Abstract
BACKGROUND: The prognosis of early breast cancer is linked to clinic-pathological stage and the molecular characteristics of intrinsic tumor cells. In some patients, the amount and quality of tumor-infiltrating immune cells appear to affect long term outcome. We aimed to propose a new tool to estimate immune infiltrate, and link these factors to patient prognosis according to breast cancer molecular subtypes. METHODS: We performed in silico analyses in more than 2800 early breast cancer transcriptomes with corresponding clinical annotations. We first developed a new gene expression deconvolution algorithm that accurately estimates the quantity of immune cell populations (tumor immune contexture, TIC) in tumors. Then, we studied associations between these immune profiles and relapse-free and overall survival among the different intrinsic molecular subtypes of breast cancer defined by PAM50 classification. RESULTS: TIC estimates the abundance of 15 immune cell subsets. Both myeloid and lymphoid subpopulations show different spread among intrinsic molecular breast cancer subtypes. A high abundance of myeloid cells was associated with poor outcome, while lymphoid cells were associated with favorable prognosis. Unsupervised clustering describing the 15 immune cell subsets revealed four subgroups of breast tumors associated with distinct patient survival, but independent from PAM50. Adding this information to clinical stage and PAM50 strongly improves the prediction of relapse or death. CONCLUSIONS: Our findings make it possible to refine the survival stratification of early patients with breast cancer by incorporating TIC in addition to PAM50 and clinical tumor burden in a prognostic model validated in training and validation cohorts.