Advertisement

Utility of plasma cell-free DNA next-generation sequencing for diagnosis of infectious diseases in patients with hematological disorders

  • Author Footnotes
    1 These authors have contributed equally to this work and share the first authorship.
    Chunhui Xu
    Footnotes
    1 These authors have contributed equally to this work and share the first authorship.
    Affiliations
    State Key Laboratory of Experimental Hematology, National Clinical Research Center for Blood Diseases, Haihe Laboratory of Cell Ecosystem, Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Tianjin, China

    Microbiology laboratory, Tianjin Union Precision Medical Diagnostic Co., Ltd, 301617, Tianjin, China
    Search for articles by this author
  • Author Footnotes
    1 These authors have contributed equally to this work and share the first authorship.
    Xin Chen
    Footnotes
    1 These authors have contributed equally to this work and share the first authorship.
    Affiliations
    State Key Laboratory of Experimental Hematology, National Clinical Research Center for Blood Diseases, Haihe Laboratory of Cell Ecosystem, Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Tianjin, China
    Search for articles by this author
  • Guoqing Zhu
    Affiliations
    State Key Laboratory of Experimental Hematology, National Clinical Research Center for Blood Diseases, Haihe Laboratory of Cell Ecosystem, Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Tianjin, China
    Search for articles by this author
  • Huiming Yi
    Affiliations
    State Key Laboratory of Experimental Hematology, National Clinical Research Center for Blood Diseases, Haihe Laboratory of Cell Ecosystem, Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Tianjin, China
    Search for articles by this author
  • Shulian Chen
    Affiliations
    State Key Laboratory of Experimental Hematology, National Clinical Research Center for Blood Diseases, Haihe Laboratory of Cell Ecosystem, Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Tianjin, China
    Search for articles by this author
  • Yuetian Yu
    Affiliations
    Department of Critical Care Medicine, Ren Ji Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    Search for articles by this author
  • Erlie Jiang
    Affiliations
    State Key Laboratory of Experimental Hematology, National Clinical Research Center for Blood Diseases, Haihe Laboratory of Cell Ecosystem, Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Tianjin, China
    Search for articles by this author
  • Yizhou Zheng
    Affiliations
    State Key Laboratory of Experimental Hematology, National Clinical Research Center for Blood Diseases, Haihe Laboratory of Cell Ecosystem, Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Tianjin, China
    Search for articles by this author
  • Fengkui Zhang
    Affiliations
    State Key Laboratory of Experimental Hematology, National Clinical Research Center for Blood Diseases, Haihe Laboratory of Cell Ecosystem, Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Tianjin, China
    Search for articles by this author
  • Jianxiang Wang
    Affiliations
    State Key Laboratory of Experimental Hematology, National Clinical Research Center for Blood Diseases, Haihe Laboratory of Cell Ecosystem, Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Tianjin, China
    Search for articles by this author
  • Sizhou Feng
    Correspondence
    Corresponding author at: Chinese Academy of Medical Sciences, Institute of Hematology and Blood Diseases Hospital, No. 288 Nanjing Road, Tianjin, China.
    Affiliations
    State Key Laboratory of Experimental Hematology, National Clinical Research Center for Blood Diseases, Haihe Laboratory of Cell Ecosystem, Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Tianjin, China
    Search for articles by this author
  • Author Footnotes
    1 These authors have contributed equally to this work and share the first authorship.
Open AccessPublished:November 30, 2022DOI:https://doi.org/10.1016/j.jinf.2022.11.020

      Highlights

      • Plasma NGS performed well in diagnosing infection in hematological patients.
      • A detailed grading system was devised to assess the clinical impact of NGS.
      • Combining NGS and conventional tests yielded a positivity rate of 75%.

      Abstract

      Background

      Plasma cell-free DNA Next-Generation Sequencing has been used as a non-invasive and comprehensive method for the etiological diagnosis of infectious diseases. However, only a handful of studies have described the real-world utility of this technique in patients with hematological disorders, a cohort of patients that are distinctive due to neutropenia and weakened immune functions.

      Methods

      We retrospectively analyzed the results of plasma cell-free DNA sequencing performed on 184 and 163 specimens collected from hematological patients suspected of infections with (Group I) or without (Group II) neutropenia, respectively. The diagnostic performance and the clinical impact of plasma sequencing were comparatively evaluated to conventional microbiological tests and a composite reference standard (conventional tests combined with the clinical assessment).

      Results

      The overall positive detection rate of plasma cell-free DNA sequencing was significantly higher than that of conventional microbiological tests (72.6% vs.31.4%, P < 0.001). The positive rate of conventional microbiological tests in Group I was lower than that in Group II (25.5% vs. 38.0%, P = 0.012). Combining plasma sequencing with conventional tests yielded a positive detection rate of 75.0% and 74.8% for these two groups, respectively. Using the composite reference standard, the sensitivity and specificity of plasma sequencing were 89.1% and 65.1%, respectively. The proportions of the positive impact of cell-free DNA sequencing results in the Group I were higher than in the Group II in terms of both diagnosis and treatment (diagnosis: 54.3% vs. 40.5%, P = 0.013; treatment: 45.7% vs.30.7%, P = 0.004). A total of 73 patients (21.0%) benefited from plasma sequencing through adjustment of the antibiotic regimen.

      Conclusions

      The diagnostic yield of conventional microbiological tests was low in patients with neutropenia. Combining conventional tests with plasma cell-free DNA sequencing significantly improved the detection rate for pathogens and optimized antibiotic treatment. Our findings on the clinical impact warrant confirmation through larger, multicenter, randomized controlled trials. Moreover, the cost-effectiveness of this testing strategy remains unknown and requires further exploration.

