Probe-based metagenomic sentinel surveillance of viral respiratory infections in primary care: a prospective, national, pilot study.

Publication date: Aug 06, 2026

With the introduction of metagenomics in clinical diagnostics unfolding and the expanding role of pathogen genomics in national surveillance, conditions are favourable for the further maturation of these approaches in public health surveillance. In this study, we aimed to pilot the use of probe-based metagenomics for nationwide sentinel surveillance through general practitioner (GP) networks and for the genomic characterisation of both anticipated and emerging respiratory viruses in primary care. This prospective, pilot study included patients with acute respiratory illness attending GP practices participating in nationwide sentinel virological surveillance in the Netherlands, from Jan 10, 2025, to April 25, 2025. On predefined selection days, 90-100 combined nasopharyngeal and oropharyngeal swab specimens were analysed through parallel metagenomic testing using probes targeting 15 488 strains of human and animal viruses. Results were compared with a standard two-tiered surveillance strategy comprising PCR targeting 17 viruses, followed by amplicon-based nanopore whole-genome sequencing of influenza viruses, SARS-CoV-2, and respiratory syncytial virus (RSV). The primary outcome was detection and in-depth genomic characterisation of viruses within and beyond the scope of standard screening. Sensitivity, specificity, positive and negative predictive values, and genome coverage were analysed. 93 patients were included, with a median age of 51 years (IQR 33-67); 55 (59%) were women and 38 (41%) were men. Overall, 74 (80%) specimens tested positive through standard PCR-based screening. Metagenomic surveillance detected viruses in 88 (95%) specimens, including DNA viruses known to establish latent infections. For viruses targeted by routine PCR, metagenomics showed a pooled sensitivity of 93.8% (95% CI 88.1-98.7), specificity of 99.8% (95% CI 99.5-100.0), positive predictive value of 96.2% (95% CI 91.8-100.0), and negative predictive value of 99.7% (95% CI 91.8-100.0), with a median genome coverage of 99.4% (IQR 96.9-99.9%). Metagenomic data enabled simultaneous full genomic characterisation of circulating viruses targeted by current amplicon-based surveillance, including influenza viruses (success rate 30 [86%] of 35 detections) and RSV (six [86%] of seven), and of viruses that were post hoc characterised by whole-genome sequencing in response to epidemiological findings during the study period (human metapneumoviruses [hMPV], five [50.0%] of ten), and non-targeted viruses such as adenoviruses. The data facilitated characterisation of a reassortant A(H3N2) influenza virus, potential vaccine escape mutants, markers of susceptibility to influenza antiviral drugs and RSV monoclonal antibodies, and a human A(H1N2)v influenza virus infection. The technically robust and comprehensive performance across heterogeneous circulating viruses shown here supports evaluation in larger prospective studies in high-prevalence settings, where implementation might be most cost-efficient. Further optimisation of probe-based enrichment strategies could reduce turnaround time and facilitate integration into routine public health surveillance. The Netherlands Organisation for Health Research and Development (ZonMw), and the Ministry of Health, Welfare and Sport (VWS).

Semantics

Original Article