Detecting fecal microbiota transplantation–associated infection transmission using shotgun metagenomic sequencing and clonality analysis

Abstract

Background: Fecal microbiota transplantation (FMT) is a widely used modality for safe and effective treatment of recurrent Clostridium difficile infections, and FMT is being explored for the treatment of additional indications including gastrointestinal diseases and neurological disorders. Although microbiota-based therapies like FMT utilize rigorous donor screening procedures, these procedures are limited in resolution and scope, and there remains a risk of transmission of FMT-associated infectious agents from donor stool to a FMT recipient. Critically, these health concerns led the FDA to issue a 2019 safety alert for the transmission risks associated with FMT and to update its guidelines for screening and reporting. In a suspected transmission event, there is uncertainty around the source of infection; thus, methods are needed to rapidly determine whether a patient’s infection is linked to the donor stool product. Methods: Here, we developed a laboratory service sequencing and bioinformatics pipeline within our CLIA-certified laboratory for investigating suspected FMT infection transmission by measuring genomic relatedness. Our pipeline performs deep sequencing of a metagenomic sample, whole-genome sequencing (WGS) of an isolate derived from the implicated patient infection and determines the genomic relatedness between the 2 using a SNP-based analysis. The workflow was validated in silico with synthetic metagenomic samples spiked-in with WGS of clinically relevant isolate strains at varying abundance. Results: The sample and sequencing library preparation workflow was optimized across a panel of metagenomic and mock fecal microbiome samples demonstrating reproducible and reduced-bias sequencing of metagenomic samples. Our pipeline demonstrates high sensitivity and specificity for clonality calls when a spiked in isolate genome achieves 5× depth for >50% of the genome. We also demonstrated an interplay between abundance rate and sequencing depth for determining a clonality limit of detection. Conclusions: Taken together, our pipeline represents a new method that can support the clinical efforts of FMT and other microbiota-based therapies.