Multi-Modality Quantitative Imaging for Evaluation of Response to Cancer Therapy

骨转移在包括前列腺癌在内的晚期癌症中很常见， 超过70%的晚期疾病患者存在骨转移的证据。尽管它们普遍存在 和重要的临床意义，目前还没有广泛接受的定量成像 评估对治疗的反应。缓解或进展的评估通常基于 结合CT和血清标志物对骨密度成像进行目视判读。这个项目的目标 因此，开发、验证和推进临床实践多模态定量度量， 骨转移对治疗的反应。SPECT-CT组合系统及其鲁棒性实现 补偿方法已经使得能够开发定量SPECT成像方法， 最近已在临床上可用。在这里，我们建议利用这些发展的联系，通过 三个目标。(1)使用模拟和体模研究，我们将首先优化和评估方法， Tc-99 m MDP的定量SPECT/CT（QSPECT）成像。我们建议扩大幻影研究， QIN合作项目，以评估不同扫描仪和中心的方法。（二） 基于这些QSPECT方法，我们将开发、优化和表征定量骨SPECT 用于评估对治疗的反应的指标。计算这些指数将涉及半自动 骨和骨病变的分割，在集成的软件工具中实现。计算的指数 将包括病变数量、骨病变负荷和治疗后病变摄取的变化。我们将 使用真实模拟和治疗前重测研究表征指数的变异性。（三） 最后，我们将应用这些方法，并在目前的新治疗方案的临床试验中验证它们。 骨转移性去势抵抗性前列腺癌在这些研究中，我们将评估这些方法， 并将比较体内方法与从以下估计的类似定量指标： Na18F PET/CT。我们还将研究全身MRI提供的额外信息量 和PSMA PET/CT。我们将向QIN提供方法和图像，并建议组织 细分和响应指标的挑战，以利用QIN范围内对这些 地区总之，这些定量成像工具和研究将提供一种有效的方法， 基于定量成像的骨转移缓解评价， 其相对于临床标准骨造影术仅在成本上有微小增加。

Bone metastases are common in advanced cancers including prostate adenocarcinoma, where there is evidence of bone metastases in more than 70% of patients with advanced disease. Despite their prevalence and the important clinical implications, there is currently no widely accepted quantitative imaging assessment for response to therapy. Assessment of response or progression is typically made based on visual interpretation of bone scintigraphy in conjunction with CT and serum markers. The goal of this project is thus to develop, validate, and advance to clinical practice multi-modality quantitative metrics for assessing response of bone metastases to therapy. Combined SPECT-CT systems and the implementation of robust compensation methods has enabled the development of quantitative SPECT imaging methods, which have recently become clinically available. Here we propose to exploit the nexus of these developments via the following 3 aims. (1) Using simulation and phantom studies, we will first optimize and evaluate methods for quantitative SPECT/CT (QSPECT) imaging of Tc-99m MDP. We propose to expand the phantom studies to a QIN cooperative project in order to evaluate the methods across different scanners and centers. (2) Based on these QSPECT methods, we will develop, optimize, and characterize quantitative bone SPECT indices for assessing response to therapy. Computing these indices will involve semi-automatic segmentation of bone and bone lesions, implemented in an integrated software tool. The indices calculated will include the number of lesions, bone lesion burden, and change in lesion uptake with therapy. We will characterize the variability of the indices using realistic simulations and in pre-therapy test-retest studies. (3) Finally, we will apply the methods and validate them in current clinical trials of new therapeutic regimens for bone metastatic castrate-resistant prostate cancer. In these studies, we will evaluate the methods in terms of time to progression and will compare the methods in vivo to similar quantitative metrics estimated from Na18F PET/CT. We will also investigate the amount of additional information provided by whole-body MRI and PSMA PET/CT. We will make the methods and images available to the QIN and propose to organize challenges of segmentation and response metrics to take advantage of the QIN-wide interest in these areas. Together, these quantitative imaging tools and studies will provide a validated approach for quantitative imaging-based evaluation of response in bone metastases that is readily clinically implementable with only a minor increase in cost relative to clinical standard bone scintigraphy.