Application of mNGS in the diagnosis and treatment of central nervous system infection

Reprinted from little orange lantern network

On June 12, 2019, C.Y. Chiu and his team published their multicenter research paper on the clinical application of pathogenic metagenomics (mngs) in central nervous system infection in the New England Journal of medicine. Two days earlier, on June 10, the encephalitis assistance group led by Professor Guan Hongzhi of the Department of Neurology of Peking Union Medical College Hospital published the results of the multicenter study of central nervous infection disease mngs in China on biorxiv. The two studies included 204 patients in the United States and 276 patients in China to evaluate the utility of mngs in the clinical diagnosis and treatment of central nervous system infection.

research design 

US team: This study lasted for one year and included 204 patients with primary encephalitis, meningitis and myelitis in 8 hospitals in the United States. The performance of cerebrospinal fluid mngs in the diagnosis of central nervous system (CNS) infection was comprehensively evaluated and compared with traditional detection methods such as culture, PCR and serological detection.

Chinese team: This study lasted for one year and included 276 patients with encephalitis and meningitis in 20 hospitals in China. The performance of cerebrospinal fluid mngs technology in the diagnosis of central nervous system (CNS) infection was comprehensively evaluated and compared with traditional detection methods such as staining, culture, PCR and serological detection.

Patient characteristics

US team: 55.9% of 204 patients were male, with an average age of 39.6 years, and 46 patients (22.5%) were aged 18 years and below. The cohort mainly included 130 patients (63.7%) with encephalitis, 70 patients (34.3%) with meningitis and 4 patients (2.0%) with myelitis, including 176 patients (86.3%) in the acute stage, 28 patients (13.7%) with acute exacerbation in the chronic stage and 83 patients (40.7%) with low immune function. The average length of stay was 27.9 days. 99 patients (48.5%) were admitted to ICU due to severe symptoms, and the total number of deaths (hospitalization and discharge) was 23 (11.3%).

Chinese team: 176 (63.8%) of 276 patients were male, with a median age of 42 years. The median time from onset to cerebrospinal fluid sampling was 10 days. The median CSF leukocyte count was 80 / mm3. The median CSF monocyte count was 36 / mm3. Nine patients died during 30 days of follow-up.

detection result 

US team: 58 samples of 57 patients (27.9%) were identified as infection, of which 32 were diagnosed by mngs and 27 by traditional detection methods. Among 58 infected samples, 19 cases (32.8%) were consistent with the traditional detection; 13 cases (22.4%) were only identified by mngs, which were verified by orthogonal experiment; 26 cases (44.8%) were identified only by traditional detection, of which 11 cases were diagnosed only by serological methods (other traditional detection methods (such as culture, PCR and antigen detection) were negative), 7 cases were detected by brain tissue (non cerebrospinal fluid samples, no cerebrospinal fluid mngs detection), and 8 cases had a low number of detected sequences and did not meet the reporting standard. Among the 13 patients who only improved the diagnosis of pathogens by mngs, clinicians said that the results of mngs had a positive impact on the clinical treatment of 8 patients (62%).

Chinese team: taking traditional detection methods as the gold standard, according to relevant guidelines and consensus, the results of mngs are divided into true positive, false positive and false negative. Among the test results, 122 cases were positive for mngs, 12 cases were false positive, 16 cases were false negative, 126 cases were positive by traditional test methods, of which 114 cases were "pathogenic". Therefore, the true positive result of mngs was 101 cases, and the positive rate was 36.6%, which was higher than that of the American team (15.7%).

Interpretation standard

The interpretation standard applied by the US team is that Professor C.Y. Chiu published in genome research in May 2019: for viruses, if the detected reads can cover ≥ 3 non overlapping regions on the genome, it is judged as positive; For bacteria, fungi and parasites, rpm-r ≥ 10, that is, when the RPM (reads per million) value of clinical samples is more than 10 times that of negative quality control (NTC), it is determined to be positive.

The interpretation criteria applied by the Chinese team are: exclude the species detected in NTC or the species detected in ≥ 25% of the samples in the first 30 days, and other species; ① For some bacteria, fungi and parasites, when the species-specific reading SSRN ≥ 30 (RPM ≥ 1.5) and the reading ranks in the top 10 among bacteria, fungi or parasites, it is determined to be positive; ② For intracellular bacteria (except Mycobacterium tuberculosis and Brucella) and Cryptococcus, when SSRN ≥ 10 (RPM ≥ 0.5) and the reading is in the top 10 of bacteria and fungi, it is determined to be positive; ③ For viruses, Brucella and Mycobacterium tuberculosis, the results are considered positive when the species (or Mycobacterium tuberculosis (Mycobacterium tuberculosis complex, MTC) and Brucella) SSRN ≥ 3 (RPM ≥ 0.15).

The US team detected 23 different pathogens through mngs, including 14 viruses (20 cases), 6 bacteria (6 cases), 2 fungi (4 cases) and 1 parasite (2 cases). 22 pathogens were detected in China, including 11 kinds of bacteria (45 cases), 7 kinds of viruses (39 cases), 2 kinds of fungi (8 cases) and 2 kinds of parasites (6 cases). Among them, Mycobacterium tuberculosis (14 cases), Listeria monocytogenes (8 cases), varicella zoster virus (16 cases), herpes simplex virus type 1 (11 cases), Cryptococcus neoformans (7 cases) in fungi and pork tapeworm (5 cases) in parasites Accounting for the main proportion.

Analysis and discussion

Two research teams in China and the United States conducted multicenter research on the clinical application of mngs in central nervous system infection, and published the research results one after another. From the detection results, the clinical application of mngs in central nervous system infection widened the detection spectrum of pathogens and improved the overall detection rate of pathogens.

The U.S. team analyzed the drug resistance and typing of multiple pathogens detected, followed up the cases detected only by mngs, and collected clinical feedback. The Chinese team is superior to the American team in the amount and timeliness of test data (20m vs 5-10m, 48hr vs 96hr), and the interpretation of positive results is subdivided. At present, there is no unified standard for the interpretation of positive results at home and abroad.

The results of two major studies show that cerebrospinal fluid mngs is a very useful detection method for patients with suspected central nervous system infection. It can identify emerging infection and disease phenotypes, and speed up the examination and treatment of non infectious causes. It can be considered as a quasi first-line method for the diagnosis of central nervous system infection.

reference

M.R. Wilson, H.A. Sample, K.C. Zorn, et al. Clinical Metagenomic Sequencing for Diagnosis of Meningitis and Encephalitis. N Engl J Med. 2019 Jun 13; 380(24):2327-2340.

S.Y. Fan, X.J. Wang, Y.F. Hu, et al. Metagenomic Next-generation Sequencing of Cerebrospinal Fluid for the Diagnosis of Central Nervous System Infections: A Multicentre Prospective Study. bioRxiv, 658047.

S. Miller, S. N. Naccache, E. Samayoa, et al. Laboratory validation of a clinical metagenomic sequencing assay for pathogen detection in cerebrospinal fluid. Genome Res. 2019 May;  29(5): 831-842.