      Keywords

      Introduction

      Neutropenia is an independent risk factor for infection in patients with hematological disorders who have undergone hematopoietic stem cell transplantation (HSCT) and received cytotoxic chemotherapy or immunosuppressive therapy.
      • Aguilar-Guisado M.
      • Espigado I.
      • Martin-Pena A.
      • Gudiol C.
      • Royo-Cebrecos C.
      • Falantes J.
      • et al.
      Optimisation of empirical antimicrobial therapy in patients with haematological malignancies and febrile neutropenia (How Long study): an open-label, randomised, controlled phase 4 trial.
      • Kochanek M.
      • Schalk E.
      • von Bergwelt-Baildon M.
      • Beutel G.
      • Buchheidt D.
      • Hentrich M.
      • et al.
      Management of sepsis in neutropenic cancer patients: 2018 guidelines from the Infectious Diseases Working Party (AGIHO) and Intensive Care Working Party (iCHOP) of the German Society of Hematology and Medical Oncology (DGHO).
      • Ortega M.
      • Marco F.
      • Soriano A.
      • Almela M.
      • Martinez J.A.
      • Rovira M.
      • et al.
      Epidemiology and outcome of bacteraemia in neutropenic patients in a single institution from 1991 to 2012.
      Infection is one of the most common complications for patients with neutropenia and can lead to high mortality if inappropriate empirical antibiotic treatment is administered.
      • Martinez-Nadal G.
      • Puerta-Alcalde P.
      • Gudiol C.
      • Cardozo C.
      • Albasanz-Puig A.
      • Marco F.
      • et al.
      Inappropriate empirical antibiotic treatment in high-risk neutropenic patients with bacteremia in the era of multidrug resistance.
      ,
      • White L.
      • Ybarra M.
      Neutropenic fever.
      Accurate and timely pathogen detection is critical in optimizing antibiotic use; however, it is challenging in patients with neutropenia because of nonspecific clinical symptoms, signs, and the low positivity rate of conventional microbiological tests (CMT).
      • Schmidt-Hieber M.
      • Teschner D.
      • Maschmeyer G.
      • Schalk E.
      Management of febrile neutropenia in the perspective of antimicrobial de-escalation and discontinuation.
      ,
      • Chen S.C.
      • Kontoyiannis D.P.
      New molecular and surrogate biomarker-based tests in the diagnosis of bacterial and fungal infection in febrile neutropenic patients.
      CMT commonly includes culture, direct microscopic examination (DME), nucleic acid amplification tests (NAAT), and serological tests. However, time-consuming and low detection rates are the limitations of culture and DME, and only a few suspected pathogens can be detected by NAAT and serological tests.
      • Wang S.
      • Ai J.
      • Cui P.
      • Zhu Y.
      • Wu H.
      • Zhang W.
      Diagnostic value and clinical application of next-generation sequencing for infections in immunosuppressed patients with corticosteroid therapy.
      ,
      • Gu W.
      • Miller S.
      • Chiu C.Y.
      Clinical metagenomic next-generation sequencing for pathogen detection.
      A rapid and accurate method for pathogen detection is urgently needed.
      As a fast non-culture-based and unbiased method, microbial next-generation sequencing (NGS) has garnered considerable scientific attention in the field of etiological diagnosis since the first diagnosis of neuroleptospirosis.
      • Gu W.
      • Miller S.
      • Chiu C.Y.
      Clinical metagenomic next-generation sequencing for pathogen detection.
      • Wilson M.R.
      • Naccache S.N.
      • Samayoa E.
      • Biagtan M.
      • Bashir H.
      • Yu G.
      • et al.
      Actionable diagnosis of neuroleptospirosis by next-generation sequencing.
      • Yu X.
      • Jiang W.
      • Shi Y.
      • Ye H.
      • Lin J.
      Applications of sequencing technology in clinical microbial infection.
      Invasive specimens such as bronchoalveolar lavage fluid (BALF) and tissues are often unavoidable. However, an invasive sampling procedure can lead to the development of complications such as hypoxemia and endobronchial bleeding in immunocompromised patients, with complication rates ranging from 1% to 52%.
      • Choo R.
      • Naser N.S.H.
      • Nadkarni N.V.
      Anantham D. Utility of bronchoalveolar lavage in the management of immunocompromised patients presenting with lung infiltrates.
      According to the recommendations for the management of neutropenic patients in the intensive care unit, pneumothorax occurs in 4%∼20% of patients after transbronchial biopsies, and respiratory worsening is a common complication after bronchoscopy.
      • Schnell D.
      • Azoulay E.
      • Benoit D.
      • Clouzeau B.
      • Demaret P.
      • Ducassou S.
      • et al.
      Management of neutropenic patients in the intensive care unit (NEWBORNS EXCLUDED) recommendations from an expert panel from the French Intensive Care Society (SRLF) with the French Group for Pediatric Intensive Care Emergencies (GFRUP), the French Society of Anesthesia and Intensive Care (SFAR), the French Society of Hematology (SFH), the French Society for Hospital Hygiene (SF2H), and the French Infectious Diseases Society (SPILF).
      In recent years, plasma cell-free DNA (cfDNA) NGS, a non-invasive microbial identification technology, has shown great potential in the diagnosis of clinical infectious diseases such as bloodstream infection (BSI),
      • Goggin K.P.
      • Gonzalez-Pena V.
      • Inaba Y.
      • Allison K.J.
      • Hong D.K.
      • Ahmed A.A.
      • et al.
      Evaluation of plasma microbial cell-free DNA sequencing to predict bloodstream infection in pediatric patients with relapsed or refractory cancer.
      ,
      • Eichenberger E.M.
      • de Vries C.R.
      • Ruffin F.
      • Sharma-Kuinkel B.
      • Park L.
      • Hong D.
      • et al.
      Microbial cell-free DNA identifies etiology of bloodstream infections, persists longer than conventional blood cultures, and its duration of detection is associated with metastatic infection in patients with staphylococcus aureus and gram-negative bacteremia.
      invasive fungal infection,
      • Hong D.K.
      • Blauwkamp T.A.
      • Kertesz M.
      • Bercovici S.
      • Truong C.
      • Banaei N.
      Liquid biopsy for infectious diseases: sequencing of cell-free plasma to detect pathogen DNA in patients with invasive fungal disease.
      endocarditis,
      • To R.K.
      • Ramchandar N.
      • Gupta A.
      • Pong A.
      • Cannavino C.
      • Foley J.
      • et al.
      Use of plasma metagenomic next-generation sequencing for pathogen identification in pediatric endocarditis.
      ,
      • Lieberman J.A.
      • C Naureckas Li
      • Lamb G.S.
      • Kane D.A.
      • Stewart M.K.
      • Mamedov R.A.
      • et al.
      Case report: comparison of plasma metagenomics to bacterial PCR in a case of prosthetic valve endocarditis.
      sepsis,
      • Gosiewski T.
      • Ludwig-Galezowska A.H.
      • Huminska K.
      • Sroka-Oleksiak A.
      • Radkowski P.
      • Salamon D.
      • et al.
      Comprehensive detection and identification of bacterial DNA in the blood of patients with sepsis and healthy volunteers using next-generation sequencing method - the observation of DNAemia.
      and infection associated with chimeric antigen receptor-T.
      • Nie J.
      • Yang L.
      • Huang L.
      • Gao L.
      • Young K.H.
      • Le Grange J.M.
      • et al.
      Infection complications in febrile chimeric antigen receptor (CAR)-T recipients during the peri-CAR-T cell treatment period examined using metagenomic next-generation sequencing (mNGS).
      However, only a few studies have focused on the application of plasma cfDNA NGS in patients with hematological diseases, and the clinical impact of cfDNA NGS is unclear.
      • Horiba K.
      • Torii Y.
      • Okumura T.
      • Takeuchi S.
      • Suzuki T.
      • Kawada J.I.
      • et al.
      Next-generation sequencing to detect pathogens in pediatric febrile neutropenia: a single-center retrospective study of 112 cases.
      ,
      • Benamu E.
      • Gajurel K.
      • Anderson J.N.
      • Lieb T.
      • Gomez C.A.
      • Seng H.
      • et al.
      Plasma microbial cell-free DNA next generation sequencing in the diagnosis and management of febrile neutropenia.
      Therefore, in this study, we aimed to assess pathogen diagnostic ability and clinical impacts of plasma cfDNA NGS in patients with hematological diseases, particularly with neutropenia.

      Methods

      Study design

      We retrospectively reviewed the records of patients who undertook the plasma cfDNA NGS test for suspected infections at the Institute of Hematology and Blood Diseases Hospital—a specialist tertiary hematology center in Tianjin, China—between 28 December 2020 and 31 December 2021, in accordance with our inclusion and exclusion criteria (Fig. 1). This study was approved by the Institutional Review Board and Ethics Committee of the Institute of Hematology and Blood Diseases Hospital (IRB number: QTJC2022008-EC-1).
      The diagnosis of different types of infection was retrospectively made with reference to the CDC/NHSN surveillance definitions
      Prevention CfDCa
      CDC/NHSN Surveillance definitioins for specific types of infections.
      by the clinical committee individually after comprehensive consideration of the clinical symptoms, laboratory tests, radiological manifestations, CMT, and treatment response. Neutropenia was defined as an absolute neutrophil count (ANC) of <500/mm3.
      • Freifeld A.G.
      • Bow E.J.
      • Sepkowitz K.A.
      • Boeckh M.J.
      • Ito J.I.
      • Mullen C.A.
      • et al.
      Clinical practice guideline for the use of antimicrobial agents in neutropenic patients with cancer: 2010 Update by the Infectious Diseases Society of America.
      Previous antibiotic exposure was defined as the use of antibiotics at least 48 h within 3 days before the NGS test. Patients who received CMT ordered by their clinicians, including culture, sample DME, (1,3)-β-d-glucan test (BDG test), galactomannan test (GM test), cryptococcus antigen detection, and virus NAAT (Supplementary Table S1). Diagnostic performance and concordance were evaluated by comparison between NGS and CMT.

