Search
Close this search box.

A device for dried blood microsampling in quantitative bioanalysis: overcoming the issues associated blood hematocrit

Neil SpoonerPhilip DenniffLuc MichielsenRonald De VriesQin C JiMark E ArnoldKaren WoodsEric J WoolfYang XuValérie BoutetPatricia ZaneStuart KushonJames B Rudg

Abstract

Aims: A cross-laboratory experiment has been performed on a novel dried blood sampler in order to investigate whether it overcomes issues associated with blood volume and hematocrit (HCT) that are observed when taking a subpunch from dried blood spot samples. Materials & methods: An average blood volume of 10.6 μl was absorbed by the samplers across the different HCTs investigated (20–65%). Results: No notable change of volume absorbed was noted across the HCT range. Furthermore, the variation in blood sample volumes across six different laboratories was within acceptable limits. Conclusion: The novel volumetric absorptive microsampling device has the potential to deliver the advantages of dried blood spot sampling while overcoming some of the issues associated with the technology.

Figure 1.  Volumetric absorbtive microsampler before (left) and after (right) filling with blood.
Figure 2.  Gravimetric determination of the volume of blood absorbed by volumetric absorbtive microsampler tips at various hematocrit values.Determinations were performed at six different laboratories. Each data point is the average of six determinations made at each laboratory.
Figure 3.  Variation in the density of blood with hematocrit determined by each of the six participating laboratories.Laboratories 1, 2, 3 and 6 used human blood while laboratories 4 and 5 used rat blood. Comparisons to published values for human blood density are also illustrated [27]. Each data point is the average of six determinations made at each laboratory.

The use of dried blood spot (DBS) sampling for the collection of samples for the determination of drug concentrations has received a lot of attention recently within the pharmaceutical industry [1,2]. This approach has been widely used for the determination and discovery animal pharmacokinetic (PK), preclinical toxicokinetic, clinical PK and human therapeutic drug monitoring data [3–11]. The interest in the technique has been driven by its advantages over conventional plasma sampling for these study designs. These advantages include;

  • • Reduced blood volumes (<20 μl compared with >200 μl), leading to ethical benefits in rodent PK and toxicokinetic studies (reductions in the numbers of animals used and refinements in the warming procedures required to encourage blood flow) and its applicability to pediatric study designs;
  • • Simplified sample collection workflows, obviating the requirements for centrifugation, plasma transfer and frozen sample storage and transfer;
  • • Increased or suitable stability for some analytes without requiring frozen storage [12–16];
  • • Ability to obtain high-quality samples in locations not previously readily amenable for collection (e.g., patient homes and remote locations);
  • • Cost savings associated with the shipping and storage of study samples at ambient room temperature rather than frozen, and a requirement for reduced amounts of experimental drug substances, particularly for discovery studies.

However, recent publications have highlighted a number of issues that have the potential to adversely affect the quality of the quantitative data obtained from DBS samples when a subpunch is taken from a sample [17–21]. Most practitioners prefer to use the subpunch method for DBS sampling as it simplifies the initial collection and spotting of the sample by removing the need to use accurate volume spotting techniques. Instead, the animal technician or clinician can spot an approximate volume from which a fixed-diameter subpunch can be taken in order to realize an accurate volume at the point of analysis. However, this approach relies on the blood spreading evenly when initially spotted. Recent studies have clearly demonstrated that the blood does not spread homogenously on the DBS material (substrate). Furthermore, the hematocrit (HCT) of the blood affects its viscosity and so gives rise to different-sized DBS spots, with high-HCT blood giving smaller blood spots and low-HCT blood giving larger spots. Thus, the volume of blood in a fixed-diameter subpunch taken from these samples is difficult to determine, resulting in lower-quality drug concentration data for samples, particularly in which the blood HCT varies markedly from that of the control blood used to prepare the calibrant and quality control (QC) samples.

A number of laboratories have published manuscripts describing novel approaches to overcoming these issues regarding HCT and homogeneity with DBS [22–25]. Unfortunately, these approaches rely on spotting an accurate volume of blood, measuring another blood component or are not readily commercially available. Thus, their implementation for day-to-day study support may be impractical.

