Alpenglow Biosciences, Inc.

Researchers present Tripath: new, deep learning models that can use 3D pathology datasets to make clinical outcome predictions. The research team imaged curated prostate cancer specimens, using two 3D high-resolution imaging techniques. The models were then trained to predict prostate cancer recurrence risk on volumetric human tissue biopsies. By comprehensively capturing 3D morphologies from the entire tissue volume, Tripath performed better than pathologists and outperformed deep learning models that rely on 2D morphology and thin tissue slices. Human tissue is intricate, complex and, of course, three dimensional. But the thin slices of tissue that pathologists most often use to diagnose disease are two dimensional, offering only a limited glimpse at the tissue's true complexity. There is a growing push in the field of pathology toward examining tissue in its three-dimensional form. But 3D pathology datasets can contain hundreds of times more data than their 2D counterparts, making manual examination infeasible. In a new study, researchers from Mass General Brigham and their collaborators present Tripath: new, deep learning models that can use 3D pathology datasets to make clinical outcome predictions. In collaboration with the University of Washington, the research team imaged curated prostate cancer specimens, using two 3D high-resolution imaging techniques. The models were then trained to predict prostate cancer recurrence risk on volumetric human tissue biopsies. By comprehensively capturing 3D morphologies from the entire tissue volume, Tripath performed better than pathologists and outperformed deep learning models that rely on 2D morphology and thin tissue slices. Results are published in Cell. While the new approach needs to be validated in larger datasets before it can be further developed for clinical use, the researchers are optimistic about its potential to help inform clinical decision making. "Our approach underscores the importance of comprehensively analyzing the whole volume of a tissue sample for accurate patient risk prediction, which is the hallmark of the models we developed and only possible with the 3D pathology paradigm," said lead author Andrew H. Song, PhD, of the Division of Computational Pathology in the Department of Pathology at Mass General Brigham. "Using advancements in AI and 3D spatial biology techniques, Tripath provides a framework for clinical decision support and may help reveal novel biomarkers for prognosis and therapeutic response," said co-corresponding author Faisal Mahmood, PhD, of the Division of Computational Pathology in the Department of Pathology at Mass General Brigham. "In our prior work in computational 3D pathology, we looked at specific structures such as the prostate gland network, but Tripath is our first attempt to use deep learning to extract sub-visual 3D features for risk stratification, which shows promising potential for guiding critical treatment decisions," said co-corresponding author Jonathan Liu, PhD, at the University of Washington. Disclosures: Song and Mahmood are inventors on a provisional patent that corresponds to the technical and methodological aspects of this study. Liu is a co-founder and board member of Alpenglow Biosciences, Inc., which has licensed the OTLS microscopy portfolio developed in his lab at the University of Washington. Funding: Authors report funding support from the Brigham and Women's Hospital (BWH) President's Fund, Mass General Hospital (MGH) Pathology, the National Institute of General Medical Sciences (R35GM138216), Department of Defense (DoD) Prostate Cancer Research Program (W81WH-18-10358 and W81XWH-20-1-0851), the National Cancer Institute (R01CA268207), the National Institute of Biomedical Imaging and Bioengineering (R01EB031002), the Canary Foundation, the NCI Ruth L. Kirschstein National Service Award (T32CA251062), the Leon Troper Professorship in Computational Pathology at Johns Hopkins University, UKRI, mdxhealth, NHSX, and Clarendon Fund.

