多参数影像标志鉴别头颈肿瘤复发与治疗后改变的诊断效能
本研究回顾性纳入73例头颈癌患者共75个局灶强化病灶,比较DCE-MRI肿瘤血容量、ADC及PET/CT SUVmax的鉴别诊断价值。结果显示ADCmean和SUVmax诊断效能优秀,两者联合AUROC最高。
要点
- ADCmean是区分复发与治疗后改变最稳健的预测指标
- ADC与PET/CT联合可进一步提高诊断准确性
- ADC图像受磁敏感伪影影响时,肿瘤血容量图可提供补充信息
核心结论
- ADCmean和SUVmax在区分头颈癌复发与治疗后改变中表现优异,联合应用AUROC达0.99
研究设计与样本
- 回顾性研究,纳入73例患者共75个MRI局灶强化病灶
- 复发病灶38个,治疗后改变37个
- 采用高时间分辨率DCE-MRI计算标准化肿瘤血容量(nTBVmean),以脊髓为参照
- DWI获取ADCmean,18F-FDG-PET/CT获取SUVmax
- 采用多变量logistic回归分析和AUROC比较评估诊断效能
关键影像表现/征象
- ADCmean在复发组与治疗后改变组间差异有统计学意义(p < 0.001)
- SUVmax在两组间差异有统计学意义(p < 0.001)
- nTBVmean在复发组高于治疗后改变组,但差异无统计学意义(p = 0.11)
- ADCmean、SUVmax和nTBVmean的AUROC分别为0.94、0.88和0.60
诊断标准与关键数值
- ADCmean与SUVmax联合的AUROC为0.99
- ADCmean是鉴别复发与治疗后改变最稳健的预测指标
临床应用与阅片要点
- 当ADC图因磁敏感伪影质量下降时,肿瘤血容量图可提供补充诊断信息
局限性
- 回顾性研究设计
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
PURPOSE: To assess the diagnostic performance of tumor blood volume (TBV) derived from T1 dynamic contrast-enhanced (DCE) MRI and compare it with ADC and SUVmax in distinguishing recurrence from post-treatment changes.
METHODS: This retrospective study included 73 patients with 75 focal contrast-enhancing lesions identified on surveillance MRI. Normalized TBV (nTBVmean) was calculated from high temporal resolution DCE-MRI using the spinal cord as a reference. The 3D ROI analysis of ADCmean and SUVmax values were obtained from DWI and 18F-FDG-PET/CT, respectively. Diagnostic performances were evaluated using multivariate logistic regression analyses, with area under the receiver operating characteristic curve (AUROC) comparisons.
RESULTS: Among the 75 lesions, 38 were recurrence and 37 were post-treatment changes. ADCmean and SUVmax were significantly different between groups (p < 0.001), while nTBVmean was higher in recurrence without statistical significance (p = 0.11). The AUROC of ADCmean, SUVmax, and nTBVmean were 0.94, 0.88, and 0.60 respectively. The combination of ADCmean and SUVmax yielded the highest AUROC (0.99), with ADCmean remaining the most robust predictor. TBV offered complementary value when ADC maps were degraded by susceptibility artifacts.
CONCLUSIONS: ADCmean and SUVmax showed excellent performance in differentiating recurrence from post-treatment changes. TBV maps may offer complementary diagnostic value in cases with compromised ADC quality due to artifacts.
引用
- Chang, Lee, Ahn, et al. Diagnostic performance of multiparametric imaging markers in differentiating local recurrence from post-treatment change in head and neck cancer surveillance. Neuroradiology. 2026. PMID: 41546724. doi:10.1007/s00234-025-03882-w
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。