Integrated analysis of transcriptome, methylome and copy number aberrations data of marginal zone lymphoma and follicular lymphoma in dog

StellarPath: Hierarchical-vertical multi-omics classifier synergizes stable markers and interpretable similarity networks for patient profiling

The Patient Similarity Network paradigm implies modeling the similarity between patients based on specific data. The similarity can summarize patients' relationships from high-dimensional data, such as biological omics. The end PSN can undergo un/supervised learning tasks while being strongly interpretable, tailored for precision medicine, and ready to be analyzed with graph-theory methods. However, these benefits are not guaranteed and depend on the granularity of the summarized data, the clarity of the similarity measure, the complexity of the network's topology, and the implemented methods for analysis. To date, no patient classifier fully leverages the paradigm's inherent benefits. PSNs remain complex, unexploited, and meaningless. We present StellarPath, a hierarchical-vertical patient classifier that leverages pathway analysis and patient similarity concepts to find meaningful features for both classes and individuals. StellarPath processes omics data, hierarchically integrates them into pathways, and uses a novel similarity to measure how patients' pathway activity is alike. It selects biologically relevant molecules, pathways, and networks, considering molecule stability and topology. A graph convolutional neural network then predicts unknown patients based on known cases. StellarPath excels in classification performances and computational resources across sixteen datasets. It demonstrates proficiency in inferring the class of new patients described in external independent studies, following its initial training and testing phases on a local dataset. It advances the PSN paradigm and provides new markers, insights, and tools for in-depth patient profiling.

Hypermethylation-Mediated Silencing of CIDEA, MAL and PCDH17 Tumour Suppressor Genes in Canine DLBCL: From Multi-Omics Analyses to Mechanistic Studies

Gene expression is controlled by epigenetic deregulation, a hallmark of cancer. The DNA methylome of canine diffuse large B-cell lymphoma (cDLBCL), the most frequent malignancy of B-lymphocytes in dog, has recently been investigated, suggesting that aberrant hypermethylation of CpG loci is associated with gene silencing. Here, we used a multi-omics approach (DNA methylome, transcriptome and copy number variations) combined with functional in vitro assays, to identify putative tumour suppressor genes subjected to DNA methylation in cDLBCL. Using four cDLBCL primary cell cultures and CLBL-1 cells, we found that CiDEA, MAL and PCDH17, which were significantly suppressed in DLBCL samples, were hypermethylated and also responsive (at the DNA, mRNA and protein level) to pharmacological unmasking with hypomethylating drugs and histone deacetylase inhibitors. The regulatory mechanism underneath the methylation-dependent inhibition of those target genes expression was then investigated through luciferase and in vitro methylation assays. In the most responsive CpG-rich regions, an in silico analysis allowed the prediction of putative transcription factor binding sites influenced by DNA methylation. Interestingly, regulatory elements for AP2, MZF1, NF-kB, PAX5 and SP1 were commonly identified in all three genes. This study provides a foundation for characterisation and experimental validation of novel epigenetically-dysregulated pathways in cDLBCL.

A bitesize introduction to canine hematologic malignancies

Hematologic malignancies are frequently diagnosed in dogs and result in a spectrum of clinical signs associated with specific disease types. The most frequently encountered hematologic tumors in dogs include lymphoma, lymphoid and myeloid leukemias, and mast cell, plasma cell, and histiocytic neoplasias. Coupled with the heterogeneous presentations of the different categories and subtypes of canine hematologic malignancies, outcomes for these tumors are also variable. Considering this, appropriate treatment options range from active surveillance to curative intent approaches harnessing surgical, chemotherapeutic, and radiation-based modalities. The underlying pathology of many of these diseases bears remarkable resemblance to that of the corresponding diagnosis made in human patients. We introduce some of the pathogenic drivers of canine hematologic cancers alongside their clinical presentations. An overview of standard-of-care therapies for each of these diseases is also provided. As comparative oncology gains recognition as a valuable setting in which to investigate the pathogenesis of neoplasia and provide powerful, clinically relevant, immunocompetent models for the evaluation of novel therapies, the number of clinicians and scientists participating in cancer research involving dogs is expected to increase. This review aims at providing an introductory overview of canine hematologic malignancies.

