Multilaboratory assessment of metagenomic next-generation sequencing for unbiased microbe detection

The effects of sequencing strategies on Metagenomic pathogen detection using bronchoalveolar lavage fluid samples

Objectives

Metagenomic next-generation sequencing (mNGS) is a powerful tool for pathogen detection. The accuracy depends on both wet lab and dry lab procedures. The objective of our study was to assess the influence of read length and dataset size on pathogen detection.

Methods

In this study, 43 clinical BALF samples, which tested positive via clinical mNGS and were consistent with the diagnosis, were subjected to re-sequencing on the Illumina NovaSeq 6000 platform. The raw re-sequencing data, consisting of 100 million (M) paired-end 150 bp (PE150) reads, were divided into simulated datasets with eight different data sizes (5 M, 10 M, 15 M, 20 M, 30 M, 50 M, 75 M, 100 M) and five different read lengths (single-end 50 bp (SE50), SE75, SE100, PE100, and PE150). Both Kraken2 and IDseq bioinformatics pipelines were employed to analyze the previously diagnosed pathogens in the simulated data. Detection of pathogens was based on read counts ranging from 1 to 10 and RPM values ranging from 0.2 to 2.

Results

Our results revealed that increasing dataset sizes and read lengths can enhance the performance of mNGS in pathogen detection. However, a larger data sizes for mNGS require higher economic costs and longer turnaround time for data analysis. Our findings indicate 20 M reads being sufficient for SE75 mode to achieve high recall rates. Additionally, high nucleic acid loads in samples can lead to increased stability in pathogen detection efficiency, reducing the impact of sequencing strategies. The choice of bioinformatics pipelines had a significant impact on recall rates achieved in pathogen detection.

Conclusions

Increasing dataset sizes and read lengths can enhance the performance of mNGS in pathogen detection but increase the economic and time costs of sequencing and data analysis. Currently, the 20 M reads in SE75 mode may be the best sequencing option.

PHDtools: A platform for pathogen detection and multi-dimensional genetic signatures decoding to realize pathogen genomics data analyses online

OBJECTIVES

Facing the emerging diseases, rapid identification of the pathogen and multi-dimensional characterization of the genomic features at both isolate-level and population-level through high-throughput sequencing data can provide invaluable information to guide the development of antiviral agents and strategies. However, a user-friendly program is in urgent need for clinical laboratories without bioinformatics background to decode the complex big genomics data.

METHODS

In this study, we developed an interactive online platform named PHDtools with a total of 15 functions to analyze metagenomics data to identify the potential pathogen and decode multi-dimensional genetic signatures including intra-/inter-host variations and lineage-level variations. The platform was applied to analyze the meta-genomic data of the samples collected from the 172 imported COVID-19 cases.

RESULTS

According to the analytical results of mNGS, 27 patients were found to have the co-infections of SARS-CoV-2 with either influenza virus (n = 9) or human picobirnavirus (n = 19). Enough coverages of all the assembled SARS-CoV-2 genomes provided the sub-lineages of Omicron variant, and the number of mutations in the non-structural genes and M gene was increased, as well as the intra-host variations occurred in E and M gene were under positive selection (Ka/Ks > 1). These findings of increased or changed mutations in the SARS-CoV-2 genome characterized the current adaptive evolution patterns of Omicron sub-lineages, and revealed the evolution speed of these sub-lineages might increase.

CONCLUSIONS

Consequently, the application of PHDtools has proved that this platform is accurate, user-friendly and convenient for clinical users who are deficient in bioinformatics, and the clinical monitor of SARS-CoV-2 genomes by PHDtools also highlighted the potential evolution features of current SARS-CoV-2 and indicated that the development of anti-SARS-CoV-2 agents and new-designed vaccines should incorporate the gene variations other than S gene.

