Comprehensive Study Of Minimal Invasive Surgical Approach And Non-invasive, Individualized Diagnosis In Gastrointestinal Stromal Tumor

Part Ⅰ Prognosis and Long-term Outcome of Surgical Resected Gastric Gastrointestinal Stromal TumorObjective To investigate the prognostic factors and long-term outcome of surgical treatment for gastric gastrointestinal stromal tumors.Methods 287 patients enrolled in this study, clinicopathological data was retrospectively analyzed, over-all survival and recurrence-free survival was applied for assessing the prognosis.Results 76.3% lesion located in gastric body. All patients underwent radical surgery and no tumor rupture occurred during operation. The mean diameter of tumor was 6.68cm while mitotic rate was 5.31 ± 6.15/50HPF. Based on the follow-up of 58.9 months in average, we found the 5-year and 10-year overall and recurrence-free survival rate was 97.7%,91.9%,91.9% and 88.4% respectively. Necrosis and mitotic rate> 10/50 HPF were independent prognostic factors associated with overall survival while tumor diameter>10cm and mitotic rate>10/50HPF were independent prognostic factors associated with recurrence-free survival.Conclusion Gastric Gastrointestinal stromal tumors reveal relatively better prognosis. Surgery remains the first-line treatment for localized lesion. Intensive follow-up should be applied to patients with high recurrence risk.Part Ⅱ Laparoscopic Wedge Resection Versus Open Wedge Resection for Gastric Gastrointestinal Stromal Tumor:A Non-Randomized Case-matched StudyObjective To investigate the safety and feasibility of Laparoscopic wedge resection for gastric gastrointestinal stromal tumors.Method One hundred GIST patients between 2010 and 2014 were retrospectively enrolled. Fifty patients underwent laparoscopic wedge resection and the other fifty was conducted by open surgery. A comparison of clinicopathological character, perioperative associated factors, postoperative nutritious status and short-term outcome was conducted between two groups.Results Statistical significance appeared only in age (61.58 ± 10.43 vs.56.40 ± 9.82, P= 0.012) of clinicopathological information. Over fifty percent of tumors occurred at gastric body. Although operation time did not reveal statistically significant, laparoscopic group showed predominant advantage in estimated blood loss (48.60 ± 48.89ml vs.137.60 ± 140.69ml, P< 0.05), average hospital stay (12.14 ± 4.32d vs.l 7.22±7.11d, P< 0.05), oral intake time (3.76 ± 1.73d vs.6.28 ± 3.73d, P< 0.05), decline of prealbumin (0.07 ± 0.04) g/L vs. (0.11 ± 0.05) g/L, P< 0.05, decline of transferrin (0.51 ± 0.29) g/L vs. (0.64±0.30) g/L, P= 0.034 and complication (2.0% vs.14.0%, P= 0.027).Conclusion Laparoscopic wedge resection is feasible and safe for GIST surgery, while conducting a more minimally invasive approach for patients.Part Ⅲ Validation of Various Nomograms and Staging Systems of Gastrointestinal Stromal Tumor and Establishing a New Nomogram for Assessing Postoperative Recurrence Risk in Localized GISTObjective This study sought to develop a new nomogram to evaluate the recurrence free survival after surgery, and validate different nomograms and compare the diagnostic accuracy against three established staging systems.Method 254 patients underwent radical surgery without adjuvant imatinib therapy were retrospectively enrolled. Recurrence free survival rate and associated factors was analyzed by Log-rank test. A nomogram to evaluate RFS based on tumor size, mitotic rate and primary site was established. All nomograms were validated by calibration curve and concordance index. Diagnostic accuracy between different staging system and nomograms was compared by ROC test.Results The mean age was 57.31 ± 11.51 years, gender ratio was 1:1.21 (male: female), average tumor size and mitotic rate was 6.16±4.33 cm and 4.72 ± 5.38/50HPF respectively. Tumor size, mitotic rate, primary site was independent risk factor associated with recurrence free survival. The 2-year,5-year RFS rate was 91.8% and 86.5% respectively.The C-index of new established nomogram of 2-year RFS was 0.908, and 5-year RFS was 0.903, which was slightly higher than MSKCC and multi-institution nomogram. The other two nomograms also had excellent diagnositic efficiency (All c-index> 0.85). In subgroup analysis of all staging systems, the intermediate risk group in AFIP and AJCC system had significant recurrence risk against low risk (P< 0.001), which had a better discrimination than modified NIH system. For the comparison of AUC in different staging systems, the new established nomogram and modified NIH system has the best accuracy (AUC= 0.935, AUC= 0.901). Significant difference was only discovered between MSKCC nomogram and ours (P= 0.042). A statistic tendency was shown between AFIP staging system and the new established nomogram (P= 0.059).Conclusion Currently, all staging systems and established nomograms have accurate diagnositic effiency. Modified NIH system is recommended as first-line diagnostic system, AJCC and AFIP could be supplymentary for assessing risk of intermediate group. For the two published momograms, diagnostic performance is not superior to the traditional staging system, and the new established nomogram still need external validation.Part Ⅳ Non-invasive Diagnosis by High-throughout Sequencing of Circulating Tumor DNA in Localized Gastic Gastrointestinal Stromal TumorObjective Identifying genotype of gastrointestinal stromal tumors is crucial for adjuvant targeted therapy. Biopsy is an invasive approach and could not be applied repeatedly. The aim of this study is to evaluate the feasibility of identifying mutation in ctDNA via next-generation sequencing for localized GISTs, and compare the consistency of sequencing results in tissue and plasma.Method We collected plasma and tumor tissue from two patients. After ctDNA and tumor genome DNA was purified, we compared the mutation in KIT exon 9,11,13, 17 and PDGFRAexon 12,18.Results The mutation analysis revealed a S501delins SAY mutation in exon 9 and a substitution of V560D in exon 11, respectively. Correspondingly, the density of circulating free DNA was 0.45ng/uL and 0.303ng/uL. Mutation sites in ctDNA were not consistent with tumor tissue. After a deep sequencing of 20,000X depth, we found F496L in exon 9 and W557R in exon 11, with a mutation rate of 0.0206% and 0.0232%, respectively.Conclusion Mutation analysis of ctDNA via high throughout sequencing is not appropriate for localized GISTs for non-invasive diagnosis.