      Plasma cfDNA NGS results in adjudication and clinical impact evaluation on diagnosis and management

      The composite clinical standard was evaluated by the clinical committee mainly according to the standardized criteria of the Karius test research (Supplementary Figure S1).
      • Blauwkamp T.A.
      • Thair S.
      • Rosen M.J.
      • Blair L.
      • Lindner M.S.
      • Vilfan I.D.
      • et al.
      Analytical and clinical validation of a microbial cell-free DNA sequencing test for infectious disease.
      The plasma NGS results were classified as either definite, probable, possible, unlikely, or false-negative causes of the cases. (1) Definite: The plasma cfDNA NGS results were found to be consistent with those of CMT performed within 7 days of the NGS test; (2) Probable: microorganisms detected by NGS were probably the cause of infection; (3) Possible: microorganisms detected by NGS showed potential to cause infection, but not as a common cause as per the evaluation of clinical experts based on the consideration of clinical medical records; (4) Unlikely: microorganisms detected by NGS was not the possible cause of infection according to other clinical results or result was inconsistent with that of CMT; (5) False negative: NGS result was negative, but the case was assessed to be that of an infection.
      As an important part of this study, we evaluated the impact of NGS results on the diagnosis and treatment for clinical work. To achieve a relatively objective and credible goal, an independent clinical committee was created, which included two directors (haematologists, who made the final judgment), a clinical microbiologist, and a radiologist. The clinical committee retrospectively evaluated the clinical impact of NGS with reference to the Grading Criteria (Supplementary Table S2). The diagnostic impact evaluation was focused on the possibility of NGS results as a pathogen of infection, while the impact of treatment was assessed based on the antibiotic adjustment according to the NGS results within 7 days after receiving the NGS report.

      Plasma cell free-DNA next-generation sequencing

      The peripheral blood samples were carefully delivered to the local laboratory, and NGS was immediately performed. The nucleic acid was extracted from 600 uL of the plasma supernatant by using the TIANamp Micro DNA Kit (Tiangen Biotech). After adding the adapters, PCR amplification DNA libraries were constructed by using an end-repair method. The quality of the library was evaluated by quantitative PCR with Agilent 2100 (Agilent Technologies) and Qubit 4.0 (Thermo Fisher). The number of sequencing reads was >20 million for each library. After removing low-quality and short reads and human host sequences, the remaining data were aligned to those of the microbial genome database. Microbial genomes were obtained from the NCBI RefSeq and GenBank database and included 5224 bacterial genomes, 13,596 viral genomes, 512 fungal genomes, 171 parasite genomes, 155 Mycobacteria genomes, and 201 other intracellular genomes, such as Mycobacteria and Chlamydia.

      Criteria for reporting NGS detection

      Considering the lack of a recognized standard experimental process and the reporting of NGS interpretation and based on the expert consensus and other studies, we employed the following criteria in this study: (1) For bacteria and fungi, relative abundance at the genus level >30%. (2) For bacteria (excluding Mycobacterium tuberculosis), viruses, and fungi (excluding Pneumocystis jiroveci), a microbe was considered positive when the stringently mapped read number (SMRN) at the species level≥3. (3) For M. tuberculosis and P. jiroveci, it was considered positive when the reads were aligned specifically. (4) Parasite: SMRN ≥100. (5) The positive microbes were required to have a higher SMRN than that of the negative controls.

      Statistical analysis

      SPSS 25.0 software (IBM) and GraphPad Prism 7.0 (GraphPad Software) were employed for statistical analysis and to draw figures. Continuous variables were presented as medians and ranges and categorical variables as counts and percentages. The Mann–Whitney U test was used for comparing the differences in the continuous variables and the Chi-square test was used for categorical variables. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the clinical composite diagnosis were calculated as the reference standard(25). Test concordance was analyzed using the kappa statistic. P < 0.05 was considered to indicate statistical significance.

      Results

      Patient baseline characteristics and sample collection

      A total of 379 tests were screened, of which 347 tests were finally enrolled in this study (Fig. 1). Among the patients, common underlying diseases were acute myeloid leukemia (42.3%), followed by acute lymphoblastic leukemia (24.6%), myelodysplastic syndromes (14.1%), aplastic anemia (13.1%), chronic myeloid leukemia (3.3%), and acute heterozygosis leukemia (1.0%). Among the 347 cases, 146 (42.1%) plasma cfDNA NGS tests were performed after allo-HSCT. The 347 cases were divided into two groups: Neutropenia (Group I: 184 cases) and Non-Neutropenia (Group II: 163 cases) based on whether the patient was neutropenic at the time of collection of plasma cfDNA NGS test sample (Table 1). No significant differences were observed in gender, age, previous antibiotic exposure, procalcitonin level, and 30-day mortality between both groups. In terms of treatment for underlying diseases, a higher proportion of patients were allo-HSCT recipients in the Non-Neutropenia group (60.1% vs. 26.1%, P<0.001).
      Table 1Samples characteristics and baseline of the two study groups.
      Samples characteristicsTotalGroup I: NeutropeniaGroup II: Non-NeutropeniaP value
      Number of tests347184163
      Median age (range)39 (9–75)39.5 (9–75)39 (10–72)0.185
      Male, n (%)194 (55.9%)95 (51.6%)99 (60.7%)0.088
       Underlying diseases, n (%)<0.001
       Acute myeloid leukemia148 (42.7%)104 (56.5%)44 (27.0%)
       Acute lymphoblastic leukemia84 (24.2%)32 (17.4%)52 (31.9%)
       Aplastic anemia45 (13.0%)25 (13.6%)20 (12.3%)
       Myelodysplastic syndromes52 (15.0%)17 (9.2%)35 (21.5%)
       Other Diseases18 (5.2%)6 (3.3%)12 (7.4%)
      Laboratory examination, median (range)
       WBC, 10^9/L0.97 (0.01–85.22)0.35 (0.01–38.56)2.82 (0.54–85.22)<0.001
       ANC, 10^9/L0.35 (0–20.98)0.03 (0–0.49)1.94 (0.50–20.98)<0.001
       CRP, mg/L61.02 (<0.50–361.43)76.01 (1.50–361.43)43.83 (<0.50–320.00)<0.001
       PCT, ng/mL0.13 (<0.05–122.27)0.12 (<0.05–122.27)0.15 (<0.05–4.91)0.232
      Therapy of underlying diseases, n (%)
       HSCT146 (42.1%)48 (26.1%)98 (60.1%)<0.001
       Chemotherapy133 (38.3%)101 (54.9%)32 (19.6%)<0.001
       Immunosuppressive therapy18 (5.2%)9 (4.9%)9 (5.5%)0.792
      Previous antibiotic exposure, n (%)304 (87.6%)156 (84.8%)148 (90.8%)0.090
      Hospital-acquired, n (%)291 (83.9%)156 (84.8%)135 (82.8%)0.640
      28-day mortality, n (%)35 (10.1%)20 (10.9%)15 (9.2%)0.607
      Abbreviation: WBC, white blood cell; ANC, absolute neutrophil count; CRP, C-reactive protein; PCT, procalcitonin; HSCT, hematopoietic stem cell transplant.
      The most common site of infection involved was the lower respiratory tract (46.4%), followed by oral and perianal mucosa (20.5%) and the gastrointestinal tract (16.4%). We found that the distribution of infection sites was different between the two groups (Fig. 2AB). In the Neutropenia Group, more mucositis, BSIs, and upper respiratory tract infections were observed than those in the Non-Neutropenia Group.
      Fig 2
      Fig. 2Distribution of the infection sites involved in this study and the comparison of the detection rates. (A), The number of infection sites involved between the groups. (B), The infection sites distribution involved between the groups. (C), The detection rate comparison. Abbreviations: SSTIs, skin and soft-tissue infections; LRTIs, lower respiratory tract infections; UTIs, urinary tract infection; CNSIs, central nervous system infections; URTIs, upper respiratory tract infections; *P < 0.05; **P < 0.01; ***and half parentheses, P < 0.001.

