It is now well accepted that intra- and peri-tumoural infiltration of T cells has prognostic significance in breast cancer. Specifically the current literature indicates that a high FOXP3+ to CD8+ infiltrating T cell ratio is prognostic of poor outcome in breast cancers. However, there is evidence that the prognostic power of T cell infiltration must take into account the specific T cell subsets present in the particular tumour types. RANK is overexpressed in a proportion of breast cancers and this is associated with the more aggressive basal phenotype associated with a poor outcome. In normal breast tissue, RANK signalling activated at the onset of pregnancy by RANK-ligand (RANKL) and/or progestins drives differention of the resting epithelium to produce the fully differentiated lactating gland. This regresses post-lactation and the RANK-positive population remains quiescent unless reactivated in subsequent pregnancies. In prostate it has been shown that FOXP3+ T cells infiltrating the tumour express RANKL and that this is associated with invasion and metastasis. In this study we have used multiplex immunohistochemistry for FOXP3+, CD4+ and CD8+ T cells and RANK+, RANKL+ and PR+ cells,to characterise the T cell subsets in the primary tumour of locally-advanced breast cancer (LABC) and metastatic breast cancer patients with respect to the tumour type. Pilot data indicates there is a strong trend between a high FOXP3+ to CD8+ ratio and poor outcome, irrespective of absolute FOXP3+ cell number. We are currently investigating this in a larger cohort of 300 patients diagnosed with Infiltrating Duct Carcinoma with respect to RANK, RANKL and PR expression.