膀胱癌及膀胱外脂肪的定量T2加权与表观扩散系数纹理特征用于经尿道电切术后局部肿瘤分期
回顾性纳入36例经尿道膀胱肿瘤电切术后行MRI并随后膀胱切除术的患者,提取膀胱癌及膀胱外脂肪的T2加权和ADC纹理特征。多因素分析显示熵值在≥T3较≤T2、≥T2较T1肿瘤中更高,MRI纹理分析可能有助于TURBT术后局部膀胱癌分期。
要点
- 膀胱癌及膀胱外脂肪的熵值在≥T3肿瘤中高于≤T2肿瘤,在≥T2肿瘤中高于T1肿瘤。
- T2加权图像上膀胱癌熵值鉴别≥T3与≤T2的AUC为0.85,膀胱外脂肪熵值AUC为0.84。
- ADC图上膀胱癌熵值鉴别≥T2与T1的AUC为0.76,膀胱外脂肪熵值AUC为0.74。
核心结论
- 膀胱癌及膀胱外脂肪的熵值在≥T3肿瘤中高于≤T2肿瘤、在≥T2肿瘤中高于T1肿瘤,MRI纹理分析可能作为更高局部分期膀胱癌的生物标志物。
研究设计与样本
- 回顾性研究纳入36例(27男、9女,平均年龄71岁)2011年8月至2016年10月间TURBT后行MRI并随后无干预行膀胱切除术的患者,提取膀胱癌及邻近膀胱外脂肪的T2加权和ADC纹理特征,采用Kruskal-Wallis或Mann-Whitney U检验比较,并行多因素logistic回归和ROC曲线分析。
诊断标准与关键数值
- 多因素分析中,鉴别≥T3与≤T2肿瘤:T2加权膀胱癌熵值p=0.006、OR=4.56(95%CI 1.49-20.41)、AUC=0.85;ADC图膀胱癌熵值p=0.019、OR=2.24(95%CI 1.13-5.31)、AUC=0.80;T2加权膀胱外脂肪熵值p=0.005、OR=17.50(95%CI 3.01-200.80)、AUC=0.84;ADC图膀胱外脂肪熵值p=0.002、OR=6.54(95%CI 1.90-32.40)、AUC=0.82。鉴别≥T2与T1肿瘤:ADC图膀胱癌熵值p=0.027、OR=2.11(95%CI 1.08-5.03)、AUC=0.76;T2加权膀胱外脂肪熵值p=0.010、OR=5.33(95%CI 1.25-3.79)、AUC=0.78;ADC图膀胱外脂肪熵值p=0.029、OR=3.80(95%CI 1.25-16.97)、AUC=0.74。
临床应用与阅片要点
- 作者认为MRI纹理分析可能有助于TURBT术后患者的膀胱癌局部分期,并可能作为更高局部分期膀胱癌的生物标志物。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
OBJECTIVE. The purpose of this study was to determine whether quantitative T2-weighted imaging and apparent diffusion coefficient (ADC) texture features of bladder cancer and extravesical fat are predictive of muscle invasive bladder cancer (category ≥ T2) and extravesical (category ≥ T3) disease after transurethral resection of a bladder tumor (TURBT). MATERIALS AND METHODS. In this retrospective study, 36 patients (27 men, nine women; mean age, 71 years) were identified who underwent post-TURBT MRI followed by cystectomy without intervening treatment from August 2011 through October 2016. Texture features of bladder cancer and extravesical fat adjacent to the tumor on T2-weighted and ADC images were extracted and compared between category ≤ T2 versus ≥ T3 and category T1 versus ≥ T2 tumors by means of Kruskal-Wallis or Mann-Whitney U test. Multivariate logistic regression analysis was performed, and ROC curves were calculated. RESULTS. Twenty-six of the 36 (72%) tumors were ≥ T2, and 53% (19/36) were ≥ T3. In multivariate analysis, bladder cancer entropy on T2-weighted images (p = 0.006; odds ratio [OR], 4.56; 95% CI, 1.49-20.41; AUC, 0.85) and ADC maps (p = 0.019; OR, 2.24; 95% CI, 1.13-5.31; AUC, 0.80) and extravesical fat entropy on T2-weighted images (p = 0.005; OR, 17.50; 95% CI, 3.01-200.80; AUC, 0.84) and ADC maps (p = 0.002; OR, 6.54; 95% CI, 1.90-32.40; AUC, 0.82) remained greater for ≥ T3 than for ≤ T2 tumors. In multivariate analysis, bladder cancer entropy on ADC maps (p = 0.027; OR, 2.11; 95% CI, 1.08-5.03; AUC, 0.76) and extravesical fat entropy on T2-weighted images (p = 0.010; OR, 5.33; 95% CI, 1.25-3.79; AUC, 0.78) and ADC maps (p = 0.029; OR, 3.80; 95% CI, 1.25-16.97; AUC, 0.74) remained greater for category ≥ T2 compared with category T1 tumors. CONCLUSION. Greater entropy of primary bladder cancers and extravesicular fat was observed in category ≥ T3 than in category ≤ T2 and in category ≥ T2 than in category T1 tumors. MRI texture analysis can help with local bladder cancer staging in patients who have undergone TURBT and may serve as a biomarker for higher local category bladder cancers.
引用
- Lim CS, Tirumani S, van der Pol CB, et al. Use of Quantitative T2-Weighted and Apparent Diffusion Coefficient Texture Features of Bladder Cancer and Extravesical Fat for Local Tumor Staging After Transurethral Resection. AJR Am J Roentgenol. 2019. PMID: 30860885. doi:10.2214/AJR.18.20718
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。