Prognostic Factors of Gastrointestinal Stromal Tumors: A Single Institutional Retrospective Experience with Surgical Management over 20 Years

Discussion

With the improvement of imaging techniques and histological diagnostic methods, the discovery and diagnosis of GISTs has increased. It is very important in clinical practice to accurately differentiate benign from malignant tumors, and predict the prognosis of malignancies; however, this is difficult for GISTs because their biological behavior is very complicated, and GISTs of any size can exhibit malignant behavior^[4]. In this study, the rates of postoperative recurrence and metastasis were 19.0% (4/21), 22.0% (24/109), and 39.0% (77/197) in the 3 groups with tumors < 2 cm, 2-5 cm, and > 5 cm in size, respectively. Therefore, it can be assumed that every GIST is potentially malignant, and should be classified according to malignant potential rather than simply being divided into benign and malignant groups. A consensus conference held at the National Institute of Health (NIH) provided both an evidence-based definition and a practical scheme for the assessment of the risk in the clinical course of this disease. The risk categorization is based on evaluation of the size and mitotic rate of the tumors as the most reliable prognostic factors, and its use is strongly advocated. However, the design of trials assessing adjuvant treatment in GIST patients with imatinib has raised debate about the accuracy of the NIH consensus criteria for the risk of disease recurrence after surgical treatment of primary tumors^[5].

The 5-year survival rate after resection of GISTs ranged from 50%-70%. The tumor size, tumor location, necrosis, mitotic activity, PCNA index > 10%, and the presence of metastasis were reported to correlate with patient survival in previous studies^[4-6]. The grading standard suggested by Fletcher who divided patients with GISTs into very low, low, intermediate, and high risk groups according to the tumor size and mitotic count. In the present study, patients were classified according to NIH categories: 18 were in very low risk; 75 were in low risk; 79 were in intermediate risk; and 155 were in high risk, and their corresponding 5-year survival rates were 100%, 73.8%, 65.8%, and 38.3%, respectively. The differences between the various groups were statistically significant (P < 0.001). This suggests that Fletcher’s criteria is effective in determining the malignant potential of GISTs and useful for guiding treatment.

Currently, most scholars consider the site of the primary tumor as a factor that influences the prognosis of GISTs^[7]. Tumors arising from the small intestine, colon, rectum or mesentery are generally associated with a less favorable outcome than those arising from the stomach, and the risk of GIST recurrence in the colon and rectum is similar to that in the small intestine. In this study, 5-year survival rates of patients with GISTs of the stomach were 67.4%, and 5-year survival rates of patients with GISTs of the small intestine were 44.1%. Univariate analysis showed that there were significant differences between the survival rates of GISTs occurring at different locations (P < 0.001), which equates with the results of other studies^[7,8].

Complete en-bloc resection of the tumor along with surrounding involved organs with negative tumor margins is the standard surgical treatment, and the resection rate of primary GISTs is approximately 85%^[8]. The complete resection rate was 91% in this study. A tumor-free surgical margin is an important factor influencing the prognosis of GISTs. DeMatteo et al.^[9] evaluated 200 GIST patients and found that the survival rate of patients who underwent total resection was 54% at 5 years with a median survival of 66 months; however, the postoperative median survival time was only 22 months for patients in whom resection was incomplete. Pierie et al.^[10] evaluated 69 cases of GISTs and found that the 5-year survival rate of patients who underwent en-bloc resection was 42%, while that of patients who underwent palliative resection was only 9%. In the current study, surgical margins were negative in 299 cases (91.4%), and positive in 28 cases (8.6%), and the median survival time was 68 months in the group with negative surgical margins and 25 months in the group with positive surgical margins. Their 5-year survival rates were 65.4% and 19.1%, respectively. It is commonly believed that routine lymph node resection is not necessary because of the low rate of lymph node metastasis in patients with GISTS^[11]. In this study, lymph node resection was performed in 16 cases, and among them, lymph node metastasis was found in 2 cases. There is increasing evidence that laparoscopic resection can be performed safely^[12]. The feasibility of laparoscopic resection is now being addressed in the patient presenting with GIST in our unit, and an analysis paper on this will be published in the future.

Most GISTs arise due to a mutation in the protooncogene C-kit^[13]. CD117 is the product of C-Kit, and CD34 is a hematopoietic precursor cell antigen, and both are widely expressed in GIST tissue. At present, most pathologists consider that identification of the presence of both CD117 and CD34 is an effective method for the diagnosis of GISTs. However, the expressions of CD117 and CD34 are not related with the risk of tumor malignancy. Ki-67 is a marker that reflects the activity of cell division and proliferation, and it is widely used for immunohistochemical examination for the diagnosis of GISTs. It has been shown that Ki-67 expression is an effective prognostic indicator for GIST patients^[14]. Our results showed that 5-year survival rate in the high Ki-67 group was 46.5% and in the low Ki-67 group was 71.4%. The differences were statistically significant, and the results of multivariate analysis demonstrated that Ki-67 was an independent prognostic factor of survival, which is in accordance with other reports^[15]. Thus, we find that it is reasonable to evaluate the prognosis of resectable GISTs and guide adjunctive therapy based on NIH categories and Ki-67 index.

Local recurrence, abdominal cavity implantation or liver metastasis occurs even after en-bloc resection in 40%-80% of GIST patients. The postoperative rate of recurrence and metastasis was 32.1% (105/327) shown in the current study. The therapeutic effect of conventional chemotherapy and cytotoxic drugs is poor and the response rate is estimated to be only 5% to 10%^[16]. Surgery also has limited effectiveness as a treatment for patients with metastatic GISTs. The molecular-targeting drug IM, a tyrosine kinase inhibitor blocking most mutated-activated KIT and PDGFRa proteins of GIST, controlled tumor growth in up to 85% of advanced GIST in the phase I, II and III trials reported to date^[17-20]. IM significantly improved the survival rate of patients with metastatic GISTs^[21]. In the present study, IM was used for 28 of 105 patients with postoperative recurrence or metastasis, and the results of univariate analysis showed that in the group with recurrence or metastasis, the median survival time of patients who did not receive IM was 30 months and that of patients who received IM was 59 months, and their 5-year survival rates were 16.4% and 39.4%, respectively. The difference was statistically significant (P = 0.017). These results suggest that IM can improve survival in GIST patients with recurrence or metastasis. IM was used as an adjuvant treatment for 4 patients in the group with high risk of malignancy. The therapeutic effect is, however, uncertain because of the small number of patients. The latest research results show that assignment to 1 year of adjuvant imatinib improved recurrence-free survival after the complete resection of primary gastrointestinal stromaltumors compared with taking a placebo. Additionally, adjuvant imatinib was safe and well tolerated^[22].

In conclusion, the limitations of this study included retrospective data analysis and the molecular classification of GISTs based on KIT and PDGFRa mutational status were not performed. However, it is reasonable to evaluate the prognosis of respectable GISTs and guide adjunctive therapy based on NIH categories and Ki-67 index. Complete resection and adjuvant imatinib use is clearly associated with improved survival in GISTs patients. Several other studies have also shown a correlation between exon 11 kit mutations and poor prognosis and suggested that exon 11 mutation may be one of the strongest prognostic factors. The molecular classification of GISTs based on KIT and PDGFRa mutational status may provide more exact prognostic estimates and with the possibility of an individually tailored imatinib treatment^[23-25].