Micro-tumor detection by quantifying tumor-induced vascular abnormalities (PQ-13)

描述(申请人提供)：目前的成像方式允许检测由大约107个细胞组成的肿瘤或1立方毫米范围内的肿瘤。成像灵敏度的任何提高都将在肿瘤检测和治疗结果方面提供有价值的进步；然而，检测灵敏度的重大提高将从根本上改变我们在临床实践中使用成像的方式。超声(US)因其安全、低成本、便于携带等优点，在未来的肿瘤学成像中可能扮演重要的角色，并不断扩大。然而，由于基本分辨率的限制，临床US只能检测到几毫米或更大的肿瘤肿块。为了提高这种检测灵敏度，需要改变美国对肿瘤成像的方法。这个项目提出了这样的转变。众所周知，肿瘤在整个血管生成过程中会严重扭曲微血管。数据显示，微血管结构的实质性变化发生在肿瘤细胞到达10s至100s之后，这些变化延伸到 相对较大(直径数百微米)，微血管变化远远超出肿瘤边缘，甚至在发病后不久也是如此。这些独特的微血管“癌症信号”为我们提供了一种克服传统分辨率限制的方法，否则就会损害微肿瘤的检测。因此，我们对提高对微小癌症的成像敏感性的创新反应是检测这些微血管的变化，而不是实体肿瘤本身。先前的研究小组已经使用光学显微镜说明了这一概念的潜力，然而光学显微镜对于这一应用来说本质上是非临床可翻译的，因此我们将利用一种新的超声方法。虽然以前超声在评估微血管结构的变化方面没有提供有用的工具，但我们的团队最近实施了一种新的美国成像技术，称为“声学血管成像”，它为成像微血管结构提供了最高的信噪比和高分辨率。因此，这种新的成像技术使得微血管分割和曲折量化成为可能。我们的第一个假设是，我们可以优化这种成像方法，以获得足够的空间分辨率和穿透深度，以便临床实施。我们的第二个假设是，当肿瘤比目前的检测极限小至少两到三个数量级时，我们可以使用声学血管造影术来检测肿瘤引起的微血管变化。我们的第三个假设是，我们可以开发当前的分割和分析算法，这些算法将表征微血管的形态，并为检测存在微肿瘤风险的组织提供一种特定和敏感的分类方法。令人鼓舞的初步数据已经表明，我们有能力使用声学血管造影术来区分小肿瘤和健康组织，仅基于微血管形态的分析，这些进一步的研究将使这项前景光明的新技术的全面发展成为可能。我们的方法将包括使用临床相关的乳腺癌基因工程模型进行的体外研究和临床前体内研究。

DESCRIPTION (provided by applicant): Current imaging modalities allow detection of tumors composed of approximately 107 cells or in the range of 1 cubic millimeter. Any increase in imaging sensitivity provides valuable advances in tumor detection and also treatment outcomes; however, a major increase in detection sensitivity would provide a radical change in how we might employ imaging in clinical practice. Ultrasound (US) will likely play a significant and expanding role in oncological imaging in the future because it is safe, low-cost, and readily portable. However, due to fundamental resolution limitations, clinical US can only detect tumor masses on the order of a few millimeters, or larger. In order to improve this detection sensitivity a paradigm shift in the US approach for imaging tumors is needed. This project proposes such a shift. It is well known that tumors dramatically distort microvasculature throughout the angiogenic process. Data show that substantial changes in microvasculature structure occur after the arrival of only 10s to 100s of tumor cells, that these changes extend to vessels that are relatively large (hundreds of microns in diameter), and that microvascular changes extend well beyond tumor margins, even soon after the onset of disease. These unique microvascular "cancer signatures" provide us with a means to overcome traditional resolution limitations which otherwise impair micro-tumor detection. Thus, our innovative response to improving imaging sensitivity to micro-cancers is to detect these microvascular changes, rather than the solid tumor itself. Prior groups have illustrated the potential for this concept using optical microscopy however optical microscopy is inherently non-clinically translatable for this application, and hence we will utilize a novel ultrasound approach. Although previously, ultrasound has not provided utility in assessing changes in microvascular structure, our group has recently implemented a new US imaging technique called "Acoustic Angiography" which provides supreme signal-to-noise and high resolution for imaging microvessel structure. This new imaging technique thus enables microvessel segmentation and tortuosity quantification. Our first hypothesis is that we can optimize this imaging approach for adequate spatial resolution and depth of penetration for clinical implementation. Our second hypothesis is that we can use acoustic angiography to detect tumor-induced microvascular changes when tumors are at least two to three orders of magnitude smaller than current detection limits. Our third hypothesis is that we can develop current segmentation and analysis algorithms which will characterize microvessel morphology and provide a specific and sensitive classification approach for detecting tissue that is at risk for hosting micro-tumors. Encouraging preliminary data have already illustrated our ability use acoustic angiography to discriminate small tumors and healthy tissue based on an analysis of microvessel morphology alone, and these further studies will enable a comprehensive development of this promising new technology. Our approach will involve in-vitro studies as well as preclinical in-vivo studies using clinically-relevant geneticaly engineered models of breast cancer.