New Mayo Clinic research may eventually help physicians determine whether pulmonary nodules detected on CT scans are cancerous and require treatment.
The researchers’ goal was to develop a tool that could help determine whether detected nodules in the lungs are benign or not, to cut down on the number of false positives when checking for lung cancer.
The National Lung Screening Trialstudy found that up to 96 percent of detected pulmonary nodules — even in high-risk individuals — are actually benign.
"Only a small minority turn out to be malignant," said Tobias Peikert, a pulmonologist with Mayo Clinic.
Of course, assuming that nodules are benign can be harmful. Based on the size of the growths, physicians may choose to follow the growth of nodules with regular CT scans, or try to determine whether it is cancerous through other means.
However, there are downsides to continued CT, PET scans, or potential surgery to biopsy nodules. All of the options are costly, and radiation and surgery pose risks to the patient.
"When we see these patients, we want to make sure we’re not causing any harm," Peikert said.
Ordinarily, doctors will classify a nodule’s level of riskbased on characteristics such as location in the lungs, size of the growth, age of the patient, and whether he or she is or was a smoker.
Some nodules are high-risk and should be treated as cancerous, Peikert said. But deciding how to treat nodules that are above a 10 percent chance of being cancerous, but below 70 percent, is tricky.
"Where the big clinical dilemma is, and where we struggle, is that middle ground," he said.
Using radiomics, a field of study that extracts information from medical imaging, Dr. Peikert and Fabien Maldonado, M.D., from Vanderbilt University, along with collaborators, found variables that let them distinguish benign growths from cancerous nodules.
Those variables are different from the other classification/risk assessment systems in that they look at the surface characteristics of the nodules.
The smooth or rugged texture of the nodules, their density, and qualities in the surrounding lung tissue can be evaluated to determine whether a nodule has a high chance of being cancerous, Peikert said.
The technology still has to be validated, which will be done using DECAMP — Detection of Early Lung Cancer Among Military Personnel— data.
"The preliminary data actually looks good," Peikert said. Once it is validated, he hopes it’ll be implemented in clinical practice when pulmonary nodules are detected.