【摘要】 目的：探讨¹⁸F-成纤维活化蛋白抑制剂（fibroblastactivationproteininhibitor,FAPI）-42正电子发射体层成像（positron emission tomography,PET）/计算机体层成像（computed tomography,CT）在脑胶质瘤分级预测中的应用价值。方法：回顾并分析2020年6月—2021年5月行¹⁸F-FAPI-42PET/CT检查的初诊怀疑脑胶质瘤或怀疑胶质瘤复发的患者25例（年龄17～67岁），所有患者均经立体定向活检或手术后病理学检查明确各病灶病理学类型。在PET/CT图像上勾画胶质瘤病灶感兴趣区（region of interest,ROI），获取病灶最大标准摄取值(maximum standardized uptake value,SUV_max)及靶本比值（target to background ratio,TBR）。采用Mann-Whitney U检验及独立样本t检验比较不同病理学类型脑胶质瘤PET/CT指标差异，并用受试者工作特征（receiver operating characteristic,ROC）曲线预测脑胶质瘤分级。结果：25例患者经病理学检查证实共26个脑胶质瘤病灶，其中高级别（WHOⅢ-Ⅳ级）脑胶质瘤病灶19个，包括异柠檬酸脱氢酶（isocitrate dehydrogenase,IDH）野生型10个，IDH突变型9个；低级别（WHOⅡ级）脑胶质瘤病灶7个，包括IDH野生型2个，IDH突变型5个。高级别胶质瘤病灶SUV_max[1.75（1.25,3.19）]及TBR[（26.24±4.83）]均明显高于低级别脑胶质瘤病灶[0.32（0.31,1.14）及9.47±2.81]，两者差异有统计学意义（U=25.00,P=0.018;t=2.280,P=0.032）。SUV_max预测高级别胶质瘤最佳阈值为1.20，相应ROC曲线的曲线下面积（area under curve,AUC）为0.812（95%CI0.609～1.000），灵敏度为0.789，特异度为0.857。TBR预测高级别胶质瘤最佳阈值为9.09，相应AUC为0.850（95%CI0.686～1.000），灵敏度为0.842，特异度为0.857。对于所有胶质瘤病灶，IDH野生型病灶与IDH突变型病灶之间SUV_max及TBR差异均无统计学意义（U=71.00,P=0.520;U=80.00,P=0.860）；此外，对于高级别胶质瘤病灶，IDH野生型病灶与IDH突变型病灶之间SUV_max及TBR差异亦均无统计学意义（U=36.00,P=0.462;U=30.00,P=0.221）。结论：¹⁸F-FAPI-42PET/CT在非侵袭性预测脑胶质瘤分级中具有较好的临床应用价值。更多还原

【Abstract】 Objective: To investigate the value of ¹⁸F-fibroblast activation protein inhibitor（FAPI）-42 positron emission tomography（PET）/computed tomography（CT） imaging for clinical application in the prediction of glioma grading. Methods: From June 2020 to May 2021, 25 patients（aged 17-67 years） who underwent ¹⁸F-FAPI-42 PET/CT with a primary diagnosis of suspected glioma or suspected glioma recurrence were retrospectively analyzed. All of whom had stereotactic biopsy or surgical resection to obtain the pathological type of each lesion. The region of interest（ROI） of glioma lesions was outlined on PET/CT images, and the maximum standardized uptake value(SUV_max) and target to background ratio（TBR） of the lesions were obtained. The indices of different pathological types of gliomas were compared by the Mann-Whitney U test and independent sample t test, and the optimal cut-off values for differentiating high-grade gliomas and low-grade gliomas were obtained by receiver operating characteristic（ROC） curves analysis. Results: A total of 26 glioma lesions were pathologically confirmed in 25 patients, 19 of them are high-grade（WHO grade Ⅲ-Ⅳ） gliomas, including 10 isocitrate dehydrogenase（IDH）wild-type and 9 IDH-mutant; and 7 of them are low-grade（WHO grade Ⅱ） gliomas, including 2 IDH wild-type and 5 IDH-mutant. Both SUV_max [1.75（1.25, 3.19）] and TBR[（26.24±4.83）] were significantly higher in high-grade gliomas than in low-grade gliomas [0.32（0.31, 1.14） and 9.47±2.81], both of which were statistically different（U=25.00, P=0.018; t=2.280, P= 0.032）.The optimal cut-off values of SUV_max to predict high-grade gliomas was 1.20, with corresponding area under curve（AUC） of 0.812（95% CI 0.609-1.000）, sensitivity of 0.789, and specificity of 0.857. The optimal cut-off values of TBR to predict high-grade gliomas was 9.09, with corresponding AUC of 0.850（95% CI 0.686-1.000）, sensitivity of 0.842 and specificity of 0.857. For all gliomas, the differences of SUV_max and TBR between IDH wildtype lesions and IDH-mutant lesions were not statistically significant（U=71.00, P=0.520; U=80.00, P=0.860）, furthermore, for high-grade gliomas, the differences of SUV_max and TBR between IDH wild-type and IDH-mutant lesions were also not statistically significant（U=36.00,P=0.462; U=30.00, P=0.221）. Conclusion: The ¹⁸F-FAPI-42 PET/CT is clinically useful in non-invasive grade prediction for glioma patients.更多还原