Blood biomarker identifying benign lung nodules performs well across subgroups

Key takeaways:

• The negative predictive value of a biomarker that detects benign vs. malignant nodules was more than 90% in men and women.
• Researchers also observed high values across smoking status and nodule size subgroups.

A blood biomarker used to identify benign vs. malignant pulmonary nodules displayed “excellent” performance regardless of sex, smoking status, screen detection and nodule size, according to a research letter published in CHEST.

The biomarker, or proteomic integrated classifier (IC), joins together plasma proteins LG3BP and C163A with smoking status, nodule size, edge and location, according to researchers.

“Incorporation of the IC into a screening population could decrease the rate of unnecessary PET/CT imaging and biopsies by reclassifying likely benign nodules,” Kathryn J. Long, MD, a third-year pulmonary and critical care fellow of the Medical University of South Carolina, and colleagues wrote.

In a subgroup analysis of patients from the Pulmonary Nodule Plasma Proteomic Classifier (PANOPTIC) trial and the Observational Registry Study to Evaluate the Performance of BDX-XL2 in Routine Clinical Use (ORACLE) trial, Long and colleagues analyzed 430 patients to see if IC performance differs when patients are divided by sex, screen detection, smoking status and nodule size.

Researchers previously found that this biomarker had a negative predictive value of 98% in identifying benign nodules from malignant nodules in the PANOPTIC trial.

The current analysis included 150 patients (mean age, 64.5 years; 46% women) from the PANOPTIC trial and 280 patients (mean age, 67.8 years; 62% women) from the ORACLE trial. Between the two groups, average nodule diameter was larger in the PANOPTIC cohort (13.2 mm vs. 12 mm).

Of the pooled cohort, more patients had benign vs. malignant (cancer) nodules (86% vs. 14%).

Researchers found similar, high negative predictive values in identifying benign nodules with the IC among men (97%) and women (95%).

In terms of screen detection, IC performance was comparable between incidental nodules and screen-detected nodules (negative predictive value, 94% vs. 100%).

Half of the total cohort reported being former tobacco smokers (n = 216), whereas the rest either reported being a current smoker (n = 103) or a never smoker (n = 111).

Similar to the findings above, IC performance did not significantly differ between tobacco users and nonusers, with negative predictive values of 97% and 94%, respectively.

Researchers did note that sensitivity of the IC was significantly higher in the set of tobacco users vs. nonusers (94% vs. 67%), whereas specificity was significantly lower (39% vs. 68%).

The last subgroup was divided based on nodule diameter, and patients fell into either the 11 mm or smaller cohort or the larger than 11 mm cohort since the median diameter was 11 mm.

Between the smaller and larger diameter cohorts, researchers found comparable negative predictive values (98% vs. 93%) and sensitivity (92% vs. 87%) with the IC. Notably, the IC had significantly higher specificity in patients with smaller vs. larger nodules (57% vs. 35%).

“This subgroup analysis demonstrated a high [negative predictive value] for the IC that did not differ based on screen detection, sex, smoking status or nodule size,” Long and colleagues wrote. “These results suggest that the IC may be used effectively in a broader range of patient populations than previously reported.”