Novel Tool Uses Deep Learning for Precision Cancer Therapy

Nearly 50 new cancer therapies are approved each year, but selecting the right one for patients with highly individual tumor characteristics remains a major challenge. Physicians struggle to navigate the growing number of options and determine which treatment will bring the most benefit. To address this complexity, researchers have developed a new artificial intelligence (AI) tool to integrate diverse data sources and guide more precise, personalized therapy decisions.

Unlike traditional methods, the AI-powered toolkit named Flexynesis, which has been developed by researchers at the Max Delbrück Center (Berlin, Germany), combines deep neural networks with multimodal data analysis, including multi-omics datasets, medical imaging, and clinical text. By processing this variety of inputs simultaneously, the system enables physicians to improve diagnosis, prognosis, and treatment strategies across multiple cancer types.

Image: Summary of the Flexynesis data integration and analysis workflow (B Uyar et al., Nature Communictions. DOI: 10.1038/s41467-025-63688-5)

Flexynesis was designed as a flexible and accessible toolkit, packaged for use across platforms such as PyPI, Guix, Docker, Bioconda, and Galaxy. Researchers validated its effectiveness through translational projects with medical doctors, identifying biomarkers that align with disease outcomes. The study, published in Nature Communications, highlights its versatility in answering diverse clinical questions, from identifying tumor types to predicting drug effectiveness and survival outcomes.

The tool can identify suitable biomarkers, improve cancer subtype classification, and even locate the primary tumor when metastases of unknown origin are present. Its design makes it a valuable complement to existing AI tools like Onconaut, offering a broader capability for multimodal integration. While widespread clinical use is limited by the availability of multi-omics data, increasing adoption in hospital tumor boards and research programs suggests this barrier may soon be overcome.

"Comparable tools so far have often been either difficult to use or only useful for answering certain questions,” said Dr. Altuna Akalin, senior author of the study. "Flexynesis, by contrast, can answer various medical questions at the same time: for example, what type of cancer is involved, what drugs are particularly effective in this case, and how these will affect the patient's chances of survival.”

Related Links:
Max Delbrück Center