LAG3 as a potential biomarker in breast cancer: integrated tissue and peripheral blood analysis reveals clinical and pathomic insights in triple-negative disease

Breast cancer is a heterogeneous disease. Triple-negative breast cancer (TNBC) is the most aggressive subtype and lacks reliable biomarkers for immunotherapy. This study aimed to evaluate lymphocyte activation gene 3 (LAG3) as a potential translational biomarker in breast cancer, assessing its expression across molecular subtypes and whether its levels in patients’ peripheral blood match those in tumour tissue, thereby enabling non-invasive monitoring. The TCGA genomic database was analysed to investigate the association between LAG3 expression and breast cancer subtypes. These findings were then validated in an independent patient cohort of 50 patients using immunohistochemistry and molecular analysis of peripheral blood mononuclear cells (PBMCs). Digital pathology approaches were also employed, incorporating intensity- and texture-based feature extraction from DAB and ODsum channels to evaluate spatial patterns of marker distribution. A significant association was identified in the TCGA dataset between high LAG3 expression and the TNBC subtype, and this was confirmed in the validation cohort. LAG3 expression was found only in immune cells and was significantly more prevalent in TNBC. PBMC-derived mRNA levels were consistent with tissue expression, supporting systemic concordance. While the DAB-positive area did not significantly differ among subtypes, digital pathology analyses identified increased spatial heterogeneity in TNBC compared with luminal tumors using intensity - and texture - based features derived from DAB and ODsum channels, highlighting the value of computational pathology in TME characterization. LAG3 is emerging as a promising immune-related biomarker with translational relevance in breast cancer. The observed concordance between tissue and peripheral blood samples suggests that liquid biopsies could be used as a non-invasive tool for immune profiling, patient stratification and longitudinal disease monitoring. This could improve the selection of immunotherapies.