Application of metagenomic next-generation sequencing technology in the etiological diagnosis of peritoneal dialysis-associated peritonitis

Pathogens detected by metagenomic next-generation sequencing (mNGS) and the laboratory blood culture flask method were compared to understand the advantages and clinical significance of mNGS assays in the etiological diagnosis of peritoneal dialysis-associated peritonitis (PDAP). The study involved a total of 37 patients from the hospital's peritoneal dialysis centre, six of whom were patients with non-peritoneal dialysis-associated peritonitis. Peritoneal dialysis samples were collected from the 37 patients, who were divided into two groups. One group's samples were cultured using conventional blood culture flasks, and the other samples underwent pathogen testing using mNGS. The results showed that the positive rate of mNGS was 96.77%, while that of the blood culture flask method was 70.97% (p < 0.05). A total of 29 pathogens were detected by mNGS, namely 24 bacteria, one fungus, and four viruses. A total of 10 pathogens were detected using the bacterial blood culture method, namely nine bacteria and one fungus. The final judgment of the PDAP's causative pathogenic microorganism was made by combining the clinical condition, response to therapy, and the whole-genome sequencing findings. For mNGS, the sensitivity was 96.77%, the specificity was 83.33%, the positive predictive value was 96.77%, and the negative predictive value was 83.33%. For the blood culture flask method, the sensitivity was 70.97%, the specificity was 100%, the positive predictive value was 100%, and the negative predictive value was 0%. In conclusion, mNGS had a shorter detection time for diagnosing peritoneal dialysis-related peritonitis pathogens, with a higher positive rate than traditional bacterial cultures, providing significant advantages in diagnosing rare pathogens.

Combination of metagenomic next-generation sequencing and conventional tests unraveled pathogen profiles in infected patients undergoing allogeneic hematopoietic stem cell transplantation in Jilin Province of China

Background

Infection is the main cause of death for patients after allogeneic hematopoietic stem cell transplantation (HSCT). However, pathogen profiles still have not been reported in detail due to their heterogeneity caused by geographic region.

Objective

To evaluate the performance of metagenomic next-generation sequencing (mNGS) and summarize regional pathogen profiles of infected patients after HSCT.

Methods

From February 2021 to August 2022, 64 patients, admitted to the Department of Hematology of The First Hospital of Jilin University for HSCT and diagnosed as suspected infections, were retrospectively enrolled.

Results

A total of 38 patients were diagnosed as having infections, including bloodstream (n =17), pulmonary (n =16), central nervous system (CNS) (n =4), and chest (n =1) infections. Human betaherpesvirus 5 (CMV) was the most common pathogen in both bloodstream (n =10) and pulmonary (n =8) infections, while CNS (n =2) and chest (n =1) infections were mainly caused by Human gammaherpesvirus 4 (EBV). For bloodstream infection, Mycobacterium tuberculosis complex (n =3), Staphylococcus epidermidis (n =1), and Candida tropicalis (n =1) were also diagnosed as causative pathogens. Furthermore, mNGS combined with conventional tests can identify more causative pathogens with high sensitivity of 82.9% (95% CI 70.4-95.3%), and the total coincidence rate can reach up to 76.7% (95% CI 64.1-89.4%).

Conclusions

Our findings emphasized the importance of mNGS in diagnosing, managing, and ruling out infections, and an era of more rapid, independent, and impartial diagnosis of infections after HSCT can be expected.

ddPCR provides a sensitive test compared with GeneXpert MTB/RIF and mNGS for suspected Mycobacterium tuberculosis infection

Introduction

The Metagenomics next-generation sequencing (mNGS) and GeneXpert MTB/RIF assay (Xpert) exhibited a sensitivity for tuberculosis (TB) diagnostic performance. Research that directly compared the clinical performance of ddPCR analysis, mNGS, and Xpert in mycobacterium tuberculosis complex (MTB) infection has not been conducted.

Methods

The study aimed to evaluate the diagnostic performance of ddPCR compared to mNGS and Xpert for the detection of MTB in multiple types of clinical samples. The final clinical diagnosis was used as the reference standard.

Results

Out of 236 patients with suspected active TB infection, 217 underwent synchronous testing for tuberculosis using ddPCR, Xpert, and mNGS on direct clinical samples. During follow-up, 100 out of 217 participants were diagnosed with MTB infection. Compared to the clinical final diagnosis, ddPCR produced the highest sensitivity of 99% compared with mNGS (86%) and Xpert (64%) for all active MTB cases.

Discussion

Twenty-two Xpert-negative samples were positive in mNGS tests, which confirmed the clinical diagnosis results from ddPCR and clinical manifestation, radiologic findings. Thirteen mNGS-negative samples were positive in ddPCR assays, which confirmed the clinical final diagnosis.ddPCR provides a higher sensitive compared to Xpert and mNGS for MTB diagnosis, as defined by the high concordance between ddPCR assay and clinical final diagnosis.

