Accuracy of Integrated Classifier to Distinguish Benign From Malignant Lung Nodules

The accuracy of an integrated classifier in patients with indeterminate pulmonary nodules with a pretest probability for malignancy (pCA) of at least 50% was confirmed, according to the results of a 2-year follow-up extended analysis published in CHEST.

Current guidelines for the management of pulmonary nodules emphasize the assessment of pretest pCA to avoid diagnostic procedures in patients with benign disease and expedite diagnosis and treatment in patients with malignancy. An integrated proteomic biomarker combining 2 plasma proteins (LG3BP and C163A) with 5 clinical/imaging factors (age, smoking status, nodule size, edge, and location) previously demonstrated potential usefulness in nodule patients with a pretest pCA of at least 50%.

Using 1-year follow-up to determine benignity, the accuracy of the biomarker reported a sensitivity of 97%, specificity of 44%, and a negative predictive value (NPV) of 98%, and the biomarker was more accurate than physician and risk calculator assessments.

In the current study, researchers reported the 2-year follow-up results of the Pulmonary Nodule Plasma Proteomic Classifier (PANOPTIC; ClinicalTrials.gov Identifier: NCT01752114) trial and an extended analysis of patients with multiple pulmonary nodules were reported by researchers. All nodules diagnosed as benign at year 1 remained benign at year 2, and that the integrated classifier performance characteristics were maintained at year 2. In addition, the researchers found that this study, although not powered to be definitive, suggests that the classifier performed similarly regardless of nodule number. Of note, more than 4 or 5 nodules have been shown to be an indicator of a benign process.

“In conclusion, this 2-year follow-up extended analysis of the PANOPTIC trial confirms the accuracy of an integrated classifier in patients with indeterminate pulmonary nodules with pCA ≤50%, whether solitary or multiple,” stated the study authors. “These results substantiate the case for a clinical utility study to determine how the integrated classifier impacts physician decision-making and patient outcomes.”

Reference

Tanner NT, Springmeyer SC, Porter A, et al. Assessment of integrated classifier’s ability to distinguish benign from malignant lung nodules: extended analyses and 2 year follow-up results of the PANOPTIC (Pulmonary Nodule Plasma Proteomic Classifier) trial. CHEST. Published online November 7, 2020. doi:10.1016/j.chest.2020.10.069