The tumor microenvironment (TME) is the complex niche surrounding cancer cells, composed of various cell types and molecules that influence tumor growth and treatment responses. In the TME, a striking diversity of immune cells, fibroblasts, endothelial cells, and other tissue-specific cells interact with phenotypically diverse cancer cells and evolve with cancer progression. The TME components and their complex interaction network determine to a large extent disease progression and response to therapies. We use high-resolution spatial and single-cell technologies to study the biological variation that occurs within and between tumors, with the goal to identify general principles of immune system regulation. This approach has the potential to reveal the predominant immune regulation mechanisms underpinning cancer progression and therapy resistance, leading to novel therapeutic targets.

Therapies that modulate the immune system, such as immune checkpoint inhibitors, have revolutionized the landscape of cancer treatment. However, only a fraction of patients experience clinical benefit across cancer. Profiling the tumor immune microenvironment offers a great opportunity for the discovery of biomarkers of therapy response and of immune-related adverse events. However, tumor biopsy-based biomarkers face limitations due to tumor heterogeneity and accessibility of various metastatic sites. In contrast, peripheral blood samples present several advantages including enhanced accessibility, feasibility to implementation in clinical practice and possibility of multiple sampling for monitoring of dynamic changes. We study how the peripheral blood is altered by the tumor characteristics at the single-cell level and exploit this knowledge to develop machine learning models to predict disease state and response to therapy.

Single-cell transcriptomics and chromatin accessibility profiling allow us to unbiasedly explore cell diversity and to discover cell types, transient cell states and their characteristic gene programs. Robust, expert-curated single-cell maps that can be created with these data modalities have become essential resources as they serve as references to interpret cellular differences across perturbations, conditions, tissues and individuals. A single-cell map is a useful abstraction that allows modeling complex cell heterogeneity as discrete or partially-overlapping cell populations defined by characteristic gene programs activity and chromatin accessibility patterns – and potentially distinct functions. Combining our expertise in bioinformatics and immunology, we generate high-quality reference maps for immune cells that are used by the research community. These are available at our SPICA portal https://spica.unil.ch/.