AI Maps Tumors Cell by Cell to Predict Cancer Spread

Imagine if doctors could look inside a tumor and see exactly which cells are dangerous and how to stop them. That future just got closer.

Researchers at Lund University in Sweden have created a groundbreaking AI system that maps cancer tumors cell by cell. Instead of looking at a tumor as one mass, the technology identifies different populations of cells and predicts which ones will spread.

The team tested their approach on a young woman with metastatic melanoma. They discovered two distinct cell populations in her original tumor. One aggressive group, called PT1, likely drove the cancer's spread to her lungs and eventually her brain.

"To be able to reconstruct and connect localized tumor cell regions at single-cell resolution was a major achievement," said lead researcher György Marko-Varga. The system combines AI-guided imaging with advanced protein analysis to reveal how different cells behave.

The technology goes beyond just mapping cells. It identified specific metabolic programs and signaling pathways that made the tumor resistant to standard treatments. This allowed doctors to recommend alternative therapies targeting the tumor's actual vulnerabilities, including metabolic inhibitors and specific pathway blockers.

The Ripple Effect

The precision medicine market for cancer is exploding, expected to grow from $141 billion in 2026 to $226 billion by 2030. This AI approach could help determine which patients benefit most from emerging treatments like personalized cancer vaccines.

Marko-Varga's team has studied thousands of patient tumors over the years. Their conclusion is clear: every tumor is biologically unique with its own individual architecture. "Ultimately, each patient will likely require a dedicated and individualized combination of therapies, guided by the specific biological characteristics of their tumor," he explained.

The researchers are already preparing to publish similar results from two additional patients. Their goal is moving beyond treating just the tumor toward truly patient-centered care where scientific evidence, clinical judgment, and the patient's experience all guide treatment decisions.

For cancer patients facing uncertain outcomes, this technology offers something invaluable: a personalized roadmap showing exactly how their unique tumor behaves and which treatments stand the best chance of working.