Assessing the Quality of Metagenomic Next-Generation Sequencing for Pathogen Detection in Lower Respiratory Infections

BACKGROUND

Laboratory-developed metagenomic next-generation sequencing (mNGS) assays are increasingly being used for the diagnosis of infectious disease. To ensure comparable results and advance the quality control for the mNGS assay, we initiated a large-scale multicenter quality assessment to scrutinize the ability of mNGS to detect pathogens in lower respiratory infections.

METHODS

A reference panel containing artificial microbial communities and real clinical samples was used to assess the performance of 122 laboratories. We comprehensively evaluated the reliability, the source of false-positive and false-negative microbes, as well as the ability to interpret the results.

RESULTS

A wide variety of weighted F1-scores was observed across 122 participants, with a range from 0.20 to 0.97. The majority of false positive microbes (68.56%, 399/582) were introduced from "wet lab." The loss of microbial sequence during wet labs was the chief cause (76.18%, 275/361) of false-negative errors. When the human context is 2 × 105 copies/mL, most DNA and RNA viruses at titers above 104 copies/mL could be detected by >80% of the participants, while >90% of the laboratories could detect bacteria and fungi at titers lower than 103 copies/mL. A total of 10.66% (13/122) to 38.52% (47/122) of the participants could detect the target pathogens but failed to reach a correct etiological diagnosis.

CONCLUSIONS

This study clarified the sources of false-positive and false-negative results and evaluated the performance of interpreting the results. This study was valuable for clinical mNGS laboratories to improve method development, avoid erroneous results being reported, and implement regulatory quality controls in the clinic.

Rapid Diagnostics of Joint Infections Using IS-Pro

Diagnosis of bone and joint infections (BJI) relies on microbiological culture which has a long turnaround time and is challenging for certain bacterial species. Rapid molecular methods may alleviate these obstacles. Here, we investigate the diagnostic performance of IS-pro, a broad-scope molecular technique that can detect and identify most bacteria to the species level. IS-pro additionally informs on the amount of human DNA present in a sample, as a measure of leukocyte levels. This test can be performed in 4 h with standard laboratory equipment. Residual material of 591 synovial fluid samples derived from native and prosthetic joints from patients suspected of joint infections that were sent for routine diagnostics was collected and subjected to the IS-pro test. Bacterial species identification as well as bacterial load and human DNA load outcomes of IS-pro were compared to those of culture. At sample level, percent positive agreement (PPA) between IS-pro and culture was 90.6% (95% CI 85.7- to 94%) and negative percent agreement (NPA) was 87.7% (95% CI 84.1 to 90.6%). At species level PPA was 80% (95% CI 74.3 to 84.7%). IS-pro yielded 83 extra bacterial detections over culture for which we found supporting evidence for true positivity in 40% of the extra detections. Missed detections by IS-pro were mostly related to common skin species in low abundance. Bacterial and human DNA signals measured by IS-pro were comparable to bacterial loads and leukocyte counts reported by routine diagnostics. We conclude that IS-pro showed an excellent performance for fast diagnostics of bacterial BJI.

