Unraveling the spatial and longitudinal heterogeneity of glioblastoma tumors and its relationship with treatment response and patient survival

In this project, we aim to unravel the inter- and intra-tumoral heterogeneity of
immune, stromal and malignant cells within distinct spatially-defined human
GBM tumor regions. We will rely on a cutting-edge multimodal approach that
will combine in vivo imaging and neuronavigation guidance surgery with singlecell
multi-omic analysis and high-parametric spatial profiling. Using Magnetic
Resonance Imaging, Positron Emission Tomography and machine learning
algorithms, GBM tumors will be subdivided into specific regions of interest,
which will then be selectively biopsied during surgery. These individual regions
will undergo single-cell sequencing, profiling both RNA and protein expression
using the CITE-Seq approach. CITE-seq will identify the subsets and cell states
within tumors and predict important pathways and cell-cell interactions. Based
on this information, we will design high-parametric panels for mapping the
spatial distribution of the key populations of interest within the intact tumor
tissue using ultra-high-plex spatial phenotyping at single-cell resolution. By
applying the same selective tissue sampling and analysis upon tumor recurrence
in the same patients, we intend to obtain longitudinal information on tumor
heterogeneity. This will show how cell states and interactions evolve over time,
as tumors relapse following standard GBM treatment. The registration of clinical
information will also allow us to correlate the obtained results to disease
progression and treatment responsiveness. By integrating the various datasets
using an extensive bio-informatics pipeline, we will build a comprehensive atlas
at unprecedented scale, capturing cellular, spatial and longitudinal
heterogeneity, across macroscopically-defined tumor regions. This will reveal
how tumor heterogeneity is correlated to intrinsic features of the tumor
microenvironment. Importantly, it will allow us to deconstruct the cellular
interactome that exists in GBM tumors across distinct tumor niches. We
particularly aim to unravel the bidirectional interplay that exists between
tumor-associated macrophages and malignant GBM cells. Combined, these
results will predict the ligands and signaling pathways that are driving key tumor
promoting processes, ultimately leading to the identification of new therapeutic
targets. This data-driven target identification will form an invaluable framework
for follow-up studies that will translate these insights into new therapeutic
breakthroughs for GBM patients.