Rapid Cancer-Treatment Efficacy Monitoring System

描述（由申请人提供）：由于缺乏快速、可靠和稳健的定量评价方案用于癌症治疗疗效的常规临床测量，通过研究患者状况的症状改善来评价癌症治疗疗效是一种流行的做法。在不影响准确性的情况下减少做出决定所需的时间和降低成本，可以提供潜在的改进治疗，特别是对患有各种类型的晚期白血病的患者。该提案的重点是快速定量评估白血病的癌症治疗效果。所提出的方法是基于量化细胞凋亡（程序性细胞死亡），并已被证明是最可靠的和最早的指标之一，用于评估药物的抗癌疗效。几种其他技术可用于检测细胞凋亡（琼脂糖凝胶电泳、半胱天冬酶-3、TUNEL测定、形态学估计、膜联蛋白V测定等）。然而，这些技术是不准确的，昂贵的或耗时的。 最近开发的两阶段DNA扩散法有望成为一种准确、相对廉价和快速的方法，用于通过凋亡指数定量测定凋亡。这两个阶段是：（1）载玻片制备和显微镜成像，其需要大约1.5小时，以及（2）图像分析，其是繁琐的，并且需要大约2小时的额外的密集手工劳动，导致细胞分类中的错误。我们建议完全自动化的第二阶段的目标，以尽量减少手动工作，从而减少人为错误。这也将使程序更具可重复性。第一阶段的工作将包括：（1）用于自动凋亡指数计算的分割和分类算法开发，以及（2）通过使用已知的凋亡诱导剂在培养的白血病细胞中诱导凋亡来表征算法性能。

DESCRIPTION (provided by applicant): Evaluating cancer therapy efficacy by studying symptomatic improvement in patient condition is a popular practice due to the lack of rapid, reliable and robust quantitative evaluation protocols for routine clinical measurement of cancer therapy efficacy. Reducing time required for arriving at decisions and reducing cost without compromising accuracy could provide potentially improved treatments especially for patients with advanced leukemia of various types. This proposal focuses on rapid quantitative assessment of cancer treatment efficacy for leukemia. The proposed approach is based on quantifying apoptosis (programmed cell death) and has been shown to be one of the most reliable and one of the earliest indicators for assessing anticancer efficacy of a drug. Several other techniques are available to detect apoptosis (Agarose gel electrophoresis, Caspase-3, TUNEL assay, Morphological estimation, Annexin V assays etc). However, these techniques are inaccurate, expensive or time consuming. A recently developed a two stage DNA diffusion assay holds promise to be an accurate, relatively inexpensive and rapid methodology for quantitative apoptosis measurement via the apoptotic index. The two stages are:(1) slide preparation and microscopy imaging, which requires approximately 1.5 hrs and (2) image analysis, which is tedious, and takes approximately 2 hours of additional intensive manual labor leading to errors in categorization of the cells. We propose to fully automate the second stage with the goal to minimize manual effort and there by reduce the human errors. This would also make the procedure more reproducible. Phase I work will involve: (1) segmentation and classification algorithm development for automatic apoptotic index calculation, and (2) characterizing the algorithm performance by inducing apoptosis in leukemia cells in culture using known apoptosis inducing agents.