Inter and intra-tumor molecular heterogeneity assessment for clinical applications: illustration in several cancer types

The understanding of molecular heterogeneity in cancer is a crucial component of precision medicine. In particular, tumor subtypes with different underlying oncogenic mechanisms could benefit from personalized care with specific targeted therapies.
Tumor pathologists classify tumors according to cell and tissue level criteria, as reported in the WHO tumor classifications. In addition, molecular subtyping based on large-scale omics profiling has been progressively integrated into this work. Despite obvious advances in tumor pathology, some cancers are still poorly characterized and a precise description of intra-tumor heterogeneity remains lacking.
I will present our recent findings on two aggressive and insufficiently characterized cancers: pancreatic adenocarcinoma and pleural mesothelioma. By unsupervised clustering approaches on multi-omics data from pancreatic tumors, we identified different clinically-relevant tumor subtypes. We further showed how data integration could pinpoints novel therapeutic targets. In a second study focusing on mesothelioma, we illustrated the importance of taking into account intra-tumor heterogeneity. Using a deconvolution approach, we showed that the cellular components describing tumors could fairly well recapitulate existing molecular classifications. More importantly the proportion of a specific tumor component was strongly associated to prognosis and drug response as immunotherapies and targeted therapies.
Overall, we demonstrate the relevance of data integration and signal deconvolution approaches to better understand cancer heterogeneity and enable personalized therapeutic strategies.

Yuna BLUM, CIT (Cartes d'Identité des Tumeurs) 

Invited by Marie-Dominique GALIBERT

>>Friday 11 October 2019 at 11:00 - IGDR conference room, ground floor, building 4 / Villejean Campus

Subscribe to our newsletter