Stability of Proteins in Dried Blood Spot Biobanks

This technical evaluation study looks at the feasibility of utilizing dried blood spot (DBS)-based biobanks to enable personalized longitudinal analysis of dynamic biomarkers that might be invaluable in identifying diseases at earlier and more easily curable stages. This places different requirements on biobanking compared to genetic studies, since markers such as proteins are likely to vary between individuals and over time, necessitating individual baselines to follow each subject. DBS collection is a very convenient and attractive option for collecting and storing such samples, and has been shown to be amenable to analysis using PEA.

The authors looked at multiple variables linked to sample drying and storage, using the Onc I v2 and Oncology II panels for different parts of the study. Firstly, they could show that drying samples had only a minor influence on detection of proteins (comparing whole blood-ETDA, plasma-EDTA and dilution series of pooled recombinant antigens in liquid and dried formats). Correlations between liquid and dried samples for all matrices were very high (~0.97), although slightly lower than between technical replicates within each matrix. As expected, the correlation between blood and plasma levels for the proteins were significantly lower for both liquid and dried samples. Similar result were seen with the pooled antigen samples. They also looked at homogeneity of the DBS samples and found that the precise location of the punches taken from the DBS has a negligible effect on protein levels.

To look at storage parameters, neonatal samples stored as DBS at either +4°C or -24°C for between 0 to 30 years were examined for individual protein levels. Clear differences among the 92 proteins were seen, with some (e.g. S100A11) unaffected by storage, while others such as TXLNA decreased significantly and steadily over time. Overall, the 10 year storage medians for protein abundance were 93% at -24C and 80% at +4C.For the affected proteins, half-lives varied between 10-50 years. Looking at the 30 year storage time (and unsurprisingly), more proteins remained unaffected by storage at -24C compared to +4C (76% & 34% respectively).

Overall the study shows that choice of sample matrix and storage temperature are much bigger sources of variability than the drying process, and given the advantages easy, convenient sample collection, DBS-based biobanks have great potential.