      Comparison of the diagnostic performance of plasma cfDNA NGS and CMT

      All patients in the study were subjected to both CMT and plasma cfDNA NGS tests, the positive rate for NGS tests was 72.6% (Group I: 73.4% vs. Group II: 71.8%), whereas the positive rate was 31.4% (Group I: 25.5% vs. Group II: 38.0%) for CMT. The positive rates for bacteria and virus for the plasma cfDNA NGS test were significantly higher than that for CMT in both groups (Fig. 2C) (P < 0.001). Notably, more bacteria were detected in Group I than in Group II by both NGS tests and CMT (NGS, Bacteria detection rate: 42.4% vs.23.9%, P<0.001; CMT, Bacteria detection rate: 18.5% vs. 9.2% P<0.001).
      The comparison of the diagnostic performance between the NGS test and CMT is shown in Table 2. Among the 347 cases, 316 patients were subjected to blood cultures within 48 h of NGS specimen collection, and the positive rate of blood culture was 7.3% (23/316). The sensitivity and specificity of NGS compared with blood culture were 82.6% and 59.0%, respectively. Plasma NGS tests and CMT were concordant for 174 of 347 (50.1%) cases (kappa = 0.124; 95% CI: 0.049 to 0.200). For the composite clinical standard, the sensitivity and specificity of NGS were 89.1% and 65.1%, respectively. The agreement rate was 80.4% (kappa = 0.561; 95% CI: 0.467 to 0.651).
      Table 2The agreement of plasma cfDNA NGS results versus those of blood culture, blood culture plus plasma virus DNA test, all conventional microbiological testing, and the composite clinical standard.
      NGS PositiveNGS NegativeSensitivity (%)Specificity (%)PPV (%)NPV (%)Kappa, agreement
      Blood culture
      The blood culture was only included within 48 h of the NGS specimen collection.
      positive
      19482.659.013.797.70.125, 60.8%
      Blood culture
      The blood culture was only included within 48 h of the NGS specimen collection.
      negative
      120173
      Blood culture
      The blood culture was only included within 48 h of the NGS specimen collection.
      or plasma virus DNA
      Plasma virus DNA tests included CMV and EBV qPCR, as these two types of viruses were detected conventionally, especially in patients after HSCT treatment.
      positive
      44589.841.021.295.80.132, 48.3%
      Blood cultur
      The blood culture was only included within 48 h of the NGS specimen collection.
      and plasma virus DNA
      Plasma virus DNA tests included CMV and EBV qPCR, as these two types of viruses were detected conventionally, especially in patients after HSCT treatment.
      negative
      164114
      CMT
      Patients who received CMT included culture, sample DME, BDG test, GM test, antigen detection, and virus NAAT within 7 days of the NGS specimen collection.
      positive
      862279.636.836.380.00.124, 50.1%
      CMT
      Patients who received CMT included culture, sample DME, BDG test, GM test, antigen detection, and virus NAAT within 7 days of the NGS specimen collection.
      negative
      15188
      Composite clinical standard
      Plasma NGS results were classified as definite, probable, possible, unlikely, or false-negative causes of the cases. In the definite, probable, and possible situations, the NGS results were evaluated to be positive in accordance with the composite clinical standard, while in the unlikely and false-negative situations, they tended to be negative.
      positive
      1972489.165.181.777.40.561, 80.4%
      Composite clinical standard
      Plasma NGS results were classified as definite, probable, possible, unlikely, or false-negative causes of the cases. In the definite, probable, and possible situations, the NGS results were evaluated to be positive in accordance with the composite clinical standard, while in the unlikely and false-negative situations, they tended to be negative.
      negative
      4482
      1 The blood culture was only included within 48 h of the NGS specimen collection.
      2 Plasma virus DNA tests included CMV and EBV qPCR, as these two types of viruses were detected conventionally, especially in patients after HSCT treatment.
      3 Patients who received CMT included culture, sample DME, BDG test, GM test, antigen detection, and virus NAAT within 7 days of the NGS specimen collection.
      4 Plasma NGS results were classified as definite, probable, possible, unlikely, or false-negative causes of the cases. In the definite, probable, and possible situations, the NGS results were evaluated to be positive in accordance with the composite clinical standard, while in the unlikely and false-negative situations, they tended to be negative.
      A greater variety of microbes was detected by NGS than CMT (Fig. 3). In terms of fungal detection by NGS, Aspergillus was the most common pathogen, followed by Mucorales, Candida, and P. jiroveci. Notably, P. jiroveci and Mucorales were only detected by the NGS method. The positive rate of CMT in Group I was lower than that in Group II (25.5% vs. 38.0%, P = 0.012). After combining with NGS, overall positive rates of 75.0% and 74.8%, respectively, can be achieved. Moreover, a difference was observed in terms of test consistency between the two groups, and the proportion of NGS and CMT both positive was higher in the Non-Neutropenia Group than in the Neutropenia Group (35.0% vs. 22.8%, P = 0.009). For the double-positive cases, the matched proportion (completely matched and partially matched) between the two groups was 77.8% and 89.5%, respectively, with no significant difference (P = 0.096).
      Fig 3
      Fig. 3Distribution of microbes and results comparison by cfDNA NGS versus CMT. (A), Distribution of microbes detected by NGS and CMT; (B), Comparison of the results between two groups by NGS and CMT; (C), Consistency comparison in double-positive results of NGS and CMT between Neutropenia and Non-Neutropenia groups.

      The clinical impact of plasma cfDNA NGS on diagnosis and treatment of infection

      The plasma cfDNA NGD positively affected the diagnosis and treatment of infection in 47.8% (166/347) and 38.6% (134/347) of cases, respectively. The negative impacts on diagnosis and treatment accounted for 0.6% (2/347) and 2.3% (8/347), respectively. Antibiotics were adjusted in 73 (21.0%) cases, whereas the pathogen was uncovered by empirical treatment in 61 (17.6%) cases.
      The proportions of the positive impact of NGS results in the Neutropenia Group were higher than in the Non-Neutropenia group in terms of both diagnosis and treatment (diagnosis: 54.4% vs. 40.5%, P = 0.013; treatment: 45.7% vs.30.7%, P = 0.004) (Fig. 4). In the Neutropenia Group, in 86 of 184 (46.7%) cases, the microbes detected by NGS were assessed as the probable or possible pathogen of the infection, whereas no microbes were detected by CMT (Table 3). This proportion in the Non-Neutropenia was 37.4% (61/163), which was lower than the Neutropenia Group (P = 0.026).
      Fig 4
      Fig. 4Comparison of clinical impact on infection diagnosis and treatment of the two groups. (A), Clinical impact on the diagnosis of infection in our study. (B), Clinical impact on antibiotic treatment of infection in our study. *P < 0.05; **P < 0.01; ***P < 0.001.
      Table 3The clinical impact of plasma cfDNA NGS on diagnosis and treatment.
      AspectClinical ImpactGrade Diagnosis: D1-D8 Treatment: T1-T8DescriptionCase number, n(%)P
      Neutropenia N = 184Non-Neutropenia N = 163
      DiagnosisPositiveD1NGS result was quicker than CMT5 (2.7%)2 (1.2%)ns
      D2Co-infection was diagnosed according to NGS.7 (3.8%)1 (0.6%)ns
      D3The detection time window was longer than that for blood culture2 (1.1%)2 (1.2%)ns
      D4Plasma NGS results contributed to pathogen identification86 (46.7%)61 (37.4%)0.026
      No impactD5Negative result of NGS50 (27.2%)45 (27.6%)ns
      D6The pathogen detected by NGS was the same as CMT, but not detected earlier than CMT.9 (4.9%)25 (15.3%)0.002
      D7The microbes detected by NGS were assessed as unlikely pathogens.23 (12.5%)27 (16.6%)ns
      NegativeD8BSI pathogen was undetected by NGS and without suspected pathogen detection2 (1.1%)0 (0.0%)ns
      TreatmentPositiveT1Initialization of the appropriate antibiotics treatment37 (20.1%)24 (14.7%)ns
      T2Antibiotic escalation4 (2.2%)3 (1.8%)ns
      T3Antibiotic de-escalation3 (1.6%)2 (1.2%)ns
      T4Confirmed empirical treatment40 (21.7%)21 (12.9%)0.016
      No impactT5No adjustment in treatment while the result was positive49 (26.6%)60 (36.8%)ns
      T6The patient was discharged or dead1 (0.5%)3 (1.8%)ns
      T7No adjustment in treatment while the result was negative49 (26.6%)43 (26.4%)ns
      NegativeT8NGS led to unnecessary treatment1 (0.5%)7 (4.3%)ns
      Abbreviation: ns, no statistically significant.

