Metagenomic Sequencing Detects Respiratory Pathogens in Hematopoietic Cellular Transplant Patients

BACKGROUND: Lower respiratory tract infections (LRTI) are a leading cause of mortality in hematopoietic cell transplant (HCT) recipients. Current microbiologic diagnostics often fail to identify etiologic pathogens, leading to diagnostic uncertainty and precluding the implementation of targeted therapies. To directly address the need for improved LRTI diagnostics, we undertook this study to evaluate the potential utility of metagenomic next generation sequencing (mNGS) approaches for detecting LRTI in the HCT population. METHODS: We enrolled 22 post-HCT adults ages 19-69 years with acute respiratory illnesses who underwent bronchoalveolar lavage (BAL) at the University of Michigan between January 2012 and May 2013. Unbiased mNGS was performed on BAL fluid to detect microbes and assess host response. Results were compared to those obtained by standard clinical microbiology testing. RESULTS: Unbiased mNGS detected all microbes identified by standard testing (human metapneumovirus, respiratory syncytial virus, Stenotrophomonas maltophilia, human herpesvirus 6 and cytomegalovirus). Previously unrecognized LRTI pathogens were identified in six patients for whom standard testing was negative (human coronavirus 229E, human rhinovirus A, Corynebacterium propinquum and Streptococcus mitis) and findings were confirmed by independent PCR testing. mNGS identified microbes of unlikely or uncertain pathogenicity in 10 patients with clinical evidence of non-infectious respiratory conditions. Patients with respiratory pathogens were found to have significantly increased expression of immunity related gene biomarkers relative to those without (p=0.022) as well as lower alpha diversity of their respiratory microbial communities (p=0.017). CONCLUSIONS: Compared to conventional diagnostics, host/pathogen mNGS enhanced detection of microbial pathogens in BAL fluid from HCT patients. Furthermore, this approach simultaneously evaluated the association between identified microbes and the expression of innate immunity gene biomarkers. Host/pathogen mNGS holds promise for precision diagnosis of post-HCT respiratory infection.