CAMBRIDGE, Mass.--(BUSINESS WIRE)-- Incendia Therapeutics, a precision oncology company discovering and developing a novel class of therapies that reprogram the tumor microenvironment (TME), today announced that it will have three poster presentations at the American Association for Cancer Research (AACR) Annual Meeting 2024, taking place April 5-10, 2024 in San Diego, California. “We look forward to presenting these posters at AACR, which highlight our work to characterize immune exclusion in the TME,” said Laura Dillon, PhD, Vice President of Translational Medicine & Bioinformatics at Incendia Therapeutics. “The data continue to validate that Discoidin Domain Receptor 1 (DDR1), the target of Incendia’s PRTH-101 program that is currently in the clinic, is highly correlated with immune exclusion in the tumor microenvironment across epithelial tumors. Additionally, they demonstrate the power of applying H&E-based biomarkers, driven by machine learning, to better interrogate the TME. We look forward to further exploring the potential of this technology, combined with our novel class of anti-cancer therapies, to better treat cancer patients by targeting the tumor microenvironment.” Presentation details for AACR 2024 are as follows: Title: Discoidin Domain Receptor 1 (DDR1) expression is associated with degree of immune exclusion across epithelial tumors Abstract Number: 2916 / 24 Presenting Author: Laura Dillon, Ph.D., Vice President of Translational Medicine & Bioinformatics at Incendia Therapeutics Session: The Tumor Microenvironment as a Drug Target Date and Time: Monday, April 8 from 1:30 - 5:00 PM PDT (4:30 – 8:00 PM EDT) Key Highlights: DDR1 mRNA and protein expression levels are correlated with immune exclusion of lymphocytes, CD8+ T cells, and CD45+ immune cells across epithelial tumor types (R: 0.31-0.56, adjusted p values: 0.04-0.07). While the degree of the correlation varied by indication and immune cell type considered, pancreatic cancer exhibited the strongest correlation between the lymphocyte-based Immune Exclusion Scores (IES) and DDR1 mRNA and protein expression (R: 0.48-0.54, adjusted p values: 0.04-0.07). This work provides additional insight into the role of DDR1 in human cancers and has utility for selecting indications and stratifying patients for DDR1-targeted therapies. Title: Machine learning-based identification of H&E-derived morphologic features associated with CD8+ T cell immune exclusion Abstract Number: 7392 / 20 Presenting Author: Yanchao Wang, Ph.D., Data Scientist at Incendia Therapeutics Session: Artificial Intelligence and Machine/Deep Learning 4 Date and Time: Wednesday, April 10 from 9:00 AM - 12:30 PM PDT (12:00 PM – 3:30 PM EDT) Key Highlights: Morphologic features derived from H&E images can be effective predictors of CD8-defined immune exclusion, providing an option for patient stratification by immune phenotype using widely available H&E images. We developed an Ens-L1-SVM model that identified 6 H&E features predictive of CD8-defined immune phenotypes. CD8-based immune phenotypes were classified by the Ens-L1-SVM model with an estimated accuracy of 97% using only the H&E-based features as input. Details on Incendia’s poster which is a collaboration with Alpenglow Biosciences: Title: 3D spatial quantification of lymphocyte infiltration and collagen features in the tumor microenvironment using a novel assay: 3D I/O Pro™ Abstract Number: 2309 / 20 Presenting Author: Bonnie Phillips, Ph.D., Director, Commercial Development at Alpenglow Biosciences Session: Liquid Biopsy and Precision Oncology Date and Time: Monday, April 8 from 9:00 AM - 12:30 PM PDT (12:00 PM – 3:30 PM EDT) Key Highlights: An end-to-end 3D spatial biology workflow, Alpenglow’s 3D I/O Pro™, was applied to colorectal cancer (CRC) tissues stained with nuclear and general protein fluorescent dyes to quantify lymphocyte density in tumor parenchyma and stroma and analyze collagen features, including orientation, within 3D regions of interest in the TME. The ratio of stromal to parenchymal lymphocytes (lymphocyte infiltration ratio) varied from 1.4 up to 9.1 in 3D volumes and 1 to 25 in 2D virtual sections taken throughout all 5 CRC samples. Qualitatively, areas with perpendicular collagen had more lymphocyte infiltration into the tumor parenchyma than areas with parallel collagen orientation. This work highlights how the 3D I/O Pro™ workflow can characterize tumors based on complex spatial relationships within the tumor microenvironment, and could have broad applicability in research and development of novel cancer therapies that target tumor fibrosis or other features of the TME. About Incendia Therapeutics Incendia Therapeutics is discovering and developing a novel class of experimental therapeutics that reprogram the tumor microenvironment (TME). Incendia’s platform is based on rigorous, groundbreaking research involving spatial characterization of the tumor microenvironment, multi-omics data integration, and extensive preclinical testing. The Company’s most advanced experimental molecule, PRTH-101, is in a Phase 1 clinical trial for the treatment of patients with advanced solid tumors. For more information visit incendia.com and LinkedIn.

