Li Xiadong, Wang Hui, Xu Lixia, Kuang Yu
Medical Imaging and Translational Medicine laboratory, Department of Radiotherapy, Affiliated Hangzhou Cancer Hospital, Westlake University School of Medicine, Hangzhou, Zhejiang, China.
Medical Physics Program, University of Nevada, Las Vegas, Nevada, USA.
Med Phys. 2024 May;51(5):3619-3634. doi: 10.1002/mp.17036. Epub 2024 Mar 22.
This study addresses the technical gap between clinical radiation therapy (RT) and preclinical small-animal RT, hindering the comprehensive validation of innovative clinical RT approaches in small-animal models of cancer and the translation of preclinical RT studies into clinical practices.
The main aim was to explore the feasibility of biologically guided RT implemented within a small-animal radiation therapy (SART) platform, with integrated quad-modal on-board positron emission tomography (PET), single-photon emission computed tomography, photon-counting spectral CT, and cone-beam CT (CBCT) imaging, in a Monte Carlo model as a proof-of-concept.
We developed a SART workflow employing quad-modal imaging guidance, integrating multimodal image-guided RT and emission-guided RT (EGRT). The EGRT algorithm was outlined using positron signals from a PET radiotracer, enabling near real-time adjustments to radiation treatment beams for precise targeting in the presence of a 2-mm setup error. Molecular image-guided RT, incorporating a dose escalation/de-escalation scheme, was demonstrated using a simulated phantom with a dose painting plan. The plan involved delivering a low dose to the CBCT-delineated planning target volume (PTV) and a high dose boosted to the highly active biological target volume (hBTV) identified by the F-PET image. Additionally, the Bayesian eigentissue decomposition method illustrated the quantitative decomposition of radiotherapy-related parameters, specifically iodine uptake fraction and virtual noncontrast (VNC) electron density, using a simulated phantom with Kidney1 and Liver2 inserts mixed with an iodine contrast agent at electron fractions of 0.01-0.02.
EGRT simulations generated over 4,000 beamlet responses in dose slice deliveries and illustrated superior dose coverage and distribution with significantly lower doses delivered to normal tissues, even with a 2-mm setup error introduced, demonstrating the robustness of the novel EGRT scheme compared to conventional image-guided RT. In the dose-painting plan, doubling the dose to the hBTV while maintaining a low dose for the PTV resulted in an organ-at-risk (OAR) dose comparable to the low-dose treatment for the PTV alone. Furthermore, the decomposition of radiotherapy-related parameters in Kidney1 and Liver2 inserts, including iodine uptake fractions and VNC electron densities, exhibited average relative errors of less than 1.0% and 2.5%, respectively.
The results demonstrated the successful implementation of biologically guided RT within the proposed quad-model image-guided SART platform, with potential applications in preclinical RT and adaptive RT studies.
本研究旨在解决临床放射治疗（RT）与临床前小动物RT之间的技术差距，这一差距阻碍了在癌症小动物模型中对创新临床RT方法进行全面验证，以及将临床前RT研究转化为临床实践。
主要目的是在蒙特卡洛模型中探索在小动物放射治疗（SART）平台内实施生物引导RT的可行性，该平台集成了四模态机载正电子发射断层扫描（PET）、单光子发射计算机断层扫描、光子计数光谱CT和锥束CT（CBCT）成像，作为概念验证。
我们开发了一种采用四模态成像引导的SART工作流程，整合了多模态图像引导RT和发射引导RT（EGRT）。EGRT算法利用PET放射性示踪剂的正电子信号进行概述，能够在存在2毫米设置误差的情况下对放射治疗束进行近实时调整，以实现精确靶向。使用带有剂量涂抹计划的模拟体模展示了结合剂量递增/递减方案的分子图像引导RT。该计划包括对CBCT划定的计划靶区（PTV）给予低剂量，对F-PET图像确定的高活性生物靶区（hBTV）给予高剂量增强。此外，贝叶斯本征组织分解方法使用了带有Kidney1和Liver2插入物并与碘造影剂混合的模拟体模，在电子分数为0.01 - 0.02的情况下，展示了放疗相关参数（特别是碘摄取分数和虚拟非增强（VNC）电子密度）的定量分解。
EGRT模拟在剂量切片递送中产生了超过4000个射束响应，并展示了卓越的剂量覆盖和分布，即使引入2毫米设置误差，输送到正常组织的剂量也显著更低，这表明与传统图像引导RT相比，新型EGRT方案具有更强的稳健性。在剂量涂抹计划中，将hBTV的剂量加倍同时保持PTV的低剂量，导致危及器官（OAR）剂量与单独对PTV进行低剂量治疗相当。此外，Kidney1和Liver2插入物中放疗相关参数的分解，包括碘摄取分数和VNC电子密度，平均相对误差分别小于1.0%和2.5%。
结果表明在所提出的四模态图像引导SART平台内成功实施了生物引导RT，在临床前RT和自适应RT研究中具有潜在应用。
