blood to plasma partitioning using dried blood samples in research

Over the last two years we have published around sixty blogs summarizing some of the amazing research papers that have been published about studies using dried blood samples collected on Mitra^® based on VAMS^® technology and the hemaPEN^® microsampling devices. These blogs have covered research focused on clinical trials, therapeutic drug monitoring, environmental exposure monitoring, omics, serology studies, and related topics.   

A Strong Body of Data Using Dried Blood Samples 

The published research papers we have reviewed have shown that successfully validated methods have been achieved analyzing a wide range of analytes. These include studies measuring small analytes such as for blood lead levels (BLL) to small biomarker molecules such as creatinine. Dried blood samples have also been used to study medium-sized peptides and large biomarkers such as IgG antibodies.

Additional validated methods using Trajan’s Neoteryx dried blood devices have also involved small drug molecules such as antiepileptics and large drug molecules, such as monoclonal antibodies (MAbs). It must be noted, however, that not all analytes can be effectively measured from dried blood extractions.

This is due to both physiological and/or instrumentational incompatibilities, as discussed in a previous blog.  As a result, extraction methodology and analysis need consideration with respect to the analyte, matrix (usually dried capillary blood), and choice of analytical instrumentation.   

What to Consider When Validating Dried Blood Assays  

It is recommended that all assays follow regulatory guidance. The vast majority of papers which we have reviewed align with one or more guidance documents issued by the US (United States) Food and Drug Administration (FDA), European Medicines Agency (EMA), and the International Association of Therapeutic Drug Monitoring and Clinical Toxicology (IATDMCT). Much of the guidance is around assessing typical bioanalytical parameters, such as stability, accuracy, imprecision, and linearity. 

However, the major difference when comparing plasma to dried blood is the presence of the packed cell volume or hematocrit (HCT) in the blood. Indeed, it is recommended to measure the impact that different % HCT has on a number of your analytical parameters. HCT typically ranges from 25-65% and the usual value for most healthy people is around 45%.   

However, the %HCT can impact the way blood spots spread on dried blood spot (DBS) cards, which can lead to analytical biases. Furthermore, %HCT can impact how easily analytes are extracted from any dried blood medium, whether collected on DBS cards or other dried blood microsampling devices.

This issue is outlined in the IATDMCT guidance document and mentioned in a previous blog reviewing a paper from SK Hall on variances in DBS observed for neonatal screening. Finally, the age of the sample also has an impact on sample extractability, which is comprehensively covered in the IATDMCT guidance document. 

Blood to Plasma Partitioning  

We can consider blood to be of two compartments, which are separated by cellular membranes. These two compartments are the intracellular compartment (the space within cells) and the extracellular compartment, known as plasma or serum.

Plasma is the straw-colored liquid that is seen when blood has been carefully centrifuged to avoid hemolysis, and serum is the clear fluid left from blood that has been left to clot and then centrifuged.   

The reason it is important to consider these two compartments (extracellular and intracellular) is because, depending on the analyte’s physicochemical properties and in some cases biochemical properties, they will either partition equally between both compartments or they will preferentially partition into one compartment over the other.   

If analytes partition equally between the intracellular and extracellular compartments, this can have little to no impact or contribution to any observed analytical bias, when compared to a standard plasma or serum assay. However, if analytes accumulate in or on the cellular portion, then whole blood extraction, rather than processing plasma or serum, is conducted irrespective of whether the blood is wet or dry.

This is the case with analytes such as phosphatidyl ethanol (PEth) or the calcineurin inhibitor Tacrolimus. Indeed, for these analytes, concordance between wet and dried blood is often highly correlative with minimal bias reported. 

When Analytes Favor the Plasma Compartment 

There are two scenarios to consider when analytes are portioning into the plasma compartment, and these include complete or partial partitioning.   

Complete partitioning often occurs with certain biomarkers, such as steroid hormones and natural antibodies, are found only in the extracellular compartment and do not partition into the HCT. However, certain drugs such as monoclonal antibodies can also be totally excluded from the cellular fraction too. When this occurs, negative biases are seen in the whole blood when compared to plasma — sometimes upwards of -50%. As a result, there are three main ways to mitigate this natural bias: 

• Validate the method using only dried whole blood, which includes using calibrator and quality controls from the same matrix. The downside of this is that it can be challenging from a practical and acceptance standpoint.   

• Assume population-based correction factor, as reported in several other studies on different analytes and discussed in a previous blog on a c-peptide validation. The upside of this is that it is easy to implement. The downside is that it may not be applicable for populations with a wide hematocrit range. 
     
• Correct per datapoint, by analyzing a sub-aliquot of an aqueous extraction and then measuring either potassium or hemoglobin to correct for the HCT fraction. This approach has been discussed in the published literature by several of research groups, one example of which is summarized in this blog about a research study in the Netherlands. The upside of this approach is that each data point is corrected per blood sample, though two measurements (one for the analyte and one for the hematocrit correction) are needed per extraction. 

Incomplete partitioning is where an uneven proportion of the analyte will partition into one compartment or the other. This is sometimes seen with small drug molecules for example and is discussed in a review on the analysis of antiepileptic drugs from volumetric absorptive microsampling) extracts. 

This phenomenon can also be concentration- and time-dependent.  Indeed, researchers at Astra Zeneca published an approach that intended to mitigate some of this dynamic observed for some molecules.

Final Thoughts from Neoteryx 

When developing an assay, it is important to spike the blood (allow enough time for any partitioning to occur), then sample with a dried blood microsampling device and separate out the plasma for comparative analysis.  If partitioning has occurred, then a negative bias will be observed from the dried blood extract. However, it is important to ensure high extraction efficiency has been achieved as % recovery will cause negative biases too.  Please refer to the Mitra Microsampling User Guide as well as the IATDMCT guidance document for advice on optimizing extractions from dried blood samples.   

It must be noted that if plasma and blood is spiked into each matrix at the same concentration and then extracted separately, then blood to plasma partitioning bias will not be observed. Blood to plasma partitioning can only be observed when the analyte is spiked into the blood and plasma is harvested from the same blood samples, after capillary microsampling.  

For an excellent deep dive into the impact of blood to plasma partitioning, the reader is encouraged to read a paper by Gary Emmon and Malcom Rowland entitled “Pharmacokinetic considerations as to when to use dried blood spot sampling.”   

This blog includes content curated from published study papers that were summarized for our readers by James Rudge, PhD, Microsampling Technical Director. To learn more about the important research discussed here, please visit our Microsampling Resource Library, where we provide access to the original, third-party papers.

Image Credits: Trajan, Neoteryx, iStock, Dreamtime