Diagnostic performance of 68Ga-PSMA PET/CT for identification of aggressive cribriform morphology in prostate cancer with whole-mount sections
Diagnostic performance of 68Ga-PSMA PET/CT for identification of aggressive cribriform morphology in prostate cancer with whole-mount sections
复制标题
Ga-68-PSMA PET/CT 对全封片切片前列腺癌侵袭性筛状形态识别的诊断性能
DOI：
10.1007/s00259-019-04320-9
复制
发表时间：
2019-07-01
影响因子：
9.1
通讯作者：
Guo, Hongqian
中科院分区：
文献类型：
--
作者：
Gao, Jie;Zhang, Chengwei;Guo, Hongqian
目的：探讨68ga - psma PET/CT对前列腺癌（PCa）病理筛状形态的诊断价值。方法：回顾性研究纳入49例术前行多参数MRI （mpMRI）和68ga - psma PET/CT检查的前列腺癌患者，根治性前列腺切除术标本具有Gleason模式（GP） 4和GP 5缺失。gp4的病变根据其筛状状态进行轮廓和分层。在匹配的mpMRI和PET图像上绘制感兴趣的体积，并推导出包括平均表观扩散系数（ADCmean）、十分位数ADC （ADC10%）和最大标准化摄取值（SUVmax）在内的参数。对连续变量采用Mann-Whitney U检验，对分类变量采用卡方检验。使用接收器工作特征分析来比较识别筛状形态的成像参数。使用逻辑回归模型进行单变量分析，评估筛孔膜阳性PCa与影像学变量之间的关系。结果：49例GP 4患者共发现62个病变。34例（69.4%）患者中37例（59.7%）病变表现为筛状形态。筛状膜阳性组和非筛状膜组的adcmean和adc10%相似（P < 0.05），而suvmax差异显著（中位SUVmax18.3 vs. 9.4 /例，P = 0.003, 18.2 vs. 7.2 /例，P < 0.001），每例患者分析的敏感性和特异性分别为76%和86%，每例病变分析的敏感性和特异性分别为77%和88%。此外，PSMA在cribriform阳性PCa中显著过表达（P = 0.003）。suvmax是PCa筛状形态的显著预测因子（优势比为8.61,95%可信区间为4.96-25.27，每例患者；优势比为11.93,95%可信区间为6.49-33.74，每个病变；均P < 0.001）。结论：68Ga-PSMA PET/CT能有效识别前列腺癌侵袭性筛网形态。
Purpose:To explore the diagnostic performance of68Ga-PSMA PET/CT for identification of pathological cribriform morphology in prostate cancer (PCa).Methods:The study retrospectively enrolled 49 PCa patients who had undergone preoperative multiparametric MRI (mpMRI) and68Ga-PSMA PET/CT, and who had Gleason pattern (GP) 4 and absence of GP 5 on radical prostatectomy specimens. Lesions with GP 4 were outlined and stratified according to their cribriform status. Volumes of interest were drawn on matched mpMRI and PET images, and parameters including average apparent diffusion coefficient (ADCmean), tenth percentile ADC (ADC10%) and maximum standardized uptake value (SUVmax) were derived. The Mann-Whitney U test was used for continuous variables and the chi-squared test for categorical variables. Receiver operating characteristic analysis was used to compare imaging parameters in identifying cribriform morphology. The associations between cribriform-positive PCa and imaging variables were evaluated in a univariate analysis using a logistic regression model.Results:A total of 62 lesions were identified in 49 patients with GP 4. Of these lesions, 37 (59.7%) in 34 patients (69.4%) showed cribriform morphology. ADCmeanand ADC10%were similar between cribriform-positive and non-cribriform groups (P > 0.05), while SUVmaxwas significantly different (median SUVmax18.3 vs. 9.4 per patient, P = 0.003, 18.2 vs. 7.2 per lesion, P < 0.001), yielding sensitivities and specificities of 76% and 86% in a per-patient analysis, and 77% and 88% in a per-lesion analysis, respectively. Further, PSMA was significantly overexpressed in cribriform-positive PCa (P = 0.003). SUVmaxwas a significant predictor of cribriform morphology in PCa (odds ratio 8.61, 95% confidence interval 4.96-25.27, per patient; odds ratio 11.93, 95% confidence interval 6.49-33.74, per lesion; both P < 0.001).Conclusion:68Ga-PSMA PET/CT effectively identifies the aggressive cribriform morphology in PCa.