Clinical Implementation and Value of Plasma Metagenomic Sequencing in Immunocompromised Pediatric Patients

Abstract

Abstract Background Metagenomic next-generation sequencing (mNGS) of cell-free DNA from plasma is emerging as a promising diagnostic tool in immunocompromised hosts but its performance characteristics in specific patient populations and clinical syndromes have not been systematically compared. Here, we describe the implementation and clinically utility of plasma mNGS in a large cohort of pediatric patients who are immunocompromised. Methods We performed retrospective review of all plasma mNGS testing (Karius®) performed at University of Minnesota between December 2017 – December 2022. Study design was reviewed by UMN IRB and determined to be nonhuman subjects research. For this study, we performed a nested subgroup secondary analysis of all patients aged <18 years at time of mNGS test who had underlying conditions that included hematologic malignancy, solid organ malignancy, hematopoietic stem cell transplantation or solid organ transplantations. Electronic medical records were reviewed for each mNGS test, recording patient demographics, underlying conditions, indications, and conventional microbiological results within 30 days of mNGS. The primary endpoint was change in clinical management as defined by Hogan et al and adjudicated by 2 independent reviewers. Secondary end points included sensitivity, time to mNGS order and time to diagnosis. Results We identified 71 unique pediatric patients with an immune compromised state who had 104 plasma mNGS tests performed between them. Fever was the most common indication for plasma mNGS (66%, 69/104), followed by Pulmonary Syndrome/Infiltrate/Lesion (23%, 22/104). The median time to plasma mNGS ordering from symptom onset was 8 days (IQR 3-17). Pediatric Infectious Diseases was consulted 70% (73/104) of the time. In cases where a microbiological cause(s) were identified, plasma mNGS identified the organism 48% of the time compared to 74% in conventional testing. Plasma mNGS led to a change in clinical management in 14 cases (13%,14/104) ordered in 13 unique patients. In 8 cases this led to a new diagnosis not confirmed by conventional microbiological testing or an earlier diagnosis, later confirmed by conventional testing. The new or earlier diagnoses cases included detection of invasive fungal/mold infections, anaerobic infections, and parasitic infections and were ordered on average within 2 days of symptom onset. Negative or DNA virus results on the plasma mNGS led to de-escalation of antimicrobials in 28% (4/13) of the positive impact cases. Conclusions Here, we describe the implementation of plasma mNGS in large immunocompromised pediatric cohort. Plasma mNGS led to a change in clinical management in a minority of cases (13%). However, these cases demonstrate that when ordered early in the clinical syndromes, plasma mNGS testing can detect difficult to diagnose organisms and promote antimicrobial stewardship. More prospective studies with early implementation of plasma mNGS in immunocompromised pediatric patients will be informative of the clinical value of this powerful diagnostic tool.