A novel dried blood sampler, termed the volumetric absorbtive microsampler (VAMS), has been designed in order to deliver the benefits of DBS sampling while overcoming the issues associated with HCT and homogeneity and also enabling further simplification of the sample collection and processing/extraction workflows [26]. The VAMS consists of an absorbent polymeric tip designed to take up a fixed volume of blood (nominally 10 μl) by capillary action. The tip is attached to a handle by a plastic pin (Figure 1). The handle has been designed to be ergonomic to hold and has fins to help locate it within 96-well blocks during extraction and to minimize the possibility of the sampler tip coming into contact with surfaces during storage and shipping. The sampler is filled by holding the handle part of the sampler and dipping only the leading surface of the tip into a pool of blood and allowing it to fill. The tip of the sampler must not be completely submerged into the blood sampler, as this may cause overfilling. When the tip has turned completely red, it is full, which takes 2–5 s, making the device self-indicating. The last part of the tip to turn red is the shoulder. In addition, the sampler is engineered to fit a standard laboratory air-displacement pipette or automated liquid handler in order to allow for automation during extraction of the blood sample.

This article describes results from a cross-company and -laboratory gravimetric experiment to determine the volume of blood absorbed by the sampler at various blood HCT values (˜20, 45 and 65%). Furthermore, in order to convert the weights of the blood absorbed by the sampler into volumes, the density of the blood at each HCT was determined by each laboratory.

Materials & methods

Equipment & reagents

VAMSs were supplied by Phenomenex, Inc. (CA, USA; exclusive distributor for the Mitra™ microsampler, manufactured by Neotryx, LLC, CA, USA). Control EDTA blood was obtained from the following suppliers in accordance with each company’s current policies on informed consent and ethical approval: human blood was obtained from: GlaxoSmithKline (GSK Stevenage, UK) from in-house facilities and used within 7 h of draw; Merck from Biological Specialties (PA, USA) and used within 10 h of draw; Janssen from the Clinical Pharmacology Unit (Belgium) and in-house and used within 4 h of draw; and Sanofi from Biopredic International (Saint-Gregoire, France) and used within 48 h of draw. Rat blood was obtained from Bioreclamation (Brussels, Belgium) and used within 2 weeks of draw (Bristol-Meyers Squibb [BMS]) or from in-house facilities used within 7 h of draw (AstraZeneca [AZ]). Blood samples with different nominal HCT values of 20, 45 and 65% (and additionally at 60 and 69% for two laboratories) were prepared at each participating laboratory by determining the HCT of the control blood by centrifugation and then either removing the appropriate volumes of plasma from centrifuged blood or adding plasma to blood that had not been centrifuged, followed by gentle mixing. The HCT of the modified blood samples was then confirmed using centrifugation [19].

All experimental sites used microbalances that measured to at least five decimal places of a gram. GSK used a Sartorius R200D, Janssen used a Sartorius CPA225D, BMS used a Sartorius Cubis Precision Lab Balance, AZ used a Sartorius MC210P, Merck used a Mettler Toledo SAG 285 and Sanofi used a Mettler Toledo AT261. The pipettes used were Gilson Microman M10 (GSK), Eppendorf Reference (Janssen), Biohit mline (10–100 μl; BMS), Gilson Microman M25 (AZ), Eppendorf positive displacement repeaters equipped with a 0.1-ml ‘Combitip’ (Merck) and Biohit Proline (0.5–10 μl; Sanofi). Scintillation vials were obtained from Thermo Fisher (20 ml: GSK and BMS; 7 ml: Merck), Fiolax (20 ml: Janssen) and Perkin Elmer (20 ml: AZ and Sanofi).

Gravimetric determination of blood volume absorbed

Prior to performing the blood absorption experiments, the balance and weighing areas were cleaned and dried to ensure there was no loose liquid on the balance plate, as evaporation would lead to poor balance stability. Moreover, to improve precision during weighing, vials were placed in the same place on the balance. Finally, to increase humidity and reduce the rate of evaporation, a wet tissue was placed in a beaker and placed within the balance enclosure, but not on the balance plate.

