Loss of p53 dictates Wnt-dependent systemic inflammation in breast cancer

Cancer-associated inflammatory processes in the tumour microenvironment, as well as systemically, are strongly linked with poor disease outcome in cancer patients. For most human solid tumour types, high systemic neutrophil-to-lymphocyte ratios (NLR) are associated with increased metastasis and poor overall survival and recent experimental studies have demonstrated a causal relationship between neutrophils and metastasis formation. However, to date, the cancer cell-intrinsic mechanisms dictating the substantial heterogeneity in systemic neutrophilic inflammation between tumour-bearing hosts are largely unresolved. Using a panel of 16 distinct genetically engineered mouse models (GEMMs) for breast cancer, we demonstrate that tumour cell-intrinsic loss of p53 changes the phenotype and function of macrophages in the microenvironment, leading to activation of a systemic inflammatory cascade that drives neutrophil expansion. Mechanistically, p53 loss in cancer cells induces secretion of Wnt ligands that act in a paracrine fashion to stimulate IL-1b production from tumour-associated macrophages. Intratumoural IL-1β production stimulates an inflammatory cascade leading to the systemic accumulation of neutrophils. Pharmacological and genetic blockade of cancer cell-derived Wnt secretion reverses IL-1β expression by macrophages and subsequent systemic neutrophilic inflammation. Collectively, using pre-clinical mouse models for breast cancer, we demonstrate a novel mechanistic link between loss of p53 in cancer cells, Wnt ligand secretion and systemic immune activation. This illustrates the importance of cancer cell-intrinsic genetic aberrations in dictating cancer-associated inflammation. These insights set the stage for personalized immune intervention strategies for cancer patients. In this study, gene expression profiles of tumours from genetically engineered mouse models (GEMMs) were analysed using RNA sequencing. Analysis was performed on bulk tumours of 10 GEMMs with different tissue-specific mutations driving tumorigenesis, totalling to 125 different tumours (n=5 or more per group). Subsequently, samples were grouped according to p53 status of the tumour (models containing Trp53 floxed alleles, or not) and comparisons were made between p53-KO and p53-WT tumours.