Dynamic-CT-based biomarker for predicting clinical outcome in CRC

基于动态 CT 的生物标志物用于预测 CRC 的临床结果

基本信息

批准号：
8893927
负责人：
HIROYUKI YOSHIDA
金额：
$ 18.92万
依托单位：
MASSACHUSETTS GENERAL HOSPITAL
依托单位国家：
美国
项目类别：
财政年份：
2014
资助国家：
美国
起止时间：
2014-08-01 至 2018-07-31
项目状态：
已结题

来源：
https://reporter.nih.gov/project-details/8893927
关键词：
Accounting Address Adverse effects Area Biological Markers Bolus Infusion Cancer Etiology Carcinoma Cessation of life Clinical Colon Colorectal Cancer Computational algorithm Cost Savings Databases Decision Making Development Diagnosis Diagnostic Diagnostic Neoplasm Staging Dose Electromagnetic Energy Evaluation Exposure to Fourier Transform Frequencies Generations Growth Health Image Individual Kinetics Machine Learning Modeling Morbidity - disease rate Neoplasm Metastasis Noise Operative Surgical Procedures Outcome Patients Performance Perfusion Phase Physiological Play Process Prognostic Marker Property Protocols documentation Radiation Recurrence Regimen Reporting Residual state Risk Role Solutions Source Staging Stratification Survival Rate Techniques Therapeutic Time Tissues Tracer Tumor Angiogenesis Tumor stage United States X-Ray Computed Tomography abdominal aorta angiogenesis base clinical practice colon cancer patients effective therapy follow-up hemodynamics imaging biomarker improved in vivo in vivo Model mortality novel response temporal measurement

项目摘要

DESCRIPTION (provided by applicant): Colorectal cancer (CRC) is the second leading cause of cancer death in the United States and is responsible for significant morbidity and mortality. A patient's five-year survival rate depend on the tumor stage at the time of diagnosis, and stage of the tumor plays a substantial role in decision-making regarding treatment. Thus, precise pre-treatment diagnostic evaluation and staging of colorectal cancer are important. In addition, angiogenesis plays an important role in the process of growth and metastasis in CRC and is reported as a useful prognostic marker, similar to many other carcinomas. Thus, in vivo quantification of the tumor angiogenesis rate holds promise in improving the management of CRC. Perfusion computed tomography (PCT) acquires high temporal resolution images, thus enabling evaluation of hemodynamic changes of tissue in vivo by modeling tracer kinetics. PCT has been reported to characterize tumor angiogenesis, and to be a more sensitive imaging biomarker for predicting of overall survival (OS) of CRC patients than conventional tumor staging. In clinical practice, however, the PCT protocol is a trade-off between high-temporal resolution and the total radiation dose required. Thus, the use of dynamic CT imaging derived from four temporal phases, which include pre-contrast, arterial, portal, and delayed phases, is highly desirable, because it is more readily available and yields substantially lower radiation exposure to the patients than that of PCT. However, low temporal resolution in four-phase dynamic CT presents several barriers in modeling tracer kinetics, primarily because of the lack of temporal enhancement information, which limits the ability to obtain reliable physiological information. We will thus develop a novel continuous-time modeling of tracer kinetics without any discretization of the enhancement curves. Such an approach will enable estimation of the time lag between onset time points of input and response enhancements as well as other kinetic parameters in four-phase dynamic CT. We hypothesize that the proposed tracer kinetic model can be an effective imaging biomarker for the risk stratification of recurrence of CRC and for prediction of OS. To explore these hypotheses, the specific aims of the proposed project are (1) Develop a novel single-input continuous-time tracer kinetic model without any discretization to fit temporal enhancement curves in four-phase dynamic CT of the colon, and (2) develop kinetic-model-based imaging biomarkers from four-phase dynamic CT and evaluate their performance in predicting clinical outcome in CRC patients. Successful development of a novel imaging biomarker based on four-phase dynamic CT holds high promise for the development of tailor-made optimal therapy without excessive radiation exposure to the patient.

描述（由申请人提供）：结直肠癌 (CRC) 是美国癌症死亡的第二大原因，导致较高的发病率和死亡率。患者的五年生存率取决于诊断时的肿瘤分期，肿瘤分期在治疗决策中起着重要作用。因此，精确的治疗前诊断评估和结直肠癌分期非常重要。此外，血管生成在结直肠癌的生长和转移过程中发挥着重要作用，据报道，与许多其他癌症类似，血管生成是一种有用的预后标志物。因此，肿瘤血管生成率的体内定量有望改善结直肠癌的治疗。灌注计算机断层扫描 (PCT) 获取高时间分辨率图像，从而能够通过模拟示踪动力学来评估体内组织的血流动力学变化。据报道，PCT 可以表征肿瘤血管生成，并且是比传统肿瘤分期更敏感的成像生物标志物，用于预测 CRC 患者的总生存期 (OS)。然而，在临床实践中，PCT 方案是高时间分辨率和所需总辐射剂量之间的权衡。因此，非常需要使用源自四个时间阶段（包括预造影期、动脉期、门静脉期和延迟期）的动态 CT 成像，因为它更容易获得，并且对患者产生的辐射暴露比常规成像要低得多。 PCT。然而，四相动态 CT 的低时间分辨率在示踪动力学建模中存在一些障碍，主要是因为缺乏时间增强信息，这限制了获得可靠的生理信息的能力。因此，我们将开发一种新颖的示踪动力学连续时间建模，而无需对增强曲线进行任何离散化。这种方法将能够估计输入和响应增强的开始时间点之间的时间滞后以及四相动态 CT 中的其他动力学参数。我们假设所提出的示踪动力学模型可以作为 CRC 复发风险分层和 OS 预测的有效成像生物标志物。为了探索这些假设，该项目的具体目标是（1）开发一种新颖的单输入连续时间示踪动力学模型，无需任何离散化来拟合结肠四相动态 CT 的时间增强曲线，以及（2）从四相动态 CT 开发基于动力学模型的成像生物标志物，并评估其在预测 CRC 患者临床结果方面的性能。基于四相动态 CT 的新型成像生物标志物的成功开发，为开发量身定制的最佳治疗方案带来了巨大的希望，而无需对患者进行过多的辐射暴露。