      Application of plasma cfDNA NGS in patients with suspected pulmonary fungal infection

      In our study, 161 (46.4%) patients had lower respiratory tract infections (LRTIs), among whom, the detection rates of Aspergillus spp., Mucorales, and P. jiroveci were 14.2% (23/161), 8.1% (13/161), and 5.0% (8/161), respectively, which were the most commonly isolated fungal pathogens of LRTIs. In patients with negative results from CMT, the etiological diagnosis of invasive fungal infection remains a substantial challenge. The NGS results of patients with fungal infections were consistent with their clinical features and radiological manifestations (Fig. 5). Notably, in 46 cases, fungus detection by NGS in patients with LRTIs had a positive impact on diagnosis. Among these 46 cases, the initial empirical treatment was adjusted according to NGS results in 58.7% (27/46) cases.
      Fig 5
      Fig. 5Radiographic signs of suspected invasive pulmonary fungal infection in this study. (A–D), Signs on CT for invasive pulmonary aspergillosis include one or more micronodules, halo sign, cavity, and an air-crescent sign. Halo, as a nodular with ground-glass opacity surrounding, can be typically seen in invasive aspergillosis. (E-H), Sighs on CT of mucormycosis including halo sign and reversed halo sign. Reversed halo sign—a center of glass opacity surrounded by a ring of denser consolidation, as typically seen in mucormycosis. (I-L), Typical radiographic features of pneumocystis pneumonia are bilateral, diffuse, and interstitial infiltrates. PJP, P. jiroveci.
      Compared with the Non-Neutropenia Group, more Aspergillus spp. was detected in patients in the Neutropenia Group (25.3% vs. 5.3%, P = 0.009) (Fig. 6A). Among 23 patients with Aspergillus spp. detected by the NGS method, only 10 patients with CMT were positive for fungal infection. Several microbiological tests were performed to diagnose invasive Aspergillus infection (IAI), including culture, DME, BDG, and GM tests. GM was the most common positive test, followed by BDG and DME (Fig. 6B).
      Fig 6
      Fig. 6Fungi detected by NGS and CMT in patients with pulmonary infection. (A), Aspergillus, Mucorales, and PJP detection rate in patients with a pulmonary infection between the two groups. (B), The methods of yielding positive results of CMT in patients infected with Aspergillus, as detected by NGS. PJP, P. jiroveci; *P < 0.05; **P < 0.01; ***P < 0.001.

      Application of plasma cfDNA NGS in patients with BSIs

      In our study, 39 cases (11.2%) were diagnosed with BSIs. Consistent with blood culture results, the top 3 isolated pathogens of BSIs by NGS were Klebsiella pneumonia (n = 5), Pseudomonas aeruginosa (n = 5), and Escherichia coli (n = 4). NGS reported false-negative in four cases, and the pathogens were E. coli, Enterobacter cloacae, Haemophilus parainfluenzae, and Streptococcus salivarius. In five cases of BSIs, the pathogens were detectable by plasma cfDNA NGS after the blood culture results turned negative (Supplementary Figure S2). These five patients’ symptoms were not improving and plasma cfDNA NGS was recommended by the clinicians. In two cases, the pathogens were earlier detected by NGS than by blood culture. In the M. tuberculosis-detected case, the turnaround time of NGS was 27 days earlier than that of the final positive blood culture.

