Collecting drops of blood from patients and depositing them on cards, where they dry, has been used for decades as a method to screen newborns for errors of metabolism and to screen for infectious disease. Despite that long history, the technique is only now gaining momentum as a method pharmaceutical companies use to determine the fate of drugs in the bodies of patients. “This is a classic case of a new idea that’s 40 years old,” says Richard M. LeLacheur, vice president at PharmaNet USA, a contract research organization (CRO) in Princeton, N.J.
Dried blood spots had been a nonstarter for pharmaceutical analysis because analytical instruments were not sensitive enough for the quantitative analysis needed to obtain pharmacokinetic data from such tiny samples. In neonatal screening, in contrast, clinicians just need to find out whether something is present, not how much. But in the past few years, improvements in the sensitivity of analytical instruments, especially mass spectrometers, have freed scientists to exploit the many advantages of dried blood spot analysis.
Those advantages include small sample volumes and easy shipment and storage. The tiny sample volumes, typically less than 25 μL, compared with milliliters for plasma, mean that fewer animals are needed during preclinical studies and less blood needs to be drawn from humans during clinical trials relative to conventional blood analysis. Because the samples are spotted and dried on cards, they don’t need to be frozen, leading to changes in shipping practices that can save pharmaceutical companies millions of dollars.
Analysis of dried blood samples also has a number of drawbacks. For example, it is more time-consuming than the analysis of liquid samples, and components of the cards on which spots are collected can interfere with some analyses.
Dried blood analysis is a relatively simple method. Tiny amounts of whole blood from an animal or human are transferred onto a paper card and allowed to dry. A portion of the analyte spot is punched out and placed in solvent to extract the blood from the card. The extract, which can be handled from that point like any other liquid sample, is then analyzed by liquid chromatography and tandem mass spectrometry.
The amount of blood per spot varies from 10 to 100 μL, but 15 to 20 μL seems to be most common. With larger blood spots, multiple analyses can be done from each spot, but the spots are less homogeneous.
“It’s better to have three or four smaller spots” of 20 μL or less, says Neil Spooner, director of bioanalytical science and development at GlaxoSmithKline in Ware, England. “Those small spots are each more homogeneous, and we have a better chance of having at least one good spot,” he says. “If you have only one spot, and someone screws it up a bit, that’s it—that sample’s gone. We decided that it gives a higher quality sample and it’s simpler to do the analysis if we split our sample into multiple spots rather than one large spot.”
The recent push in dried blood spot analysis got its start at GSK. The pharma giant’s scientists were attracted by the fact that the method made it easier to work with the small sample volumes involved in the pediatric testing of drugs, Spooner says. Regulatory agencies are starting to require that such testing accompany an initial filing for drug approval if there’s a chance the drug will be prescribed for children. Pediatric testing requires the use of juvenile animals in supporting toxicology studies. “Imagine how small a juvenile rat is and how much blood you can get from it,” Spooner adds.
GSK might have gotten started with dried blood spots out of sheer necessity, but Spooner and his colleagues quickly realized that the technique had strengths that made it desirable even when samples weren’t volume-limited. A key advantage is the possibility of improving ethical aspects of animal studies, an important consideration in Europe, where the animal-rights community is particularly strong.
Another major benefit is cost savings in clinical trials. “In Phase II and Phase III studies, we’re basically shipping big boxes of dry ice around the world with a few tubes of blood inside,” Spooner says. Provided a compound is stable in blood, which must be demonstrated for each compound, dried blood spot samples can be shipped in an envelope at room temperature.
Industry scientists expect that dried blood spot analysis will not only provide ethical and financial benefits but also improve data quality. They typically use multiple animals to generate a given curve in conventional pharmacokinetic and toxicology studies using plasma. They must do so because they can’t take blood from an individual animal enough times to construct the entire curve, but multiple-animal use introduces a source of undesirable variation in the data.
That source of variability can be eliminated with dried blood spot analysis. The smaller volumes associated with the technique mean that serial sampling can be performed with each animal, PharmaNet’s LeLacheur says.
“If we’re looking to get six time points, we can get all six from a single animal using dried blood spot sampling,” LeLacheur says. “With plasma, we typically obtained two time points per animal, so three animals were needed to get all six points.”
Getting an entire curve from one animal instead of three reduces the inherent variability in the data set, he notes. “There’s an expectation across the industry that we will have better intrinsic preclinical data quality” with dried blood spot analysis, LeLacheur says.
Using fewer animals also means that less of the substance undergoing testing is required for preclinical studies, says Fabio Garofolo, vice president of bioanalytical services at Algorithme Pharma, a CRO in Montreal. “Usually, preclinical studies are performed to test new chemical entities,” he says, and it takes time and money to produce a large amount of these compounds as test substances for additional animals.
