Online Program Home
My Program

Abstract Details

Activity Number: 418 - SPEED: Biostatistical Methods, Application, and Education, Part 2
Type: Contributed
Date/Time: Tuesday, July 30, 2019 : 2:00 PM to 2:45 PM
Sponsor: Section on Medical Devices and Diagnostics
Abstract #307828
Title: Assessment of Biomarker Strategies in Lung Cancer Management via Net Reclassification Indices
Author(s): Piper Williams* and Alexander Kaizer and Anna Barón
Companies: University of Colorado Anschutz Medical Campus and University of Colorado Anschutz Medical Campus and University of Colorado Anschutz Medical Campus
Keywords: Lung cancer; Indeterminate pulmonary nodule; Biomarker testing strategies; Net reclassification index; Super Learner; Bayesian stacking

Improving discrimination of indeterminate pulmonary nodules (IPNs) continues to be a significant challenge in lung cancer management and treatment. While numerous lung cancer risk prediction models have been developed to predict the probability of malignancy of IPNs, many IPNs remain in what is called an indeterminate risk range. IPNs within this range can be potentially misdiagnosed, thus leading to either unnecessary medical procedures or the missed opportunity to begin treatment promptly and increase a patient’s probability of survival. The addition of biomarker information to pre-existing risk prediction models has been proposed as a method to improve prediction of malignancy. In this study, various net reclassification indices (NRIs) were used to measure the incremental value of an additional biomarker including the two-category NRI as well as the bias-corrected clinical NRI. This study also compared the NRIs calculated from logistic regression, Super Learner, and Bayesian stacking approaches. Through these various approaches, we will highlight strategies that will improve the diagnostic accuracy of IPNs and, ultimately, improve the clinical management of IPNs.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program