The Real-World Clinical Impact of Plasma mNGS Testing: an Observational Study

Plasma metagenomic next-generation sequencing (mNGS) testing is a promising diagnostic modality for infectious diseases, but its real-world clinical impact is poorly understood. We reviewed patients who had undergone plasma mNGS at a general hospital to evaluate the clinical utility of plasma mNGS testing. A total of 76.9% (113/147) of plasma mNGS tests had a positive result. A total of 196 microorganisms (58) were identified and reported, of which 75.6% (148/196) were clinically relevant. The median stringent mapped read number (SMRN) of clinically relevant organisms was 88 versus 22 for irrelevant organisms (P = 0.04). Based on the clinically adjudicated diagnosis, the positive and negative percent agreements of plasma mNGS testing for identifying a clinically defined infection were 95.2% and 67.4%, respectively. The plasma mNGS results led to a positive impact in 83 (57.1%) patients by diagnosing or ruling out infection and initiating targeted therapy. However, only 32.4% (11/34) of negative mNGS tests showed a positive impact, suggesting that plasma mNGS testing alone may not be a powerful tool to rule out infection in clinical practice. In the subset of 37 patients positive for both plasma mNGS and conventional testing, mNGS identified the pathogen(s) 2 days (IQR = 0.75 to 4.25) earlier than conventional testing. mNGS enables pathogen identification within 24 h, but given that the detection of clinically irrelevant organisms and nearly half of the tests result in no or a negative clinical impact, more clinical practice and studies are required to better understand who and when to test and how to optimally integrate mNGS into the infectious disease diagnostic workup. IMPORTANCE In this study, we show that although plasma mNGS testing significantly improved the detection rate of tested samples, nearly one in four (24.5%, 48/196) mNGS tests reported organisms were not clinically relevant, emphasizing the importance of cautious interpretation and infectious disease consultation. Moreover, based on clinical adjudication, plasma mNGS testing resulted in no or a negative impact in nearly half (43.5%, 64/147) of patients in the current study, indicating that how best to integrate this advanced method into current infectious disease diagnostic frameworks to maximize its clinical utility in real-world practice is an important question. Therefore, recommending plasma mNGS testing as a routine supplement to first-line diagnostic tests for infectious diseases faces great challenges. The decision to conduct mNGS testing should take into account the diagnostic performance, turnaround time and cost-effectiveness of mNGS, as well as the availability of conventional tests.

Pulmonary nocardiosis with bloodstream infection diagnosed by metagenomic next-generation sequencing in a kidney transplant recipient: A case report

BACKGROUND

Pulmonary nocardiosis is difficult to diagnose by culture and other conventional testing, and is often associated with lethal disseminated infections. This difficulty poses a great challenge to the timeliness and accuracy of clinical detection, especially in susceptible immunosuppressed individuals. Metagenomic next-generation sequencing (mNGS) has transformed the conventional diagnosis pattern by providing a rapid and precise method to assess all microorganisms in a sample.

CASE SUMMARY

A 45-year-old male was hospitalized for cough, chest tightness and fatigue for 3 consecutive days. He had received a kidney transplant 42 d prior to admission. No pathogens were detected at admission. Chest computed tomography showed nodules, streak shadows and fiber lesions in both lung lobes as well as right pleural effusion. Pulmonary tuberculosis with pleural effusion was highly suspected based on the symptoms, imaging and residence in a high tuberculosis-burden area. However, anti-tuberculosis treatment was ineffective, showing no improvement in computed tomography imaging. Pleural effusion and blood samples were subsequently sent for mNGS. The results indicated Nocardia farcinica as the major pathogen. After switching to sulphamethoxazole combined with minocycline for anti-nocardiosis treatment, the patient gradually improved and was finally discharged.

CONCLUSION

A case of pulmonary nocardiosis with an accompanying bloodstream infection was diagnosed and promptly treated before the dissemination of the infection. This report emphasizes the value of mNGS in the diagnosis of nocardiosis. mNGS may be an effective method for facilitating early diagnosis and prompt treatment in infectious diseases, which overcomes the shortcomings of conventional testing.

Validation of a Metagenomic Next-Generation Sequencing Assay for Lower Respiratory Pathogen Detection

Lower respiratory infection (LRI) is the most fatal communicable disease, with only a few pathogens identified. Metagenomic next-generation sequencing (mNGS), as an unbiased, hypothesis-free, and culture-independent method, theoretically enables the detection of all pathogens in a single test. In this study, we developed and validated a DNA-based mNGS method for the diagnosis of LRIs from bronchoalveolar lavage fluid (BALF). We prepared simulated in silico data sets and published raw data sets from patients to evaluate the performance of our in-house bioinformatics pipeline and compared it with the popular metagenomics pipeline Kraken2-Bracken. In addition, a series of biological microbial communities were used to comprehensively validate the performance of our mNGS assay. Sixty-nine clinical BALF samples were used for clinical validation to determine the accuracy. The in-house bioinformatics pipeline validation showed a recall of 88.03%, precision of 99.14%, and F1 score of 92.26% via single-genome simulated data. Mock in silico microbial community and clinical metagenomic data showed that the in-house pipeline has a stricter cutoff value than Kraken2-Bracken, which could prevent false-positive detection by the bioinformatics pipeline. The validation for the whole mNGS pipeline revealed that overwhelming human DNA, long-term storage at 4°C, and repeated freezing-thawing reduced the analytical sensitivity of the assay. The mNGS assay showed a sensitivity of 95.18% and specificity of 91.30% for pathogen detection from BALF samples. This study comprehensively demonstrated the analytical performance of this laboratory-developed mNGS assay for pathogen detection from BALF, which contributed to the standardization of this technology. IMPORTANCE To our knowledge, this study is the first to comprehensively validate the mNGS assay for the diagnosis of LRIs from BALF. This study exhibited a ready-made example for clinical laboratories to prepare reference materials and develop comprehensive validation schemes for their in-house mNGS assays, which would accelerate the standardization of mNGS testing.