Before weighing, each balance was set to zero and a 2-ml aliquot of the 20, 45 or 65% (including 60 and 69% for two laboratories) HCT blood was placed in a scintillation vial, which was capped and placed in the center of the balance plate.

To determine the weight of an accurately pipetted volume (control protocol) at each HCT level, the initial weight of a capped vial containing 2 ml of blood was recorded. The vial was then uncapped and 10 μl of blood was removed using a pipette. The vial was then recapped, reweighed and the weight was recorded. This procedure was conducted six times per HCT at each laboratory. Once completed, the average density of each HCT at each laboratory was calculated as follows:

To determine the weight of the blood absorbed by the VAMS tip at each HCT level (VAMS protocol), the initial weight of a capped vial containing 2 ml of blood was recorded. The vial was then uncapped and a fresh VAMS was carefully ‘dipped’ into the blood such that only the tip was exposed to the blood. The tip was held in the blood until it appeared to be full (i.e., no white portions were observed) and then for an extra approximately 2 s before being removed. Care was taken when filling the tips not to immerse the tip past the shoulder, as this could result in excess blood being retained and hence introduce a positive bias to any measurements. Furthermore, upon removing the tip after filling, care was taken to ensure the sampler or tip had not contacted the walls of the vial, as this could have caused errors in the result. The VAMS was then discarded. As in the control experiment, each vial was recapped, reweighed and the weight was recorded. This procedure was conducted six times per HCT at each laboratory. The blood volume absorbed on the VAMS tip for each HCT at each laboratory was calculated as follows:

Laboratories 1, 4, 5 and 6 performed the operation to determine the mean weight of blood in a 10-μl aliquot for the above experiments as described earlier, while laboratories 2 and 3 performed the operation by weighing a vial containing 2 ml of blood and then dispensed 10-μl aliquots of blood into it from a separate vial containing blood at the same HCT as that being weighed, and reweighing the target vial after each addition.

Results & discussion

Volume of blood absorbed at different HCTs

The volume of human and rat blood with different HCTs absorbed by the VAMS was investigated in six different laboratories by determining changes in weight after dipping the device into control blood (Figure 2). The average volume of human or rat blood collected across the approximate HCT range of 20–65% was 10.6 ± 0.4 μl (error determined as [2 × standard deviation]/square root of the count). This is in good agreement with the 10.5 ± 0.1 μl determined using radioactive 14C caffeine spiked into human blood followed by measurement of CO2 after oxidation of the dried tip [26]. Furthermore, the slope of the blood volume against HCT plot is close to zero, demonstrating that the blood volume collected by the tip showed little variance with different blood HCT values.

The variability in the blood volume collected across laboratories and HCTs was minimal (CV: 8.7%) and well within the 15% that is routinely accepted for the validation of quantitative bioanalyical methods. However, there was notable variance in the minimum and maximum volumes absorbed (between 9.1 and 13.1 μl). While the reasons for this variance are not known, one possible explanation may originate from the approaches taken by the different participating laboratories to pipette 10 μl of blood, the weight of which is required for the calculation of both the volume of blood in the VAMS tip and the blood density. Laboratories 1, 4, 5 and 6 derived the value for the weight of the blood by removal with a pipette, while laboratories 2 and 3 derived it by the addition of blood with a pipette. It is possible that a small amount of blood may remain in the pipette tip when dispensing, resulting in the weights for the determination of the volume values being marginally higher in the laboratories using blood removal compared with those using addition. This is in turn reflected in the lower density (Figure 3) and higher blood volume (Figure 2) values for laboratories 2 and 3, where addition was used, than those for the other laboratories, where removal was used. Alternative explanations for these differences may be either a variation in manufacturing of the tip or differences in which the VAMS device was used for sampling (i.e., dipping the tip too deeply into the blood pool). Based upon this, it is important that laboratories perform additional experiments as part of the development and validation of a quantitative bioanalytical assay when using these blood samplers. These may include the inclusion of tips from various production batches and that more than one operator should prepare QC samples. It is also important that users of the technology be thoroughly trained before using them to collect study samples.

