Unclassified sarcomas: a study to improve classification in a cohort of Golden Retriever dogs

Integrated analysis of canine soft tissue sarcomas identifies recurrent mutations in TP53, KMT genes and PDGFB fusions

Soft tissue sarcomas (STS) are a heterogenous group of mesenchymal tumors representing over 50 distinct types with overlapping histological features and non-specific anatomical locations. Currently, localized sarcomas are treated with surgery + / -  radiation in both humans and dogs with few molecularly targeted therapeutic options. However, to improve precision-based cancer therapy through trials in pet dogs with naturally occurring STS tumors, knowledge of genomic profiling and molecular drivers in both species is essential. To this purpose, we sought to characterize the transcriptomic and genomic mutation profiles of canine STS subtypes (fibrosarcoma, undifferentiated pleomorphic sarcoma, and peripheral nerve sheath tumors), by leveraging RNAseq, whole exome sequencing, immunohistochemistry, and drug assays. The most common driver mutations were in cell cycle/DNA repair (31%, TP53-21%) and chromatin organization/binding (41%, KMT2D-21%) genes. Similar to a subset of human sarcomas, we identified fusion transcripts of platelet derived growth factor B and collagen genes that predict sensitivity to PDGFR inhibitors. Transcriptomic profiling grouped these canine STS tumors into 4 clusters, one PNST group (H1), and 3 FSA groups selectively enriched for extracellular matrix interactions and PDFGB fusions (H2), homeobox transcription factors (H3), and elevated T-cell infiltration (H4). This multi-omics approach provides insights into canine STS sub-types at a molecular level for comparison to their human counterparts, to improve diagnosis, and may provide additional targets for chemo- and immuno-therapy.

Integrated analysis of canine soft tissue sarcomas identifies recurrent mutations in TP53, KMT genes and PDGFB fusions

Canine soft tissue sarcomas (STS) are a heterogenous group of malignant tumors arising from mesenchymal cells of soft tissues. This simplified collective of tumors most commonly arise from subcutaneous tissues, are treated similar clinically, and conventionally exclude other sarcomas with more definitive anatomical, histological, or biological features. Histologically, canine STS sub-types are difficult to discern at the light microscopic level due to their overlapping features. Thus, genomic, and transcriptomic profiling of canine STS may prove valuable in differentiating the diverse sub-types of mesenchymal neoplasms within this group. To this purpose we sought to characterize the transcript expression and genomic mutation profiles of canine STS. To delineate transcriptomic sub-types, hierarchical clustering was used to identify 4 groups with district expression profiles. Using the RNAseq data, we identified three samples carrying driver fusions of platelet derived growth factor B ( PDGFB ) and collagen genes. Sensitivity to imatinib was evaluated in a canine STS cell line also bearing a PDGFB fusion. Using whole exome sequencing, recurrent driver variants were identified in the cancer genes KMT2D (21% of the samples) and TP53 (21%) along with copy number losses of RB1 and CDKN2A. Gene amplifications and resulting transcript increases were identified in genes on chromosomes 13, 14, and 36. A subset of STS was identified with high T-cell infiltration. This multi-omics approach has defined canine STS sub-types at a molecular level for comparison to their human counterparts, to improve diagnosis, and may provide additional targets for therapy.

Comparative oncology reveals DNMT3B as a molecular vulnerability in undifferentiated pleomorphic sarcoma

PURPOSE

Undifferentiated pleomorphic sarcoma (UPS), an aggressive subtype of soft-tissue sarcoma (STS), is exceedingly rare in humans and lacks effective, well-tolerated therapies. In contrast, STS are relatively common in canine companion animals. Thus, incorporation of veterinary patients into studies of UPS offers an exciting opportunity to develop novel therapeutic strategies for this rare human disease. Genome-wide studies have demonstrated that UPS is characterized by aberrant patterns of DNA methylation. However, the mechanisms and impact of this epigenetic modification on UPS biology and clinical behavior are poorly understood.

METHODS

DNA methylation in mammalian cells is catalyzed by the canonical DNA methyltransferases DNMT1, DNMT3A and DNMT3B. Therefore, we leveraged cell lines and tissue specimens from human and canine patients, together with an orthotopic murine model, to probe the functional and clinical significance of DNMTs in UPS.

RESULTS

We found that the DNA methyltransferase DNMT3B is overexpressed in UPS relative to normal mesenchymal tissues and is associated with a poor prognosis. Consistent with these findings, genetic DNMT3B depletion strongly inhibited UPS cell proliferation and tumor progression. However, existing hypomethylating agents, including the clinically approved drug 5-aza-2'-deoxycytidine (DAC) and the DNMT3B-inhibiting tool compound nanaomycin A, were ineffective in UPS due to cellular uptake and toxicity issues.

CONCLUSIONS

DNMT3B represents a promising molecular susceptibility in UPS, but further development of DNMT3B-targeting strategies for these patients is required.

Comparative whole transcriptome analysis of gene expression in three canine soft tissue sarcoma types

Soft tissue sarcomas are pleiotropic tumors of mesenchymal cell origin. These tumors are rare in humans but common in veterinary practice, where they comprise up to 15% of canine skin and subcutaneous cancers. Because they present similar morphologies, primary sites, and growth characteristics, they are treated similarly, generally by surgical resection followed by radiation therapy. Previous studies have examined a variety of genetic changes as potential drivers of tumorigenesis and progression in soft tissue sarcomas as well as their use as markers for soft tissue sarcoma subtypes. However, few studies employing next generation sequencing approaches have been published. Here, we have examined gene expression patterns in canine soft tissue sarcomas using RNA-seq analysis of samples obtained from archived formalin-fixed and paraffin-embedded tumors. We provide a computational framework for using resulting data to categorize tumors, perform cross species comparisons and identify genetic changes associated with tumorigenesis. Functional overrepresentation analysis of differentially expressed genes further implicate both common and tumor-type specific transcription factors as potential mediators of tumorigenesis and aggression. Implications for tumor-type specific therapies are discussed. Our results illustrate the potential utility of this approach for the discovery of new therapeutic approaches to the management of canine soft tissue sarcomas and support the view that both common and tumor-type specific mechanisms drive the development of these tumors.