171P Immune landscape and CLEVER 1 expression in hepatoblastoma

Hepatoblastoma (HB) is the most common primary liver malignancy among infants and young children. HBs are regarded as embryonal tumors arising from hepatic progenitor cells. Current treatment modalities for HB produce unsatisfactory results (20-30% respond poorly to interventions). To decrease adverse effects and suboptimal outcomes in patient care, research in immuno-oncology is investigating safer and more effective treatment options. Finding new targets for HB immunotherapeutics requires improved knowledge of the tumor immune microenvironment (TIME). In this study, our objective is to elucidate the TIME of HBs and investigate known targets of novel treatment options such as CLEVER-1, inhibition of which triggers M2 macrophage conversion towards M1 type and provokes an antitumoral response in the immune system. Whole-slide multiplex immunofluorescence staining (mIHC) was utilized to assess the immunological characteristics of HB patient specimens (n=25) and normal liver samples (n=2). The mIHC panel chosen included antibodies against all macrophages, M2 macrophages, M1 macrophages, T-cells, and all leukocytes, as well as stellate cells, and epithelial cells. Antibody validation was performed to assess the consistency of staining of the targeted proteins in HB tissue. Image analyses were carried out using machine learning-assisted software (QuPath) to calculate the proportions of each immune cell type and assess the expression of potentially targetable proteins. QuPath was used to train an artificial intelligence model (AIM) for each marker. After automatic cell detection, the AIMs were applied to selected stained specimens, and cells positive for different markers were uncovered. Subsequently, spatial analyses regarding the proximity of immune cells to the tumor were performed. CLEVER-1-positive M2 macrophages are detected in HB tissue. Furthermore, preliminary results indicate that the staining is adequate for detecting other immune cell types within the specimens. Moreover, the AIMs show promising accuracy in automatically classifying immune cells. HB is a cancer type with CLEVER-1-positive M2 macrophages and is thus a possible candidate for novel immunotherapies.