Time-series blood cytokine profiles correlate with treatment responses in triple-negative breast cancer patients.

Abstract

Triple-negative breast cancer (TNBC) is the most heterogeneous breast cancer subtype. Partly due to its heterogeneity, it is currently challenging to stratify TNBC patients and predict treatment outcomes. In this study, we examined blood cytokine profiles of TNBC patients throughout treatments (pre-treatment, during chemotherapy, pre-surgery, and 1 year after the surgery in a total of 294 samples). We analyzed the obtained cytokine datasets using weighted correlation network analyses, protein-protein interaction analyses, and logistic regression analyses. We identified five cytokines that correlate with good clinical outcomes: interleukin (IL)-1α, TNF-related apoptosis-inducing ligand (TRAIL), Stem Cell Factor (SCF), Chemokine ligand 5 (CCL5 also known as RANTES), and IL-16. The expression of these cytokines was decreased during chemotherapy and then restored after the treatment. Importantly, patients with good clinical outcomes had constitutively high expression of these cytokines during treatments. Protein-protein interaction analyses implicated that these five cytokines promote an immune response. Logistic regression analyses revealed that IL-1α and TRAIL expression levels at pre-treatment could predict treatment outcomes in our cohort. We concluded that time-series cytokine profiles in breast cancer patients may be useful for understanding immune cell activity during treatment and for predicting treatment outcomes, supporting precision medicine. The study has been registered with the University Hospital Medical Information Network Clinical Trials Registry ( http://www.umin.ac.jp/ctr/index-j.htm ) with the unique trial number UMIN000023162. The association Japan Breast Cancer Research Group trial number is JBCRG-22. The clinical outcome of the JBCRG-22 study was published in Breast Cancer Research and Treatment on 25 March 2021. https://doi.org/10.1007/s10549-021-06184-w .