Comparison of metagenomic next-generation sequencing using cell-free DNA and whole-cell DNA for the diagnoses of pulmonary infections

Although the fast-growing metagenomic next-generation sequencing (mNGS) has been used in diagnosing infectious diseases, low detection rate of mNGS in detecting pathogens with low loads limits its extensive application. In this study, 130 patients with suspected pulmonary infections were enrolled, from whom bronchoalveolar lavage fluid (BALF) samples were collected. The conventional tests and mNGS of cell-free DNA (cfDNA) and whole-cell DNA (wcDNA) using BALF were simultaneously performed. mNGS of cfDNA showed higher detection rate (91.5%) and total coincidence rate (73.8%) than mNGS of wcDNA (83.1% and 63.9%) and conventional methods (26.9% and 30.8%). A total of 70 microbes were detected by mNGS of cfDNA, and most of them (60) were also identified by mNGS of wcDNA. The 31.8% (21/66) of fungi, 38.6% (27/70) of viruses, and 26.7% (8/30) of intracellular microbes can be only detected by mNGS of cfDNA, much higher than those [19.7% (13/66), 14.3% (10/70), and 6.7% (2/30)] only detected by mNGS of wcDNA. After in-depth analysis on these microbes with low loads set by reads per million (RPM), we found that more RPM and fungi/viruses/intracellular microbes were detected by mNGS of cfDNA than by mNGS of wcDNA. Besides, the abilities of mNGS using both cfDNA and wcDNA to detect microbes with high loads were similar. We highlighted the advantage of mNGS using cfDNA in detecting fungi, viruses, and intracellular microbes with low loads, and suggested that mNGS of cfDNA could be considered as the first choice for diagnosing pulmonary infections.

Clinical evaluation of metagenomic next-generation sequencing for detecting pathogens in bronchoalveolar lavage fluid collected from children with community-acquired pneumonia

This study is to evaluate the usefulness of pathogen detection using metagenomic next-generation sequencing (mNGS) on bronchoalveolar lavage fluid (BALF) specimens from children with community-acquired pneumonia (CAP). We retrospectively collected BALF specimens from 121 children with CAP at Tianjin Children's Hospital from February 2021 to December 2021. The diagnostic performances of mNGS and conventional tests (CT) (culture and targeted polymerase chain reaction tests) were compared, using composite diagnosis as the reference standard. The results of mNGS and CT were compared based on pathogenic and non-pathogenic organisms. Pathogen profiles and co-infections between the mild CAP and severe CAP groups were also analyzed. The overall positive coincidence rate was 86.78% (105/121) for mNGS and 66.94% (81/121) for CT. The proportion of patients diagnosed using mNGS plus CT increased to 99.18%. Among the patients, 17.36% were confirmed only by mNGS; Streptococcus pneumoniae accounted for 52.38% and 23.8% of the patients were co-infected. Moreover, Bordetella pertussis and Human bocavirus (HBoV) were detected only using mNGS. Mycoplasma pneumoniae, which was identified in 89 (73.55%) of 121 children with CAP, was the most frequent pathogen detected using mNGS. The infection rate of M. pneumoniae in the severe CAP group was significantly higher than that in the mild CAP group (P = 0.007). The symptoms of single bacterial infections (except for mycoplasma) were milder than those of mycoplasma infections. mNGS identified more bacterial infections when compared to the CT methods and was able to identify co-infections which were initially missed on CT. Additionally, it was able to identify pathogens that were beyond the scope of the CT methods. The mNGS method is a powerful supplement to clinical diagnostic tools in respiratory infections, as it can increase the precision of diagnosis and guide the use of antibiotics.