ABSTRACT
Objective:To analyze the imaging features of hepatic metastases in gastrointestinal stromal tumors (GIST) on 18F-fluorodexyglucose (FDG) PET/CT in order to improve the accuracy of diagnosis. Methods:Clinical and imaging data of 33 patients (18 males, 15 females, age 34-70 years) with hepatic metastases in GIST who underwent PET/CT examination between May 2013 and July 2019 in Fujian Cancer Hospital were analyzed retrospectively. All patients underwent 18F-FDG PET/CT early imaging, and nine of them underwent delayed imaging. Visual analysis was performed on the PET/CT images by comparing FDG uptake of hepatic lesions and liver background, and all lesions were classified as significant hypermetabolism, slightly higher metabolism and equal or lower metabolism. Maximum standardized uptake value (SUV max) of primary GIST lesions and hepatic metastases were calculated and compared, and the relationship between them was analyzed. Wilcoxon rank sum test and Spearman rank correlation analysis were used to analyze the data. Results:Among 33 GIST patients, 9 patients had solitary hepatic metastasis, and 24 patients had multiple hepatic metastases (104 lesions). The diameter of metastases was 0.8-14.6(2.2(1.5, 3.9)) cm, and SUV max was 1.4-21.5(3.6(2.4, 5.7)). Of the 104 hepatic metastases, 94.2%(98/104) lesions had clear boundaries, 65.4%(68/104) lesions had uniform density (2 lesions with cystic density), 34.6%(36/104) lesions had uneven density in which hemorrhage, cystic change or necrosis could be found. On visual analysis of PET images, 38.5%(40/104) lesions were with significant hypermetabolism, 26.0%(27/104) were with slightly higher metabolism and 35.6%(37/104) were with equal or lower metabolism. In 24 patients with multiple hepatic metastases, 79.2%(19/24) showed different metabolic levels synchronously. Among 67 hypermetabolic metastases, 34.3%(23/67) were with homogeneous metabolism, of which 13 lesions with diameter<2.0 cm; 65.7%(44/67) were with heterogeneous metabolism, of which 36 lesions with diameter≥2.0 cm. There was a moderate correlation of SUV max between GIST primary tumors and hepatic metastases ( n=15; 9.2(6.8, 14.5) vs 3.8(2.1, 6.0), rs=0.556, P<0.01). The difference of SUV max between GIST primary tumors and hepatic metastases was statistically significant ( z=-5.098, P<0.01). In delayed imaging, 13/15 hepatic metastases with equal or lower metabolism changed to slightly higher metabolism. Conclusions:Hepatic metastases in GIST on 18F-FDG PET/CT imaging usually have clear boundary, and often associate with cystic degeneration, hemorrhage or necrosis. The metabolic patterns of hepatic metastases in GIST are varied. Delayed PET/CT imaging is helpful for the diagnosis of hypometabolic hepatic metastases in GIST.
ABSTRACT
Objective@#To evaluate the value of 18F-FDG PET-CT in predicting the malignant potential of Gastrointestinal Stromal Tumors (GIST).@*Methods@#The clinical and pathological features of 31 patients with GIST confirmed by surgery or biopsy were retrospectively analyzed. The malignant potential of GIST before treatment was assessed by 18F-FDG PET-CT. The GIST risk classification was graded according to the Standard revised by the National Institutes of Health (NIH) in 2008. The relationship between the maximal standard uptake value (SUVmax) and GIST risk classification, tumor diameter, Ki-67 index, and mitotic count were analyzed respectively. The cut-off level of SUVmax for the diagnosis of malignant GIST was calculated from the Receiver Operating Characteristic (ROC) curve.@*Results@#Among the 31 cases of GIST patients, 14 cases were gastric primary (stomach group) and 17 cases were nongastric primary (outside stomach group). The SUVmax, tumor diameter, Ki-67 index and mitotic count of the 31 patients were 8.21±4.68, (7.82±5.12)cm, (10.03±11.07)% and (12.29±10.55)/50 HPF, respectively. SUVmax was significantly correlated with GIST risk classification (r=0.727, P<0.01), but not with tumor diameter, Ki-67 index and mitotic count (r=0.348, r=0.284, r=0.290, P=0.055, P=0.121, P=0.114). The SUVmax, tumor diameter, Ki-67 index and mitotic count in the stomach group were 4.36±2.36, (6.08±4.31)cm, (3.43±3.03)% and (5.71±2.20)/50 HPF, respectively. SUVmax was significantly correlated with tumor diameter, GIST risk classification and Ki-67 index (r=0.682, r=0.868, r=0.732, P<0.01) but not with mitotic count (r=0.510, P=0.063). The SUVmax of the GIST in the gastric group and the outside gastric group were 4.36±2.36 and 10.68±5.50, respectively. The difference was statistically significant (P=0.001). The SUVmax in the malignant group of GIST (middle or high risk grade) was 8.90±4.89, which was significantly higher than 2.22±0.86 in the benign group (low or very low risk grade). The difference was statistically significant between the two group (P<0.01). ROC curve analysis showed that a SUVmax cut-off of 3.75 was the most sensitive for predicting malignant GIST. When the area under the curve of 0.969, the sensitivity was 84.6% and the specificity was 100%.@*Conclusions@#The SUVmax was strongly correlated with the GIST risk category and also with the tumor diameter and Ki-67 index in the gastric primary GIST, so it can be used as an effective indicator in predicting malignant potential of GIST before treatment.
ABSTRACT
Objective To compare the diagnostic values of 18F-FDG and 18F-FLT PET/CT in patients with suspicious recurrence of glioma after multimodal treatment.Methods A total of 20 patients (13 males,7 females;age range:12-73 years) with glioma who underwent 18F-FDG and 18F-FLT PET/CT due to abnormal enhancement on MRI from January 2012 to June 2015 were enrolled in this retrospective study.According to the pathological or follow-up results,patients were divided into therapy-related benign changes (TRBC) group and recurrent glioma group,the later was subdivided into initial low-grade glioma (LGG) group and initial high-grade glioma(HGG) group.T/NT ratios of 18F-FDG and 18F-FLT between HGG (LGG) group and TRBC group were compared using one-way analysis of variance and the least significant difference t test.ROC curve analysis was conducted to calculate the differential diagnostic efficiency of 18F-FDG and 18F-FLT between TRBC and recurrent glioma.Results A total of 14 patients were proved as recurrent glioma and 6 patients as TRBC.The mean 18F-FDG T/NT ratios of HGG group,LGG group and TRBC group were 2.31±0.86,1.32±0.86 and 1.32±0.64,respectively.The 18F-FDG T/NT ratio of the HGG group was significantly higher than that of the TRBC group(F=3.671,t=-2.471,P＜0.05).The mean 18F-FLT T/NT ratios of HGG group,LGG group and TRBC group were 8.94±3.14,7.18±3.29 and 1.92±1.20,respectively (F=13.301,t values:-5.150 and-2.360,both P＜0.05).The optimal T/NT cutoff values for 18 F-FDG and 18F-FLT PET/CT were 1.62 and 4.58,respectively.The sensitivity,specificity and accuracy of detecting recurrent glioma with optimal T/NT cutoff value were 11/14,5/6 and 16/20 for 18F-FDG PET/CT,and those for 18F-FLT PET/CT were 13/14,6/6 and 19/20,respectively.No significant difference was observed between the diagnostic efficiencies of the two imaging modalities (x2 values:1.167,1.091 and 2.057;all P＞0.05).Conclusion There were no statistical significances between 18F-FDG and 18F-FLT PET/CT on the differential diagnosis of glioma recurrence.