Despite being commonly used to collect upper airway epithelial lining fluid, nasal washes are poorly reproducible, not suitable for serial sampling, and limited by a dilution effect. In contrast, nasal filters lack these limitations and are an attractive alternative. To examine whether nasal filters are superior to nasal washes as a sampling method for the characterization of the upper airway microbiome and immune response, we collected paired nasal filters and washes from a group of 40 healthy children and adults. To characterize the upper airway microbiome, we used 16S ribosomal RNA and shotgun metagenomic sequencing. To characterize the immune response, we measured total protein using a BCA assay and 53 immune mediators using multiplex magnetic bead-based assays. We conducted statistical analyses to compare common microbial ecology indices and immune-mediator median fluorescence intensities (MFIs) between sample types. In general, nasal filters were more likely to pass quality control in both children and adults. There were no significant differences in microbiome community richness, α-diversity, or structure between pediatric samples types; however, these were all highly dissimilar between adult sample types. In addition, there were significant differences in the abundance of amplicon sequence variants between sample types in children and adults. In adults, total proteins were significantly higher in nasal filters than nasal washes; consequently, the immune-mediator MFIs were not well detected in nasal washes. Based on better quality control sequencing metrics and higher immunoassay sensitivity, our results suggest that nasal filters are a superior sampling method to characterize the upper airway microbiome and immune response in both children and adults.

Respiratory syncytial virus (RSV) is a significant cause of lower respiratory tract infection in infants and elderly. To understand the evolution of neutralizing epitopes on the RSV glycoprotein (G) and fusion (F) proteins, we conducted a multi-year surveillance program (OUTSMART-RSV) in the US. Analysis of 1,146 RSV samples from 2015-2017 revealed a slight shift in prevalence from RSV A (58.7%) to B (53.7%) between the two seasons. RSV B was more prevalent in elderly (52.9% and 73.4%). Approximately 1% of the samples contained both RSV A and B viruses. All RSV A isolates were ON1 and almost all the B isolates were BA9 genotypes. Compared with the 2013 reference sequences, changes at the F antigenic sites of RSV B were greater than RSV A, which mainly occurred at antigenic sites V (L172Q/S173L at 99.6%), Ø (I206M/Q209K at 18.6%) and IV (E463D at 7%) of RSV B F. Sequence diversities in the G protein second hypervariable region were observed in the duplicated regions for RSV A and B, and at the G stop codon resulting in extension of 7 amino acids (22.1%) for RSV B in 2016-17. Thus, RSV surface glycoproteins are continuously evolving, and continued surveillance is important for the clinical evaluation of immunoprophylactic products.