Automated On-Line Extraction and Analysis of Dried Blood Spot Cards

The use of dried blood spot cards for screening neonates for metabolic and genetic disorders was conceived and proposed almost 50 years ago. Since then, newborns in many countries have had their heels pricked with a needle in order to collect a single drop of blood onto a piece of blotter paper. The resulting spot provides enough material for a battery of genetic disorder tests. However, the use of dried blood spot (DBS) cards has only recently made its way into preclinical and clinical testing of pharmaceuticals.

Most clinical tests and virtually all preclinical animal studies have traditionally required much larger volumes of blood or body fluids. The use of such large volumes creates a range of logistical, technical, and ethical problems. The need to ship whole blood and serum on dry ice dramatically inflates the costs of global clinical trials. Blood shipments must also be labeled as biohazardous.

Benefits of dried blood spot cards

Preclinical studies involve laboratory animals with small blood volumes, and only a limited number of samples can be taken. Researchers often have to sample and sacrifice one group of animals early in the experiment while saving an additional group for sampling and sacrificing later, making it impossible to track metabolic and pharmacodynamic changes in each animal over the length of the study. There is significant pressure to reduce the use of lab animals for ethical and cost reasons. Utilizing non-biohazardous DBS cards can provide better data and reduce the number of animals required for each experiment.

The invasiveness of phlebotomy can be a challenge in recruiting subjects for clinical trials. The use of DBS cards obviates the need to insert a cannula in the arm vein of the subject, and a finger or heel prick is much less traumatic. This can enable easier patient recruitment, especially of seriously ill patients.

In addition to making trials more patient-friendly, DBS cards enable minimally trained staff in remote areas with poor medical coverage to take samples in critical global trials, and the samples themselves can be shipped at ambient temperatures in an envelope. The use of DBS cards removes the need to centrifuge, aliquot, and freeze samples at the collection point.

Use of dried blood spots for preclinical and clinical trial applications: An old solution to new challenges

Due to all of these advantages, the use of DBS cards has gained attention for preclinical and clinical trial applications. The attractiveness of DBS cards for these applications is due in large part to the availability of highly sensitive analytical techniques such as mass spectrometry (MS), which have enabled new assays on tiny volumes of blood and body fluids.

The use of DBS cards is straightforward, using manual techniques developed years ago. The steps include collection of the sample onto the card, manual punch-out of a portion of the sample, and simple extraction procedures. Well-established small-volume chromatography methods can then be used to separate the analytes of interest, often using MS as a highly sensitive and selective detector.

Automating dried blood spot card analysis

For DBS card analysis to be a practicable alternative to large blood sample analysis techniques, it must be able to handle the large number of samples produced by preclinical studies and clinical trials. However, traditional DBS card analysis techniques are highly labor intensive, requiring collecting and recording the cards, then manually punching each one, extracting the sample from the blot paper, and centrifugation and transfer of the supernatant to a vial for analysis. Each step presents opportunity for human error and variability in results. An automated approach is essential to meet throughput needs and minimize variability.

An automated DBS card extraction and analysis system must handle a large number of cards per batch, track all replicates of all samples on all cards, extract each sample efficiently, and provide the sensitivity required to work with such small samples. Most importantly, it must provide performance at least equivalent to that of the manual system.

Anatomy of an automated DBS card system

A suitable automated system would provide automated card handling and sample tracking. An on-line extraction system would obviate the need for hole-punching. A multivalve chromatography system can provide the multisolvent and multicolumn capability required to automatically extract and separate analytes contained in the blood spots. An MS system, particularly one that can perform multiple reaction monitoring (MRM) analysis, can provide high sensitivity and selectivity. Finally, sophisticated control software can provide sample tracking, instrument control, and data processing and reporting.

Figure 1 – The automated extraction component of the Agilent Automated Card Extraction system (Agilent Technologies, Santa Clara, CA, developed in conjunction with Prolab GmbH [Reinach, Switzerland]), illustrating the card storage system, the camera for documenting and tracking each card, and the robotic gripper for moving the cards to the clamp module, which interfaces with the chromatography system.

Such a system is illustrated in Figures 1 and 2. A robotic arm provides DBS card handling of large numbers of cards stored in a rack, and a camera provides a photo of each card before and after sample extraction to ensure accurate sample processing and tracking. Each card is clamped into an apparatus that interfaces with the chromatography system. Two trap columns and three pumps are used to extract the sample spot, deliver the analytes to the analytical column, and provide the solvent gradient required for efficient separation (Figure 2). The analytes are then delivered to a triple quadrupole mass spectrometer for detection and quantitation of the analytes.

