BACKGROUND AND PURPOSE In radiotherapy, treatment response is generally evaluated many weeks after end of the treatment course. If the treatment outcome could be predicted during radiotherapy better tumour control could be achieved through timely adaptation of the treatment strategy. In this study intra-treatment change based on the diffusion MRI outlined viable tumour volume (VTV) was assessed and compared to the standard GTV to study their outcome prediction capacity. MATERIALS AND METHODS Thirty-eight brain metastases from twenty-one cancer patients were analysed in this prospective trial. Diffusion and structural MRI was acquired on a 1 T machine before, during, and at follow-up 2-3 months after radiotherapy. The VTV was defined as a region with high cellularity using high b-value diffusion MRI scans. Further, the diffusivity of the VTV was derived as the apparent diffusion coefficient (ADC). Treatment outcome was determined using RECIST defined bounds in the T1W MRI follow-up scan. Longitudinal statistical analysis was performed using a linear mixed effect model. RESULTS The GTV declined in both responding and non-responding (significantly) tumours with inseparable rates during radiotherapy. The VTV volume fraction reduced significantly in the responding tumours only. The ADC of the VTV increased significantly in responding metastases whereas it decreased in non-responding metastases. Furthermore, no association between baseline tumour size or primary disease and outcome was observed. CONCLUSION GTV size change during radiotherapy is not a reliable predictor of outcome in brain metastases. On the other hand, change in the volume fraction of VTV and diffusivity of VTV shows ability to stratify treatment outcome.

中文翻译：

---

扩散 MRI 概述了可行的肿瘤体积在治疗中的结果分层中优于 GTV

背景和目的 在放射治疗中，治疗反应通常在疗程结束后数周进行评估。如果可以在放疗期间预测治疗结果，则可以通过及时调整治疗策略来更好地控制肿瘤。在这项研究中，基于扩散 MRI 概述的可行肿瘤体积 (VTV) 的治疗中变化进行了评估，并与标准 GTV 进行了比较，以研究其结果预测能力。材料与方法 本前瞻性试验分析了 21 名癌症患者的 38 处脑转移瘤。在放疗前、放疗中和放疗后 2-3 个月的随访中，在 1 T 机器上采集了扩散和结构 MRI。VTV 被定义为使用高 b 值扩散 MRI 扫描具有高细胞数量的区域。更多，VTV 的扩散系数被推导出为表观扩散系数 (ADC)。在 T1W MRI 随访扫描中使用 RECIST 定义的界限确定治疗结果。使用线性混合效应模型进行纵向统计分析。结果 放疗期间，有反应和无反应（显着）肿瘤的 GTV 均下降，且比率不可分割。VTV 体积分数仅在响应肿瘤中显着降低。VTV 的 ADC 在有反应的转移中显着增加，而在无反应的转移中则下降。此外，未观察到基线肿瘤大小或原发疾病与结果之间的关联。结论 放疗期间 GTV 大小的变化不是脑转移瘤预后的可靠预测指标。另一方面，