When a patient has a lung tumor, doctors need to know whether it carries certain genetic changes before they can choose the right treatment. Today, that process often means cutting away a piece of the tumor, sending it to a lab, and waiting weeks for results — and it can cost thousands of dollars. But researchers at the University of Edinburgh in Scotland may have found a faster, cheaper way.
They have developed a new microscope technique that can predict cancer gene changes right from a tissue sample, without the need for traditional lab testing. The method uses a tool called fluorescence lifetime imaging microscopy, or FLIM for short. Put simply, FLIM shines light on a tissue sample and captures the natural glow that comes back. An artificial intelligence system then looks for patterns in that glow to figure out whether the tumor carries a specific genetic mutation called EGFR.
Why does EGFR matter? Some lung cancers have changes in this gene, and patients with those changes respond very well to targeted treatments that attack the cancer more precisely. The problem is that detecting EGFR mutations traditionally requires expensive gene sequencing — a slow process that uses up precious tissue from small biopsy samples. Many hospitals, especially in lower-income countries, do not have access to this kind of testing at all.
The Edinburgh team tested their FLIM approach on real patient samples and found it could predict EGFR mutations with very high accuracy. It also managed to tell apart the two most common types of EGFR mutations, which is important because treatment choices can differ between them. Since the technique works on untreated tissue, the sample stays intact and can be used for other tests afterward.
The researchers are now working to bring this approach out of the lab and into real hospital settings. They also want to expand it to other cancer types and additional genetic mutations beyond EGFR.
Dr. Qiang Wang, who co-led the study, said the method could transform how hospitals diagnose and treat cancer. "This approach has the potential to take processes that currently cost thousands of pounds and require weeks of lab work and reduce them to something that takes minutes and costs hundreds," he said. Professor Ahsan Akram, the study's other co-lead, called it "a significant step towards a future where a single, non-destructive fluorescence scan of a biopsy could quickly inform clinicians whether a patient has cancer, what type of cancer they have, and now whether it is likely to respond to targeted treatment."
Lung cancer remains the leading cause of cancer-related death worldwide. As screening programs catch more tumors at an earlier stage, diagnostic services face growing pressure to deliver fast, accurate answers from ever-smaller tissue samples. Technologies like FLIM could help ease that pressure — and, more importantly, get the right treatment to the right patient sooner.