Â鶹ӳ»­

The World Health Organization splits sarcomas into over fifty unique subtypes based on their significant genetic, phenotypic and lineage-specific diversity. Many sarcomas harbor specific chromosomal translocations, oncogenes or lost tumor suppressors used as diagnostic markers and potential therapeutic targets. Although genetically heterogeneous, sarcomas are often classified based on their apparent differentiation status and cell types within the adult mesenchymal lineage that they most resemble. Importantly, the extent to which each differentiation state is susceptible to genetic perturbation in sarcomas is largely unknown. Additionally, since the differentiation state of individual cells has only recently become detectable with the advent of powerful single-cell transcriptomics and novel computational algorithms, the scientific community has just begun to answer these questions.

We developed a high-resolution scRNA-seq reference map composed of osteogenic, adipogenic and chondrogenic lineages from human primary mesenchymal stem cells, described herein as the Mesenchymal Tissue Landscape (MTL). We hypothesized that the MTL would catalog the various differentiation states possible in osteosarcoma. To quantify differentiation states in the MTL, we applied a Normalized Nonnegative Matrix Factorization (N-NMF) based archetype analysis to identify recurring gene expression profiles that accurately captured lineage-specific temporal dynamics.

Ewing's sarcoma (ES), like several other translocation-positive sarcoma subtypes, exists exclusively in a high-grade, undifferentiated state unable to undergo mesenchymal lineage commitment. Recent preclinical work has shown that the pathognomonic EWS-FLI1 fusion protein (FP) contributes to this blockade, and that EWS-FLI1 silencing allows ES cells to partially regain the capacity for neural/mesenchymal lineage commitment in certain conditions after transiently entering a metastable pluripotent stem-cell-like state. How this occurs, however, remains an enigma that, if solved, could lead to the advent of an entire class of druggable targets selected for their ability to promote ES differentiation and tumor cell death.

Recent data raise concerns that EWS-FLI1 may have dimorphic clinical effects closely tied to expression levels. High EWS-FLI1 levels block mesenchymal lineage commitment and promote unconstrained cell proliferation and tumor growth, whereas EWS-FLI1 silencing ¡ª at least in the preclinical setting ¡ª elicits harmful, metastasis-prone cells bearing stem cell features. Though the latter point implies a clinical risk for FP suppression, that¡¯s true only if the reprogrammed stem cells end their journey at that pluripotent state. Given this body of research, we aim to determine how the ES FP locks tumor cells in an undifferentiated state by disrupting key gene regulatory networks (GRNs) needed for mesenchymal cell differentiation. By systematically probing critical genes within GRNs responsible for mesenchymal differentiation and neurogenesis using pooled single-cell CRISPR interference/activation (scCRISPRi/a), we expect to identify targetable genes, proteins or pathways that allow ES cells to differentiate.

We showed that desmoplastic small round cell tumors (DSRCTs) ¡ª rare, incurable abdominal tumors that present most often in young males ¡ª depend upon the androgen receptor (AR) for sustained growth and survival (). Beyond demonstrating a connection between the AR and DSRCTs, our team revealed key, potentially targetable epigenetic changes that facilitate tumor behavior.

Our current work takes the next logical step by investigating a new finding that DSRCTs, like prostate cancer, are capable of epithelial-to-neuroendocrine programming to resist androgen deprivation therapies (ADT) such as enzalutamide. As our group is the first to describe neuroendocrine reprogramming in DSRCT cells, it remains to be determined how this phenomenon occurs, since most patients wouldn¡¯t have received ADT as standard-of-care treatment. Furthermore, at the basic science level, it is undetermined if DSRCT cells undergo

1. Dedifferentiation towards a more stem-like cell before undergoing neural lineage commitment; or 
2. Transdifferentiation directly from an epithelial cell type towards a neuroendocrine cell fate

This distinction matters because the potential therapies that delay or reverse dedifferentiation and subsequent re-differentiation towards a neural cell fate are likely to differ from those used to prevent transdifferentiation.

Multiplex immunofluorescence (mIF) is a maturing platform for identifying biomarkers and studying spatial biology that is reliable and reproducible. We are using mIF to characterize the tumor microenvironment in various sarcomas like ES and DSRCT. The Lunaphore COMET? is a quantitative mIF tool capable of imaging with 40-plex and can image slides from archival formal-fixed paraffin-embedded (FFPE) material. The equipment is located at MD Anderson, and our imaging core has validated more than 165 proteins, with several articles published that establish the feasibility of this technology.