Conclusion

A device for consistently sampling approximately 10 μl of whole blood is demonstrated. The sampler displays the same benefits as those associated with DBS sampling. Furthermore, the device offers a number of additional benefits over that sampling system:

  • • Simplified sampling: the VAMS is used for accurate volume sample collection and storage/transport. This compares well with DBS, which requires a separate sample collection device (often a capillary) from which the blood is applied to the DBS substrate surface;
  • • Simplified analysis: the whole VAMS tip is extracted without further manipulation, whereas for DBS, the spot must be punched (either in its entirety or partially) prior to extraction. Furthermore, the design of the VAMS is readily amenable to the tip-based automation systems that are present in most bioanalytical laboratories.

The effect of blood HCT on the volume of blood absorbed by the sampler was investigated in the laboratories of six different companies. This demonstrated minimal variation in the volume of blood collected with different HCTs and acceptable differences in blood volumes collected between the participating laboratories using both human and rat blood. These results indicate that the issues associated with variations in the volume of blood analyzed with different HCTs when using DBS sampling are minimized or eliminated with the novel VAMS.

While the VAMS device demonstrates much promise for the accurate collection of small blood volumes for quantitative bioanalysis and overcomes a number of issues associated with DBS sampling, further investigation is required in order to demonstrate its quantitative bioanalytical performance and practical use in busy clinical and laboratory settings. As such, a number of other characterizations of the VAMS for drug concentration measurement have been undertaken and will be reported in separate publications.

Future perspective

Further thorough investigations in order to fully understand the reality of the benefits and any possible issues associated with the bioanalytical and practical field performance of the VAMS device are required before its widespread practical application and regulatory acceptance is realized. If successful, the VAMS has the potential to deliver simplified blood collection, shipping, storage and analytical approaches for the quantitative bioanalysis of pharmaceuticals and biomarkers compared with currently accepted wet and dry processes. In particular, it is likely that the device will see rapid adoption for the collection of high-quality samples that are problematic for current processes (i.e., pediatric studies, subjects in remote geographic locations, home sampling or serial sampling from rodent studies).

Key terms

Hematocrit: Ratio of the volume of red blood cells to the volume of whole blood.

Volumetric absorbtive microsampler: A technique for the collection of dried biological samples for quantitative bioanalysis by absorbing a fixed accurate volume of blood onto an absorbent tip. The commercial name for these samplers is Mitra™.

Executive summary

  • • A blood sampler has been developed that simplifies the sample collection of a specific fixed volume and bioanalytical workflows compared to dried blood spot (DBS) and conventional plasma samples.
  • • Issues associated with sample homogeneity for DBS sampling are eliminated as the entire sample is extracted.
  • • The device minimizes the effect of hematocrit on the volume of the sample analyzed that is observed for DBS samples.
  • • While the between-laboratory variation in human and rat blood volumes collected was within acceptable limits, there is notable interoperator variability, which will need to be further investigated, and appropriate operator training will need to be conducted.

Acknowledgements

Janssen would like to thank D Van Roosbroek. BMS would like to thank S Basdeo, C D’Arienzo, G Cornelius and T Olah. Merck would like to thank I Xie, L Xue and M Wang. The Neoteryx and Phenomenex authors would like to formally acknowledge GlaxoSmithkline’s N Spooner and P Dennif for their significant contribution and support in the development of the volumetric absorbtive microsampler technology. The authors would like to formally acknowledge the scientific contributions of our colleagues in this pilot study who were involved in the development of the volumetric absorbtive microsampler technology and the Mitra™ microsampling device.

Financial & competing interests disclosure

Financial support for this work was provided by Neoteryx and Phenomenex. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.

Open access

This work is licensed under the Creative Commons Attribution-NonCommercial 3.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/3.0/

Scroll to Top