Vincent DETOURS, PhD
The computational biology group at IRIBHM focuses on the transcriptomes of, mostly but not exclusively, breast and thyroid cancers. Two major topics are currently investigated.
A-to-I editing by the enzyme ADAR substitutes inosines for adenosines at specific positions in double stranded RNAs (dsRNA) and can very substantially alter a cell’s transcriptome and function. While investigation of RNA editing in cancer is still in its infancy, recent studies point to an important role in cancer. We have shown that, overall, A-to-I editing is a pervasive, yet reproducible, source of variation that is globally controlled by amplification of chromosome and inflammation, both of which are highly prevalent among human cancers. Importantly, ADAR expression also modulates the proliferation and apoptosis of cancer cells. We currently investigate RNA editing at the level of noncoding RNA, its intratumoral heterogeneity and its relation with inflammation in cancer.
Textbooks suggest that cancers are compact balls with an inner core and an invasive front in contact with non-cancerous cells, and that tumor expansion is driven by tumor cells proliferation subsequently to oncogenic mutations in their genomes. We recently reconstructed at histological scale the 3D volume occupied by tumor cells in a BRAFV600E-mutated thyroid cancer with low tumor purity, as determined by sequencing, but initially considered high purity during pathology review.
In contrast with a compact ball, tumor cells form a sparse mesh deeply embedded within the stroma. The concepts of inner core and invasive front brakes down in this morphology: all tumor cells were within short distance from the stroma. The fibrous stroma was highly cellular and proliferative. This case is not unique: 3.5% of thyroid cancers have purities <25%. Moreover, a substantial fraction of BRAFV600E tumors are associated with extensive fibrosis, high stromal activation, and dedifferentiation.
We are currently in the process of developing this line of research about intra tumoral heterogeneity with single cell RNA-seq and by running spatial transcriptomics experiments, which makes it possible to resolve the transcriptome at thousands of locations along the three spatial dimensions of tumor blocks.