Department of Radiology, University of Texas Health Science Center at San Antonio, San Antonio, Texas 78229-3900, USA.
Med Phys. 2010 Sep;37(9):4806-16. doi: 10.1118/1.3464491.
There are growing expectations that imaging biomarkers for tumor therapeutic drug response assessment will speed up preclinical testing of anticancer drugs in rodent models. The only imaging biomarker presently approved by the U.S. Food and Drug Administration is tumor size measurement based on either World Health Organization (WHO) criteria or Response Evaluation Criteria in Solid Tumors (RECIST). Frequently, preclinical data are accumulated from multiple research centers on multiple continents using scanners from different manufacturers and sometimes even using different imaging modalities. Very expensive cancer drug response studies can be compromised by inadequate controls to assure precision and accuracy of tumor size measurements. This project was undertaken to develop standardized quality assurance (QA) procedures using a multimodality preclinical tumor response phantom to validate the accuracy of tumor size measurements based on WHO criteria, RECIST, or global tumor volume criteria for evaluation of cytostatic drugs.
A tumor response phantom containing five low contrast test objects designed to simulate animal tumor models was made of tissue-mimicking materials. Imaging of the phantom was performed using three modalities in two institutions to evaluate size measurement of tumor-simulating test objects.
Evaluation of tumor measurements from the three commonly used imaging devices in two different institutions for monitoring tumor size changes showed that a single phantom for multiple modalities was feasible. The tumor response phantom validated precision and accuracy of tumor response data input from ultrasound, computed tomography, and/or magnetic resonance imaging devices.
Measurement results show that the standardized QA procedures using the tumor response phantom can provide a rationale check of data that excludes input from poorly maintained instruments, inadequate measurement protocols, or random operator error that frequently introduce unacceptable variability or systematic error in multiple institutions trials.
人们越来越期望肿瘤治疗药物反应的成像生物标志物能够加快在啮齿动物模型中进行抗癌药物的临床前测试。目前，只有一种成像生物标志物获得了美国食品和药物管理局的批准，即基于世界卫生组织（WHO）标准或实体瘤反应评估标准（RECIST）的肿瘤大小测量。通常，使用来自不同制造商的扫描仪，甚至使用不同的成像方式，在多个大陆的多个研究中心积累了大量的临床前数据。非常昂贵的癌症药物反应研究可能因缺乏控制而受到影响，这些控制措施旨在确保肿瘤大小测量的精确性和准确性。本项目旨在使用多模态临床前肿瘤反应体模开发标准化质量保证（QA）程序，以验证基于 WHO 标准、RECIST 或全球肿瘤体积标准评估细胞抑制剂药物的肿瘤大小测量的准确性。
一个包含五个低对比度测试物体的肿瘤反应体模，旨在模拟动物肿瘤模型，由组织模拟材料制成。使用两种机构中的三种模式对体模进行成像，以评估模拟肿瘤的测试物体的大小测量。
对来自两个不同机构的三种常用成像设备的肿瘤测量评估表明，对多个模式使用单一体模是可行的。肿瘤反应体模验证了来自超声、计算机断层扫描和/或磁共振成像设备的肿瘤反应数据输入的精度和准确性。
测量结果表明，使用肿瘤反应体模的标准化 QA 程序可以提供数据的合理性检查，排除来自维护不善的仪器、不充分的测量协议或随机操作人员错误的数据，这些错误经常会在多个机构试验中引入不可接受的变异性或系统误差。
