Resistance of cancers to immune checkpoint blockade (ICB) can result from a deviated taxonomic composition of the intestinal microbiota. A surge in the Enterocloster genus, for instance following discontinuation of antibiotics or chronic inflammation caused by tumors, induces the downregulation of MAdCAM-1 in the ileal lamina propria and mesenteric lymph nodes through perturbations of biliary salts. In turn, the ileal MAdCAM-1 loss induces the exodus of immunosuppressive T lymphocytes expressing the MAdCAM-1 receptor α4β7, i.e., FoxP3+ RAR-related orphan receptor gamma t (RORγt+) regulatory (Tr17) cells, from the gut to distant tumors. Disruption of the MAdCAM-1–α4β7 axis compromises the efficacy of immunotherapy and reprograms the tumor microenvironment towards a regulatory phenotype. Moreover, serum soluble MAdCAM-1 is a proxy of intestinal dysbiosis and a robust predictor of survival in cancer patients treated with ICB. To decipher the biological significance of these findings, we will first investigate the transcriptional and post-translational mechanisms regulating MAdCAM-1 expression and function (Task 1), in particular neuroendocrine and metabolic cues. Next, a comprehensive phenotyping of the emigrating enterotropic T cells reaching the tumor by single cell transcriptomics and genomics, as well as specific genetic and immunopharmacological intervention on T lymphocytes and cancer cells will lead to the identification of the molecular mechanisms regulating the gut-tumor axis (Task 2). We will investigate how enterotropic T cells homing to cancers maintain their proliferative and suppressive capacities, in particular in the light of the recognition of tumor and/or commensal antigens (Task 3). Altogether, this approach will lay the molecular and metabolic foundations governing the MAdCAM-1–α4β7 gut immune checkpoint

Anaplastic thyroid carcinoma (ATC) is a rare and very aggressive disease. Even with multimodality treatment, the prognosis remains very poor with overall survival between 6 and 12 months. Following seminal work from our team, we hypothesize that ATC aggressiveness and resistance to treatments are related to highly immunosuppressive tumoral microenvironment consisting of immunosuppressive myeloid cells and, in particular, tumor-associated macrophages (TAMs). The goal of the PRIMAThyC project is to investigate the profile of circulating and tumor-associated immune cells in the blood and tumor samples of patients with ATC. The expected output of the study is the identification of disease markers that can be used to improve patient’s management and survival, to suggest pathophysiological hypotheses and to identify putative therapeutic targets in order to design new therapeutic strategies. To achieve this goal, the project is composed of 3 tasks: - Mapping: to describe the circulating and tumoral immune landscape of ATC by hogh throuput techniques - Referencing: to identify biomarkers of prognostic and/or therapeutic importance by searching for correlations between candidate biomarkers/immune cell populations and mutational status, pathological characteristics, response to treatments, and overall survival - Validation/modeling: to validate the relevance of the identified biomarkers in vitro and in vivo and by immunohistochemistry on archival tumoral samples. The PRIMAThyC project will deliver an integrative immunological and tumoral map of anaplastic thyroid carcinoma and has the power to identify biomarkers of predictive and/or prognostic interest and new therapeutic targets to design new effective treatment strategies for this very rare deadly disease.