Nicholas Reder ，医学博士Alpenglow Biosciences首席执行官随着药物开发商面临越来越大的压力，需要快速有效地开发新药来满足广泛的未满足的医疗需求，病理学仍然是了解药物安全性和有效性的重要活动，但它本质上受到手动操作、模拟 2D 分析、和主观解释的影响。为了克服这些限制并改变病理学以更好地满足药物开发商的需求，Alpenglow Biosciences 正在开拓 3D 空间生物学，该技术将公司独特的成像技术与云计算和人工智能分析相结合。在本次问答中，Alpenglow 创始人兼首席执行官 Nick Reder 医学博士与 Pharma's Almanac 主编David Alvaro (DA)讨论了药物开发中病理学的历史、3D 空间生物学的前景，以及公司为实现其愿景而进行的合作和其他工作。David Alvaro (DA)：首先，您能向我介绍一下 Alpenglow Biosciences 的起源以及公司成立所依据的技术和愿景吗？Nicholas Reder (NR)： Alpenglow 是一家 3D 空间生物学公司。我们正在将病理学领域从 2D 模拟载玻片转变为利用 AI 进行 3D 直接数字完整组织成像，并保存组织用于下游测序分析。当我还是华盛顿大学病理学住院医师时，这个项目就开始了。我找到了一位刚刚来到校园的机械工程师（乔纳森·刘博士），询问他是否可以制造一台用于组织的平板扫描仪，这样我们就可以将组织直接放在扫描仪上，进行扫描并获得数字图像，而无需使用经历制作幻灯片然后将其数字化的漫长而艰巨的过程。我们从那里开始与 Adam Glaser 博士合作，他是华盛顿大学机械工程系的博士后，在 Liu 博士的指导下工作。我们正在构建一个显微镜和相关软件来实现这一愿景并提供新的组织 3D 视图。随后我们就意识到，如果我们不将其商业化，那么华盛顿大学就只有一个实验室知道这个技术有多好用了。我们可能会发表很多优秀的出版物，但我们不会真正地改变这个领域。2021年6月，我们从Dynamk Capital筹集了400万美元的融资。从 2021 年到 2022 年，我们的收入增长了十倍，并从 3 人团队扩大到 21 人团队。我们最近与 Mayo Clinic 建立了合作伙伴关系，我们开始将我们的技术应用于临床试验。Dynamk 的资金为我们公司提供了很大的助力，我们开始看到我们的技术在世界各地和临床环境中得到采用。DA：您能否讨论一下迄今为止空间生物学在药物开发中的作用以及仅使用 2D 成像的局限性或痛点？NR：几十年前，在显微镜下观察组织几乎是药物开发中必须要做的部分，但绝对不是最重要的部分。之后测序浪潮带来了变革，它引发了人们对进一步技术的更多期待，以期从组织中了解更多信息。空间生物学领域真正形成是在大约 10 年前，当时我们开始在 2D 切片中进行多重处理并添加越来越多的生物标记。在过去的几年里，它确实获得了很大的关注，人们已经开始意识到组织正在讲述一个故事，但我们只看到了这本书的第一章。我们认为我们公司对该领域的贡献使研究人员能够阅读整本书。DA：您能否详细介绍一下人工智能和云计算如何在此发挥作用吗？NR：对我而言人工智能只是算法，可以帮助你完成你不想手动完成的工作。有时，我们每天会从一块组织中生成数 TB 的数据，这会完全压垮当前的系统，并且许多任务无法手动执行。因此，我们必须开发这些算法来处理图像和数据标签，或者分割组织内的不同结构、不同类型的细胞或血管或纤维化。我以前分析前列腺活检时，我需要坐下来看着这些图像，然后圈出了所有的癌症，之后重建了组织。十小时后，我才完成了第一个数据集，但我知道我不可能以同样的方式完成另外 1,199 个数据集。从那时起我们就开始开发人工智能方法。DA：您能谈谈您的技术的成本以及让所有不同的潜在用户都能使用该技术的考虑因素吗？NR：我们的产品是一台非常复杂的显微镜。我们的目标用户是高端实验室的工作人员。但我提到的与波多黎各 CorePlus 的合作伙伴关系正在探索其他一些问题，包括我们的技术如何改善才能对高通量、资源有限的实验室有用。我们一直与他们密切合作，探索如何更广泛地采用该技术。价格可能会下降，但更重要的是流程变得如何简化。如果你能让实验室变得更有效率，经济效益就会随之而来。DA：由于最近几乎每次采访都会提到人工智能，我希望您能分享一下您如何看待这项技术的潜力以及它将在未来发挥的作用。NR：在 Alpenglow，我们相信人工智能是价值的。虽然应用人工智能的方法有很多，我们只专注于那些真正能增加很多价值的方法。另外，我们想要拥有可解释的人工智能；在数字病理学中，当你无法解释算法在做什么时，就会存在很多的风险。参考资料：https://www.pharmasalmanac.com/articles/accelerating-drug-development-with-ai-enabled-3d-spatial-biology版权信息本文系AIDD Pro接受的外部投稿，文中所述观点仅代表作者本人观点，不代表AIDD Pro平台，如您发现发布内容有任何版权侵扰或者其他信息错误解读，请及时联系AIDD Pro (请添加微信号sixiali_fox59)进行删改处理。原创内容未经授权，禁止转载至其他平台。有问题可发邮件至sixiali@stonewise.cn关注我，更多资讯早知道↓↓↓