Clinicopathological features and prognosis of mesenteric gastrointestinal stromal tumor: evaluation of a pooled case series

Background Due to the extremely rare incidence, data of clinicopathological features and prognosis of mesenteric gastrointestinal stromal tumors (GISTs) are limited. Therefore, the aim of the present study was to investigate the clinicopathological features and prognosis of mesenteric GISTs. Patients and Methods Mesenteric GISTs cases were obtained from our center and from case reports and clinical series extracted from MEDLINE. Clinicopathological features and survivals were analyzed. Results A total of 114 mesenteric GISTs were enrolled in present study. The most common symptom was abdominal pain (20/72, 27.8%), followed by abdominal mass (13/72, 18.1%) and distention (9/72, 12.5%). Most tumors exceeded 10 cm in diameter (71/112, 63.4%), exceeded 5/50HPF in mitotic index (50/85, 58.8%), and were high risk (82/90, 91.1%). The five-year disease free survival (DFS) and disease specific survival (DSS) was 57.7% and 60.1%, respectively. Tumor size and mitotic index were associated with DFS and DSS. The distribution of tumor size, histological type, mitotic index and NIH risk category were significantly different between mesenteric and gastric GISTs. Prognosis of mesenteric GISTs was worse than that of gastric GISTs. However, multivariate analysis showed that location was not an independent prognostic factor for mesenteric and gastric GISTs. Conclusions Most mesenteric GISTs exceeded 10 cm in diameter, exceeded 5/50HPF in mitotic index and were high risk. Mesenteric GISTs differed significantly from gastric GISTs in respect to clinicopathologic features. Mitotic index and tumor size were prognostic factors for mesenteric GISTs. The prognosis were comparable between mesenteric and gastric GISTs.

Due to the extremely rare incidence, reporting of mesenteric GISTs has been limited to individual case reports and case series of small numbers. Studies involving large numbers of mesenteric GISTs are lacking. Thus, several questions remain unanswered, including clinical and pathological characteristics and prognosis. Therefore, the aim of the present study was to investigate the clinicopathological features and prognosis of mesenteric GISTs.
Prognostic factors for DFS and DSS of mesenteric GISTs were shown in Table 3. Univariate analysis showed that tumor size and mitotic index were prognostic factors for mesenteric GISTs. However, multivariate analysis showed that tumor size was the only independent risk  The clinicopathological features of 114 mesenteric GISTs including age, gender, tumor size, histological type, mitotic index and NIH risk category were compared with 297 gastric GISTs in our institution ( Table 4). The results showed that the distribution of tumor size, histological type, mitotic index and NIH risk category were significantly different between mesenteric and gastric GISTs (all P < 0.05).
In order to compare the prognosis of mesenteric GISTs with gastric GISTs, 57 cases of mesenteric GISTs and 217 cases of gastric GISTs with follow up data were analyzed. The results showed that DFS and DSS of mesenteric GISTs were significantly lower than that of gastric GISTs ( Figure 5). Further, multivariate analysis was performed to evaluate the prognostic value       (Table 5). However, the results showed that location was not an independent risk factor for prognosis of mesenteric and gastric GISTs (both P > 0.05).

