Using machine learning and extensive genomic and pharmacological data, researchers at Lund University in Sweden have identified a promising drug combination—statins and phenothiazines—for treating high-risk neuroblastoma, an aggressive childhood cancer. The findings, published in EMBO Molecular Medicine, show that the combination significantly slowed tumor growth and improved survival rates in preclinical models. Neuroblastoma affects approximately 15 to 20 children in Sweden each year, with most cases diagnosed before the age of five. While some forms are mild, the high-risk variant remains the most challenging, with the lowest survival rate among childhood cancers. A major obstacle in treating this disease is its tendency to relapse, often with tumors that have become resistant to chemotherapy. Daniel Bexell, head of the Molecular Pediatric Oncology research team at Lund University, has led efforts to identify effective treatments, particularly through drug repurposing—using existing drugs approved for other conditions. In this new study, his team partnered with Healx, a UK-based AI and biotech company, researchers from Karolinska Institutet, and two childhood cancer advocacy organizations: the aPODD Foundation (UK) and ENEA (European Neuroblastoma Association). By applying machine learning to vast datasets on gene activity and drug mechanisms, the researchers pinpointed a combination of two drug classes: statins, commonly used to lower cholesterol, and phenothiazines, which are used to treat conditions like migraine and nausea. When tested on patient-derived neuroblastoma tumors, both drugs showed individual effects, but their combination produced a powerful synergistic response. The study revealed that while statins block cholesterol production, phenothiazines reduce cholesterol through a different, complementary pathway. Together, they drastically lowered cholesterol levels within tumor cells, leading to widespread cell death. Surviving cells became more sensitive to chemotherapy, suggesting the combination could enhance the effectiveness of standard treatments. In mouse models of aggressive neuroblastoma, the drug combination significantly slowed tumor progression and increased survival. The results were especially encouraging for tumors that had developed resistance to conventional therapies. “The effect was stronger than we expected,” said Bexell. “We knew cholesterol was important for tumor growth, but we didn’t anticipate such a dramatic impact from disrupting it through this dual approach.” The researchers now plan to optimize the chemical properties of the drugs to improve their delivery and efficacy. While further studies are needed before clinical trials in humans, the findings offer a promising new direction for treating high-risk neuroblastoma, particularly in cases where current therapies fail.