NEW YORK – A team led by researchers at Sweden's Science for Life Laboratory (SciLifeLab) has used a panel of Olink proximity extension assays (PEAs) to measure hundreds of proteins in home-collected dried blood spot (DBS) samples.
Detailed in a study published this week in Nature Communications Medicine, the effort demonstrates the feasibility of measuring large numbers of proteins in self-collected DBS samples, pointing toward potential approaches to large-scale human health studies and improved patient monitoring, said Jochen Schwenk, scientific director of the affinity proteomics unit at the SciLifeLab and senior author on the paper.
In the study, the researchers analyzed DBS samples from 228 Swedish individuals who self-collected using quantitative sample cards from Capitainer during the spring of 2020 and the summer of 2021. They grouped the subjects based on COVID-19 serology data as well as information on their COVID-19 exposure obtained through questionnaires and measured 276 proteins involved in cardiovascular and metabolic processes, looking for associations between protein expression and COVID-19 exposure.
The samples were collected as part of a prior study led by Schwenk and his collaborator Niclas Roxhed, associate professor at KTH Royal Institute of Technology, focused on multi-analyte serology testing of home-collected blood samples during the COVID-19 pandemic. For that project, the researchers purchased a list of addresses from an advertising firm and mailed out Capitainer DBS sample cards with a questionnaire and note requesting that the recipient self-collect a sample and return it.
To Schwenk's surprise, roughly 55 percent of the cards they sent out were returned, a rate he attributes in part to the fact that there was little or no COVID-19 testing available at the time, possibly making people more interested in participating in the study.
The researchers followed this initial round of sample collection with two additional rounds, one during the first wave of the pandemic and another of non-vaccinated individuals during the third wave of the pandemic. Having collected and analyzed the serology data, the researchers then used the Olink assays to measure a broader collection of proteins, looking for differences in the proteomic profiles between seropositive and seronegative individuals as well as between individuals in the early phase of infection and those in the post-infection phase.
The analyses identified several proteins associated with aspects of COVID-19 infection. The protein MBL2, for instance, which has been linked to COVID-19 mortality, was higher in seropositive individuals than seronegative individuals, as was the protein IL2RA, which the authors noted has been linked to longer illness in severe COVID-19 cases. The study also identified protein signatures linked to cell-mediated immune response and tissue damage, which, the authors wrote, are mechanisms behind severe COVID-19.
More than identifying particular protein markers of COVID-19 infection, the researchers aimed to demonstrate the feasibility of using Olink's platform for large-scale proteomic measurements in self-collected DBS samples, Schwenk said.
Such an effort faced several potential challenges, he said, noting that Olink's kits have been developed for plasma analysis but that DBS samples are a somewhat different material, which could pose difficulties. For instance, Schwenk said, as proteins dry out in the DBS sample, it is possible that they may unfold or otherwise change their conformation in ways that could affect antibody binding.
There is also the question of how well protein measurements in DBS samples correlate with measurements in conventional venous blood draws. To explore this issue, the researchers looked at a set of 12 subjects who provided both kinds of samples, finding, Schwenk said, that proteins secreted into the blood by other tissues were "fairly well correlated between the two sample types," while proteins expressed by blood cells were present in larger proportions in DBS samples than in conventional venous samples.
"So you have some components that are simply intrinsically different in abundance," he said, adding that "it's important for us that this is a sort of anchor publication where we can show that this works."
He further noted that since completing the recently published work he and his colleagues have shown they can measure several thousand proteins in DBS samples using Olink's technology.
Leigh Anderson, whose protein analysis firm SISCAPA Assay Technologies has been exploring DBS sampling through its LongitudeDx business, said that while differences between DBS and venous draws could make it difficult to correlate results between the two for some proteins, the approach show promise as a tool for proteomic discovery work.
Anderson, who was not involved in the study, has largely used targeted mass spectrometry for his DBS work but said he and his colleagues are also exploring technologies like Olink that could enable discovery-style work looking at a broader range of proteins.
Anderson said he sees great potential for DBS protein analyses for dense longitudinal measurements of patients. He suggested, for instance, that clinical trial participants could provide DBS samples as frequently as every day, allowing researchers to more closely monitor the effects of therapies being studied.
"There are like 400,000 people in clinical trials in the US at any given time, and those people are spending 99.95 percent of their time outside the clinic and that's essentially unmonitored," he said. "There is some adoption gradually of things like Apple watches to collect external physical data, but they aren't getting molecular data."
Schwenk said he envisioned a variety of uses for the sort of DBS-based approach used in the Nature Communications Medicine paper. He cited as examples population studies of individuals throughout peak allergy seasons to develop profiles linked to allergy and asthma or for remote monitoring of patients following surgery.
Schwenk and his colleagues are currently writing up two additional DBS studies, one analyzing samples collected by residents of Stockholm and Gothenburg during the COVID-19 vaccine rollout and another looking at proteomic profiles and pandemic health outcomes in a Swedish cohort whose health information has been tracked for thirty years.