Molecular Pathology Research for Cancer Diagnostics and Biomarkers

在这个项目中，进行研究是为了确定和开发新的人类疾病动物模型，并开发更好地确定模型与人类疾病相关性的手段，解决研究进展的关键障碍。其他目标包括开发新的研究技术，以评估和应用疾病生物标志物。在开发癌症诊断方法和开发和表征人类癌症新模型的研究资源方面取得了进展。该研究项目包括开发癌症模型的分子诊断能力，开发用于定量病理的癌症标本的自动形态图像分析方法，研究S100在癌症中的作用，开发用于有限组织（如活组织检查或模型动物）的质谱新方法，以及Ras癌基因激活的影响。并对组织生物库质量保证方法的研究进展及应用进行了展望。生物储存库支持的转化研究依赖于高质量，良好注释的标本。组织病理学评估有助于深入了解具有代表性的病变对研究目标的影响。研究了记录肿瘤和间质组织学比例的可行性，以增强有关生物库组织异质性的信息。开发了独特的空间光谱图像分析算法，用于应用自动模式识别形态学图像分析来量化生物标本组织切片中的组织学肿瘤和非肿瘤组织区域。成功获得的淋巴瘤、骨肉瘤和黑色素瘤的测量扩展到包括血液和血管组织、肺和结缔组织间充质组织（软组织肉瘤）癌症的开发和验证算法的额外进展。定量图像分析自动化有望最大限度地减少与常规生物库病理评估相关的可变性，并通过帮助指导选择适合研究的标本来增强生物标志物的发现。模式识别图像分析（PRIA）是一种人工智能自动化技术，当应用于组织病理学时，它有能力对病理学家的工作产生重大影响。计算机辅助诊断模式识别图像分析在多大程度上与公认的组织病理学方法一致尚未明确建立。来自两个来源的数字扫描组织形态学全幻灯片图像用于商业模式识别图像分析平台的评估，以解决可实现的诊断协议。在肺转移癌区域，模式识别图像分析和人工形态学图像分割之间获得了实质性的一致，缺乏显著的常数或比例误差（Passing/Bablok回归）。Bland-Altman分析显示异方差测量和方差随肿瘤负荷增加而增加的趋势，但平均偏倚无显著趋势。两种方法间肿瘤含量的平均差异为-0.64。对两种方法测量差异与肿瘤大小百分比的分析显示，方法差异主要影响最小的测量值(肿瘤负荷0.988，表明两种方法的重复性高，但模式识别图像分析的重复性更好（C.V: PRIA = 7.4, manual = 17.1）。对形态复杂畸胎瘤的模式识别图像分析的评估导致诊断与畸胎瘤亚群的病理学多能性评估一致。适应畸胎瘤的组织学特征的多样性往往导致有害的权衡，增加模式识别图像分析错误在其他地方的图像。模式识别图像分析误差是非随机的，受组织形态学变化的影响。在训练算法中遇到的文件大小限制和光谱图像处理优势的后果导致了一些畸胎瘤的诊断不准确。模式识别图像分析似乎更适合于表型多样性有限的组织。技术的改进可以提高诊断的一致性，一致的病理输入将有利于模式识别图像分析的进一步发展和应用。采用定量图像分析病理方法研究氯离子胞内通道（CLIC） 4。CLIC 4是一个氧化还原调控的，变质的多功能蛋白家族的成员，最初被描述为细胞内氯离子通道。目前的研究表明，CLICs参与多种组织的信号传导、细胞骨架完整性和分化功能。我们开发并应用了图像分析算法，以获得核定位的证据和定量结果，支持CLIC4抑制鳞状肿瘤生长的迹象，癌细胞中检测到的CLIC4表达减少和核驻留与肿瘤细胞氧化还原状态的改变有关，癌症中检测不到核CLIC4有助于tgf - β抵抗并促进肿瘤发展。体内图像分析提供了进一步的手段来证明使用小鼠IL-15与激动性抗cd40 Ab （FGK4.5）联合治疗方案增强CD8 T细胞介导的治疗效果，这导致在同基因建立的trump - c2肿瘤模型中树突状细胞（dc）以及其他细胞类型上IL-15Ralpha表达增加。IL-15是NK细胞和CD8(+) T细胞增殖和活化的关键因子，因此有潜力成为癌症治疗的免疫治疗剂。在接受联合治疗的trump - c2荷瘤小鼠中观察到的il -15相关持续缓解中，抗cd40介导的IL-15Ralpha增强表达至关重要。本项目使用的重要材料、设备或方法包括重组DNA技术、体外细胞培养、DNA序列分析、免疫诊断、分子成像、形态测量学、计算机辅助图像分析、光学成像、质谱、分子病理学和兽医医学诊断。

In this project research is conducted to characterize and develop new animal models of human disease and to develop the means to better characterize a model's relevance for human disease, addressing critical barriers to research progress. Additional aims include the the development of new research technologies for the evaluation and application of disease biomarkers. Progress was made in developing cancer diagnostics and in research resources useful in developing and characterizing new models of human cancer. This research project included developing capabilities in molecular diagnostics for cancer models, developing methods for automated morphmetric image analysis of cancer specimens for quantitative pathology, investigating the role of S100 in cancer, developing new methods in mass spectrometry for limited tissue such as biopsies or model animals, and effects of Ras oncogene activation. Continued advances and applications in developing quality assurance methods for tissue biobanking were also made. Biorepository supported translational research depends upon high-quality, well-annotated specimens. Histopathology assessment contributes insight into how representative lesions are for research objectives. Feasibility of documenting histological proportions of tumor and stroma was studied in an effort to enhance information regarding biorepository tissue heterogeneity. Unique spatial-spectral image analysis algorithms were developed for applying automated pattern recognition morphometric image analysis to quantify histologic tumor and non-tumor tissue areas in biospecimen tissue sections. Successfully acquired measurements for lymphomas, osteosarcomas and melanomas were extended to include additional progress in developing and validating algorithms for cancers of the blood and vascular tissues, lung, and connective mesenchymal tissues (soft tissue sarcoma). Quantitative image analysis automation is anticipated to minimize variability associated with routine biorepository pathologic evaluations and enhance biomarker discovery by helping to guide the selection of study-appropriate specimens. Pattern recognition image analysis (PRIA) is artificial intelligence automation technology that has the capacity to significantly impact the