Model Can Predict Lung Cancer in Patient With Visible Nodule on CT

For patients with visible lung nodules, a model combining clinical and radiologic factors can predict risk for incident lung cancer, according to a study published online in Cancer Prevention Research.

Barbara Nemesure, Ph.D., from Stony Brook University in New York, and colleagues referred 2,924 eligible patients for evaluation of pulmonary nodules between Jan. 1, 2002 and Dec. 31, 2015. During the observation period, 171 patients developed incident lung cancer. The sample was randomly divided into discovery and replication samples (1,469 and 1,455 patients, respectively). Concordance was computed in the replication sample to indicate predictive accuracy; risk scores were calculated using linear predictions. Youden index was used to identify high- and low-risk patients; lung cancer incidence was examined for these groups.

The researchers identified a combination of clinical and radiologic predictors for incident lung cancer, including ln-age, ln-pack-years smoking, a history of cancer, chronic obstructive pulmonary disease, spiculation, ground glass opacity, and nodule size in multivariable analyses. In the replication sample, the final model reliably detected patients who developed lung cancer (C = 0.86; sensitivity/specificity = 0.73/0.81). In high- versus low-risk groups, the cumulative incidence of lung cancer was elevated (hazard ratio, 14.34).

"Differentiating pulmonary nodules that will progress into cancer from those that will remain benign has high clinical utility," the authors write. "The final model may assist physicians in managing the care of patients who may be at increased risk of developing lung cancer."

-----------------------------------------------------------------------
Editor's Note:

If you have any suggestion to the content, 

please email: Julia.Zhang@ubmsinoexpo.com