1213. A TRAIL/IP-10/CRP Signature Distinguishes between Viral and Bacterial Infection in Chronic Obstructive Pulmonary Disease Patients

Abstract

BackgroundChallenges in determining the etiology of acute exacerbations of chronic obstructive pulmonary disease (COPD) lead to significant overuse of antibiotics. A new host-response assay that integrates the levels of three proteins (TRAIL, IP-10, and CRP) was shown to exhibit high performance in distinguishing between bacterial and viral disease in two double-blind pediatric validation studies. Here we sought to evaluate its ability to differentiate bacterial from viral infection in adult COPD patients with suspicion of lower respiratory tract infection (LRTI).MethodsThe study population included 492 febrile adult patients prospectively recruited in “Observer”, an EU Horizon 2020 funded study (grant #684589). Patient etiology was determined by majority expert panel based on clinical, laboratory, multiplex PCR, radiological and follow-up data. We compared the expert panel diagnosis with the assay that gives three possible outcomes: viral, bacterial (including viral with bacterial coinfection) or equivocal.Results45 out 492 adult patients prospectively recruited with suspicion of LRTI had a medical history of COPD. Of these, 20 cases were assigned as suspected viral infections and 19 as suspected bacterial infections (Figure 1). Antibiotics were prescribed to 19/19 bacterial infections and 16/20 viral infections. The assay correctly classified 19/19 bacterial infections and 12/20 viral infections, with 2 viral cases classified by the assay as bacterial and 6 receiving an equivocal outcome. These data support the assay’s potential to reduce antibiotic overuse from 16/20=80% to 8/20=40% (P=0.01).FIgure 1: Flow through of COPD patients in prospective performance validation study “Observer” ConclusionA new TRAIL/IP-10/CRP signature has potential to significantly reduce antibiotic overuse for patients with suspected LRTI and a history of COPD without missing bacterial infection.DisclosuresMeital Paz, MD, MeMed Ltd. (Employee) Noa Avni, PhD, MeMed (Employee) Michal Stein, MeMed Ltd. (Employee) Liran Shani, MD, MeMed Ltd. (Employee) Tanya Gottlieb, PhD, MeMed (Employee) Kfir Oved, MD, PhD, MeMed (Employee) Eran Eden, PhD, MeMed (Employee)