Search

Cancer is the leading cause of death worldwide. Unfortunately, efforts to understand this disease are confounded by the complex, heterogenous tumor microenvironment (TME). Better understanding of the TME could lead to novel diagnostic, prognostic, and therapeutic discoveries. One way to achieve this involves in vitro tumor models that recapitulate the in vivo TME composition and spatial arrangement. Here, we review the potential of harnessing in vitro tumor models and artificial intelligence to delineate the TME. This includes (i) identification of novel features, (ii) investigation of higher-order relationships, and (iii) analysis and interpretation of multiomics data in a (iv) holistic, objective, reproducible, and efficient manner, which surpasses previous methods of TME analysis. We also discuss limitations of this approach, namely inadequate datasets, indeterminate biological correlations, ethical concerns, and logistical constraints; finally, we speculate on future avenues of research that could overcome these limitations, ultimately translating to improved clinical outcomes.

BackgroundRecent advances in multiplex immunohistochemistry/immunofluorescence (mIHC/IF) technologies have enabled simultaneous measurements of large numbers of markers on the same tissue sections, and more comprehensive views of the cellular compositions and immune responses of the tumor microenvironment. However, the reproducibility and interpretation of the complex staining patterns and analysis results obtained from these markers remain major barriers to more general adoptions of these technologies. Here, we report the availability of an online public portal, “ImmunoAtlas”, which would enable researchers/clinicians to present or share their published mIHC/IF images or results; international workgroups to create and share standard marker panels or assay guidelines; end users to validate antibodies or protocols; or computational scientists to benchmark different analysis methods on standard reference images (figure 1).MethodsImmunoAtlas is based on a HistoPathological image management and Analysis (HPA) software platform developed by us, and hosted in the data center of Bioinformatics Institute. The platform uses the cellXpress software,1 which is written in C++ and supports parallel processing, to efficiently process and manage large numbers of huge mIHC/IF or brightfield images. The web interface of ImmunoAtlas is also completely developed by us in PHP and JavaScript to address the specific needs and requirements in managing these images.ResultsImmunoAtlas is a user-friendly online portal for sharing, visualizing, and referencing original tissue images and analysis results (https://ImmunoAtlas.org). We have completed the first phase of development of the portal. Users can now perform image uploading, annotation, publishing, sharing, and viewing with standard web browsers on desktop computers or mobile devices/phones (figure 2). The portal supports image files from common microscopes and slide scanners, and can process mIHC/IF or brightfield images from selected areas, tissues microarrays, or whole slides. It can handle up to 1000 multiplexed markers, and whole-slide images that are >20GB/image. Several internal and external immuno-oncology studies have deposited and shared their images via ImmunoAtlas. They include a study of multiplexed PD-L1 markers in breast cancers2; the development of a panel of 56 highly-multiplexed markers for cutaneous T cell lymphoma3; and a study of CD38 scoring in hepatocellular carcinoma.4ConclusionsImmunoAtlas promotes open science and collaborations that can accelerate the adoptions of mIHC/IF technologies in immuno-oncology. The next phase of development will focus on adding image searching and comparison functions to the portal. The community is welcome to use and share their images and analysis results via the portal.ReferencesLaksameethanasan D, Tan RZ, Toh GW-L, et al. cellXpress: a fast and user-friendly software platform for profiling cellular phenotypes. BMC Bioinformatics 2013;14:S4. doi:10.1186/1471-2105-14-S16-S4.Yeong J, Tan T, Chow ZL, et al. Multiplex immunohistochemistry/immunofluorescence (mIHC/IF) for PD-L1 testing in triple-negative breast cancer: a translational assay compared with conventional IHC. J Clin Pathol 2020;73:557–62. doi:10.1136/jclinpath-2019-206252.Phillips D, Schürch CM, Khodadoust MS, et al. Highly multiplexed phenotyping of immunoregulatory proteins in the tumor microenvironment by CODEX tissue imaging. Front Immunol 2021;0. doi:10.3389/fimmu.2021.687673.Ng HHM, Lee RY, Goh S, et al. Immunohistochemical scoring of CD38 in the tumor microenvironment predicts responsiveness to anti-PD-1/PD-L1 immunotherapy in hepatocellular carcinoma. J Immunother Cancer 2020;8:e000987. doi:10.1136/jitc-2020-000987Abstract 627 Figure 1Key target users and applications of ImmunoAtlasAbstract 627 Figure 2ImmunoAtlas' web interface for sharing and visualizing mIHC

Human lungs are susceptible to the toxicity induced by soluble xenobiotics. However, the direct cellular effects of many pulmonotoxic chemicals are not always clear, and thus, a general in vitro assay for testing pulmonotoxicity applicable to a wide variety of chemicals is not currently available. Here, we report a study that uses high-throughput imaging and artificial intelligence to build an in vitro pulmonotoxicity assay by automatically comparing and selecting human lung-cell lines and their associated quantitative phenotypic features most predictive of in vivo pulmonotoxicity. This approach is called “High-throughput In vitro Phenotypic Profiling for Toxicity Prediction” (HIPPTox). We found that the resulting assay based on two phenotypic features of a human bronchial epithelial cell line, BEAS-2B, can accurately classify 33 reference chemicals with human pulmonotoxicity information (88.8% balance accuracy, 84.6% sensitivity, and 93.0% specificity). In comparison, the predictivity of a standard cell-viability assay on the same set of chemicals is much lower (77.1% balanced accuracy, 84.6% sensitivity, and 69.5% specificity). We also used the assay to evaluate 17 additional test chemicals with unknown/unclear human pulmonotoxicity, and experimentally confirmed that many of the pulmonotoxic reference and predicted-positive test chemicals induce DNA strand breaks and/or activation of the DNA-damage response (DDR) pathway. Therefore, HIPPTox helps us to uncover these common modes-of-action of pulmonotoxic chemicals. HIPPTox may also be applied to other cell types or models, and accelerate the development of predictive in vitro assays for other cell-type- or organ-specific toxicities. Electronic supplementary material The online version of this article (10.1007/s00204-018-2213-0) contains supplementary material, which is available to authorized users.