Genomic and Transcriptomic Characterization of Canine Osteosarcoma Cell Lines: A Valuable Resource in Translational Medicine

Osteosarcoma (OSA) represents the most common primary bone tumor in dogs and is characterized by a highly aggressive behavior. Cell lines represent one of the most suitable and reproducible pre-clinical models, and therefore the knowledge of their molecular landscape is mandatory to investigate oncogenic mechanisms and drug response. The present study aims at determining variants, putative driver genes, and gene expression aberrations by integrating whole-exome and RNA sequencing. For this purpose, eight canine OSA cell lines and one matched pair of primary tumor and normal tissue were analyzed. Overall, cell lines revealed a mean tumor mutational burden of 9.6 mutations/Mb (range 3.9–16.8). Several known oncogenes and tumor suppressor genes, such as ALK, MYC, and MET, were prioritized as having a likely role in canine OSA. Mutations in eight genes, previously described as human OSA drivers and including TP53, PTCH1, MED12, and PI3KCA, were retrieved in our cell lines. When variants were cross-referenced with human OSA driver mutations, the E273K mutation of TP53 was identified in the Wall cell line and tumor sample. The transcriptome profiling detected two possible p53 inactivation mechanisms in the Wall cell line on the one hand, and in D17 and D22 on the other. Moreover, MET overexpression, potentially leading to MAPK/ERK pathway activation, was observed in D17 and D22 cell lines. In conclusion, our data provide the molecular characterization of a large number of canine OSA cell lines, allowing future investigations on potential therapeutic targets and associated biomarkers. Notably, the Wall cell line represents a valuable model to empower prospective in vitro studies both in human and in dogs, since the TP53 driver mutation was maintained during cell line establishment and was widely reported as a mutation hotspot in several human cancers.

Phenotypical Characterization and Clinical Outcome of Canine Burkitt-Like Lymphoma

In dogs, Burkitt-like lymphoma (B-LL) is rare tumor and it is classified as a high-grade B-cell malignancy. The diagnosis is challenging because of the similar histologic appearance with other histotypes, no defined phenotypical criteria and poorly described clinical aspects. The aim of the study was to provide a detailed description of clinical and morphological features, as well as immunophenotypical profile of B-LL in comparison with the human counterpart. Thirteen dogs with histologically proven B-LL, for which a complete staging and follow-up were available, were retrospectively selected. Immunohistochemical expression of CD20, PAX5, CD3, CD10, BCL2, BCL6, MYC, and caspase-3 was evaluated. Histologically, all B-LLs showed a diffuse architecture with medium to large-sized cells, high mitotic rate and diffuse starry sky appearance. B-phenotype of neoplastic cells was confirmed both by flow-cytometry and immunohistochemistry. Conversely, B-LLs were negative for BCL2 and MYC, whereas some cases co-expressed BCL6 and CD10, suggesting a germinal center B-cell origin. Disease stage was advanced in the majority of cases. All dogs received CHOP-based chemotherapy with or without immunotherapy. Despite treatment, prognosis was poor, with a median time to progression and survival of 130 and 228 days, respectively. Nevertheless, ~30% of dogs survived more than 1 year. An increased apoptotic index, a high turnover index and caspase-3 index correlated with shorter survival. In conclusion, canine B-LL shows phenotypical differences with the human counterpart along with features that might help to differentiate this entity from diffuse large B-cell lymphoma.

Machine learning application identifies novel gene signatures from transcriptomic data of spontaneous canine hemangiosarcoma

Angiosarcomas are soft-tissue sarcomas that form malignant vascular tissues. Angiosarcomas are very rare, and due to their aggressive behavior and high metastatic propensity, they have poor clinical outcomes. Hemangiosarcomas commonly occur in domestic dogs, and share pathological and clinical features with human angiosarcomas. Typical pathognomonic features of this tumor are irregular vascular channels that are filled with blood and are lined by a mixture of malignant and nonmalignant endothelial cells. The current gold standard is the histological diagnosis of angiosarcoma; however, microscopic evaluation may be complicated, particularly when tumor cells are undetectable due to the presence of excessive amounts of nontumor cells or when tissue specimens have insufficient tumor content. In this study, we implemented machine learning applications from next-generation transcriptomic data of canine hemangiosarcoma tumor samples (n = 76) and nonmalignant tissues (n = 10) to evaluate their training performance for diagnostic utility. The 10-fold cross-validation test and multiple feature selection methods were applied. We found that extra trees and random forest learning models were the best classifiers for hemangiosarcoma in our testing datasets. We also identified novel gene signatures using the mutual information and Monte Carlo feature selection method. The extra trees model revealed high classification accuracy for hemangiosarcoma in validation sets. We demonstrate that high-throughput sequencing data of canine hemangiosarcoma are trainable for machine learning applications. Furthermore, our approach enables us to identify novel gene signatures as reliable determinants of hemangiosarcoma, providing significant insights into the development of potential applications for this vascular malignancy.