Use of single-cell RNA sequencing and comparative immuno-oncology to gain insights into spontaneous canine cancers
Date
2023
Journal Title
Journal ISSN
Volume Title
Abstract
Advances in human clinical medicine stem from discoveries and reports in model systems, therefore the use of biologically relevant models in essential for developing effective human therapeutics. Traditionally, small mammals, such as mice and rats, have been used to address basic science questions and they have contributed substantially to our understanding of biology. Despite widespread use and accessibility of rodent models, there is a growing awareness that findings in rodents frequently fail to translate to human medicine. In recent years, pet dogs have been proposed as an ideal model system to facilitate translational research. As such, the overarching themes of this dissertation are to (1) build upon the dog as a model by providing novel cell type transcriptomic references for immuno-oncology research and (2) investigate immunological correlates with treatment responses in clinical trials using dogs with spontaneously arising tumors. First, the introductory chapter discusses the dog as a model for human disease with a focus on the application in glioma and osteosarcoma (OS). The biological and molecular features of each tumor type are described, then current therapeutic approaches in dogs and human are discussed. After introducing the tumor types, two cell types, myeloid-derived suppressor cells (MDSCs) and tumor-associated macrophages (TAMs), are discussed in detail as they are key cell types throughout the dissertation. In the final section of the introduction, single-cell RNA (scRNA) sequencing, the technology foundational to the work presented here, is discussed in detail. In chapters 2 through 5 we focus on OS, a malignant tumor of the bone with minimal therapeutic options. In chapter 2 we generated a reference scRNA dataset of canine circulating leukocytes, then applied the dataset to investigate how the presence of a primary OS tumor impacts systemic immune cell transcriptomes. Through evaluation of 74,067 cells from 17 dogs (7 healthy, 10 OS) we identified relative increases in the abundances of polymorphonuclear (PMN-) and monocytic (M-) MDSCs and provided their transcriptomic signatures for further study. The reference aspect of the work constituents a comprehensive database with gene signatures for each of the 36 cell types identified in canine blood. This work provides key insights into OS induced changes to circulating immune cells while also providing a broadly applicable reference that can be applied to many different areas of canine research. In chapter 3 we generate another comprehensive database, this time focusing on characterizing the heterogeneity within canine OS tumors. Through analysis of 35,310 cells we identified exhausted T cells, mature regulatory dendritic cells (mregDCs), and 8 transcriptomically distinct macrophage/monocyte populations and provide their transcriptomic signatures. We used cell-cell interaction inference approaches to investigate active immune suppressive pathways in OS and found TAMs and mregDCs to be major contributors to T cell suppression. Lastly, we obtained an external human OS scRNA dataset to evaluate cell type homologies between dogs and human which suggested a high degree of similarities between the species. We hope the data generated in this chapter can be applied to enhance canine OS research and shed light on conserved immune suppressive pathways in OS. In chapter 4 we apply the datasets generated in chapters 2 and 3 to investigate how the tumor microenvironment (TME) impacts the transcriptional programs of infiltrating immune cells. To complete the analysis, we used data from circulating leukocytes of the 10 OS dogs in chapter 2 and the OS tumor-infiltrating immune cells identified in chapter 3. Through direct comparison of infiltrating and circulating immune cells we were able to confirm several tumor-induced changes reported in humans are also apparent in the dog. Key confirmatory findings in infiltrating immune cells included the upregulation of activation markers on T cells, increased relative abundance in exhausted T cells, and increased expression of immune suppressive molecules on myeloid cells. Overall, the analysis suggests overarching tumor-induced immunological changes are conserved between human and dogs. In chapter 5 we apply scRNA sequencing to investigate how a myeloid targeted combination therapeutic (losartan, ladarixin, and toceranib) impacts intratumoral and systemic immune responses. Analysis revealed broad immune cell depletion in the tumor and increases in circulating M-MDSCs in dogs receiving treatment. We identified modulation to multiple chemokine signaling axes which shed light on mechanisms associated with treatment-induced immune cell depletion. Finally, the analysis revealed profound impacts to tumor cells and fibroblasts, with treatment skewing transcriptomic profiles toward a hypoxic phenotype and increased insulin-like growth factor associated gene expression. Ultimately, this study represents the first insights into how any therapeutic modulates the OS tumor microenvironment at the single-cell level. Finally, in chapter 6 we conducted a canine glioma clinical trial to investigate the utility of another myeloid targeted therapy (vaccination, losartan, and propranolol). We observed treatment to induce partial tumor regression in 2 and stable disease in 6 of 10 dogs, for an overall clinical benefit rate of 80%. Through evaluation of antibody responses to vaccination, we identified a subset of patients to be immunological responders, which we found exhibited enhanced overall survival times relative to dogs that did not generate antibody responses. The findings from the clinical study suggest that myeloid targeted therapy for treatment of glioma may be a valuable approach that warrants further investigation in canine and human glioma patients. In conclusion, our work applying single-cell RNA sequencing resulted in the generation of valuable canine-specific cell type reference datasets and revealed key insights in osteosarcoma immunobiology. The work evaluating myeloid therapeutics in the setting of osteosarcoma and glioma provide mechanistic and clinical insight that can be applied to further study of the therapeutic approach. Overall, we hope the body of work presented here strengthens the foundation of the dog as a model for translational biomedical research.
Description
Zip file contains supplementary Power Points and data CSVs.
Rights Access
Embargo expires: 09/10/2025.
Subject
glioma
osteosarcoma
canine
single-cell RNA
immunotherapy