Utility of a Molecular Classifier as a Complement to High-Resolution Computed Tomography to Identify Usual Interstitial Pneumonia

• DOI: 10.1164/rccm.202003-0877oc
• Publication Date: 2020-07-28T21:38:06Z
• Authors: Luca Richeldi; Mary Beth Scholand; David A. Lynch; Thomas V. Colby; Jeffrey L. Myers; Steve D. Groshong; Jonathan H. Chung; Sadia Benzaquen; Steven D. Nathan; J. Russell Davis; Shelley L. Schmidt; Lars Hagmeyer; David Sonetti; Jurgen Hetzel; Gerard J. Criner; Amy H. Case; Murali Ramaswamy; Karel Calero; Umair A. Gauhar; Nina M. Patel; Lisa Lancaster; Yoonha Choi; Daniel G. Pankratz; P. Sean Walsh; Lori R. Lofaro; Jing Huang; Sangeeta M. Bhorade; Giulia C. Kennedy; Fernando J. Martinez; Ganesh Raghu
• Keywords: High-resolution computed tomography

Rationale: Usual interstitial pneumonia (UIP) is the defining morphology of idiopathic pulmonary fibrosis (IPF). Guidelines for IPF diagnosis conditionally recommend surgical lung biopsy histopathology UIP when radiology and clinical context are not definitive. A "molecular UIP" in transbronchial biopsy, Envisia Genomic Classifier, accurately predicted histopathologic UIP.Objectives: We evaluated combined accuracy Classifier local detection pattern.Methods: Ninety-six patients who had diagnostic pathology as well a molecular testing with were included this analysis. The classifier results scored against reference pathology. identified on high-resolution computed tomography (HRCT) documented by features radiologists' reports was compared histopathology.Measurements Main Results: In 96 patients, achieved specificity 92.1% (confidence interval [CI],78.6-98.3%) sensitivity 60.3% (CI, 46.6-73.0%) histology-proven pattern. Local radiologists 18 53 histopathology, 34.0% 21.5-48.3%) 96.9% 83.8-100%). conjunction HRCT patterns UIP, 24 additional (sensitivity 79.2%; 90.6%).Conclusions: suspected disease, regardless These suggest that recognition pattern factors multidisciplinary discussion may assist clinicians making an disease (especially IPF) without need biopsy.