Researchers found that incorporating lung nodule features and patient-specific characteristics improved the positive predictive ability of a cancer screening model while maintaining its sensitivity of low-dose CT scans, according to research published Aug. 22 in the Journal of the American College of Radiology.
A lung screening model that can distinguish between malignant and benign lung nodules may reduce false-positive results. But official selection criteria for appropriate variables or cutoff points is lacking, according to authors Reshad Hassannezhad, MD, and Nafiseh Vahed, MSc, from Tabriz University in Tabriz, Iran.
“Increment in positive predictive value of lung cancer screening with sensitivity same as National Lung Screening Trial is feasible, and inclusion of other nodule size dimensions plus longest diameter to the model significantly improves the predictive ability of models,” they wrote.
For the study, the researchers used data from the National Lung Screening Trial containing a multilevel design with nodules nested within rounds and rounds nested within individuals, with information regarding 9,728 patients and a total of 32,746 lung nodules.
To incorporate nodule level features and patients characteristics into the model construction, the researchers used multilevel logistic regression.
“Model construction was based on improvement in predictive ability of the model, and there were no restrictions to any significance level on variable inclusion,” the researchers wrote.
With a sensitivity value equal to that of the National Lung Screening Trial (93.6 percent), the study model demonstrated a positive predictive value of 7.94 percent, or twice that of the National Lung Screening Trial (3.6 percent). Researchers also found that including other nodule dimensions and the longest diameter improved the model's predictive ability.
The research may help clinicians better spot high-risk nodules and reduce false-positive screening results, according to the researchers.
“The ultimate result of the current study is the reduced number of false-positive results of lung cancer screening with no reduction in sensitivity,” according to the researchers. “This first maintains the observed mortality benefits in the original trial, and second, it reduces the costs and complications resulting from unnecessary diagnostic follow-ups.”
A recent graduate from Dominican University (IL) with a bachelor’s in journalism, Melissa joined TriMed’s Chicago team in 2017 covering all aspects of health imaging. She’s a fan of singing and playing guitar, elephants, a good cup of tea, and her golden retriever Cooper.