed to develop and validate a prognostic nomogram for recurrence-free survival (RFS) after surgery in the absence of adjuvant therapy to guide the selection for adjuvant imatinib therapy based on Residual Neural Network (ResNet). The ResNet model was developed based on contrast-enhanced computed tomography (CE-CT) in a training cohort consisted of 80 patients pathologically diagnosed gastrointestinal sromal tumors (GISTs) and validated in internal and external validation cohort respectively. Independent clinicopathologic factors were integrated with the ResNet model to construct the individualized nomogram. The performance of the nomogram was evaluated in regard to discrimination, calibration, and clinical usefulness. The ResNet model was significantly associated with RFS. Integrable predictors in the individualized ResNet nomogram included the tumor site, size, and mitotic count. Compared with modified NIH, AFIP, and clinicopathologic nomogram, both ResNet nomogram and ResNet model showed a better discrimination capability with AUCs of 0·947(95%CI, 0·910-0·984) for 3-year-RFS, 0·918(0·852-0·984) for 5-year-RFS, and AUCs of 0·912 (0·851-0·973) for 3-year-RFS, 0·887(0·816-0·960) for 5-year-RFS, respectively. Calibration curve shows the good calibration of the nomogram in terms of the agreement between the estimated and the observed 3- and 5- year outcomes. Decision curve analysis showed that the ResNet nomogram had a higher overall net benefit. In conclusion, we presented a deep learning-based prognostic nomogram to predict RFS after resection of localized primary GISTs with excellent performance and could be a potential tool to select patients for adjuvant imatinib therapy., (Copyright © 2018 The Authors. Published by Elsevier B.V. All rights reserved.)

Background: To determine the better risk stratification based on surgical pathology and to assess the clinical outcomes after curative resection with a new scoring system in high risk gastrointestinal stromal tumor (GIST) patients., Methods: We retrospectively evaluated 506 high-risk GIST patients who underwent curative resection as initial treatment at four centers from 2001 to 2015., Results: Multivariate analysis revealed that only Ki-67 labeling index (LI) and mitotic index were independent prognostic factors of overall survival (OS). For the two tumor-related pathological factors, Ki-67 LI > 7% and mitotic index ≥ 7/50 high power fields were allocated 1 point each. The total score was defined as the Pathological Prognostic Score (PPS). When Ki-67 LI and mitotic index were replaced by PPS, a multivariate analysis still identified PPS as an independent predictor of OS (HR 2.719; 95% CI 1.309-5.650; P = 0.007). Patients with a PPS of 0, 1, or 2 had a 5-year survival of 91.8, 79.8, and 51.0%, respectively (P = 0.001). Furthermore, an elevated PPS (PPS = 2) was associated with larger tumor size, non-stomach tumor, and open resection (all P < 0.05)., Conclusion: The PPS independently predicted postoperative survival in high-risk GIST, and it might facilitate the selection of appropriate treatment strategy for these patients.

We sought to determine whether Ki-67 labeling index (LI) was an independent prognostic factor for gastrointestinal stromal tumor (GIST). A multicenter cohort of 1022 patients undergoing surgical resection of primary GIST between August 2004 and October 2015 was retrospectively analyzed. Immunohistochemical analysis was performed to evaluate expression of Ki-67 in their paraffin-embedded tissue samples. The optimal cutoff value of Ki-67 LI was determined as 6% by receiver operating characteristics curve analysis. Multivariate analysis showed that Ki-67 LI was a significant predictor of overall survival (OS) (hazard ratio: 1.793; 95% confidence interval, 1.240-2.593; P=.002). When stratified by modified National Institutes of Health classification, it was still independently associated with OS in high-risk and non-high-risk patients (P=.001 and P=.055, respectively). Of note, the prognostic significance of Ki-67 LI was also maintained when stratified by tumor size, mitotic index, tumor site, and histological subtype (all Ps<.05). In addition, high-risk patients with Ki-67 LI >6% exhibited a significantly poorer OS rate than those with Ki-67 LI ≤6% (53.6% versus 88.7%, respectively; P=.001). The area under the receiver operating characteristics curve for Ki-67 LI was higher than that of modified National Institutes of Health classification component in high-risk patients (P=.029). Therefore, Ki-67 LI is a promising predictor of outcome in GIST, especially in high-risk patients, and it may have important clinical utility in identifying "very high-risk" patients for rational targeted therapy., (Copyright © 2017 Elsevier Inc. All rights reserved.)