GSK scientists have been quite open about their work with dried blood spots, inviting interested scientists from other companies to spend time in GSK labs learning the technique. The entire industry will benefit from additional scientists “thinking about the problems, solving those problems, publishing on those things, and making the technique more acceptable,” Spooner says. “When regulators see the data, they’ll be more aware because there have been lots of publications by a lot of different companies, all increasing the understanding of the technology.”
Indeed, GSK’s competitors want to make sure they aren’t left behind. “We have to do due diligence to make sure the technologies out there don’t give our competitors an advantage,” says Kevin P. Bateman, a distinguished senior investigator in the drug metabolism and pharmacokinetics department at Merck Research Laboratories (MRL). “We put in the time to make sure we’re not missing the boat.”
Researchers at Bristol-Myers Squibb have found that the relatively high stability of compounds in dried blood spots, especially prodrugs and their metabolites, is a key advantage of the technology. “In some cases, instability can be so pronounced that in the time it takes you to draw the sample, generate plasma, and freeze it, you can lose a great deal of the compound you intend to measure,” says Celia J. D’Arienzo, a scientist in drug discovery at BMS. “Even if dried blood spots extend the stability by a few hours, that’s enough for us in discovery to feel we’ve added another level of accuracy to our determination.”
Merck researchers tried using dried blood spots in discovery several years ago but got faster results with liquid samples. “In discovery, it’s all about speed,” Bateman says. “From an analytical perspective, there’s no benefit to using dried blood spots in discovery.”
However, the company does see the benefit of applying this approach to drug development. Merck is using dried blood spots in ongoing pediatric trials and is preparing to use them in multicenter clinical studies in an undisclosed disease area, says Eric J. Woolf, senior director at MRL.
Despite the enthusiasm of GSK scientists for dried blood sampling, even they see little point in using the method in discovery, where “you’re not really taking advantage of the traits it offers,” says Christopher A. Evans, a scientist with GSK in King of Prussia, Pa. In the discovery process, “you’re not sending samples around the world for analysis, and you’re not storing them for a long time period. I know that people in discovery in other companies are using dried blood spots and having great success, but I don’t think it’s worth the extra time.”
Even with dried blood sampling’s array of benefits, convincing management to give the technology a chance has been a “difficult and multistage task,” Spooner says. His department at GSK was perhaps the hardest to win over. “All the advantages are for the rest of the organization, not for bioanalysis,” says Paul Abu-Rabie, Spooner’s coworker at the Ware, England, site. The safety assessment group jumped aboard quickly because the methods simplify bleeding procedures for obtaining samples from animals and minimize the number of animals required, Spooner says.
Convincing the clinical organization has also been a tough job, Spooner notes. “The clinical organization and the project teams are large and changing,” he says. “You have to fight the battle one project team at a time. We’ve not yet reached a tipping point where this has become the standard technology for clinical studies.”
Instead, at GSK, the decision about whether to use dried blood spots is currently being made on a program-by-program basis as drug candidates move from discovery into early-stage development, Spooner says. One holdup has been the impracticability of switching late-stage compounds with a long history of analyses in plasma over to dried blood spot analysis. The pharmacokinetic values obtained from liquid plasma and from dried blood are not directly comparable, and “bridging” studies are required to switch from one matrix to the other.
“Even though you can generate an in vitro number for converting between blood and plasma, it doesn’t always work,” Spooner says. So far, dried blood spot analysis has been used as a number of GSK’s drug candidates have gone through toxicological and Phase I human clinical studies, but it has not yet been used in Phase II trials.
One sector that has been quick to adopt dried blood spot sampling has been CROs, which offer dried blood spot analysis as one of their services. “It’s a technology that everybody is looking into,” says Zoe Cobb, a client manager at Quotient Bioresearch, a CRO based in Cambridgeshire, England. “As a competitive CRO, we need to be at the forefront of new technology.”
Another CRO, Princeton, N.J.-based Covance, already has more than 20 validated dried blood spot assays and seven clients that use them. “Even small companies are very interested in dried blood spots,” says Steven M. Michael, vice president and chief scientific officer of global bioanalytical services at Covance. But they often hesitate to make the leap to the new technology. “Some of them might have only one compound in the pipeline, so they don’t want to start in dried blood spots in their current position—maybe next time around,” Michael says.
In addition to firms’ initial reluctance to get involved, several other hurdles need to be overcome for dried blood spot analysis to be more widely adopted. Perhaps the most pressing need is for improved automation. “With samples in fluids, we have a lot of techniques that are fully automated and allow very high throughput during the analysis of thousands of samples,” Algorithme’s Garofolo says. That’s not the case with dried blood spots, although some automation is available.
“Punching out the spots by hand is slow and results in blistering, at least in my hands,” D’Arienzo of BMS says.
Cobb notes that “automation is getting there, but it’s not quite there yet.” Dried blood spot automation methods “still need to prove themselves,” she says.
