Metagenomic identification of pathogens and antimicrobial-resistant genes in bacterial positive blood cultures by nanopore sequencing

Next-generation diagnostics of bloodstream infections enabled by rapid whole-genome sequencing of bacterial cells purified from blood cultures

BACKGROUND

Blood culture (BC) remains the cornerstone for diagnosis of bloodstream infections (BSI), but the long turn-around time (TAT) hampers timely selection of appropriate chemotherapy. Novel molecular approaches have been developed to provide faster results but are also affected by limitations. We developed a analytical workflow named LC-WGS (Whole-Genome Sequencing of Liquid Colony) for rapid whole-genome sequencing-based diagnosis of BSI, evaluating its accuracy performance over standard of care (SoC) diagnostic procedures.

METHODS

A total of 85 prospectively collected positive BC were processed in parallel with SoC (subculturing, identification by MALDI-ToF, antimicrobial susceptibility testing by reference broth microdilution, usage of syndromic panels) and LC-WGS, which relied on automated purification of microbial cells (Qvella FAST system, Qvella Corp.), DNA purification, and real-time sequencing with the Oxford Nanopore MinION. A streamlined analysis pipeline was designed for pathogen identification (Kraken2), detection of resistance markers (KmerResistance, AMRFinderPlus), virulome profiling (abricate, VFDB), phylogenetic analysis (snippy, IQ-TREE), and pathogen subtyping (Meningotype).

FINDINGS

Compared with SoC, LC-WGS returned accurate species-level identification for 98% (65/66) of monomicrobial and 88% (14/16) of polymicrobial BCs, with a TAT as short as ∼2·6 h. Accurate resistome profiling (allelic variants) was achieved for 94% (58/62) of the most clinically-relevant resistance profiles in ∼4·2 h. In silico serotying (Neisseria meningitidis), virulotyping (Escherichia coli, Klebsiella pneumoniae) and comparative phylogenomics for outbreak investigation (K. pneumoniae) proved also feasible.

INTERPRETATION

In this proof-of-concept study, we proved that diagnosis of BSI can be significantly shortened using an optimised workflow based on real-time sequencing, providing rapid, actionable clinical microbiological data in support of timely selection of appropriate chemotherapy. LC-WGS proved also useful as molecular epidemiology tool for public health and infection control applications.

FUNDING

This study was partially supported by an investigator-initiated grant from Qvella Corporation.

Integrating sequencing methods with machine learning for antimicrobial susceptibility testing in pediatric infections: current advances and future insights

Antimicrobial resistance (AMR) presents a critical challenge in clinical settings, particularly among pediatric patients with life-threatening conditions such as sepsis, meningitis, and neonatal infections. The increasing prevalence of multi- and pan-resistant pathogens is strongly associated with adverse clinical outcomes. Recent technological advances in sequencing methods, including metagenomic next-generation sequencing (mNGS), Oxford Nanopore Technologies (ONT), and targeted sequencing (TS), have significantly enhanced the detection of both pathogens and their associated resistance genes. However, discrepancies between resistance gene detection and antimicrobial susceptibility testing (AST) often hinder the direct clinical application of sequencing results. These inconsistencies may arise from factors such as genetic mutations or variants in resistance genes, differences in the phenotypic expression of resistance, and the influence of environmental conditions on resistance levels, which can lead to variations in the observed resistance patterns. Machine learning (ML) provides a promising solution by integrating large-scale resistance data with sequencing outcomes, enabling more accurate predictions of pathogen drug susceptibility. This review explores the application of sequencing technologies and ML in the context of pediatric infections, with a focus on their potential to track the evolution of resistance genes and predict antibiotic susceptibility. The goal of this review is to promote the incorporation of ML-based predictions into clinical practice, thereby improving the management of AMR in pediatric populations.

Applying nanopore sequencing in the etiological diagnosis of bloodstream infection

Bloodstream infection (BSI) is a systemic infectious disease that can lead to shock, disseminated intravascular coagulation, multiorgan failure, and even death. Blood culture is considered the gold standard for the etiological diagnosis of BSI; however, blood culture is time-consuming and has a low positivity rate, which has limited its utility for early and rapid clinical diagnosis. Nanopore sequencing technology (NST), a third-generation sequencing method, offers rapid detection, real-time single-molecule sequencing, and ultra-long reads. These features enable the prompt detection of pathogens and the analysis of drug-resistant genes and genomic characteristics, thereby optimizing the clinical diagnosis and treatment of BSI. In this article, we summarize the application of NST in the etiological diagnosis of BSI.

