Quantitative molecular and cellular MRI of hepatocyte transplantation

 描述(申请人提供)：对于许多严重的肝病，唯一有效的治疗方法是肝移植。不幸的是，由于可用的供体器官短缺，或者由于患者的年龄，并不是所有的患者都可以进行肝移植。肝细胞移植(HTX)是一种替代的实验性治疗方法，在人类身上的长期成功有限。一个尚未解决的主要问题是，我们如何动态监测和量化HTX中的细胞移植，以帮助改善HTX并密切监测临床实施情况？为了回答这个问题，我们结合了创新的分子和细胞磁共振方法，以及机器学习和计算机视觉，以非侵入性量化HTX后肝脏中的细胞植入和长期植入和再繁殖(LTER)。使用便于HTX后LTER的小鼠模型，我们将使用这种组合方法来量化HTX后第1天和第7天肝脏中的移植细胞，反映最初细胞输送的时间(第1天)和 实际细胞在组织中的植入(第7天)。然后，我们将在移植后30-90天测量这些细胞在肝脏中的LTER，创新地使用FDA批准的专门用于健康肝细胞的MRI造影剂Eovist。这些实验将评估小鼠供体细胞，以及猪原代肝细胞和猪胚胎干细胞来源的肝细胞系。这项变革性的工作将是第一次在任何动物模型中利用再生医学的分子和细胞MRI实现这种水平的量化。此外，我们将在早期阶段测试旨在预测LTER结果的模式识别算法。预测LTER结果的能力将是范式的转变，因为它将使医生能够考虑额外的HTX方案或在HTX失败的情况下进行二线治疗。这几乎是不可能的。虽然这个项目将在老鼠身上开发，但成像方案的临床翻译将是直截了当的，因为我们用来测量小鼠HTX的准确成像和数据分析方案也可以用于测量人类的HTX。Eovist被FDA批准用于患有肝病的人类，基于MRI的细胞跟踪铁标记细胞正在进行临床试验。我们的初步数据有力地表明，磁共振成像和数据分析可以在200微米的分辨率下区分单个细胞，这意味着磁共振成像很可能在任何高场人体磁共振系统上进行。此外，干细胞来源的肝细胞甚至达到部分LTER的发现，对于探索人类和/或猪干细胞来源的肝细胞供人类使用将是非常令人鼓舞的，因为这些细胞可能会缓解细胞供应不足的关键问题，类似于猪胰岛移植的进展。 这项拟议的研究采取了多学科的方法，拥有肝细胞移植和生物学、分子成像、机器学习/计算机视觉和小鼠肝脏疾病模型方面的专业知识。研究人员之间的合作正在进行中，在拟议工作的所有方面都有广泛的初步数据。

 DESCRIPTION (provided by applicant): For many severe liver diseases, the only effective treatment is liver transplantation. Unfortunately, due to the shortage of available donor organs, or the patient's age, liver transplantation is not available to all patients. Hepatocyte transplantation (HTx) is an alternative, experimental treatment, with limited long term success in humans. A major unsolved question is how can we dynamically monitor and quantify cell transplantation in HTx to help improve HTx and to closely monitor clinical implementation? To answer this question, we combine an innovative molecular and cellular MRI approach, with machine learning and computer vision, to non-invasively quantify cell engraftment and long term engraftment and repopulation (LTER) in the liver following HTx. Using a mouse model that facilitates LTER following HTx, we will use this combined approach to quantify transplanted cells in the liver at Days 1 and 7 after HTx, reflecting the timings of initial cell delivery (Day 1) and actual cellular engraftment in the tissue (Day 7). Then, we will measure LTER of these cells in the liver 30 - 90 days post-transplant, making innovative use of Eovist, an FDA approved MRI contrast agent specific for healthy hepatocytes. These experiments will evaluate mouse donor cells, as well as pig primary hepatocytes and pig embryonic stem cell-derived hepatocyte cell line. This transformative work will be the first study to achieve this level of quantification wit molecular and cellular MRI of regenerative medicine, in any animal model. Additionally, we will test pattern recognition algorithms aimed at predicting the outcome of LTER, at an early stage. The capability to predict LTER outcome would be paradigm shifting as it would enable physicians to consider additional HTx regimens or second line treatments if HTx fails. This is seldom possible. Though this project will be developed on mice, clinical translation of the imaging protocol would be straightforward because the exact imaging and data analysis scheme that we use to measure HTx in mouse, can be used to measure HTx in humans. Eovist is FDA approved for use in humans with liver disease, and MRI-based cell tracking of iron labeled cells is in clinical trials. Our preliminary data strongly suggests that MRI and data analysis can discriminate single cells at 200 µm resolution, meaning the MRI could likely be performed on any high field human MRI system. Additionally, the discovery that a stem cell-derived hepatocyte achieved even partial LTER would be extremely encouraging for exploring human and/or pig stem cell-derived hepatocytes for human use, because these cells can potentially alleviate the crucial issue of poor cell supply, similar to progress seen in pig islet transplant. The proposed research takes a multidisciplinary approach with expertise in hepatocyte transplant and biology, molecular imaging, machine learning/computer vision, and mouse liver disease models. Collaboration among the researchers is ongoing with extensive preliminary data across all aspects of the proposed work.