Faculty of Nuclear Sciences and Physical Engineering, Czech Technical University in Prague, Prague, Czechia.
Department of Radiation Oncology, H. Lee Moffitt Cancer Center and Research Institute, Tampa, FL, USA.
Radiat Oncol. 2024 Sep 27;19(1):127. doi: 10.1186/s13014-024-02519-1.
Recent papers suggested a correlation between the risk of distant metastasis (DM) and dose outside the PTV, though conclusions in different publications conflicted. This study resolves these conflicts and provides a compelling explanation of prognostic factors.
A dataset of 478 NSCLC patients treated with SBRT (IMRT or VMAT) was analyzed. We developed a deep learning model for DM prediction and explainable AI was used to identify the most significant prognostic factors. Subsequently, the prognostic power of the extracted features and clinical details were analyzed using conventional statistical methods.
Treatment technique, tumor features, and dosiomic features in a 3 cm wide ring around the PTV (PTV) were identified as the strongest predictors of DM. The Hazard Ratio (HR) for D was significantly above 1 (p < 0.001). There was no significance of the PTV dose after treatment technique stratification. However, the dose in PTV was found to be a highly significant DM predictor (HR > 1, p = 0.004) when analyzing only VMAT patients with small and spherical tumors (i.e., sphericity > 0.5).
The main reason for conflicting conclusions in previous papers was inconsistent datasets and insufficient consideration of confounding variables. No causal correlation between the risk of DM and dose outside the PTV was found. However, the mean dose to PTV can be a significant predictor of DM in small spherical targets treated with VMAT, which might clinically imply considering larger PTV margins for smaller, more spherical tumors (e.g., if IGTV > 2 cm, then margin ≤ 7 mm, else margin > 7 mm).
最近的一些论文表明，肿瘤靶区（PTV）外剂量与远处转移（DM）风险之间存在相关性，但不同文献的结论存在冲突。本研究旨在解决这些冲突，并对预后因素提供有说服力的解释。
对 478 例接受 SBRT（调强放疗或容积旋转调强放疗）治疗的非小细胞肺癌患者的数据集进行了分析。我们开发了一种用于 DM 预测的深度学习模型，并使用可解释人工智能来确定最重要的预后因素。随后，使用常规统计方法分析提取特征和临床细节的预后能力。
治疗技术、肿瘤特征和 PTV 周围 3cm 宽环内的剂量特征被确定为 DM 的最强预测因素。DM 的危险比（HR）显著大于 1（p<0.001）。在分层治疗技术后，PTV 剂量无显著意义。然而，当仅分析球形度大于 0.5 的小且球形肿瘤的 VMAT 患者时，PTV 中的剂量被发现是 DM 的高度显著预测因素（HR>1，p=0.004）。
以往文献中结论不一致的主要原因是数据集不一致，以及对混杂变量考虑不足。未发现 PTV 外剂量与 DM 风险之间存在因果关系。然而，在使用 VMAT 治疗小而球形靶区时，PTV 的平均剂量可能是 DM 的重要预测因素，这可能意味着对于较小、更球形的肿瘤，临床可能需要考虑更大的 PTV 边缘（例如，如果 IGTV>2cm，则边缘≤7mm，否则边缘>7mm）。