Tumor necrosis is associated with poor clinical outcomes in many malignancies. We aimed to determine whether tumor necrosis was an independent predictor of outcomes in gastric gastrointestinal stromal tumors (GISTs). We retrospectively analyzed data from 740 patients undergoing curative resection for gastric GIST at four centers between 2001 and 2015. Disease-free survival (DFS) was estimated with the Kaplan-Meier method, and associations with prognosis were assessed with Cox regression models. Tumor necrosis was present in 122 cases (16.5%). The prevalence of tumor necrosis increased with higher risk-stratification, including 0.7%, 7.4%, 17.3%, and 39.3% for very low-, low-, intermediate- and high-risk tumors, respectively (P < 0.001). Tumor necrosis was associated with aggressive tumor biology, such as larger tumor size, higher mitotic index, tumor rupture, and presence of nuclear atypia (all P < 0.05). Multivariate analysis revealed that tumor necrosis was an independent predictor of unfavorable DFS (HR: 2.641; 95% CI: 1.359-5.131; P = 0.004). When stratified by the modified National Institutes of Health (NIH) classification, tumor necrosis still independently predicted DFS in high-risk patients (P = 0.001) but not in non-high-risk patients (P = 0.349). The 5-year DFS rate in high-risk patients with and without tumor necrosis was 56.5% and 82.9%, respectively (P = 0.004). Notably, the prognostic significance of tumor necrosis was maintained when the patients were stratified by age, sex, tumor location, tumor size, and mitotic index (All P < 0.05). Tumor necrosis is a useful predictor of outcomes in gastric GIST, especially in high-risk patients. Based on these results, we recommend that the current NIH classification should be further improved and expanded to include tumor necrosis as a valuable prognostic indicator., (© 2017 The Authors. Cancer Medicine published by John Wiley & Sons Ltd.)

Objective: To evaluate and compare the value of Modified NIH criteria and AFIP criteria for the risk classification of gastrointestinal stromal tumors (GIST)., Methods: Clinicopathological and follow-up data of 539 patients diagnosed as primary GIST with or without irregular tyrosine kinase inhibitors in the Nanfang Hospital(n=143), Sun Yat-sen University Cancer Center (n=138), Guangdong Provincial People's Hospital (n=102) and Wuhan Union Hospital (n=156) from January 2012 to December 2015 were retrospectively analyzed. Recurrence risks of these 539 patients were classified by the modified NIH criteria and AFIP criteria. Overall survival and tumor-free survival of patients with different risks were compared by Log-rank test and the accuracy of the two criteria in predicting postoperative recurrence was compared by receiver operating characteristic(ROC) curves., Results: Of 539 GIST patients, 283 were male and 256 were female; the age was (56.5±12.5) years old; tumors of 390 cases (72.4%) located in the stomach; tumor diameter of 178 cases (33.0%) was more than 5 cm; nuclear division number of 164 cases(30.4%) was more than 5/50 high magnification. The mean follow-up time was (37.5±13.6) months. According to the modified NIH criteria, the mean overall survival time of patients with very low, low, intermediate, and high risk was 52.0, 57.0, 56.9 and 53.6 months respectively (P=0.002), and the mean tumor-free survival time was 56.0, 58.1, 58.2 and 51.2 months respectively (P=0.000). According to the AFIP criteria, the mean overall survival time of patients with very low, low, intermediate, and high risk was 54.1, 57.8, 55.5 and 52.0 months respectively(P=0.015), and the mean tumor-free survival time was 57.3, 56.6, 54.9 and 50.4 months respectively(P=0.000). While predicting the risk of postoperative recurrence, the ROC curve of AFIP criteria has a larger area under the curve compared to the curve of the modified NIH criteria(0.689 vs 0.641, P<0.05)., Conclusion: Compared with the modified NIH criteria, AFIP criteria predicts the risk postoperative recurrence more accurately in GIST patients.

Objective: To retrospectively analyze the clinicopathology of patients with gastric gastrointestinal stromal tumor(gGIST) who underwent radical excision within 18 years in 10 domestic medical centers in order to understand the status of domestic surgical treatment of gGIST., Methods: Clinicopathological data of gGIST patients undergoing radical excision in 10 medical centers from January 1998 to January 2016 were collected, and their operational conditions, postoperative adjuvant therapy, gene detection and survival were analyzed retrospectively., Results: A total of 1 846 cases were recruited in this study, including 246 cases from Guangdong General Hospital, 331 cases from Sun Yat-sen University Cancer Center, 374 cases from Union Hospital, Tongji Medical College of Huazhong University of Science and Technology, 342 cases from Nanfang Hospital of Southern Medical University, 265 cases from Fujian Medical University Union Hospital, 148 cases from Fudan University Shanghai Cancer Center, 49 cases from West China Hospital of Sichuan University, 43 cases from Peking University Cancer Hospital and Institute, 28 cases from the 81st Hospital of Pepole's Liberation Army(PLA), 20 cases from Cancer Hospital of China Medical University, Liaoning Cancer Hospital and Institute. There were 918 male (49.7%) and 928 female patients (50.3%) with median onset age of 59(18 to 95) years old. Fundus(735 cases, 39.8%) and body (781 cases, 42.3%) of stomach were the common sites of lesions. The average size of tumor was (5.3±4.6) cm. There were 1 421 cases with mitotic count ≤5(77.0%). According to the operation procedure, 924 cases (50.1%) underwent laparoscopic surgery, 759 cases (41.1%) laparotomy, 120 cases (6.5%) endoscopic surgery, and 20 cases (1.1%) laparoscopic combined with endoscopic surgery, 6 cases (0.3%) laparoscopic excision surgery through gastric wall and cavity, and 17 cases (0.9%) laparoscopy and then were transferred to laparotomy. Wedge excision were performed in 1 308 cases (70.9%), proximal gastric excision in 226 cases(12.2%), distal gastric excision in 92 cases (5.0%), total gastrectomy in 94 cases (5.1%), and local gastrectomy in 126 cases(6.8%). Multi-visceral excision was performed in 138 cases, and the splenectomy was performed in 83 cases(60.1%)with the highest ratio. According to modified NIH classification, 399 cases(21.6%) were extreme low risk, 580 cases(31.4%) were low risk, 424 cases(23.0%) were moderate risk, 443 cases (24.0%) were high risk. A total of 461 cases received postoperative imatinib adjuvant therapy, accounting for 53.2%(461/867) of patients with moderate and high risk. Among 1 846 cases, 1 402 cases (75.9%) had complete follow-up data and the median follow-up time was 33.6 (0.1 to 158) months. The 5-year survival rates of extreme low risk, low risk, moderate risk and high risk were 100%, 98.5%, 92.5%, and 79.2% with significant difference(P=0.000)., Conclusions: Gastric GIST occurs mostly in fundus and body of stomach in China. Wedge excision is the main operational procedure and laparoscopic operation is over 50%. General prognosis of gastric GIST is quite good.

Objective: To elucidate the historic and current diagnosis and treatment status of gastrointestinal stromal tumor (GIST) in the Chinese population based on four high volume databases., Methods: Clinicopathological data of GIST patients with follow-up information between January 1998 and December 2015 from Sun Yat-sen University Cancer Center, Union Hospital of Huazhong University of Science and Technology, Southern Medical University Nanfang Hospital and Guangdong General Hospital were retrospectively analyzed. Kaplan-Meire method was used to draw survival curve. The accumulative survival rate was calculated by life table method. Comparison of survival rate among groups was examined by Log-rank test., Results: A total of 2 610 cases were enrolled into the study, including 667(25.6%) cases from Sun Yat-sen University Cancer Center, 754(28.9%) cases from Union Hospital of Huazhong University of Science and Technology, 692(26.5%) cases from Southern Medical University Nanfang Hospital and 497 (19.0%) cases from Guangdong General Hospital. There were 1 394 male and 1 216 female cases with the ratio of 1.15 to 1.00. The age of patients was from 18 to 95 (median 58.0) years old. Three-year was used as a time stage, then 18 years were divided into 6 stages. New GIST patients increased gradually year by year. There were 13(0.5%) cases during 1998 to 2000, 68(2.6%) cases during 2001 to 2003, 256(9.8%) cases during 2004 to 2006, 517 (19.8%) cases during 2007 to 2009, 814(31.2%) cases during 2010 to 2012, and 942 (36.1%) cases during 2013 to 2015. Primary GIST sites were esophagus in 50(1.9%) cases, stomach in 1 686(64.6%) cases, duodenum in 206 (7.9%) cases, jejunum and ileum in 446 (17.1%) cases, colon and rectum in 133 (5.1%) cases, and non-gastrointestinal tract in 89 (3.4%) cases. GIST lesions of 2 404(92.1%) cases located in the primary sites and relapse/metastasis occurred in 206 cases when consulting. Among 206 relapse/metastasis cases, liver metastasis was found in 126 (61.2%) cases, abdominal cavity/pelvic cavity metastasis in 64 (31.1%) cases, liver plus abdominal cavity/pelvic cavity metastasis in 12 (5.8%) cases, and other site metastasis in 4 (1.9%) cases. Among all the patients, 352 received gene detection, including 1 (0.4%) during 2004 to 2006, 7 (1.4%) during 2007 to 2009, 150 (18.4%) during 2010 to 2012, and 194 (20.6%) during 2013 to 2015. Most of the primary oncogenic mutational site occurred in c-Kit, including 30 (8.5%) cases in exon 9, 242 (68.8%) cases in exon 11, 4 (1.1%) cases in exon 13, 2 (0.6%) cases in exon 17, while 3 (0.9%) cases in PDGFRA exon 12 and 20 (5.7%) cases in PDGFRA 18, besides, no mutations of KIT and PDGFRA were detected in 51 (14.5%) cases. A total of 2 202 cases underwent operation, including 2 038 (92.6%) of radical resection and 164 (7.4%) of palliative resection. Among 2 038 patients undergoing radical resection, 450 (22.1%) cases were very low risk, 593 (29.1%) cases were low risk, 283 (13.9%) cases were moderate risk and 712 (34.9%) cases were high risk according to NIH risk classification. Of 995 patients with moderate and high risk, 550(55.3%) cases received postoperative imatinib adjuvant therapy, whose ratio in above 6 time stages was as follows: 0, 42.8%(12/28), 19.8%(20/101), 9.8% (21/215), 65.7% (176/268) and 85.6% (321/375). Of 206 relapse/metastasis patients, 200 (97.1%) cases received imatinib as the first-line therapy, and 22 (10.7%) received sunitinib as the second-line therapy. A total of 1 743 patients had complete follow-up data and median follow-up time was 35.9 (0.1 to 173.8) months. The 5-year overall survival rates in very low, low, moderate and high risk patients were 100%, 97%, 95% and 78% respectively., Conclusion: This retrospective study provides the largest data of GIST and indicates the historic changes of clinicopathological characteristics, diagnosis and treatment of GIST for further domestic GIST research.

Importance. Extragastrointestinal stromal tumor (EGIST) is a rare tumor, and its diagnosis and treatment strategy lack clinical guideline and relative literature evidence. In clinical practice, EGIST only misuses the pattern of GIST of diagnosis and treatment. The study hopes to find evidence of the treatment pattern for EGIST. Objective. This study aimed to compare the tumor characteristics and long-term outcomes between EGIST and GIST. The confounding function was applied to improve the result credibility in the case of small sample size. Design, Setting, and Participants. This cohort study enrolled 55 patients with EGIST who underwent surgery and were selected from four high-volume hospitals in China and 221 GIST patients who were collected from one of the four hospitals between January 2006 and September 2017. We used propensity score matching (PSM) and subgroup analysis to compare EGIST with GIST in terms of prognosis. The confounding function was used for sensitivity analysis to reduce unmeasured confounding. Results. We matched 43 patients in each of the GIST and EGIST groups by PSM. We compared EGIST data with GIST data to explore the prognostic factors between them. In the multivariate Cox regression model, tumor location of EGIST was negatively correlated with overall survival (after PSM: HR, 4.32; 95% CI, 1.22–15.26) or disease-free survival (after PSM: HR, 9.79; 95% CI, 2.22–43.31), which was also intuitively shown in the Kaplan–Meier survival curves (all P values < 0.05). In the subgroup analysis, EGIST with high risk factors had a worse prognosis than GIST. In unmeasured confounding analysis, the overall curve tends to show all combinations of c(0) of c(1) up to 2.0, none of which would bring the corrected relative risk to 1 for OS and DFS. Conclusions and Relevance. EGIST was associated with worse prognosis compared with GIST patients, particularly in EGIST patients with high risk factors, while there was a similar prognosis without those high risk factors. [ABSTRACT FROM AUTHOR]

Purpose: Gastrointestinal stromal tumors (GISTs) are commonly known to be derived from the gastrointestinal (GI) tract, but recently there have been more and more literature describing lesions with similar pathological and immunohistochemical resembling GISTs but located outside the GI tract, and they have been termed as extra-GISTs (eGISTs). However, due to the rare incidence of eGISTs, its association with survival outcomes is poorly understood, especially in the Chinese population. Here, we aimed to identify the risk factors of eGISTs and to assess their association with overall survival (OS) and disease-free survival (DFS). Patients and Methods: Data of pathologically confirmed eGISTs cases, without radiological and perioperative evidence of other primary lesions, and with no microscopically identified adhesion between the tumor and the gastrointestinal serosa, which were surgically treated between January 2006 and September 2017 were retrieved from the database of four high-volume hospitals. Immunohistochemical and genetic testing were performed on the postoperative lesions and were staged using the National Institutes of Health (NIH) criteria. Results: A total of 55 cases were retrieved. eGISTs were identified from the retroperitoneum (36.4%), mesocolon (25.5%), small bowel mesentery (12.7%), abdominopelvic cavity (12.7%), lesser omental sac (5.5%), ovary (3.6%), pancreatic capsule (1.8%), or urinary bladder (1.8%). Based on the NIH risk classification, majority of the lesion were classified as high risk (85.5%). KIT 11 was the most common mutation site (76.5%) and 25.0% of the cases were wild-type eGISTs. Multivariate analyses showed that tumor location and size were independent factors affecting prognoses. Patients with tumors in the retroperitoneum had significantly poorer OS and DFS as compared to those in the non-retroperitoneum (HR [95% CI] for OS and DFS: 2.546 [1.023– 6.337] [P = 0.037] and 2.475 [0.975– 6.273] [P = 0.049], respectively). Similar findings were found for tumors of size > 15 cm, compared to ≤ 15 cm (HR [95% CI] for OS and DFS: 5.350 [2.022– 14.156] [P < 0.001] and 3.861 [1.493– 9.988] [P = 0.003], respectively). Conclusion: eGISTs were predominantly found from the retroperitoneum and mostly classified as high risk. Those located in the retroperitoneum and of size > 15 cm had the poorer OS and DFS as compared to those in the non-retroperitoneum and of size < 15 cm. [ABSTRACT FROM AUTHOR]