Association of transitional gradient from clear cell to sarcomatoid renal cell carcinoma with macrophage/tumor cell crosstalk.
Allison May,Claire Williams,Stephanie The,Jake McGue,Matthew Wainstein,Greg Shelley,Tyler Robinson,Simpa Samuel Salami,Rohit Mehra,Timothy Frankel,Aaron M. Udager,Evan T. Keller
DOI: https://doi.org/10.1200/jco.2024.42.4_suppl.480
IF: 45.3
2024-02-01
Journal of Clinical Oncology
Abstract:480 Background: Sarcomatoid de-differentiation in renal cell carcinoma (sRCC) leads to aggressive tumors that are uniquely responsive to adjuvant immunotherapy. sRCC is thought to arise by an epithelial to mesenchymal transition (EMT) of the parental tumor, most commonly clear cell RCC (ccRCC). However, factors that drive sRCC are unknown and no biomarkers exist to predict the transition. Furthermore, the immune microenvironment in sRCC is not well understood and may provide additional therapeutic targets. Here, we use spatial biology and in vitro studies to explore EMT/immune cell crosstalk in sRCC. Methods: Single cell spatial transcriptomics was performed on a human sRCC specimen via the NanoString CosMx platform and semi-supervised clustering referenced to scRNAseq data was used to identify a spatial transition from ccRCC to sRCC. Segmented bulk spatial transcriptomics was performed on 4 sRCC specimens, mapping the CosMx cell signatures. Multiplex immunofluorescent staining (mIF) was performed on 29 sRCC specimens. In vitro work was performed using 3 RCC cell lines and THP-1 derived macrophages to explore mechanisms between cell types and genes of interest. Results: Single cell spatial transcriptomic data revealed a ccRCC population, 2 sRCC cell types, and a novel transitional cell type along the EMT continuum and spatially between ccRCC and sRCC. These signatures were mapped onto segmented bulk spatial transcriptomic data confirming the same spatial transition pattern in 4 sRCC specimens. Importantly, matched H&E shows the transitional cell type present in areas histologically defined as ccRCC, demonstrating molecular evidence of transition towards sRCC prior to histologic changes. Cell to cell distance analysis and mIF in 29 sRCC specimens revealed a strong correlation of M2 macrophages with the transition and sRCC cell types. CCL20, a factor known to assist in macrophage recruitment was upregulated on transition cells and in vitro led to polarization to an M2-like phenotype. FZD4 was the highest fold changed gene lost from the ccRCC to transitional cells and knockdown of FZD4 in vitro led to EMT in RCC cells. In vitro culture of RCC cells with M2 macrophages led to CCL20 upregulation and loss of FZD4. The TCGA KIRC database was used to validate the ability to identify the transition signature in ccRCC and its association with poorer disease specific survival. Conclusions: We report the first detection of a transitional cell type along the de-differentiation pathway from ccRCC to sRCC. This cell type is detectable in areas histologically defined as ccRCC, demonstrating strong potential as a transcriptional biomarker to inform adjuvant immunotherapy. We show that CCL20 on tumor cells leads to macrophage recruitment and polarization, resulting in FZD4 loss and CCL20 upregulation, which induce EMT and propagates a vicious cycle of EMT/macrophage crosstalk in sRCC.
oncology