#28: Rapid, Non-invasive Detection of Invasive Bartonella Infections in Pediatric Patients Using the Karius Test, A Next-Generation Sequencing Test for Microbial Cell-free DNA in Plasma

Abstract

Abstract Background Bartonella henselae and Bartonella quintana are the etiologic agents of cat scratch disease (CSD) and “trench fever”, respectively. Both are important causes of culture-negative endocarditis and fever of unknown origin (FUO). The diagnosis of Bartonella infections is limited by (1) the nonspecific, protean manifestations of the disease and its broad differential diagnosis; (2) the fastidious nature of Bartonella spp., leading to rare detections with traditional culture based methods; (3) the insensitivity and poor specificity of Bartonella serologies. Rapid, non-invasive diagnosis of Bartonella through next-generation sequencing (NGS) of plasma microbial cell-free DNA (mcfDNA) offers a means to overcome these limitations. Here we describe the diagnosis of 23 Bartonella infections in children from August 2017 – December 2020 using plasma mcfDNA NGS. Methods The Karius Test (KT), developed and validated in Karius’ CLIA certified/CAP accredited lab, detects mcfDNA in plasma. After mcfDNA is extracted and NGS performed, human reads are removed, and remaining sequences are aligned to a curated database of > 1400 organisms. McfDNA from organisms present above a statistical threshold are reported and quantified in molecules/μL (MPM). Clinical information was included from data submitted with the requisition or obtained at the time of reporting from clinical consultations with the provider. Results KT detected Bartonella henselae mcfDNA in 22 cases and Bartonella quintana in 1. Detections included 10 cases of endocarditis (7 prosthetic valve), 12 cases of CSD/FUO, and a single case of osteomyelitis. Glomerulonephritis was reported in 5 the cases of endocarditis. Six cases had splenic involvement; three had hepatic involvement. History of cat exposure was elicited in 8 cases. The mean MPMs was highest for prosthetic valve endocarditis (mean 47,272 +/- 67,526) followed by native valve endocarditis (3,881 +/- 2,458), FUO/CSD (1,922 +/- 3,416), and osteomyelitis (119 +/- 0) (p<0.05). Three subjects had serial mcfDNA monitoring. Predictable declines in Bartonella mcfDNA were observed in response to therapy in all three patients. The duration of positive Bartonella mcfDNA signal ranged from 22–42 days (30.7, +/- 10.3). Conclusion Open-ended, plasma-based NGS for mcfDNA provides a rapid, non-invasive method to diagnose diverse clinical manifestations of invasive pediatric Bartonella infection against a competing broad differential diagnosis. The quantification of mcfDNA may further help in differentiating the various clinical syndromes caused by Bartonella. Finally, serial monitoring to trend MPMs may serve as an indicator of burden of infection, provide a means to monitor treatment efficacy and ultimately help define the length of therapy for optimal outcomes. All detections are Bartonella henselae except for one case of *Bartonella quintana (prosthetic valve endocarditis). MPM=molecules/μL; GN=glomerulonephritis, LN=lymph node involvement, FUO=fever of unknown origin, CSD=cat scratch disease