work that pathologists do when applied to histopathology. To what degree computer-assisted diagnostic pattern recognition image analysis agrees with accepted histopathology approaches has not been clearly established. Digitally scanned histomorphological whole-slide images from two sources served for evaluation of a commercially available pattern recognition image analysis platform, to address diagnostic agreement achievable. Substantial agreement, lacking significant constant or proportional errors, between pattern recognition image analysis and manual morphometric image segmentation was obtained for pulmonary metastatic cancer areas (Passing/Bablok regression). Bland-Altman analysis indicated heteroscedastic measurements and tendency toward increasing variance with increasing tumor burden, but no significant trend in mean bias. The average between-methods percent tumor content difference was -0.64. Analysis of between-methods measurement differences relative to the percent tumor magnitude revealed that method disagreement had an impact primarily in the smallest measurements (tumor burden 0.988, indicating high reproducibility for both methods, yet pattern recognition image analysis reproducibility was superior (C.V.: PRIA = 7.4, manual = 17.1). Evaluation of pattern recognition image analysis on morphologically complex teratomas led to diagnostic agreement with pathologist assessments of pluripotency on subsets of teratomas. Accommodation of the diversity of teratoma histologic features frequently resulted in detrimental trade-offs, increasing pattern recognition image analysis error elsewhere in images. Pattern recognition image analysis error was nonrandom and influenced by variations in histomorphology. File-size limitations encountered while training algorithms and consequences of spectral image processing dominance contributed to diagnostic inaccuracies experienced for some teratomas. Pattern recognition image analysis appeared better suited for tissues with limited phenotypic diversity. Technical improvements may enhance diagnostic agreement, and consistent pathologist input will benefit further development and application of pattern recognition image analysis.Quantitative image analysis pathology was employed to study chloride intracellular channel (CLIC) 4. CLIC 4 is a member of a redox-regulated, metamorphic multifunctional protein family, first characterized as intracellular chloride channels. Current knowledge indicates that CLICs participate in signaling, cytoskeleton integrity and differentiation functions of multiple tissues. Image analysis algorithms were developed and applied to obtain evidence of nuclear localization and quantitation results supporting the indication that CLIC4 suppresses the growth of squamous cancers, that reduced CLIC4 expression and nuclear residence detected in cancer cells is associated with the altered redox state of tumor cells, and the absence of detectable nuclear CLIC4 in cancers contributes to TGF-beta resistance and enhances tumor development.In vivo image analysis provided further means to document enhanced CD8 T cell-mediated therapeutic efficacy using a combination regimen of murine IL-15 administered with an agonistic anti-CD40 Ab (FGK4.5), which led to increased IL-15Ralpha expression on dendritic cells (DCs), as well as other cell types, in a syngeneic established TRAMP-C2 tumor model. IL-15 has potential as an immunotherapeutic agent for cancer treatment because it is a critical factor for the proliferation and activation of NK and CD8(+) T cells. Anti-CD40-mediated augmented IL-15Ralpha expression was critical in IL-15-associated sustained remissions observed in TRAMP-C2 tumor-bearing mice receiving combination therapy.The significant materials, equipment or methods in this project include use of recombinant DNA technology, in vitro cell culture, DNA sequence analysis, immunodiagnostics, molecular imaging, morphometrics, computer assisted image analysis, optical imaging, mass spectrometry, molecular pathology, and veterinary medical diagnosis.