Scientists have developed an artificial intelligence powered imaging technique that can identify key genetic mutations linked to lung cancer without relying on traditional genetic sequencing, potentially giving doctors faster and less expensive information to guide treatment decisions.
The findings were published this week in Cancer Research. Researchers from the University of Edinburgh and NHS Lothian combined fluorescence lifetime imaging microscopy, known as FLIM, with artificial intelligence to analyze natural light signals from untreated tissue samples. The system accurately predicted mutations in the EGFR gene while preserving valuable biopsy material for additional testing.
Doctors often test lung tumors for EGFR mutations because the results help determine whether patients may benefit from targeted therapies. However, current laboratory methods typically require gene sequencing, which can take time, increase costs and consume limited tissue collected during biopsies.
Additionally, obtaining enough tissue can prove difficult because many lung biopsies produce only small samples. That limitation has driven researchers to search for less invasive ways to identify clinically important mutations.
The new technique captures natural fluorescent signals from untreated tissue instead of relying on chemical stains or genetic analysis. Artificial intelligence then analyzes those signals to identify patterns associated with EGFR mutations.
Researchers reported that the system detected EGFR mutations with very high accuracy. It also distinguished between the two most common EGFR mutation types that influence treatment decisions for patients with lung cancer.
Meanwhile, expanded lung cancer screening programs continue identifying suspicious tumors at earlier stages. Earlier detection places greater pressure on hospitals to deliver accurate diagnoses using increasingly limited tissue samples.
Read more: Prestigious medtech intelligence firm recognizes Breath Diagnostics for innovation
Read more: Breath Diagnostics completes install of advanced mass spectrometry system
The study builds on earlier research from the same team
The researchers said the approach leaves biopsy tissue intact because it does not alter the sample during testing. Consequently, clinicians can preserve the remaining material for additional laboratory analysis if needed.
The study builds on earlier research from the same team showing that FLIM could distinguish major forms of non-small cell lung cancer from healthy tissue without destructive testing methods.
Professor Ahsan Akram, a co-leader of the research at the Institute for Regeneration and Repair, said the findings move medicine closer to using a single fluorescence scan to identify cancer, classify the disease and estimate whether patients will respond to targeted therapies. He said that approach could help physicians match patients with appropriate treatments more quickly.
Furthermore, Akram said the technology could reduce delays by delivering several pieces of diagnostic information from one non-destructive examination rather than requiring multiple laboratory procedures.
Co-lead researcher Dr. Qiang Wang said the platform could dramatically reduce both testing costs and turnaround times. He said procedures that currently cost thousands of pounds and require weeks of laboratory work could eventually take only minutes while costing hundreds of pounds instead.
Researchers believe the technology could prove especially valuable for hospitals and health systems with limited access to advanced molecular testing. However, the team said additional clinical validation remains necessary before doctors can adopt the method routinely.
The researchers now plan to expand the platform beyond lung cancer by evaluating additional targetable mutations and other cancer types. They also intend to integrate the technology into clinical workflows while continuing larger validation studies before seeking broader clinical adoption.
.
joseph@mugglehead.com
