Imaging Based Dosimetry for Individualized Internal Emitter Therapy

摘要/摘要目前内射治疗的标准做法很少涉及治疗前计划，以优化肿瘤的辐射吸收剂量，同时避免关键器官（通常是骨髓）的毒性。这与计划外束治疗形成鲜明对比，外束治疗现在必须精确计算肿瘤和周围器官的吸收剂量。这种治疗优化尚未用于放射免疫治疗（RIT）等治疗，主要是因为准确估计内部发射器剂量相对困难，而且迄今为止的临床研究未能显示辐射吸收剂量与效果之间的一致关系。这些过去的研究通常依赖于次优的剂量估计方法和/或没有考虑预计也会影响结果的生物因素。长期目标是针对非霍奇金淋巴瘤（NHL）的I-131 tositumumab RIT等放射性核素疗法进行有效的个体化治疗计划的临床实施。本应用的目的是为肿瘤和骨髓剂量学开发准确的基于成像的方法，并开发肿瘤反应和骨髓毒性的预测模型，包括关键的剂量学因素以及生物因素，如差异增殖、放射敏感性和对未标记抗体的敏感性，这些将由生物标志物研究决定。中心假设是，与通常使用的严格剂量测量相比，结合剂量学和生物学可以更好地预测RIT结果。该假设将在标准临床护理中接受I-131 RIT的难治性NHL患者和II期临床试验的一线患者中进行检验。为了实现本应用的目标，将追求以下具体目标：1)开发SPECT重建方法，利用ct图像信息（不需要明确分割目标边界）更准确地估计目标的三维活性分布，因为剂量异质性会影响治疗效果；2)发展基于成像的骨髓剂量学，结合SPECT/CT和蒙特卡罗辐射传输，并计算定量CT确定的骨髓成分的变化；3)利用患者数据建立多元回归模型，基于剂量学因素和生物标志物研究中的生物因素（如免疫组织化学/免疫测定中的Ki-67、p53和FLT3-L）预测治疗结果（反应、毒性）；4)建立一个机制模型（对低剂量超放射敏感性进行或不进行修改），以确定预测治疗结果的等效生物学效应。这项工作具有创新性，因为与以往的研究不同，这项工作将剂量学因素与生物因素结合起来，得出了个性化治疗计划的最佳模型。这一贡献是非常重要的，因为一旦治疗优化的方法和预测模型建立起来，它们将在未来被医生用于患者选择，并在逐个患者的基础上定制RIT，以显着提高治疗效果。

DESCRIPTION (provided by applicant): Imaging based dosimetry for individualized internal emitter therapy Summary/Abstract The current standard practice for internal emitter therapy rarely involves pre-treatment planning to optimize the radiation absorbed dose to the tumor while avoiding critical organ (typically bone marrow) toxicity. This is in stark contrast to planning for external beam therapy where precise absorbed dose calculations to tumor and surrounding organs are now mandatory. Such therapy optimization has not been used in therapies such as radioimmunotherapy (RIT) primarily because accurate internal emitter dose estimation is relatively difficult and because clinical studies thus far have failed to show a consistent relationship between radiation absorbed dose and effect. These past studies typically relied on suboptimal methods of dose estimation and/or did not account for biologic factors that are also expected to affect outcome. The long-term goal is clinical implementation of effective individualized treatment planning for radionuclide therapies such as I-131 tositumomab RIT in non- Hodgkin's lymphoma (NHL). The objective in the present application is to develop accurate imaging based methods for tumor and bone marrow dosimetry and to develop predictive models for tumor response and bone marrow toxicity incorporating key dosimetric factors as well as biologic factors, such as differential proliferation, radiosensitivity and sensitivity to the unlabeled antibody, that will be determined by biomarker studies. The central hypothesis is that integrating dosimetry and biology will enable better prediction of RIT outcome than is obtained with the strictly dosimetric measures commonly used. The hypothesis will be tested in refractory NHL patients undergoing I-131 RIT in standard clinical care and in frontline patients on a phase II clinical trial. To accomplish the objective of this application the following specific aims will be pursued: 1) Develop SPECT reconstruction methods that utilize CT-image information (without needing explicit segmentation of target boundaries) to more accurately estimate the 3-D activity distribution in targets, because dose heterogeneity impacts the effect of the therapy; 2) Develop imaging based bone marrow dosimetry coupling SPECT/CT with Monte Carlo radiation transport and accounting for variations in marrow composition as determined by quantitative CT; 3) Using patient data develop a multivariate regression model for predicting therapy outcome (response, toxicity) based on dosimetric factors and biologic factors from biomarker studies (e.g., Ki-67, p53 and FLT3-L from immunohistochemistry/immunoassay); and 4) Develop a mechanistic model (with and without modification for low dose hyper-radiosensitivity) to determine the equivalent biologic effect for predicting therapy outcome. The proposed work is innovative because unlike past studies focusing purely on dosimetry this work combines dosimetric factors with biologic factors to arrive at the optimal model for individualized treatment planning. The contribution is highly significant because once the methodologies and predictive models for treatment optimization are established, they will be used in the future by physicians for patient selection and to tailor RIT on a patient-by-patient basis to considerably improve the efficacy of the treatment. PUBLIC HEALTH RELEVANCE: The proposed research is relevant to public health because it is expected to provide the clinicians with the methodology and predictive models essential for future radioimmunotherapy (RIT) treatment optimization based on accurate pre-therapy dosimetric calculations and results from widely available biomarkers. The efficacy of RIT is likely to improve substantially with such individualized planning, hence the proposed research is relevant to the part of NIH's mission that pertains to developing fundamental knowledge and the application of that knowledge to extend healthy life and reduce the burdens of illness and disability.