Figure 2 – Schematic of the fluidics for extracting, capturing, separating, and quantitating the analytes in the blood spots. Included are the multiport valves, three pumps, two trap columns, one analytical column, and a triple quadrupole mass spectrometer for MRM analysis.

Proof of concept

An automated system must demonstrate its ability to deliver efficient and accurate extraction and analysis of analytes relevant to preclinical studies and clinical trials. Clozapine is a drug essential to the treatment of schizophrenia patients who have not been helped by other medications or who have tried to commit suicide. It has some significant side effects and is typical of the kind of drugs that could be the subject of toxicology and pharmacokinetic studies in preclinical and clinical situations.

Clozapine analysis from DBS samples was conducted with the illustrated system to demonstrate its sensitivity, linearity, accuracy, reproducibility, and precision relative to traditional off-line extraction of the same set of samples. Clozapine is metabolized to norclozapine and clozapine-N-oxide in rats, and this study provided data for all three analytes.

Automated extraction and analysis begins with the input of sample information into the control software, specification of acquisition method details, and starting the run. A camera captures card images, both pre- and postextraction, as well as the bar code for each card. Multiple extractions can be performed on the same blood spot.

Detection and quantitation of the analytes was performed in replicate for each sample using a triple quadrupole mass spectrometer in positive electrospray ionization (ESI) mode and MRM. Two transitions were used to monitor each analyte. Calibration curves were constructed for clozapine and its two metabolites, from 0.5 to 1000 ng/mL, giving a dynamic range >4 orders of magnitude and coefficients of linearity (R2) for clozapine >0.999 for both the automated system and the manual method (Figure 3).

Figure 3 – Calibration curves for clozapine using both the automated and manual methods, illustrating a 0.5 ng/mL lower limit of quantitation, >4 orders of magnitude of linear dynamic range, and R2 values >0.999.

Even at the limit of quantitation (LOQ) for the automated extraction of 0.5 ng/mL, analysis of the three metabolites provides highly reproducible quantitation (Figure 4). The LOQ for the manual method was also 0.5 ng/mL. Accuracy was measured as the average of three replicates of percent recovery measurements performed at each concentration across the calibration curve. Reproducibility was determined as the percent relative standard deviation (%RSD) of the three replicate measurements at each concentration across the calibration curve. Precision was measured as the %RSD of the response factors, averaged across all concentrations. All three values were determined using both automated and manual extraction, and are shown in Figure 5.

Figure 4 – Extracted ion chromatograms of the quantifier ions for clozapine and its metabolites, performed in triplicate on the automated system. Peak area is displayed on each trace and is reproducible for each analyte.

It can be seen that the automated system generated equivalent or better values for all of these parameters for all three analytes, when compared to manual extraction of DBS cards. The range of accuracy values was usually narrower for the automated determinations, while the range of reproducibility and precision values is comparable for the two methods (Figure 5).

Figure 5 – Comparison of the analysis results for clozapine and its metabolites, using both an automated and a manual DBS card extraction method.

The significant difference between the two methods is the total analysis time. The automated extraction and analysis system completes an analysis in about 1 hr per analysis, while the manual extraction method requires 5 hr per analysis.


Dried blood spots offer the opportunity to transform preclinical and clinical trial testing by substantially reducing the need for laboratory animals and the cost of shipping samples, facilitating recruitment of clinical trial subjects, and expanding the ability to collect samples in remote areas. However, to fully meet its potential, the analysis of DBS cards must be automated to meet the throughput demands of preclinical and clinical trial testing and minimize the opportunity for human error. An automated DBS card extraction and analysis system has been demonstrated to meet or exceed the performance of manual methods for sensitivity, dynamic range, accuracy, reproducibility, and precision in the analysis of clinically relevant analytes. In addition, it requires only one-fifth of the time required to perform a manual extraction and analysis, with less opportunity for human error.

Lester Taylor, Ph.D., is Director of LC/MS Product Marketing; Na Pi Parra, Ph.D., is a Product Manager; and Doug McIntyre is a Product Manager, Agilent Technologies, 5301 Stevens Creek Blvd., Santa Clara, CA 95051, U.S.A.; e-mail: