Abstract
Introduction
Gastrointestinal stromal tumor (GIST) is the most common primary mesenchymal tumor of the gastrointestinal tract. The clinical presentation of GIST varies widely and ranges from being asymptomatic to being a life-threatening emergency in the form of gastrointestinal bleeding or bowel obstruction. Multiple prognostic factors have been identified for GIST, including, most importantly, larger tumor size (>5 cm), higher mitotic activity (>5 per 50 high-power fields), rupture of the mass, site of the mass, and personal history of GIST.
Objective
In this paper, we aim to study bleeding in gastrointestinal tumors as a predictor for morbidity and mortality and investigate possible factors influencing bleeding in GIST.
Method
A retrospective study design was used. Electronic medical records of 39 patients diagnosed with GIST in Prince Sultan Military Medical City in Riyadh between January 2015 and December 2020 were retrieved. Normal variables were presented as mean and standard deviation, while non-normal variables were presented as median and interquartile range. Student t-test or Mann-Whitney test was used to compare quantitative data when appropriate. The distribution of recurrence and survival was plotted with the Kaplan-Meier curve and compared between the two groups (bleeding and non-bleeding groups) with the log-rank test. Factors affecting recurrence and mortality were assessed using univariable Cox regression.
Result
A total of 39 cases of GIST tumors were included in this study. Patients were categorized according to the presentation into two groups: patients with no bleeding (n= 28) and those presented with bleeding (n= 11). The gender distribution was equal. The mean age was 58.14± 12.46 in patients with non-bleeding GIST and 58.18± 16.01 in patients with bleeding GIST. The most common location of GIST was the stomach (22 cases, 56%). Neither group had any significant differences regarding age (P-value = 0.993), gender (P-value = 0.648), tumor location (P-value = 0.057), size (P-value = 0.250), cluster of differentiation (CD) markers, histological types (P-value = 0.692), and shape (P-value = 0.079). There was no significant difference in recurrence between both groups (log-rank P = 0.972). There was no significant difference in survival between both groups (Log-rank P = 0.506).
Conclusion
We concluded that bleeding GIST was not a significant predictor for recurrence rate or higher mortality. This can help in the debate of whether bleeding GIST should be added as an independent factor in risk stratification. Despite that, further studies are needed to identify more variables that can add more accuracy to the pre-existing risk stratification systems.
Keywords: gastrointestinal bleeding, non-bleeding gist, bleeding gist, gist, gastrointestinal stromal tumors
Introduction
Gastrointestinal stromal tumor (GIST) is the most common primary mesenchymal tumor of the gastrointestinal tract. GIST arises from the interstitial cells of Cajal, which are considered the pacemakers of the intestinal wall's musculature [1]. It is believed that GIST results from a mutated KIT gene, which is a proto-oncogene encoding the receptor tyrosine kinase protein known as tyrosine-protein kinase KIT, cluster of differentiation 117 (CD117), or mast/stem cell growth factor receptor (SCFR), and/or platelet-derived growth factor receptor alpha (PDGFRA) polypeptide, leading to higher rates of cell division and eventually tumor [2]. GIST can develop anywhere along the gastrointestinal tract but is found most commonly in the stomach, small bowel, and colon, and rarely in the omentum, mesentery, and behind the peritoneum (retroperitoneal origin) [3]. The clinical presentation of GIST varies widely and ranges from being an incidental finding on the radiological or endoscopic work-up to being an emergency in the form of gastrointestinal bleeding or bowel obstruction. Gastrointestinal bleeding is a life-threatening clinical presentation of GIST. To establish a diagnosis of GIST, several imaging modalities can be used, including computed tomography (CT) scan, magnetic resonance imaging (MRI), and positron emission tomography (PET), but CT scan remains the preferred modality as it is the most sensitive tool for visualizing local and distant metastasis. Under histological examination, a positive CD117 marker is diagnostic in more than 94% of cases. The mainstay of treatment is surgical resection, but the use of imatinib, which is a small molecule inhibitor targeting multiple tyrosine kinases, including c-KIT, might be considered in patients with poor prognostic factors and those with metastasis as it improves survival and decreases recurrence when used as neoadjuvant and/or adjuvant therapy. Multiple prognostic factors have been identified for GIST, including most importantly larger tumor size (>5 cm), higher mitotic activity (>5 per 50 high-power fields), rupture of the mass, site of the mass, and personal history of GIST [4, 5]. Tumor rupture can result not only in direct spread of the tumor to the abdomen but also in bleeding, which increases the overall morbidity and mortality. In this paper, we aim to study bleeding in gastrointestinal tumors as a predictor for morbidity and mortality and investigate possible factors influencing bleeding in GIST.
Materials and methods
Study design and settings
This study is a retrospective descriptive study that included a total of 39 adult patients diagnosed with GIST at Prince Sultan Military Medical City between January 2015 and December 2020.
The inclusion criteria were being older than 18 years of age, having provided consent to participate in the study, being residents of the Riyadh region, and having received a GIST diagnosis with conclusive histopathological findings.
The exclusion criteria were being younger than 18 years of age, not providing consent to participate in the study, residing outside the Riyadh region, and receiving a GIST diagnosis with inconclusive histopathological findings.
Data collection
Electronic medical records of patients diagnosed with GIST at Prince Sultan Military Medical City in Riyadh were retrieved for the period between January 2015 and December 2020. The variables included were patients' age, gender, clinical presentation (bleeding and non-bleeding GIST), radiological findings, histopathological findings with immunohistochemical markers such as CD117, DOG-1, DOG-7, CD34, SMA, desmin, and S-100 proteins, as well as surgical procedures. In November 2022, patients were followed up for recurrence or mortality. All patients of various nationalities who received a GIST diagnosis through imaging techniques, endoscopy, or excisional biopsy were included. Meanwhile, patients with inconclusive histopathological examinations were excluded.
Data management and statistical analysis
Statistical analysis was conducted using Stata 17 (Stata Corp, College Station, Texas). Quantitative data were assessed for normal distribution, and normal variables were reported as mean and standard deviation, while non-normal variables were presented as median and interquartile range. To compare quantitative data, the Student t-test or Mann-Whitney test was employed as appropriate. The distribution of recurrence and survival was illustrated using Kaplan-Meier curves and compared between the two groups (bleeding and non-bleeding) with the log-rank test. Factors influencing recurrence and mortality were examined through univariable Cox regression analysis. A P-value less than 0.05 was considered statistically significant.
Ethical consideration
All participants were volunteers. All data were kept confidential and used only for research purposes. Privacy was ensured for all of the included patients. The ethical approval was obtained from the ethical committee of Prince Sultan Military Medical City, Riyadh, Saudi Arabia.
Results
A total of 39 cases of GIST tumors were included in this study, with an almost equal gender distribution. The age at diagnosis was between 40 and 60 years for 59% of the patients. The most common location for GIST was the stomach (22 cases, 56%), followed by the small bowel (8 cases, 20.5%). Other cases were found in the duodenum, rectum, retroperitoneum, liver, lower esophagus, and intra-abdominal locations with an unknown origin. The most prevalent manifestations included abdominal pain (46.2%), followed by upper or lower gastrointestinal (GI) bleeding (28.2%). In 5 patients, GIST was incidentally discovered during radiological or endoscopic investigations or intra-operatively. Additionally, 2 cases were identified histologically post-sleeve gastrectomy. An intraoperative case was found in the jejunum during pancreatic head resection as part of the Whipple procedure. The largest tumor diameter was observed in the esophagus, measuring 18 cm. Demographic and clinical data are detailed in Table 1. The clinical presentation of patients varied, including bleeding (n= 11), hematuria (n= 1), upper gastrointestinal bleeding (n= 6), and lower gastrointestinal bleeding (n= 4). Patients were categorized based on their presentation into two groups: those with no bleeding (n= 28) and those who presented with bleeding (n= 11).
Table 1. Demographics and clinical presentation of GIST patients (n = 39) .
| Demographic/clinical presentation | N (%) | |
| Age | <40 years | 2 (5.1%) |
| 40–49 years | 13 (33.3%) | |
| 50-60 years | 10 (25.6%) | |
| >60 years | 14 (35.9%) | |
| Gender | Male | 19 (48.7%) |
| Female | 20 (52.4%) | |
| Tumor location | Duodenum | 2 (5.1%) |
| Stomach | 22 (56%) | |
| Small bowel | 8 (20.5%) | |
| Liver | 1 (2.6%) | |
| Intra-abdominal with unknown origin | 3 (7.7%) | |
| Retroperitoneal | 1 (2.6%) | |
| Lower esophagus | 1 (2.6%) | |
| Rectum | 1 (2.6%) | |
| Median size (cm) | Stomach | 5 |
| Duodenum | 2.5 | |
| Small bowel | 3 | |
| Liver | 1 | |
| Intra-abdominal with unknown origin | 16 | |
| Retroperitoneal | 8 | |
| Lower esophagus | 18 | |
| Rectum | 2 | |
| Presentation | GI bleeding | 11 (28.2%) |
| Abdominal pain | 18 (46.2%) | |
| Incidental (radiology, endoscopy, post-operatively) | 5 (12.8%) | |
| Other | 5 (12.8%) | |
| Type of surgery | Elective | 27 (69.2%) |
| Emergency | 4 (10.3%) | |
| Not operated | 8 (20.5%) | |
| Metastasis | Yes | 8 (20.5%) |
| No | 31 (79.5%) | |
| Recurrance | Yes | 5 (12.8%) |
| No | 34 (87.2%) | |
| Mitosis (≤5 vs >5 per 50 HPF) | Stomach | 15 vs 5 |
| Duodenum | 2 vs 0 | |
| Small bowel | 3 vs 3 (2 missing) | |
| Rectum | 0 vs 1 | |
| Missing | 8 | |
| Immune histochemical assay | DOG-1 | 28 (71.8%) |
| CD 34 | 20 (51.3%) | |
| SMA | 6 (15.3%) | |
| DOG-7 | 4 (10.2%) | |
| h-Caldesmin | 2 (5.1%) | |
| Demsin | 1 (2.6%) | |
| Type of cells in histopathology | Spindle-cell | 25 (64.1%) |
| Epitheloid | 0 (0%) | |
| Mixed | 8 (20.5%) | |
| Not conclusive | 6 (15.4%) | |
Baseline and tumor characteristics
Table 2 compares baseline and tumor characteristics between patients who presented with bleeding and other presentations. Neither group had any differences regarding age, gender, tumor location, size, cluster of differentiation (CD) markers, histological types, and shape. Emergency surgery was required more frequently in patients presented with bleeding (P= 0.001).
Table 2. Comparison of baseline characteristics between patients who presented with bleeding versus other presentations.
| Factor | Type of Presentation | P-Value | |
| Non-bleeding GIST (n=28) | Bleeding GIST (n=11) | ||
| Age (years) | 58.14± 12.46 | 58.18± 16.01 | 0.993 |
| Gender | 0.648 | ||
| Male | 14 (73.7%) | 5 (26.3%) | |
| Female | 15 (53.57%) | 5 (45.45%) | |
| Tumor location | 0.057 | ||
| Gastric | 17 (60.71%) | 6 (54.55%) | |
| Duodenal | 0 | 2 (18.18%) | |
| Jejunum | 2 (7.14%) | 2 (18.18%) | |
| Ilium | 3 (10.71%) | 1 (9.09%) | |
| Others | 6 (21.43%) | 0 | |
| Metastasis | 4 (14.29%) | 3 (27.27%) | 0.379 |
| Tumor size (cm) (largest diameter) | 3 (1.5- 9.5) | 5.3 (4.4- 7) | 0.250 |
| Mitosis (>5/high power field) | 9/24 (37.50%) | 3/9 (33.33%) | >0.99 |
| CD markers | (n= 28) | (n= 10) | |
| S100 | 18 (64.29%) | 5 (50%) | 0.428 |
| DOG | 23 (82.14%) | 6 (60%) | 0.205 |
| SMA | 14 (50%) | 5 (50%) | >0.99 |
| Desmin | 12 (42.86%) | 4 (40%) | >0.99 |
| CD34 | 17 (60.71%) | 6 (60%) | >0.99 |
| CD117 | 25 (92.59%) | 10 (100%) | >0.99 |
| Emergency surgery | 0 | 5/10 (50%) | 0.001 |
| Histological type | 0.692 | ||
| Spindle | 17/24 (70.83%) | 8/10 (80%) | |
| Mixed | 7/24 (29.17%) | 2/10 (20%) | |
| Tumor shape | 0.079 | ||
| Exophytic | 10/15 (66.67%) | 2/7 (28.57%) | |
| Endophytic | 3/15 (20%) | 5/7 (71.43%) | |
| Mixed | 2 (13.33%) | 0 | |
The operation was performed endoscopically (one patient in each group), laparoscopic (13 vs. 7 in no bleeding vs. bleeding), open (7 vs. 1 in no bleeding vs. bleeding), and Whipple was performed in two patients (one in each group). There was no difference in the type of surgical procedure between groups (P = 0.440).
Outcomes
Esophagogastroduodenoscopy (EGD) evidence of bleeding occurred in two patients: one in non-bleeding (3.85%) versus one in bleeding (10%) (P = 0.484). Postoperative hemoglobin was 11.8 ± 1.87 mg/dl in the non-bleeding group versus 9.93 ± 1.85 in the bleeding group (P = 0.024).
Long-term outcomes:
Figure 1 shows the Kaplan-Meier curve for the freedom from recurrence in patients presented with bleeding versus no bleeding. The median follow-up time was 50 months (range: 36-70 months). Recurrence occurred in five patients (four in non-bleeding presentation and one in bleeding presentation). There was no significant difference in recurrence between both groups (log-rank P = 0.972). Freedom from recurrence at seven years was non-significantly higher in patients with no bleeding (93% in the bleeding group vs. 88% in the non-bleeding group).
Figure 1. Kaplan-Meier curve for the freedom from recurrence in patients presented with bleeding versus no bleeding.
Figure 2 shows the Kaplan-Meier survival curve in patients presented with bleeding versus no bleeding. Death at follow-up occurred in four patients, two in each group. There was no difference in survival between both groups (log-rank P = 0.506). Survival at seven years was not significantly lower in patients with bleeding (92% in the non-bleeding group vs. 61% in the bleeding group).
Figure 2. Kaplan-Meier survival curve in patients presented with bleeding vs. no bleeding.
Table 3 analyzes the factors associated with recurrence and mortality. There was no significant association between the baseline characteristics and mortality. Increased tumor size was associated with a higher recurrence rate (HR: 1.29 (95% CI: 1.02-1.62); P = 0.031).
Table 3. Univariable Cox regression analysis for factors associated with recurrence and mortality.
| Risk Factor | Recurrence | Mortality | ||
| HR (95% CI) | P-Value | HR (95% CI) | P-Value | |
| Age (years) | 0.90 (0.78-1.04) | 0.161 | 1.08 (0.98-1.19) | 0.116 |
| Female | 0.47 (0.04-5.16) | 0.534 | 2.26 (0.23-22.32) | 0.485 |
| Tumor location | 1.26 (0.63-2.53) | 0.512 | 1.42 (0.75-2.71) | 0.283 |
| Metastasis | 2.74 (0.25-30.35) | 0.412 | 4.99 (0.69-35.72) | 0.109 |
| Tumor size (cm) | 1.29 (1.02-1.62) | 0.031 | 0.80 (0.53-1.20) | 0.273 |
| CD markers | ||||
| S100 | - | 1.41 (0.13-15.56) | - | 0.779 |
| DOG | - | 0.20 (0.02-2.25) | - | 0.190 |
| SMA | - | 0.63 (0.06-7.13) | - | 0.710 |
| Desmin | 0.67 (0.06-7.35) | 0.59 (0.05-6.50) | - | 0.664 |
| CD34 | 1.50 (0.14-16.64) | 2.06 (0.17-24.42) | - | 0.567 |
| CD117 | - | - | - | - |
| Emergency surgery | 4.65 (0.29-74.89) | 0.279 | - | - |
| Histological type | 2.47 (0.74-8.20) | 0.140 | - | - |
| Tumor shape | - | - | 1.06 (0.08-13.57) | 0.963 |
| Bleeding | 1.03 (0.09-11.56) | 0.980 | 1.93 (0.27-13.87) | 0.513 |
Discussion
Several parameters are used to predict the outcome and prognosis of GIST, such as KIT gene and PDGFRA polypeptide mutations, size and location of the mass, and mitotic rate. However, these parameters are not consistently efficient predictors, which highlights the need to study more prognostic factors [6-9]. In our study, tumor size was significantly associated with a higher recurrence rate, consistent with the literature. On the other hand, the location of the tumor was not significantly associated with a higher recurrence rate. Although not statistically significant, bleeding GIST had higher recurrence and mortality rates. This finding is not consistent with the results of Liu et al. [10]. This may be correlated with our smaller sample size and the fact that patients with bleeding GIST presumably undergo emergency surgery more frequently, which might influence the ability to achieve an R0 resection, leading to a relatively higher recurrence rate. The incidence of bleeding GIST was consistent with the literature. Bleeding was more frequent in larger tumor sizes. In terms of metastasis incidence, there was almost no difference between bleeding and non-bleeding groups, despite the belief that bleeding results from the rupture of the tumor, subsequently leading to the invasion of nearby structures and eventually to metastasis. We concluded that bleeding GIST was higher in gastric GIST, which might be explained by the relatively richer blood supply in the stomach. In contrast to our conclusion, multiple studies suggest that the smaller lumen of the intestine is associated with more mucosal injury, tumor rupture, and bleeding [11,12]. In the literature, there was no gender risk for bleeding, which is consistent with the findings of our study. In a study done by Holmebakk et al., tumor rupture was categorized as major and minor. Major rupture was defined as mass rupture or bowel perforation, while minor rupture was defined as tumor penetration into the peritoneum. In their studies, only major rupture was associated with a higher recurrence rate. [13, 14]. This classification can be taken into consideration in future studies to evaluate rupture and bleeding in GIST as risk factors for both recurrence and mortality.
Even if R0 resection was achieved for GIST, patients tend to have a high recurrence rate (40-50%). However, this rate improved with the introduction of imatinib [12,13,15]. More efforts are needed to study additional parameters that can be involved in stratifying the risk of GIST, as this will help in tailoring the optimal medical and surgical plan for each patient.
Limitation
Our study has its own limitations that should be avoided by future researchers. Firstly, This study only included patients from a single center belonging to the Saudi Ministry of Defense. Therefore, the results yielded in this paper should not be generalized. Secondly, despite the limited prevalence of GIST, a larger sample size than ours (n = 39) would have yielded more accurate results.
Conclusions
In our study, bleeding GIST was not a significant predictor of recurrence rate or higher mortality. This can help in the debate of whether bleeding GIST should be added as an independent factor in risk stratification. Despite saying that, further studies are needed to study more variables that can add more accuracy to the pre-existing risk stratification systems.
The authors have declared that no competing interests exist.
Human Ethics
Consent was obtained or waived by all participants in this study. The Ethical Committee of Prince Sultan Military Medical City issued approval E-2146
Animal Ethics
Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue.
References
- 1.Small bowel gastrointestinal stromal tumor: a retrospective study of 32 cases at a single center and review of the literature. Zhou L, Liao Y, Wu J, Yang J, Zhang H, Wang X, Sun S. Ther Clin Risk Manag. 2018;14:1467–1481. doi: 10.2147/TCRM.S167248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Case report: jejunal gastrointestinal stromal tumour, a rare tumour, with a challenging diagnosis and a successful treatment. Sankey RE, Maatouk M, Mahmood A, Raja M. J Surg Case Rep. 2015;2015 doi: 10.1093/jscr/rjv050. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Gastrointestinal perforation due to gastrointestinal stromal tumour (GIST) in small intestine-a case report. Sagar S, lftekhar U, Sharfuzzaman A, lslam SN, Hasan MM. J Surg Sci. 2019;20:65–67. [Google Scholar]
- 4.Diagnosis of gastrointestinal stromal tumors: a consensus approach. Fletcher CD, Berman JJ, Corless C, et al. Hum Pathol. 2002;33:459–465. doi: 10.1053/hupa.2002.123545. [DOI] [PubMed] [Google Scholar]
- 5.Risk stratification of patients diagnosed with gastrointestinal stromal tumor. Joensuu H. Hum Pathol. 2008;39:1411–1419. doi: 10.1016/j.humpath.2008.06.025. [DOI] [PubMed] [Google Scholar]
- 6.Mutational status of KIT and PDGFRA and expression of PDGFRA are not associated with prognosis after curative resection of primary gastrointestinal stromal tumors (GISTs) Kern A, Görgens H, Dittert DD, et al. J Surg Oncol. 2011;104:59–65. doi: 10.1002/jso.21905. [DOI] [PubMed] [Google Scholar]
- 7.Gain-of-function mutations of c-kit in human gastrointestinal stromal tumors. Hirota S, Isozaki K, Moriyama Y, et al. Science. 1998;279:577–580. doi: 10.1126/science.279.5350.577. [DOI] [PubMed] [Google Scholar]
- 8.Gastrointestinal stromal tumors: pathology and prognosis at different sites. Miettinen M, Lasota J. Semin Diagn Pathol. 2006;23:70–83. doi: 10.1053/j.semdp.2006.09.001. [DOI] [PubMed] [Google Scholar]
- 9.Characteristics of emergency gastrointestinal stromal tumor (GIST) Uçar AD, Oymaci E, Carti EB, Yakan S, Vardar E, Erkan N, Mehmet Y. https://pubmed.ncbi.nlm.nih.gov/26897944/ Hepatogastroenterology. 2015;62:635–640. [PubMed] [Google Scholar]
- 10.Gastrointestinal bleeding is an independent risk factor for poor prognosis in GIST patients. Liu Q, Li Y, Dong M, Kong F, Dong Q. Biomed Res Int. 2017;2017:7152406. doi: 10.1155/2017/7152406. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gastrointestinal wall thickness measured with transabdominal ultrasonography and its relationship to demographic factors in healthy subjects. Nylund K, Hausken T, Ødegaard S, Eide GE, Gilja OH. Ultraschall Med. 2012;33:0–32. doi: 10.1055/s-0031-1299329. [DOI] [PubMed] [Google Scholar]
- 12.How to perform gastrointestinal ultrasound: Anatomy and normal findings. Atkinson NS, Bryant RV, Dong Y, et al. World J Gastroenterol. 2017;23:6931–6941. doi: 10.3748/wjg.v23.i38.6931. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Definition and clinical significance of tumour rupture in gastrointestinal stromal tumours of the small intestine. Hølmebakk T, Bjerkehagen B, Boye K, Bruland Ø, Stoldt S, Sundby Hall K. Br J Surg. 2016;103:684–691. doi: 10.1002/bjs.10104. [DOI] [PubMed] [Google Scholar]
- 14.Surgical options for localized and advanced gastrointestinal stromal tumors. Frankel TL, Chang AE, Wong SL. J Surg Oncol. 2011;104:882–887. doi: 10.1002/jso.21892. [DOI] [PubMed] [Google Scholar]
- 15.Surgery and imatinib in the management of GIST: emerging approaches to adjuvant and neoadjuvant therapy. Eisenberg BL, Judson I. http://10.1245/ASO.2004.09.011. Ann Surg Oncol. 2004;11:465–475. doi: 10.1245/ASO.2004.09.011. [DOI] [PubMed] [Google Scholar]