Quantitative SPECT for Targeted Radionuclide Therapy

描述(申请人提供)：靶向放射性核素疗法(TRT)在包括甲状腺癌和非霍奇金淋巴瘤在内的多种癌症的治疗中发挥着越来越重要的作用。治疗其他癌症的TRT试剂正处于不同的开发阶段。例如，使用Y-90标记的微球局部治疗不可切除的肝肿瘤似乎很有希望。对于这些疗法，剂量学是其开发、批准和验证的重要组成部分。剂量学可以帮助减少试验中的不良反应，并提供对治疗药物在个人和人群中失败或成功的原因的洞察。它在针对患者的治疗计划中也发挥着重要作用。TRT中的剂量估计基于定量平面或SPECT成像研究的结果。定量平面成像虽然被广泛使用，但涉及对各种图像退化效果的补偿的特别组合，导致不同的准确度和精确度。商业SPECT系统上可用的图像重建方法通常是为涉及视觉解释的诊断程序而不是为了量化而设计和优化的。在这笔赠款的前一个资助期，我们开发了定量SPECT重建方法，并在整个器官剂量学的背景下进行了验证。然而，对于肿瘤、局部治疗或放射性标记的多肽，3D剂量测量是必不可少的，这需要在亚器官水平上估计器官中的3D活性分布。活度分布估计的准确性受到图像退化因素、噪声和部分体积效应的限制。对不发射伽马射线的TRT试剂的韧致辐射成像的能力可能会有许多重要的应用，但由于初级光子的连续能谱和由此产生的高水平光子散射，这一能力变得复杂。在这一竞争更新中，我们建议开发和优化用于TRT剂量测量应用的定量SPECT采集和重建方法。这些方法将包括对伽马射线和韧致辐射发射体的图像退化效应进行建模的SPECT重建方法。他们将使用3D和4D最大后验概率(MAP)方法来提供降噪，并结合解剖学信息来减少部分体积效应。4D方法还将通过在时间维度上优化平滑并将配准纳入重建算法来提供进一步的降噪。我们建议使用物理模型、真实模拟和动物研究相结合的方法来优化和验证这些新的定量SPECT方法，并将它们应用于几种TRT药物的临床试验。最后，我们将严格评估这些方法的准确度和精确度，并与传统方法在模拟幻影群体中的比较。这项研究的结果将是一套经过充分验证的定量SPECT重建方法，具有良好的准确性和精确度。这些方法将大大改进3D剂量估计，从而提高预测和理解生物反应以及优化TRT治疗剂量的能力。

DESCRIPTION (provided by applicant): Targeted radionuclide therapy (TRT) plays an increasingly important role in treatment of a number of cancers including thyroid cancer and non-Hodgkins lymphoma. TRT agents for other cancers are at various stages of development. For example, locoregional therapy using Y-90 labeled microspheres for non-resectable liver tumors appears promising. For these therapies, dosimetry is an essential part of their development, approval, and validation. Dosimetry can help reduce adverse reactions in trials and provides insight into the reasons for failure or success of the therapeutic agents both in individuals and in populations. It also plays an important role in patient-specific treatment planning. Dose estimates in TRT are based on results of quantitative planar or SPECT imaging studies. Quantitative planar imaging, though widely used, involves ad hoc combinations of compensations for various image degrading effects, resulting in variable accuracy and precision. Image reconstruction methods available on commercial SPECT systems are typically designed and optimized for diagnostic procedures involving visual interpretation and not for quantification. In the previous funding period of this grant we developed quantitative SPECT reconstruction methods and validated them in the context of whole organ dosimetry. However, for tumors, locoregional therapy or radiolabeled peptides, 3D dosimetry is essential, which requires estimates of the 3D activity distribution in organs at the sub-organ level. The accuracy of activity distribution estimates is limited by image degrading factors, noise, and partial volume effects. The ability to image bremsstrahlung radiation from TRT agents that do not emit gamma rays could have a number of important applications, but is complicated by the continuous energy spectrum of primary photons and the resulting high levels of photon scatter. In this competing renewal, we propose to develop and optimize quantitative SPECT acquisition and reconstruction methods for TRT dosimetry applications. These will include SPECT reconstruction methods that model the image degrading effects for both gamma ray and bremsstrahlung radiation emitters. They will use 3D and 4D maximum a posteriori (MAP) methods to provide noise reduction and incorporate anatomical information to reduce partial volume effects. The 4D methods will also provide further noise reduction through optimized smoothing in the time dimension and incorporate registration into the reconstruction algorithm. We propose to optimize and validate these new quantitative SPECT methods using a combination of physical phantom, realistic simulation and animal studies and to apply them in clinical trials of several TRT agents. Finally, we will rigorously evaluate the accuracy and precision of these methods in comparison with conventional methods in simulated populations of phantoms. The result of this research will be a set of well-validated quantitative SPECT reconstruction methods with well-characterized accuracies and precisions. These methods would provide substantial improvements in 3D dose estimates, and thus in the ability to predict and understand biological response and optimize therapeutic doses for TRT.