Longitudinal clinical studies are an important way to understand how people’s immune responses change over time. But repeatedly asking participants to travel to a clinic for blood draws can make studies difficult to recruit for and hard to maintain. At-home blood collection could help solve this problem, but before replacing traditional blood draws, researchers need to test whether the two approaches produce comparable results. A new study in Microbiology Spectrum led by Dr. Linda Sircy with the Boeckh and Waghmare labs at Fred Hutch put the two collection methods to the test.
To compare the methods, the researchers used VirScan, a comprehensive antibody-screening assay that can test a single blood sample against over 113,000 unique peptides representing more than 450 different pathogens, including viruses, bacteria, fungi, parasites, and allergens. VirScan works by displaying these peptides on the surface of bacteriophages (viruses that infect bacteria). When mixed with a blood sample, antibodies present in the patients’ blood bind to any peptide fragments they recognize, indicating previous immune system exposure to the relevant pathogen. Sequencing which peptides get captured reveals a detailed map of a person's entire history of pathogen exposure, from childhood chickenpox to recent respiratory infections. VirScan requires only a very small amount of blood: as little as 5 microliters of serum or plasma.
“For our respiratory virus surveillance studies, we’re focused on using remote and decentralized sample collection methods that participants can perform at home,” Lead author Linda Sircy shares. “By providing clinical study volunteers with options that are less burdensome than traveling to a clinical site to provide blood samples, we can ideally expand our volunteer network to a larger regional area or even nationwide. Because these devices are also easy-to-use and relatively low cost compared to paying for clinical staff time and facility fees for larger studies, we can more frequently collect samples during a single study, which is especially useful for longitudinal studies.”
Because VirScan needs so little blood, it's an appealing candidate for at-home collection devices like Tasso, a capillary blood self-sampler that attaches to the upper arm and can collect up to 600 microliters of blood without a clinic visit. To determine whether Tasso-collected blood gives the same VirScan results as a traditional venous blood draw, the researchers took advantage of an existing bank of samples collected from the same person on the same day by the different methods. They analyzed 36 pairs of samples from 34 healthy adults who had participated in a longitudinal SARS-CoV-2 study. One sample was collected through conventional venipuncture at the clinic, producing plasma, while the other was collected using a Tasso device attached to the upper arm, producing serum (since an anticoagulant is not used with Tasso collection). Running both sample types through VirScan side-by-side let them directly compare the number of distinct antibody-binding peptides detected, known as "epitope hits", and the strength of the epitope-binding signals across hundreds of pathogens.
Overall, the two collection methods produced extremely similar results and the team found no systematic bias between collection methods. Across the broad pathogen library, 95% of paired samples differed by no more than three detected antibody-binding epitopes. When the researchers focused specifically on 35 clinically relevant respiratory viruses, herpesviruses, and bacteria, most pathogens showed moderate-to-strong agreement between the two blood sample types.
Importantly, there was no consistent tendency for either collection method to produce higher or lower antibody measurements. Even when some individual samples showed larger differences, those differences were not explained by how much blood the Tasso device collected or by delays in processing the mailed samples.
To figure out why some measurements still varied between paired samples, the team tested the exact same blood sample from a single healthy donor 10 times across three independent VirScan runs. Even with zero variation in the sample itself, some pathogens showed substantial detection variability, revealing that VirScan itself has meaningful inherent inter-assay variability. The technical noise, not the blood collection method, explained most of the discrepancies observed between Tasso and venipuncture samples.