Utilizing machine learning in analysing the anti-cancer effect of a novel immunotherapy approach

Breast cancer is one of the most common types of cancer worldwide, with more than two million new cases diagnosed annually. Traditionally, breast cancer has been treated with chemotherapy and antibody-based therapies; however, research into and the use of immuno-oncological treatments have increased over recent decades. Cancer immunotherapies aim to activate the patient’s own immune system to fight proliferating cancer cells. One form of active immunotherapy, immune checkpoint blockade (ICB) therapy, which is the focus of this study, seeks to activate immune cells to combat cancer. A novel potential target for ICB therapy is the Clever-1 receptor molecule, which is expressed on the surface of immunosuppressive macrophages. An anti-Clever-1 antibody targeting this receptor has already demonstrated significant responses in phase I/II clinical trials in cancer patients. Anti-Clever-1 therapy aims to direct immune cells against tumors and to induce an immune response in immunologically inactive tumors. In this study, we investigated the molecular biological mechanisms of action and the ex vivo efficacy of the anti-Clever-1 antibody. The study was conducted using patient samples collected from breast cancer patients at Turku University Hospital. Breast cancer tissue samples as well as blood samples were collected from the patients. Tissue samples were stained using immunohistochemical methods. In addition, peripheral blood mononuclear cells from blood samples—which include the majority of human immune cells—were treated with the anti-Clever-1 antibody. The antibody-treated cells were then used to treat the corresponding patients’ tissue samples. This approach enabled the analysis of immune cell activity and the quantitative assessment of cancer cell apoptosis. The analysis was performed on fluorescence microscopy images using an artificial intelligence–based image analysis software called Aivia. This study, conducted using a novel image analysis program, enabled a more precise and reliable investigation of the effects of the anti-Clever-1 antibody. The results showed that the anti-Clever-1 antibody helps immune cells infiltrate tumor tissue and induces apoptosis of cancer cells within the tissue. A favorable response was observed in approximately thirty percent of patients.