Bacterial enrichment prior to third-generation metagenomic sequencing improves detection of BRD pathogens and genetic determinants of antimicrobial resistance in feedlot cattle

Introduction

Bovine respiratory disease (BRD) is one of the most important animal health problems in the beef industry. While bacterial culture and antimicrobial susceptibility testing have been used for diagnostic testing, the common practice of examining one isolate per species does not fully reflect the bacterial population in the sample. In contrast, a recent study with metagenomic sequencing of nasal swabs from feedlot cattle is promising in terms of bacterial pathogen identification and detection of antimicrobial resistance genes (ARGs). However, the sensitivity of metagenomic sequencing was impeded by the high proportion of host biomass in the nasal swab samples.

Methods

This pilot study employed a non-selective bacterial enrichment step before nucleic acid extraction to increase the relative proportion of bacterial DNA for sequencing.

Results

Non-selective bacterial enrichment increased the proportion of bacteria relative to host sequence data, allowing increased detection of BRD pathogens compared with unenriched samples. This process also allowed for enhanced detection of ARGs with species-level resolution, including detection of ARGs for bacterial species of interest that were not targeted for culture and susceptibility testing. The long-read sequencing approach enabled ARG detection on individual bacterial reads without the need for assembly. Metagenomics following non-selective bacterial enrichment resulted in substantial agreement for four of six comparisons with culture for respiratory bacteria and substantial or better correlation with qPCR. Comparison between isolate susceptibility results and detection of ARGs was best for macrolide ARGs in Mannheimia haemolytica reads but was also substantial for sulfonamide ARGs within M. haemolytica and Pasteurella multocida reads and tetracycline ARGs in Histophilus somni reads.

Discussion

By increasing the proportion of bacterial DNA relative to host DNA through non-selective enrichment, we demonstrated a corresponding increase in the proportion of sequencing data identifying BRD-associated pathogens and ARGs in deep nasopharyngeal swabs from feedlot cattle using long-read metagenomic sequencing. This method shows promise as a detection strategy for BRD pathogens and ARGs and strikes a balance between processing time, input costs, and generation of on-target data. This approach could serve as a valuable tool to inform antimicrobial management for BRD and support antimicrobial stewardship.

Illumina and Nanopore sequencing in culture-negative samples from suspected lower respiratory tract infection patients

Objective

To evaluate the diagnostic value of metagenomic sequencing technology based on Illumina and Nanopore sequencing platforms for patients with suspected lower respiratory tract infection whose pathogen could not be identified by conventional microbiological tests.

Methods

Patients admitted to the Respiratory and Critical Care Medicine in Shanghai Ruijin Hospital were retrospectively studied from August 2021 to March 2022. Alveolar lavage or sputum was retained in patients with clinically suspected lower respiratory tract infection who were negative in conventional tests. Bronchoalveolar lavage fluid (BALF) samples were obtained using bronchoscopy. Sputum samples were collected, while BALF samples were not available due to bronchoscopy contraindications. Samples collected from enrolled patients were simultaneously sent for metagenomic sequencing on both platforms.

Results

Thirty-eight patients with suspected LRTI were enrolled in this study, consisting of 36 parts of alveolar lavage and 2 parts of sputum. According to the infection diagnosis, 31 patients were confirmed to be infected with pathogens, while 7 patients were diagnosed with non-infectious disease. With regard to the diagnosis of infectious diseases, the sensitivity and specificity of Illumina and Nanopore to diagnose infection in patients were 80.6% vs. 93.5% and 42.9 vs. 28.6%, respectively. In patients diagnosed with bacterial, Mycobacterium, and fungal infections, the positive rates of Illumina and Nanopore sequencer were 71.4% vs. 78.6%, 36.4% vs. 90.9%, and 50% vs. 62.5%, respectively. In terms of pathogen diagnosis, the sensitivity and specificity of pathogens detected by Illumina and Nanopore were 55.6% vs. 77.8% and 42.9% vs. 28.6%, respectively. Among the patients treated with antibiotics in the last 2 weeks, 61.1% (11/18) and 77.8% (14/18) cases of pathogens were accurately detected by Illumina and Nanopore, respectively, among which 8 cases were detected jointly. The consistency between Illumina and diagnosis was 63.9% (23/36), while the consistency between Nanopore and diagnosis was 83.3% (30/36). Between Illumina and Nanopore sequencing methods, the consistency ratio was 55% (22/42) based on pathogen diagnosis.

Conclusion

Both platforms play a certain value in infection diagnosis and pathogen diagnosis of CMT-negative suspected LRTI patients, providing a theoretical basis for clinical accurate diagnosis and symptomatic treatment. The Nanopore platform demonstrated potential advantages in the identification of Mycobacterium and could further provide another powerful approach for patients with suspected Mycobacterium infection.