Liver cancer often returns after surgery, but doctors currently lack reliable tools to identify which patients face the greatest risk.

Singapore researchers have uncovered two distinct ways that hepatocellular carcinoma, the most common type of liver cancer, recurs at the molecular level and have used these insights to develop a new machine-learning prediction tool.

Validated across independent patient groups, the tool leverages on both genomics and clinical data and outperformed the commonly used TNM staging system – which assesses disease stage based on tumour burden alone – in predicting recurrence risk and could aid surveillance, treatment decision-making and clinical trials.

New research reveals that cells of the same chronological age can follow dramatically different biological aging paths, offering new insights into cancer, neurodegeneration, and the fundamental biology of aging. Rather than aging uniformly, tissues become a mosaic of biologically younger and older cells, with some individual cells aging much faster than others—a discovery that could reshape how scientists understand aging and the earliest origins of age-related disease.

An evidence review of international research suggests widespread misconceptions within the medical community about early milk production, pacifiers and breast pumps may be undermining breastfeeding success in the critical first days after birth.

Researchers at the MRC Laboratory of Medical Sciences (LMS), Imperial College London and UCL Genetics Institute have created a detailed map of cells within breast tumours and identified distinct neighbourhoods of proliferating and dormant cells. The study, published in Genome Medicine, reveals that protective immune and connective cells surround the hibernating cancer cells, potentially shielding them from treatments. It suggests future therapies may need to target not only different cancer cell states, but also the local environments that help them survive, to both halt tumour growth and prevent cancer coming back later.