      Discussion

      In this retrospective study, we compared plasma cfDNA NGS with CMT in terms of performance in causative microbial detection. NGS offers great potential in pathogen detection because of its characteristics such as non-invasiveness, sensitivity, and broad-spectrum pathogen detection. Although a few studies have been reported on the clinical application of plasma cfDNA NGS in patients with hematological diseases, mainly focusing on the field of febrile neutropenia, studies on pathogenic diagnosis by NGS in patients without neutropenia are still rare.
      • Horiba K.
      • Torii Y.
      • Okumura T.
      • Takeuchi S.
      • Suzuki T.
      • Kawada J.I.
      • et al.
      Next-generation sequencing to detect pathogens in pediatric febrile neutropenia: a single-center retrospective study of 112 cases.
      • Benamu E.
      • Gajurel K.
      • Anderson J.N.
      • Lieb T.
      • Gomez C.A.
      • Seng H.
      • et al.
      Plasma microbial cell-free DNA next generation sequencing in the diagnosis and management of febrile neutropenia.
      Our study included larger sample size and represents a comprehensive analysis of the analytical and diagnostic performance of plasma cfDNA sequencing in hematological patients, both without and with neutropenia, which is of clinical significance as these two groups were distinctive in terms of the infection type, immune status, and empirical antibiotic usage. Furthermore, we conducted a systematic assessment of the clinical impact through a detailed grading system considering both the diagnosis and treatment aspects.
      The positive rate of NGS was approximately 50% higher than that of CMT in patients with neutropenia. In patients without neutropenia, the difference in the positive rate between NGS and CMT was 33.8%, which was lower than the neutropenia group. The possible reason is that more patients with neutropenia had mucositis and gastroenteritis. In such a case, mucositis and gastroenteritis may facilitate the entry of microbes and/or microbial nucleic acid into the circulation through the epithelial mucosa as reported by Han et al. and our previous studies.
      • Han D.
      • Li R.
      • Shi J.
      • Tan P.
      • Zhang R.
      • Li J.
      Liquid biopsy for infectious diseases: a focus on microbial cell-free DNA sequencing.
      ,
      • Xu C.H.
      • Zhu G.Q.
      • Lin Q.S.
      • Wang L.L.
      • Wang X.X.
      • Gong J.Y.
      • et al.
      A single-center study on the distribution and antibiotic resistance of pathogens causing bloodstream infection in adult patients with hematological disease during the period 2014-2018.
      We noticed that the blood culture positive rate (7.3%) was lower than that in the other studies, ranging from 16% to 20%.
      • Okamoto A.
      • Kanda Y.
      • Kimura S.I.
      • Oyake T.
      • Tamura K.
      • from the Japan Febrile Neutropenia Study G
      Predictive and risk factor analysis for bloodstream infection in high-risk hematological patients with febrile neutropenia: post-hoc analysis from a prospective, large-scale clinical study.
      ,
      • Kern W.V.
      • Roth J.A.
      • Bertz H.
      • Gotting T.
      • Dettenkofer M.
      • Widmer A.F.
      • et al.
      Contribution of specific pathogens to bloodstream infection mortality in neutropenic patients with hematologic malignancies: results from a multicentric surveillance cohort study.
      This is partly because of the high cost of the NGS test (3800 yuan). Mostly, the NGS test is performed only when the conventional microbial cultures are negative and the empirical treatment does not work. With the development of NGS technology and cost reduction, more patients with suspected infections may benefit from this method in the future.
      Moreover, we attempted to evaluate the real-world clinical impact of the NGS test in terms of diagnosis and treatment. In our study, 134 of 347 (38.6%) NGS test results positively affected clinical treatment in 73 cases (21.0%). The proportion with positive impacts was higher than another recent study, which reported that only 6 of 82 (7.3%) NGS results had a positive impact.
      • Hogan C.A.
      • Yang S.
      • Garner O.B.
      • Green D.A.
      • Gomez C.A.
      • Dien Bard J.
      • et al.
      Clinical impact of metagenomic next-generation sequencing of plasma cell-free DNA for the diagnosis of infectious diseases: a multicenter retrospective cohort study.
      This difference may be related to fewer etiological detection methods of CMT being available to guide antibiotic treatment in this study. In terms of CMT, limited CMT is available for the clinical diagnosis of infection in developing countries, particularly molecular microbiological detection methods such as plasma 16S bacterial and fungal rRNA sequencing, tissue virus NAAT, respiratory panel NAAT, which are used in the plasma and cerebrospinal fluid cfDNA NGS studies.
      • Hogan C.A.
      • Yang S.
      • Garner O.B.
      • Green D.A.
      • Gomez C.A.
      • Dien Bard J.
      • et al.
      Clinical impact of metagenomic next-generation sequencing of plasma cell-free DNA for the diagnosis of infectious diseases: a multicenter retrospective cohort study.
      ,
      • Wilson M.R.
      • Sample H.A.
      • Zorn K.C.
      • Arevalo S.
      • Yu G.
      • Neuhaus J.
      • et al.
      Clinical metagenomic sequencing for diagnosis of meningitis and encephalitis.
      Notably, we first established a standard system for the evaluation of diagnostic impacts because earlier detection, mixed pathogen infection detection, new etiology diagnosis, and longer detection time window had a positive impact on infection diagnosis. For example, Toxoplasma gondii nucleic acid was detected in the plasma of a patient. After 20 days of HSCT, the patient had a fever, CNSI, and pulmonary infection without receiving trimethoprim/sulfamethoxazole prophylaxis. The clinical characteristics and radiological manifestations were consistent with Toxoplasma infection after HSCT.
      • Cavattoni I.
      • Ayuk F.
      • Zander A.R.
      • Zabelina T.
      • Bacher A.
      • Cayroglu E.
      • et al.
      Diagnosis of Toxoplasma gondii infection after allogeneic stem cell transplant can be difficult and requires intensive scrutiny.
      ,
      • Sumi M.
      • Aosai F.
      • Norose K.
      • Takeda W.
      • Kirihara T.
      • Sato K.
      • et al.
      Acute exacerbation of Toxoplasma gondii infection after hematopoietic stem cell transplantation: five case reports among 279 recipients.
      The disease in the patient progressed rapidly and he died on the day the result had been available. Although there was no adjustment of treatment, plasma cfDNA NGS identified the pathogen and had a positive impact on the etiological diagnosis of infection.
      Additionally, pulmonary invasive fungal infections are common complications of immunocompromised patients. Without timely diagnosis and appropriate treatment, morbidity and mortality remain high.
      • Chen C.Y.
      • Sheng W.H.
      • Tien F.M.
      • Lee P.C.
      • Huang S.Y.
      • Tang J.L.
      • et al.
      Clinical characteristics and treatment outcomes of pulmonary invasive fungal infection among adult patients with hematological malignancy in a medical centre in Taiwan, 2008-2013.
      ,
      • Formanek P.E.
      • Dilling D.F.
      Advances in the diagnosis and management of invasive fungal disease.
      The availability of appropriate samples is essential for the diagnosis of invasive mycoses, such as lung tissues and BALF. However, once the patient cannot withstand invasive sampling, it will restrict the application of conventional diagnostic methods.
      • Han D.
      • Li R.
      • Shi J.
      • Tan P.
      • Zhang R.
      • Li J.
      Liquid biopsy for infectious diseases: a focus on microbial cell-free DNA sequencing.
      In our study, Aspergillus spp., Mucorales, and P. jiroveci were the most commonly isolated microorganisms. In terms of IAI, CMT was positive only in 10 of 23 (43.5%) cases. GM was the most common method; however, plasma GM had a low sensitivity in IAI diagnosis.
      • Donnelly J.P.
      • Chen S.C.
      • Kauffman C.A.
      • Steinbach W.J.
      • Baddley J.W.
      • Verweij P.E.
      • et al.
      Revision and update of the consensus definitions of invasive fungal disease from the European Organization for Research and Treatment of Cancer and the Mycoses Study Group Education and Research Consortium.
      In contrast, NGS results could provide pathogen information at the species level, which is also important in the clinical management of IFI.
      • Ullmann A.J.
      • Aguado J.M.
      • Arikan-Akdagli S.
      • Denning D.W.
      • Groll A.H.
      • Lagrou K.
      • et al.
      Diagnosis and management of Aspergillus diseases: executive summary of the 2017 ESCMID-ECMM-ERS guideline.
      Mucormycosis could rapidly progress in immunocompromised patients.
      • Cornely O.A.
      • Alastruey-Izquierdo A.
      • Arenz D.
      • Chen S.C.A.
      • Dannaoui E.
      • Hochhegger B.
      • et al.
      Global guideline for the diagnosis and management of mucormycosis: an initiative of the European Confederation of Medical Mycology in cooperation with the Mycoses Study Group Education and Research Consortium.
      The low sensitivity of cultures and DME of tissues and BALF has made the diagnosis of mucormycosis a great challenge. Mucorales were detected in 13 patients by the NGS test, and radiological manifestations were consistent with the results of the NGS. The reversed halo sign was typical of invasive pulmonary mucormycosis. There can be two reasons for the high frequency of Mucorales detection in our study. First, the sensitivity for Mucorales (non-Aspergillus) by plasma cfDNA NGS was higher than that for Aspergillus. According to Hill's research,
      • Hong D.K.
      • Blauwkamp T.A.
      • Kertesz M.
      • Bercovici S.
      • Truong C.
      • Banaei N.
      Liquid biopsy for infectious diseases: sequencing of cell-free plasma to detect pathogen DNA in patients with invasive fungal disease.
      plasma cfDNA NGS had a higher sensitivity for proven/probable non-Aspergillus IMI (Mucorales accounted for 75%, sensitivity of 79%) when compared with Aspergillus IMI (sensitivity of 31%). Second, several plasma NGS tests were submitted because the infection failed to resolve after empiric treatment in this study. When the empiric treatment does not work after the use of voriconazole, which is the common choice for the treatment of suspected pulmonary fungal infection, the likelihood of Mucorales as the pathogen increases.
      According to a previous study, longer persistence of pathogenic cfDNA detected by NGS in the plasma was related to infection severity and increased risk of metastatic infection.
      • Eichenberger E.M.
      • de Vries C.R.
      • Ruffin F.
      • Sharma-Kuinkel B.
      • Park L.
      • Hong D.
      • et al.
      Microbial cell-free DNA identifies etiology of bloodstream infections, persists longer than conventional blood cultures, and its duration of detection is associated with metastatic infection in patients with staphylococcus aureus and gram-negative bacteremia.
      In our study, 5 cases of the BSI pathogen could be detected after blood culture test results were negative. All the cases were poorly controlled, the median time from positive blood culture to NGS test was 6 days. After antibiotic administration, blood cultures had turned negative in these situations. If the patient with BSI does not improve after appropriate treatment for a few days, it is difficult to distinguish whether the cause was the progress of BSI or a new pathogen. NGS can be useful in such a condition. In addition, in 2 cases, NGS results were available before positive blood culture. It facilitated early diagnosis and treatment of BSI. Especially in the case of #T298, tuberculosis was reported 28 days after the blood sample was collected for blood culture. X-pert. MTB/RIF test of BALF was positive 2 days later, which further confirmed the tuberculosis of this patient. The patient improved after anti-tuberculosis treatment according to NGS results.
      Our study has some limitations. First, this study is a retrospective study. Many NGS tests were performed because of the negative results of CMT, which affected the comparison of NGS with CMT. Second, most microorganisms detected by NGS have not had a NAAT for molecular verification, except for a few viruses and tuberculosis. Molecular microbiological testing is still not widely available in developing countries, and culture and smear microscopy are still the main tools to identify pathogens. Third, most patients received treatment before the NGS test, which could affect the sensitivity of CMT and NGS
      • Xing X.W.
      • Zhang J.T.
      • Ma Y.B.
      • Chen X.Y.
      • Wu L.
      • Wang X.L.
      • et al.
      Evaluation of next-generation sequencing for the diagnosis of infections of the central nervous system caused by the neurotropic herpes viruses: a pilot study.
      ,
      • Miao Q.
      • Ma Y.
      • Wang Q.
      • Pan J.
      • Zhang Y.
      • Jin W.
      • et al.
      Microbiological diagnostic performance of metagenomic next-generation sequencing when applied to clinical practice.
      ; hence, further validation is necessary. Finally, the impact of cfDNA NGS on antibiotic use is uncertain as we did not maintain any patient control group for direct comparison.
      In conclusion, our findings suggest that cell-free DNA sequencing holds great potential for the detection of pathogens in patients with hematological disorders with a suspected infection. Combining conventional tests with plasma cell-free DNA sequencing can significantly improve the detection rate for pathogens and partially optimize antibiotic treatment. Despite this, our findings on clinical impact warrant confirmation through larger, multicenter randomized controlled trials. Moreover, as the cost-effectiveness of this testing strategy remains unknown, this aspect warrants further exploration.

