Ionization Detail - Biologically based treatment planning for particle therapy beyond LET-RBE

项目摘要 当前的质子和离子治疗计划程序使用物理量线性 能量转移(LET)作为生物有效性的替代品或利用相关的生物效应 有效性(RBE)模型，将吸收剂量转换为生物加权剂量，假设为等效剂量。 对光子有效。对于计划带电粒子的治疗，LET在临床上确实很重要，但 有一些已知的问题。具有相同LET的离子束可以具有不同的RBE，具体取决于粒子类型 和能量。因此，LET本身并不是用于放射治疗计划(RTP)的理想参数。 针对碳疗的临床应用，RBE模型应运而生。然而，与 用于碳治疗的不同RBE模型表明，剂量处方与 欧洲局部效应模型或日本国立放射科学研究所混合束模型 差异最高可达15%。我们使用术语电离细节(ID)来表示 在纳米尺度上沿着粒子轨迹的电离事件。我们的主要假设是 有强有力的先验证据支持，ID可以预测，比LET和现有的RBE更好 模型，与高LET辐射相关的生物效应。我们之前已经展示了如何 ID可以与这些模型一起使用来提高它们的性能，为集成提供了一条途径 基于ID的RTP进入临床实践。我们的方法可能导致在质子和离子治疗RTP方面达成共识。 有了四个具体的目标，我们选择了一种转换和逐步的方法来构建基于ID的 预测模型。我们将在不同的终端和模型系统中测试此模型，范围从体外 细胞和分子数据，通过照射烧瓶和解剖模型中的人类癌细胞而获得，以 活体小鼠/人类肿瘤数据。我们将开发先进的算法和计算GPU- 基于方法，并使用它们进行有效的反治疗计划与主动扫描的质子和离子 波束。这项技术将被应用于演示我们的实用和评估我们的临床疗效 前列腺癌和脊索瘤的治疗方法，首先是在人体大小的骨盆和头部幻影中， 最后，回顾治疗过这些疾病的患者。我们已经组建了一支强大的团队， 这一项目需要补充专业知识。我们团队的成员都成功地 通力合作。完成后，我们将提供经过严格测试和验证的方法 基于ID的粒子RTP将可用于与现有临床数据的相互关联，并用于 在预期的粒子治疗临床试验中进行仔细的测试。

Project Summary Current proton and ion therapy treatment planning procedures utilize either the physical quantity linear energy transfer (LET) as a surrogate for biological effectiveness or make use of relative biological effectiveness (RBE) models that convert absorbed dose to biologically weighted dose, assumed to be iso- effective to photons. LET is indeed important clinically for planning treatments with charged particles, but there are known problems. Ion beams with the same LET can have different RBE, depending on particle type and energy. Therefore, LET by itself is not an ideal parameter to use in radiation treatment planning (RTP). For clinical application of carbon therapy, RBE-models have been developed. However, comparisons of different RBE models used for carbon therapy have shown that dose prescriptions implemented with the European local effect model or the Japanese National Institute of Radiological Sciences mixed beam model can be up to 15% different. We use the term ionization detail (ID) to mean the detailed distribution of ionizing events along a particle track on the nanometer scale. Our chief hypothesis, which is supported by strong prior evidence, is that ID can predict, better than LET and existing RBE models, the biological effects associated with high-LET radiation. We have previously shown how ID can be used together with these models to improve their performance, providing a path for integrating ID-based RTP into clinical practice. Our approach could lead to a consensus in proton and ion therapy RTP. With four Specific Aims, we have chosen a translational and stepwise approach to build an ID-based prediction model. We will test this model for different endpoints and model systems ranging from in vitro cell and molecular data, obtained by irradiating human cancer cells in flasks and anatomical phantoms, to in vivo mice/human tumor data. We will develop advanced algorithms and computational GPU- based methods and use them for effective inverse treatment planning with actively scanned proton and ion beams. This technology will be applied to demonstrate the practicality and evaluate the clinical efficacy of our approach in prostate and chordoma treatments, first prospectively in human-size pelvis and head phantoms, and finally, retrospectively in patients treated for these diseases. We have assembled a strong team with the complementary expertise needed for this project. Members of our team have all successfully collaborated together. Upon completion, we will provide a rigorously tested and validated approach to ID-based particle RTP that will be available for cross-correlation with existing clinical data and for careful testing in prospective clinical particle therapy trials.