For many years, single-cell sequencing has been used to tackle tumour heterogeneity because this technology enables scientists to look at individual cells, DNA, and RNA molecules. Miao-Ping Chien, Principal Investigator & Assistant Professor, Erasmus University Medical Center, Oncode Institute, works on developing and applying multidisciplinary technologies to investigate the mechanisms of tumorigenesis in rare cancer-driving cells.
To kick things off, Chien began discussing a technique developed in her group called Microscopy-Based Functional Single Cell Sequencing. The hope is to adopt a single-cell multiomics approach to study rare cancer cells and use computational analysis to investigate their underlying mechanics to identify some therapeutic targets that could be translated back to patients later down the line.
Chien explained that the current limitation of existing single-cell technology is that it cannot directly link genotypes with tumorigenic phenotypes, making it complicated to study rare and aggressive cancer cells. Sometimes clustering analysis on these aggressive cells does not work because the cells are so rare and share a very similar profile to non-aggressive cancer cells so the clusters may not be related to the tumourigenic phenotypes.
If one cannot isolate the cells accordingly, it is hard to discern underlying driving mechanisms. To tackle this, microscopy-based functional single-cell sequencing sequences individual cells based on functional features observed under a microscope. FUNseq puts this principle into practice. In short, FUNseq screens, photo-labels, and sequences cells based on their functional features, thus allowing for the identification of metastasis-related pathways in aggressive cancer cells.
Chien briefly commented that in order to take this technology to the next level, FUNseq has been combined with proteomics to create FUNpro. Next, she presented a proof of concept study showing how FUNseq identifies known metastasis related to pathways in aggressive breast cancer cells. The tool uncovered a couple of pathways and several epithelial to mesenchymal genes that are known to be related to tumour aggressiveness.
Alongside this, Chien and her team discovered a lipid metabolism pathway upregulated in glioblastoma and head and neck squamous cell carcinoma. She said this was validated through inhibitor treatment and siRNA knockdown. Chien developed predictive gene signatures for metastasis using supervised machine learning on FunSeq data.
Spurred on by this success, the FUNseq technology is being extended to spatial profiling. This enables the study of tumour microenvironments by labelling and profiling cells based on their spatial locations within tissues. The extension profiling was applied to tumour immunology: isolated CD4+ T cells from clustered vs. non-clustered regions were used to study immune-tumour interactions. IGF signalling was identified as a potential driver.