      Authors' contributions

      Author Chunhui Xu and Xin Chen were responsible for the data curation methodology and writing-orifinal draft, and they contributed equally to this article. Authors Guoqing Zhu, Huiming Yi, Shulian Chen, Yuetian Yu, Erlie Jiang, Yizhou Zheng, Fengkui Zhang, Jianxiang Wang finished the formal analysis and supervision. Author Sizhou Feng supplied the conceptualization, funding acquisition, resources, supervision, and writing - review & editing. All authors reviewed the manuscript.

      Ethics approval and consent to participate

      This study was approved by the Institutional Review Board and Ethics Committee of the Institute of Hematology and Blood Diseases Hospital. The ethics committee approved the waiver of informed consent owing to the retrospective nature of the review. All human data of patients’ records were confirmed for collection in accordance with the relevant guidelines and regulations.

      Availability of data and materials

      The datasets generated and/or analyzed during the current study are not publicly available considering the privacy or ethical restrictions but are available from the corresponding author on a reasonable request.

      Declaration of Competing Interest

      The authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest.

      Funding

      This work was supported by the CAMS Innovation Fund for Medical Sciences (CIFMS) [2021-I2M-1-017] & [2021-I2M-C&T-B-080], Tianjin Municipal Science and Technology Commission Grant (21JCZDJC01170) and Haihe Laboratory of Cell Ecosystem Innovation Fund [HH22KYZX0036].

      Acknowledgements

      The authors would like to thank all the reviewers who participated in the review, as well as Chao Liu for providing English editing services during the preparation of this manuscript.