Analyzing the sample cards directly—bypassing the punching entirely—would be a huge step forward, and companies are making some progress on this front. For example, Robert S. Plumb of Waters Corp., in Milford, Mass., is collaborating with scientists at GSK to evaluate direct modes of LC/MS analysis of the cards.
GSK’s Abu-Rabie and Spooner are performing direct analysis by interfacing thin-layer chromatography with MS to analyze drugs in dried blood spots (Anal. Chem., DOI: 10.1021/ac901985e). They can further separate dried blood extracts with LC or analyze them directly by MS.
Another direct analysis method is being developed by Advion BioSciences, an instrument manufacturer and CRO located in Ithaca, N.Y. Cofounder and CSO Jack Henion and his team are working on a technique called liquid extraction surface analysis, or LESA, first reported by Vilmos Kertesz and Gary J. Van Berkel at Oak Ridge National Laboratory (J. Mass Spectrom., DOI: 10.1002/jms.1709). In this method, a drop of solvent at the tip of a pipette is brought to the card surface, where it extracts analytes from the blood spot. The solvent droplet is sucked back into the pipette and sprayed into a mass spectrometer.
With LESA, the challenge is keeping the solvent droplet from dispersing on the paper. Henion and coworkers prevent this by creating a rim around the blood spot that prevents the solvent from spreading on the card. Although the method shows promise, it is currently manual and laborious. To be adopted, “it’s got to be easy and simple and as good as or better than extracting blood or plasma from tubes,” Henion says.
But the cards themselves present a challenge for MS analysis. Some of the cards are treated to enhance compound stability, and such additives can suppress ionization of the target compound, says Jonathan McNally, marketing manager for bioanalysis at Thermo Fisher Scientific. The cards—treated or untreated—can also be a source of background ions, Plumb says. “There are inherent chemicals within the paper or cellulose that you just can’t get rid of,” he says.
There’s also room for improvement in sample collection methods. Regardless of whether the subject is an animal or a human, “a vein stick, followed by squeezing the area punctured to collect into a capillary tube, followed by spotting onto paper looks incredibly crude,” says Lucinda H. Cohen, director of the New Jersey discovery bioanalytical group at MRL. She would like to see nanoengineered materials that allow subcutaneous sampling with direct transfer to an adsorbent medium, such as cellulose. This type of material would transform the blood into a dried sample.
And quantitative analysis with dried blood spots can present logistical challenges to bioanalytical labs. For example, preparing standard curves for dried blood spot analysis requires “that we have a supply of whole blood that is fresh and refrigerated,” which necessitates careful management, BMS’s D’Arienzo says. In addition, drying time has to be factored in when whole blood is used to prepare dried blood spot standards, Quotient’s Cobb notes. Compounds used to prepare the standards must also remain stable during that drying time.
Despite its wide applicability, dried blood spot analysis isn’t appropriate for all drugs. Low-exposure drugs, such as inhaled drugs, push the limits of today’s mass spectrometers even in conventional blood plasma analyses, Spooner says. With dried blood spots, such analyses are nearly impossible with current technology.
“If the peak level is only 10 or 20 pg/mL, you’re going to need to get down to single or even subpicogram-per-milliliter levels,” Plumb says. “In a liquid format, that’s pushing the limits of our mass spectrometry technology. In a dried blood spot format, it requires us to think differently.” Plumb is currently collaborating with Evans at GSK to develop nanofluidic technology for dried blood spot analysis.
The biggest unknown for dried blood spots is how regulators will respond to the data obtained from them. Researchers get the sense that European regulators are more accepting than their U.S. counterparts. The U.S. Food & Drug Administration declined to comment for this article, citing “insufficient experience with the technology.”
“Right now there is no guidance” from FDA, Evans says. Various FDA documents provide guidance on how to deal with wet samples, but similar documents don’t yet exist for dry samples. “In the absence of that guidance, we’re doing the best science that we can,” Evans says. “We’ve received some encouraging feedback” from various regulatory agencies, “but nothing has been formally communicated.”
Regulatory acceptance shouldn’t be a problem, Spooner posits, because international guidelines state that kinetics can be measured in blood, plasma, or serum. “Provided you can demonstrate that blood spots equal blood, which we and others have done, there’s no regulatory issue whatsoever,” he asserts.
Initial work with dried blood spots has been so promising that scientists are starting to explore whether they can achieve similar benefits from other dried matrices, such as plasma and urine. With dried plasma spots, Spooner says, “you have more complex sampling and centrifugation” than with dried blood, “but we’d still have huge cost savings by shipping samples at room temperature in smaller packages” for Phase II and III studies.
The drug industry is moving slowly but surely toward the adoption of dried blood spot analysis. “We’ll see a lot more pharmaceutical development performed using dried blood spots in 2011,” PharmaNet’s LeLacheur says. “As the comfort level, regulatory experience, and infrastructure grow, people will realize it’s not a big leap to go into dried blood spots, and the benefits are worth it.”