242. Comprehensive Pathogen Detection for Pediatric Febrile Neutropenia by Metagenomic Next-Generation Sequencing

Abstract

Abstract Background Febrile neutropenia (FN) is a common complication in patients with solid tumors and hematologic malignancies. Identification of the causative microorganisms would contribute to optimization of antimicrobial treatment and thus improve the outcome of FN. However, causative microorganisms are detected in only 10% to 20% of FN patients. Next-generation sequencing (NGS) allows us to comprehensively analyze all microorganisms present in a clinical sample. In this study, we aimed to utilize NGS for the detection of microbial pathogens in infectious diseases and elucidate the infection source in FN. Methods FN is defined by two characteristics: (1) neutrophils count < 500/µL, and (2) fever ≥38.0°C. From 2016 to 2018, 112 plasma/serum samples of pediatric FN patients (11 positive blood cultures) were analyzed. Serum samples from 10 neutropenic patients without fever were also analyzed as controls. Shotgun sequencing method was applied for these samples. The metagenomic analyses were performed through the pipeline PATHDET, which has been newly established in our laboratory. Diagnosis based on NGS results was made based on the following criteria: (1) number of reads from all pathogens per million reads (PR) >650, (2) a specific pathogen’s reads per million reads (RPM) >200, and (3) diversity index >3.0. The NGS results were compared with those from blood culture. Results Sequencing reads of bacteria isolated through blood culture were identified by NGS in all 11 plasma/serum samples leading to the diagnosis of FN. The causative pathogens were diagnosed by NGS using the above criteria in 11 patients. However, the results were consistent with those of blood culture in only 4 samples. Of 101 cases with negative blood culture results, the causative pathogens were detected in 17 cases: Acinetobacter soli (2 cases), Burkholderia cepacian (1 case), Klebsiella variicola (1 case), and Roseomonas sp. (1 case) were identified at the species level. In addition, 7 cases (e.g., Acinetobacter) were identified at the genus level, and 5 cases (e.g., Enterobacteriaceae) were identified at the family level. Conclusion Metagenomic NGS technique has great potential for detecting causative pathogens with greater efficiency than the conventional methods. Disclosures All authors: No reported disclosures.