      Appendix. Supplementary materials

      References

        • Aguilar-Guisado M.
        • Espigado I.
        • Martin-Pena A.
        • Gudiol C.
        • Royo-Cebrecos C.
        • Falantes J.
        • et al.
        Optimisation of empirical antimicrobial therapy in patients with haematological malignancies and febrile neutropenia (How Long study): an open-label, randomised, controlled phase 4 trial.
        Lancet Haematol. 2017; 4: e573-ee83
        • Kochanek M.
        • Schalk E.
        • von Bergwelt-Baildon M.
        • Beutel G.
        • Buchheidt D.
        • Hentrich M.
        • et al.
        Management of sepsis in neutropenic cancer patients: 2018 guidelines from the Infectious Diseases Working Party (AGIHO) and Intensive Care Working Party (iCHOP) of the German Society of Hematology and Medical Oncology (DGHO).
        Ann Hematol. 2019; 98: 1051-1069
        • Ortega M.
        • Marco F.
        • Soriano A.
        • Almela M.
        • Martinez J.A.
        • Rovira M.
        • et al.
        Epidemiology and outcome of bacteraemia in neutropenic patients in a single institution from 1991 to 2012.
        Epidemiol Infect. 2015; 143: 734-740
        • Martinez-Nadal G.
        • Puerta-Alcalde P.
        • Gudiol C.
        • Cardozo C.
        • Albasanz-Puig A.
        • Marco F.
        • et al.
        Inappropriate empirical antibiotic treatment in high-risk neutropenic patients with bacteremia in the era of multidrug resistance.
        Clin Infect Dis. 2020; 70: 1068-1074
        • White L.
        • Ybarra M.
        Neutropenic fever.
        Hematol Oncol Clin North Am. 2017; 31: 981-993
        • Schmidt-Hieber M.
        • Teschner D.
        • Maschmeyer G.
        • Schalk E.
        Management of febrile neutropenia in the perspective of antimicrobial de-escalation and discontinuation.
        Expert Rev Anti Infect Ther. 2019; 17: 983-995
        • Chen S.C.
        • Kontoyiannis D.P.
        New molecular and surrogate biomarker-based tests in the diagnosis of bacterial and fungal infection in febrile neutropenic patients.
        Curr Opin Infect Dis. 2010; 23: 567-577
        • Wang S.
        • Ai J.
        • Cui P.
        • Zhu Y.
        • Wu H.
        • Zhang W.
        Diagnostic value and clinical application of next-generation sequencing for infections in immunosuppressed patients with corticosteroid therapy.
        Ann Transl Med. 2020; 8: 227
        • Gu W.
        • Miller S.
        • Chiu C.Y.
        Clinical metagenomic next-generation sequencing for pathogen detection.
        Annu Rev Pathol. 2019; 14: 319-338
        • Wilson M.R.
        • Naccache S.N.
        • Samayoa E.
        • Biagtan M.
        • Bashir H.
        • Yu G.
        • et al.
        Actionable diagnosis of neuroleptospirosis by next-generation sequencing.
        N Engl J Med. 2014; 370: 2408-2417
        • Yu X.
        • Jiang W.
        • Shi Y.
        • Ye H.
        • Lin J.
        Applications of sequencing technology in clinical microbial infection.
        J Cell Mol Med. 2019; 23: 7143-7150
        • Choo R.
        • Naser N.S.H.
        • Nadkarni N.V.
        Anantham D. Utility of bronchoalveolar lavage in the management of immunocompromised patients presenting with lung infiltrates.
        BMC Pulm Med. 2019; 19: 51
        • Schnell D.
        • Azoulay E.
        • Benoit D.
        • Clouzeau B.
        • Demaret P.
        • Ducassou S.
        • et al.
        Management of neutropenic patients in the intensive care unit (NEWBORNS EXCLUDED) recommendations from an expert panel from the French Intensive Care Society (SRLF) with the French Group for Pediatric Intensive Care Emergencies (GFRUP), the French Society of Anesthesia and Intensive Care (SFAR), the French Society of Hematology (SFH), the French Society for Hospital Hygiene (SF2H), and the French Infectious Diseases Society (SPILF).
        Ann Intensive Care. 2016; 6: 90
        • Goggin K.P.
        • Gonzalez-Pena V.
        • Inaba Y.
        • Allison K.J.
        • Hong D.K.
        • Ahmed A.A.
        • et al.
        Evaluation of plasma microbial cell-free DNA sequencing to predict bloodstream infection in pediatric patients with relapsed or refractory cancer.
        JAMA Oncol. 2020; 6: 552-556
        • Eichenberger E.M.
        • de Vries C.R.
        • Ruffin F.
        • Sharma-Kuinkel B.
        • Park L.
        • Hong D.
        • et al.
        Microbial cell-free DNA identifies etiology of bloodstream infections, persists longer than conventional blood cultures, and its duration of detection is associated with metastatic infection in patients with staphylococcus aureus and gram-negative bacteremia.
        Clinic Infect Dis. 2021;
        • Hong D.K.
        • Blauwkamp T.A.
        • Kertesz M.
        • Bercovici S.
        • Truong C.
        • Banaei N.
        Liquid biopsy for infectious diseases: sequencing of cell-free plasma to detect pathogen DNA in patients with invasive fungal disease.
        Diagn Microbiol Infect Dis. 2018; 92: 210-213
        • To R.K.
        • Ramchandar N.
        • Gupta A.
        • Pong A.
        • Cannavino C.
        • Foley J.
        • et al.
        Use of plasma metagenomic next-generation sequencing for pathogen identification in pediatric endocarditis.
        Pediatr Infect Dis J. 2021; 40: 486-488
        • Lieberman J.A.
        • C Naureckas Li
        • Lamb G.S.
        • Kane D.A.
        • Stewart M.K.
        • Mamedov R.A.
        • et al.
        Case report: comparison of plasma metagenomics to bacterial PCR in a case of prosthetic valve endocarditis.
        Front Pediatr. 2020; 8575674
        • Gosiewski T.
        • Ludwig-Galezowska A.H.
        • Huminska K.
        • Sroka-Oleksiak A.
        • Radkowski P.
        • Salamon D.
        • et al.
        Comprehensive detection and identification of bacterial DNA in the blood of patients with sepsis and healthy volunteers using next-generation sequencing method - the observation of DNAemia.
        Eur J Clin Microbiol Infect Dis. 2017; 36: 329-336
        • Nie J.
        • Yang L.
        • Huang L.
        • Gao L.
        • Young K.H.
        • Le Grange J.M.
        • et al.
        Infection complications in febrile chimeric antigen receptor (CAR)-T recipients during the peri-CAR-T cell treatment period examined using metagenomic next-generation sequencing (mNGS).
        Cancer Commun. 2022;
        • Horiba K.
        • Torii Y.
        • Okumura T.
        • Takeuchi S.
        • Suzuki T.
        • Kawada J.I.
        • et al.
        Next-generation sequencing to detect pathogens in pediatric febrile neutropenia: a single-center retrospective study of 112 cases.
        Open Forum Infect Dis. 2021; 8: ofab223
        • Benamu E.
        • Gajurel K.
        • Anderson J.N.
        • Lieb T.
        • Gomez C.A.
        • Seng H.
        • et al.
        Plasma microbial cell-free DNA next generation sequencing in the diagnosis and management of febrile neutropenia.
        Clinic Infect Dis. 2021;
        • Prevention CfDCa
        CDC/NHSN Surveillance definitioins for specific types of infections.
        Surveill Def. 2019; 17: 1-29
        • Freifeld A.G.
        • Bow E.J.
        • Sepkowitz K.A.
        • Boeckh M.J.
        • Ito J.I.
        • Mullen C.A.
        • et al.
        Clinical practice guideline for the use of antimicrobial agents in neutropenic patients with cancer: 2010 Update by the Infectious Diseases Society of America.
        Clinic Infect Dis. 2011; 52: 427-431
        • Blauwkamp T.A.
        • Thair S.
        • Rosen M.J.
        • Blair L.
        • Lindner M.S.
        • Vilfan I.D.
        • et al.
        Analytical and clinical validation of a microbial cell-free DNA sequencing test for infectious disease.
        Nat Microbiol. 2019; 4: 663-674
        • Han D.
        • Li R.
        • Shi J.
        • Tan P.
        • Zhang R.
        • Li J.
        Liquid biopsy for infectious diseases: a focus on microbial cell-free DNA sequencing.
        Theranostics. 2020; 10: 5501-5513
        • Xu C.H.
        • Zhu G.Q.
        • Lin Q.S.
        • Wang L.L.
        • Wang X.X.
        • Gong J.Y.
        • et al.
        A single-center study on the distribution and antibiotic resistance of pathogens causing bloodstream infection in adult patients with hematological disease during the period 2014-2018.
        Zhonghua Xue Ye Xue Za Zhi. 2020; 41: 643-648
        • Okamoto A.
        • Kanda Y.
        • Kimura S.I.
        • Oyake T.
        • Tamura K.
        • from the Japan Febrile Neutropenia Study G
        Predictive and risk factor analysis for bloodstream infection in high-risk hematological patients with febrile neutropenia: post-hoc analysis from a prospective, large-scale clinical study.
        Int J Hematol. 2021; 114: 472-482
        • Kern W.V.
        • Roth J.A.
        • Bertz H.
        • Gotting T.
        • Dettenkofer M.
        • Widmer A.F.
        • et al.
        Contribution of specific pathogens to bloodstream infection mortality in neutropenic patients with hematologic malignancies: results from a multicentric surveillance cohort study.
        Transpl Infect Dis. 2019; 21: e13186
        • Hogan C.A.
        • Yang S.
        • Garner O.B.
        • Green D.A.
        • Gomez C.A.
        • Dien Bard J.
        • et al.
        Clinical impact of metagenomic next-generation sequencing of plasma cell-free DNA for the diagnosis of infectious diseases: a multicenter retrospective cohort study.
        Clinic Infect Dis. 2021; 72: 239-245
        • Wilson M.R.
        • Sample H.A.
        • Zorn K.C.
        • Arevalo S.
        • Yu G.
        • Neuhaus J.
        • et al.
        Clinical metagenomic sequencing for diagnosis of meningitis and encephalitis.
        N Engl J Med. 2019; 380: 2327-2340
        • Cavattoni I.
        • Ayuk F.
        • Zander A.R.
        • Zabelina T.
        • Bacher A.
        • Cayroglu E.
        • et al.
        Diagnosis of Toxoplasma gondii infection after allogeneic stem cell transplant can be difficult and requires intensive scrutiny.
        Leuk Lymphoma. 2010; 51: 1530-1535
        • Sumi M.
        • Aosai F.
        • Norose K.
        • Takeda W.
        • Kirihara T.
        • Sato K.
        • et al.
        Acute exacerbation of Toxoplasma gondii infection after hematopoietic stem cell transplantation: five case reports among 279 recipients.
        Int J Hematol. 2013; 98: 214-222
        • Chen C.Y.
        • Sheng W.H.
        • Tien F.M.
        • Lee P.C.
        • Huang S.Y.
        • Tang J.L.
        • et al.
        Clinical characteristics and treatment outcomes of pulmonary invasive fungal infection among adult patients with hematological malignancy in a medical centre in Taiwan, 2008-2013.
        J Microbiol Immunol Infect. 2020; 53: 106-114
        • Formanek P.E.
        • Dilling D.F.
        Advances in the diagnosis and management of invasive fungal disease.
        Chest. 2019; 156: 834-842
        • Donnelly J.P.
        • Chen S.C.
        • Kauffman C.A.
        • Steinbach W.J.
        • Baddley J.W.
        • Verweij P.E.
        • et al.
        Revision and update of the consensus definitions of invasive fungal disease from the European Organization for Research and Treatment of Cancer and the Mycoses Study Group Education and Research Consortium.
        Clinic Infect Dis. 2020; 71: 1367-1376
        • Ullmann A.J.
        • Aguado J.M.
        • Arikan-Akdagli S.
        • Denning D.W.
        • Groll A.H.
        • Lagrou K.
        • et al.
        Diagnosis and management of Aspergillus diseases: executive summary of the 2017 ESCMID-ECMM-ERS guideline.
        Clinic Microbiol Infect. 2018; 24: e1-e38
        • Cornely O.A.
        • Alastruey-Izquierdo A.
        • Arenz D.
        • Chen S.C.A.
        • Dannaoui E.
        • Hochhegger B.
        • et al.
        Global guideline for the diagnosis and management of mucormycosis: an initiative of the European Confederation of Medical Mycology in cooperation with the Mycoses Study Group Education and Research Consortium.
        Lancet Infect Dis. 2019; 19: e405-ee21
        • Xing X.W.
        • Zhang J.T.
        • Ma Y.B.
        • Chen X.Y.
        • Wu L.
        • Wang X.L.
        • et al.
        Evaluation of next-generation sequencing for the diagnosis of infections of the central nervous system caused by the neurotropic herpes viruses: a pilot study.
        Eur Neurol. 2018; 80: 283-288
        • Miao Q.
        • Ma Y.
        • Wang Q.
        • Pan J.
        • Zhang Y.
        • Jin W.
        • et al.
        Microbiological diagnostic performance of metagenomic next-generation sequencing when applied to clinical practice.
        Clinic Infect Dis. 2018; 67: S231-SS40

      Linked Article