DISCUSSION
Due to the extremely rare incidence, studies involving large numbers of mesenteric GISTs are lacking. Therefore, the aim of the present study was to investigate the clinicopathological features and prognosis of mesenteric GISTs. The present study represents the largest analysis of mesenteric GISTs.
The precise etiology of mesenteric GISTs remains to be clarified. Some investigators have proposed that EGISTs are mural tumors with extensive extramural growth resulting in eventual loss of connection with gut wall [6]. However, this hypothesis lacks evidence. On the other hand, other researchers have proposed that GISTs may arise from a common precursor cell of ICC and smooth muscle cell, which may account for their growth from and outside the GI tract [7]. Terada et al. have demonstrated the existence of scattered KIT-positive ICC like cells in surface of the normal mesocolon [8]. This further provide evidence for the hypothesis that EGISTs may arise from precursor cell of ICC outside the GI tract. However, the existence of ICC like cells in the mesostenium have not been identified.
The spectrum of clinical presentation of GISTs is broad and depends on tumor location and tumor size. For mesenteric GISTs, tumors appear to have enough space to grow and may present clinical symptoms after a significant period of time with a considerable tumor size. In our present study, most tumors exceeded 10 cm in diameter. Thus, early diagnosis of mesenteric GISTs is very difficult. Once mesenteric GISTs reached a significant size, symptoms will appear. In our present study, the most common symptoms include abdominal pain, mass and distension.
Even with R0 resection, there is a high risk of recurrence and distant metastasis. However, no mention of mesenteric GISTs specific recurrence or metastasis was made previously. In our present study, half of patients with tumor progression after R0 resection suffered from abdominal recurrence. For distant metastasis, the most common site was liver. Metastasis to lung and brain was also occasionally found.
In 1998, Hirota et al. reported their groundbreaking discovery of KIT mutations in GISTs. It is now established that 70% to 80% of GISTs harbor KIT mutation [9], and PDGFRA mutation occur in approximately 8% to 10% of GISTs [10]. In our present study, gene mutations were recorded in only eighteen patients. Among them, 7 patients (38.9%) carrying PDGFRA mutation. The incidence of PDGFRA mutation in our present study was relatively higher than previous report. This indicated that the incidence of KIT and PDGFRA gene mutation could be various from each other depend on the location of GISTs. However, the association between the tumor location and gene mutation status needs further investigation.
In our present study, most tumors exceeded 10 cm in diameter and almost all the tumors were high risk. Therefore, the existing classification criteria which defined by a combination of mitotic index and tumor size may not be applicable to mesenteric GISTs. Reith et al. reported that high cellularity, mitotic index exceeds 2/50 HPF and presence of necrosis were factors indicative of a potentially aggressive clinical course for EGIST [11]. However, the relatively short follow up period in the study may result in bias of the data. Thus, a more appropriate grading system may be needed for the classification of EGISTs.
Tumor location is also one prognostic factor for GISTs [12], and it was considered that EGISTs were more aggressive than gastric GISTs. However, the modified NIH risk classification distinguishes only gastric from non-gastric GISTs, and the prognosis of mesenteric GISTs are not discussed. Thus, the prognosis of mesenteric and gastric GISTs were compared. We found that the prognosis of mesenteric GISTs was significantly worse than that of gastric GISTs. However, multivariate analysis showed that tumor location was not an independent risk factor for prognosis of mesenteric and gastric GISTs. We considered that the poor prognosis of mesenteric GISTs was mainly attributed to the larger tumor size and higher mitotic index, not to location.
There were a few limitations in our present study. Firstly, the present study is a retrospective analysis and the completeness of data is limited. Secondly, the sample size was not large enough, which will result in statistical bias. Thirdly, the clinicopathological features and prognosis of mesenteric GISTs were not compared with EGISTs in other locations.

CONCLUSIONS
Most mesenteric GISTs exceeded 10 cm in diameter, exceeded 5/50HPF in mitotic index and were high risk. Mesenteric GISTs differ significantly from gastric GISTs in respect to clinicopathologic features. Mitotic index and tumor size were risk factors for prognosis of mesenteric GISTs. The prognosis were comparable between mesenteric and gastric GISTs.  [7,8, including 47 cases and 6 case series [47][48][49][50][51][52] including 59 cases. As a result, a total of 114 mesenteric GISTs patients were identified ( Figure  1). In addition, the clinicopathological characteristics of 297 patients of gastric GISTs in our center were analyzed and compared with mesenteric GISTs. Among them, the prognosis of 217 gastric GISTs patients with followup data were analyzed and compared with mesenteric GISTs. This study was approved by the Ethics Committee of Xijing Hospital, and written informed consent was obtained from the eight patients in our center.

PATIENTS AND METHODS
Data including age, gender, accompanied tumor, symptoms, tumor size, imaging features, surgical intervention, histological type, immunohistochemical features, mutational status, mitotic index, NIH risk category, adjuvant therapy, tumor progression and survival data were recorded. The tumors were categorized into very low, low, intermediate and high risk groups according to the modified NIH risk classification criteria [53]. For survival analysis, the inclusion criteria were listed as follows: 1. without distant metastasis, 2. without GISTs in other locations, 3. R0 resection, 4. without other malignant tumors, 5. without neoadjuvant imatinib therapy, 6. with follow up data. Due to data acquisition, completeness of data is limited.
Data were processed using SPSS 22.0 for Windows (SPSS Inc., Chicago, IL, USA). Discrete variables were analyzed using the Chi-square test or Fisher's exact test. Numerical variables were expressed as the mean ± SD unless. Significant predictors for prognosis identified by univariate analysis were further assessed by multivariate analysis using Cox's proportional hazards regression model was employed for multivariate analysis. Evaluation of disease-free-survival (DFS) and disease-specificsurvival (DSS) were obtained by the Kaplan-Meier method. DFS was defined as the length of time from the date of surgery to the date of recurrence. DSS was defined as the length of time from the date of surgery to the date of cancer associated death. The P value was considered to be statistically significant at the 5% level.

CONFLICTS OF INTEREST
There are no financial or other relations that could lead to a conflict of interest.