Researchers report on the largest cohort of patients tested using plasma mcfDNA sequencing for pathogen detection

By Pooja Toshniwal PahariaJan 16 2023Reviewed by Danielle Ellis, B.Sc.

In a recent study posted to the medRxiv* preprint server, researchers identified a wide variety of pathogenic organisms using plasma microbial cell-free deoxyribonucleic acid (mcfDNA) sequencing.

*Important notice: medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.

Background

Sequencing mcfDNA in serum represents the combination of genome sequencing, computational analysis, and cell-free DNA detection as a clinically valuable hematological analyte. Previous studies have reported satisfactory performance of the technique in diagnosing infections, with and detection rate greater than those documented for conventional techniques.

Plasma mcfDNA sequencing enables rapid, unbiased, and non-invasive identification of pathogenic organisms, generating opportunities for improved diagnosis (and management) of fatal, deep-seated and bloodstream infections. The analysis may be especially useful for immunosuppressed individuals, who are most vulnerable to poor infection outcomes, by potentially reducing invasive procedural requirements, decreasing time to particular etiologic diagnoses, and may optimize antibiotic therapy.

However, the usage of the analysis is reportedly limited in regular clinical practices, and data on the diagnostic performance of sequencing mcfDNA from serum for diagnosing infections at a large scale in relation to positivity rates, quality metrics, taxological diversity, and reporting times are scarce.

About the study

In the present study, researchers identified pathogenic organisms using mcfDNA sequencing analysis performed using serum samples obtained from hospitalized individuals across the United States (US) to provide additional valuable insights into the depth and breadth of pathogen identification.

The commercial sample data comprised information on patient demographics, prescribing clinicians, and the international classification of diseases, tenth revision (ICD10) codes for infections], obtained from TRF (test request forms), and reports of tests conducted between 1 April 2018 and the middle of September in the following year.  

Documented pathogenic organisms and metrics for laboratory diagnostic performance were evaluated. The team performed serological mcfDNA sequencing at the Karius laboratory. DNA was extracted from the samples, and sequence libraries were prepared. The sequenced reads of microbes were aligned with those included in a genomic database comprising more than 20,000 genomic assemblies of more than 16,000 microbial species, for which more than 1,500 microbial taxa have been documented, including DNA viruses, bacteria, parasites, and fungi.

The team documented microbes detected in significantly high amounts in mcfDNA concentrations, denoted as MPM. In addition, the reports contained the range and median MPM values reported for every microbe reported in the previous 1,000 samples and the reference interval, evaluated from 675 donor individuals with asymptomatic infections, for comparative assessments.

Results

In total, 19,739 serological samples obtained from 16,172 individuals were tested. A median duration of 26.0 hours between receiving samples at the Karius clinical laboratory and report preparation was noted, and valid testing reports were available for 96.0% of the blood samples. Among the participants, 65.0% were adult individuals, and 29.0% of individuals presented with ICD10 codes during the testing period.

Fifty-eight percent (n=10,752) of reports, belonging to 8,849 individuals, yielded ≥1.0 taxon for 22,792 identifications, covering 701 distinct microbial taxa [bacteria (75%, n=526), fungi (15%, n=103), viruses (7.0%, n=47), and parasites (3.0%, n=24)]. The 50 most frequent taxa comprised environmental and commensals, including 36 bacterial organisms, nine viral organisms, and five fungal organisms.

Opportunistic fungal organisms [Aspergillus (n=374), Pneumocystis jirovecii (n=258), Mucorales (n=196), and dematiaceous fungi (n=33)] comprised 4.0% of all identifications. Pathogens that were difficult to diagnose [vector-borne and zoonotic pathogens (n=247), Mycobacteria (n=144), Legionella species (n=80), systemic dimorphic fungal organisms (n=78), Nocardia (n=69), and protozoan parasites (n=57)] comprised 3.0% of all identifications.

All co-detections were documented with identical MPM values. Analysis of 18,690 reports of 15,165 individuals, with 12% (n=1,839) of patients had ≥1.0 repeat tests during the study interval, showed that the test met all metrics concerning quality control for quantifying taxa, as MPM values, for 90% (n=9,690) of the positive samples. The main metrics for test performance in the production dataset did not significantly differ from those documented for the initial 2,000 samples tested at the laboratory and documented during the initial test validation study. Oncology/hematology and infectious diseases specialists prescribed the test most often.

In total, 11,023 bacteria were detected, including anaerobes (25%), Streptococcus (12%), Enterobacterales (18%), Staphylococcus (12%), Rothia (5.0%), Haemophilus (4.0%), Enterococcus (10%), Acinetobacter haemolyticus (2.0%), Pseudomonas aeruginosa (7.0%), Helicobacter pylori (2.0%), and Stenotrophomonas maltophilia (1.0%).

A total of 3,982 viruses were detected including cytomegalovirus (32%), Epstein-Barr virus (23%), human herpes virus 6B (12%), herpes simplex virus (12%), BK polyomavirus (9.0%), torque teno virus (3.0%), human adenovirus C (4.0%), human herpes virus 7 (3.0%), and human adenovirus B (3.0%). In total, 920 fungi were detected including Pneumocystis jirovecii (28%), Candida albicans (28%), Aspergillus fumigatus (20%), Candida glabrata (12%), and Candida tropicalis (11%).

Conclusion

Overall, the study findings showed a spectrum of pathogenic organisms identified by serological mcfDNA sequencing, reasserting pathogenic pervasiveness while also detecting pathogens not frequently identified by traditional techniques.

*Important notice: medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.