OPTIMAL: An OPTimized Imaging Mass cytometry AnaLysis framework for benchmarking segmentation and data exploration

Analysis of imaging mass cytometry (IMC) data and other low-resolution multiplexed tissue imaging technologies is often confounded by poor single-cell segmentation and suboptimal approaches for data visualization and exploration. This can lead to inaccurate identification of cell phenotypes, states, or spatial relationships compared to reference data from single-cell suspension technologies. To this end we have developed the "OPTimized Imaging Mass cytometry AnaLysis (OPTIMAL)" framework to benchmark any approaches for cell segmentation, parameter transformation, batch effect correction, data visualization/clustering, and spatial neighborhood analysis. Using a panel of 27 metal-tagged antibodies recognizing well-characterized phenotypic and functional markers to stain the same Formalin-Fixed Paraffin Embedded (FFPE) human tonsil sample tissue microarray over 12 temporally distinct batches we tested several cell segmentation models, a range of different arcsinh cofactor parameter transformation values, 5 different dimensionality reduction algorithms, and 2 clustering methods. Finally, we assessed the optimal approach for performing neighborhood analysis. We found that single-cell segmentation was improved by the use of an Ilastik-derived probability map but that issues with poor segmentation were only really evident after clustering and cell type/state identification and not always evident when using "classical" bivariate data display techniques. The optimal arcsinh cofactor for parameter transformation was 1 as it maximized the statistical separation between negative and positive signal distributions and a simple Z-score normalization step after arcsinh transformation eliminated batch effects. Of the five different dimensionality reduction approaches tested, PacMap gave the best data structure with FLOWSOM clustering out-performing phenograph in terms of cell type identification. We also found that neighborhood analysis was influenced by the method used for finding neighboring cells with a "disc" pixel expansion outperforming a "bounding box" approach combined with the need for filtering objects based on size and image-edge location. Importantly, OPTIMAL can be used to assess and integrate with any existing approach to IMC data analysis and, as it creates .FCS files from the segmentation output and allows for single-cell exploration to be conducted using a wide variety of accessible software and algorithms familiar to conventional flow cytometrists.

Location, location, location: mapping the lymphoma tumor microenvironment using spatial transcriptomics

Lymphomas are a heterogenous group of lymphoid neoplasms with a wide variety of clinical presentations. Response to treatment and prognosis differs both between and within lymphoma subtypes. Improved molecular and genetic profiling has increased our understanding of the factors which drive these clinical dynamics. Immune and non-immune cells within the lymphoma tumor microenvironment (TME) can both play a key role in antitumor immune responses and conversely also support lymphoma growth and survival. A deeper understanding of the lymphoma TME would identify key lymphoma and immune cell interactions which could be disrupted for therapeutic benefit. Single cell RNA sequencing studies have provided a more comprehensive description of the TME, however these studies are limited in that they lack spatial context. Spatial transcriptomics provides a comprehensive analysis of gene expression within tissue and is an attractive technique in lymphoma to both disentangle the complex interactions between lymphoma and TME cells and improve understanding of how lymphoma cells evade the host immune response. This article summarizes current spatial transcriptomic technologies and their use in lymphoma research to date. The resulting data has already enriched our knowledge of the mechanisms and clinical impact of an immunosuppressive TME in lymphoma and the accrual of further studies will provide a fundamental step in the march towards personalized medicine.

Generation and Integrated Analysis of Advanced Patient-Derived Orthoxenograft Models (PDOX) for the Rational Assessment of Targeted Therapies in Endometrial Cancer

Clinical management of endometrial cancer (EC) is handicapped by the limited availability of second line treatments and bona fide molecular biomarkers to predict recurrence. These limitations have hampered the treatment of these patients, whose survival rates have not improved over the last four decades. The advent of coordinated studies such as The Cancer Genome Atlas Uterine Corpus Endometrial Carcinoma (TCGA_UCEC) has partially solved this issue, but the lack of proper experimental systems still represents a bottleneck that precludes translational studies from successful clinical testing in EC patients. Within this context, the first study reporting the generation of a collection of endometrioid-EC-patient-derived orthoxenograft (PDOX) mouse models is presented that is believed to overcome these experimental constraints and pave the way toward state-of-the-art precision medicine in EC. The collection of primary tumors and derived PDOXs is characterized through an integrative approach based on transcriptomics, mutational profiles, and morphological analysis; and it is demonstrated that EC tumors engrafted in the mouse uterus retain the main molecular and morphological features from analogous tumor donors. Finally, the molecular properties of these tumors are harnessed to assess the therapeutic potential of trastuzumab, a human epidermal growth factor receptor 2 (HER2) inhibitor with growing interest in EC, using patient-derived organotypic multicellular tumor spheroids and in vivo experiments.

Collagenolytic matrix metalloproteinases antagonize proteinase-activated receptor-2 activation, providing insights into extracellular matrix turnover

The collagenase subfamily of matrix metalloproteinases (MMPs) have important roles in the remodeling of collagenous matrices. The proteinase-activated receptor (PAR) family has a unique mechanism of activation requiring proteolysis of an extracellular domain forming a neo-N terminus that acts as a tethered ligand, a process that has been associated with the development of arthritis. Canonical PAR2 activation typically occurs via a serine proteinase at Arg36-Ser37, but other proteinases can cleave PARs downstream of the tethered ligand and "disarm" the receptor. To identify additional cleavage sites within PAR2, we synthesized a 42-amino-acid peptide corresponding to the extracellular region. We observed that all three soluble MMP collagenases, MMP-1, MMP-8, and MMP-13, cleave PAR2 and discovered a novel cleavage site (Ser37-Leu38). Metalloproteinases from resorbing bovine nasal cartilage and recombinant human collagenases could cleave a quenched fluorescent peptide mimicking the canonical PAR2 activation region, and kinetic constants were determined. In PAR2-overexpressing SW1353 chondrocytes, we demonstrated that the activator peptide SLIGKV-NH2 induces rapid calcium flux, inflammatory gene expression (including MMP1 and MMP13), and the phosphorylation of extracellular signal-regulated kinase (ERK) and p38 kinase. The corresponding MMP cleavage-derived peptide (LIGKVD-NH2) exhibited no canonical activation; however, we observed phosphorylation of ERK, providing evidence of biased agonism. Importantly, we demonstrated that preincubation with active MMP-1 reduced downstream PAR2 activation by a canonical activator, matriptase, but not SLIGKV-NH2 These results support a role for collagenases as proteinases capable of disarming PAR2, revealing a mechanism that suppresses PAR2-mediated inflammatory responses.