The prognosis and successful treatments differ depending on the type of lung cancer. While it previously took several days to precisely determine the underlying lung tumor mutation, a research team has been able to perform this determination in just one step reliably. They used a combination of quantum cascade laser-based infrared microscopy and machine learning.
Researchers do not need to mark the examined tissue for this. Within half an hour, the analysis ascertains whether the tissue sample contains tumor cells, what tumor it is, and whether it contains a particular mutation.
The researchers verified the procedure on samples from over 200 lung cancer patients. When identifying mutations, they concentrated on, by far, the most common lung tumor, adenocarcinoma, which accounts for around 50 percent of tumors. It is possible to determine the most common genetic mutations with a sensitivity and specificity of 95 percent compared to laborious genetic analysis. The researchers identified spectral markers allowing a spatially resolved distinction between various molecular conditions in lung tumors.
This study presents a label-free, automated, spatially resolved, and observer/operator-independent approach to IR imaging using QCL. Thin sections of 536 formalin-fixed, paraffin-embedded (FFPE) tumor and nontumor lung tissues from 214 patients were used to validate this technique. Compared to histopathology, cancerous regions were identified with a sensitivity of 95% and a specificity of 97%.
Related Content: Microscopic Robots Aid Medical Diagnostics