The transcription factor AML1 (RUNX1) is the alpha subunit of the core binding factor (CBF) and one of the most frequent targets of chromosomal and genetic alteration in leukemias. AML1 contains a RUNT homology domain (RHD) that mediates DNA binding and heterodimerization with CBFß, and a C-terminal domain responsible for transcriptional activation. AML1 is essential for establishment of definitive hematopoiesis, has a dose dependent role in regulating the quantity of HSC and of their downstream committed progenitors, as well as a more restricted role on T cell development and platelet formation. To explore the role of AML1 mutations in the leukemic development and thrombopenia induction, a genetic hereditary pathology: Familial platelet disorder with a predisposition to acute myeloid leukaemia (FPD/AML) represents an important model. FPD/AML is an inherited autosomal dominant platelet disorder. The candidate gene is AML1 because different types of mutations were always found at the heterozygous state in 27 pedigrees. The incidence of leukemia among FPD/AML affected individuals varies from 20% to 50%. Mutations implicate always the Runt domain and lead either to the absence or decrease in binding of mutant protein to DNA but: 1 - some mutations generate proteins that retain the capacity to heterodimerize with CBFß and inhibit wt AML1 transcription activity. In these cases, the mutant may behave as a dominant-negative (DN) protein. 2 – other mutations generate AML1 haploinsufficiency. The studies performed in very few patients suggest that leukemia incidence appears to be lower in pedigrees with haploinsufficiency. The exact mechanisms by which only some genetic alterations in AML1 predispose to acute leukemia remain unknown. Our hypothesis based on our preliminary results is that dominant negative mutations in AML1 induce a pre-leukemic state by deregulating hematopoietic stem cells and progenitors and increasing the compartment of target cells (GM-progenitors) available for secondary mutations leading to full blown leukemia whereas haplo-insufficiency does not significantly deregulate hematopoiesis and therefore does not predispose to leukemia (as suggested by mouse models). The type of mutation should thus predict the risk of AML transformation. We proposed to perform a detailed in vitro study of hematopoiesis (multipotent and committed progenitors) in individuals belonging to five FPD/AML pedigrees with different types of mutations (informed consent obtained) and to correlate AML1 abnormalities with abnormal regulation of AML1 primary target genes. The transcriptome will be studied in patient hematopoietic progenitors and identification of AML1 target genes will be facilitated by AML1 knockdown and by identification of promoters binding AML1 (by ChIP-seq method) in control hematopoietic progenitors. This complex approach has already been started and our preliminary results (identification of NR4A3, NOV, p19 and ZBTB1) confirm that it is feasible. The AML1 target genes and their implication in pathology will be validated in two different types of model: murine models (conditional AML1 KO, KO of candidate gene) and a human model (pluripotent stem cells that we are actually inducting from patient skin biopsies). Contrary to the leukemic predisposition, thrombopenia, even moderate, is present in all pedigrees regardless of the mutation type suggesting a dose depending regulation of megakaryopoiesis and platelet generation. Identification of anomalies of patient megakaryopoiesis as well as identification and validation of AML1 target genes in megakaryocytes are also an important objective of these grant and should be helpful in a better understanding of molecular mechanisms regulating platelet production.

Recent large-scale genomic profiling studies have uncovered mutations in the SWI/SNF (SWItch / Sucrose Non-Fermentable) chromatin remodelling complex subunits in 20% of solid tumours. Most of these tumours resist to current therapies. To address this highly unmet medical need, it is crucial to determine how to harness vulnerabilities induced by SWI/SNF defects. Recent evidence shows that SWI/SNF influences the DNA damage response and is involved in shaping tumour immunogenicity. Though, whether there is a link between these observations, and whether the latter may be therapeutically exploited is unknown. Further, the impact of SWI/SNF defects on tumour heterogeneity, a major determinant of resistance to treatment, remains unaddressed. I therefore propose to identify and understand synthetic lethal vulnerabilities associated with two selected SWI/SNF defects of unmet need (PBRM1 and SMARCB1), and to further study how the latter can be exploited to stimulate the anti-tumour immune response. By using hypothesis-testing and -generating approaches including high-throughput screening on in-house developed isogenic and non-isogenic models, I will identify and decipher novel synthetic lethal vulnerabilities associated with PBRM1 and SMARCB1 defects. Molecular biology, high-content imaging and in vivo experiments will be performed to study the effects of drugs that cause synthetic lethality, both on intra-cellular signalling and on anti-tumour immune response. I will further characterise the impact of SMARCB1 defects on tumour heterogeneity using single cell sequencing on preclinical models and human tumour samples. Preclinical data will be integrated with tumour profiling and clinical data, and will guide the development of proof-of-concept clinical studies, as I previously did. The overall research program will identify, decipher mechanistically and evaluate clinically novel immuno-oncology therapeutic strategies for selected SWI/SNF-deficient tumours of unmet need.