Dried blood spots analysis with mass spectrometry: Potentials and pitfalls in therapeutic drug monitoring

    Dried blood spots analysis with mass spectrometry: Potentials and pitfalls in therapeutic drug monitoring Marina Venzon Antunes, Mariele Feiffer Char˜ao, Rafael Linden PII: DOI: Reference:

S0009-9120(16)30050-9 doi: 10.1016/j.clinbiochem.2016.05.004 CLB 9286

To appear in:

Clinical Biochemistry

Received date: Revised date: Accepted date:

28 March 2016 5 May 2016 8 May 2016

Please cite this article as: Antunes Marina Venzon, Char˜ ao Mariele Feiffer, Linden Rafael, Dried blood spots analysis with mass spectrometry: Potentials and pitfalls in therapeutic drug monitoring, Clinical Biochemistry (2016), doi: 10.1016/j.clinbiochem.2016.05.004

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

1

Dried blood spots analysis with mass spectrometry: potentials and

SC R

IP

T

pitfalls in therapeutic drug monitoring

MA

NU

Marina Venzon Antunes, Mariele Feiffer Charão, Rafael Linden,

D

Toxicological Analysis Laboratory, Instituto de Ciências da Saúde, Universidade

AC

CE P

TE

Feevale, Novo Hamburgo-RS, Brazil

*Corresponding author: Rafael Linden Toxicological Analysis Laboratory Instituto de Ciências da Saúde Universidade Feevale, Novo Hamburgo-RS, Brazil Rodovia ERS 239, n. 2755 CEP 93352-000 Novo Hamburgo-RS, Brazil e-mail: [email protected] Tel. 55 51 35868800

2

IP

T

ACCEPTED MANUSCRIPT

SC R

Abstract

NU

Therapeutic drug monitoring (TDM) relays in the availability of specialized laboratory assays, usually available in reference centers that are not accessible to all patients. In

MA

this context, there is a growing interest in the use of dried blood spot (DBS) sampling, usually obtained from finger pricks, which allows simple and cost-effective logistics in many settings, particularly in Developing Countries. The use of DBS assays to estimate

D

plasma concentrations is highly dependent on the hematocrit of the blood, as well as the

TE

particular characteristics of the measured analyte. DBS assays require specific validation assays, most of them are related to hematocrit effects. In the present

CE P

manuscript, the application of mass spectrometric assays for determination of drugs for TDM purposes in the last ten years is reviewed, as well as the particular validation

AC

assays for new DBS methods.

Keywords: dried blood spots; therapeutic drug monitoring; mass spectrometry

ACCEPTED MANUSCRIPT 1.

3

Introduction Therapeutic drug monitoring (TDM) relays in the availability of sophisticated

laboratory assays in order to be performed properly. Usually these assays are available

T

in specialized reference centers, not accessible to all patients. In this context, there is a

IP

growing interest in the use of dried blood spot (DBS) sampling, usually obtained from

SC R

finger pricks, which allows simple and cost effective logistics in many settings, particularly in Developing Countries. Other additional advantages of DBS for TDM, reviewed by Edelbroek et al. (2009)(1) and Wilhem et al. (2014)(2), include minimally

NU

invasive sampling, high analyte stability and the possibility of self-sampling by patients. However, DBS sampling is not free of drawbacks. Self-sampling could be

MA

associated to contaminations and samples with unacceptable quality. Clinical use of DBS data for TDM requires extensive clinical validation is once capillary blood from finger pricks could present different concentrations from venous blood. Additionally,

D

varying hematocrits affect the volume of blood in a spot with fixed diameter obtained

TE

from a supporting matrix the blood-to-plasma partition of the measured analyte. Moreover, the small amount of sample available for testing, usually in the range of 5 to

CE P

50 µL, is demanding to the testing laboratory. This latter issue is usually assessed by using analytical methods based on mass spectrometry, either gas chromatography-mass spectrometry (GC-MS) or liquid chromatography-mass spectrometry (LC-MS/MS), which associates the high sensitivity and specificity required for TDM purposes,

AC

particularly for the small DBS samples(3). The availability of GC-MS and LC-MS/MS instruments in clinical laboratories greatly improved the analysis of DBS samples in the current decade.

The aim of this manuscript is to review the application of mass spectrometric analytical methods for DBS testing in the context of TDM, discussing the clinical application of this alternative sampling strategy and the specific assay development and validation issues to be addressed in order to implement DBS assays, with a focus in published applications in the last 10 years. 2.

Use of DBS to estimate plasma concentrations The interpretation of drug concentration measurements in the context of TDM

usually is based on reference ranges established in plasma or serum samples. As DBS

ACCEPTED MANUSCRIPT

4

are essentially measurements in whole blood, there is a need to convert the information obtained in this matrix to plasma levels. The hematocrit (Hct) of the sampled blood has a major influence in this process for two major reasons: 1) the viscosity of blood affects

T

the amount of sample present in a matrix punch of fixed size, in a way that is dependent

IP

of the particular matrix used for spotting the blood, and 2) the proportion of red blood cells and plasma in the sample modifies the relative concentration of the drug on these

SC R

blood compartments.

The majority of published applications of DBS sampling for TDM had evaluated Hct ranges where accuracy and precision of the assays were acceptable and made no

NU

correction for the varying amount of sample due to different Hct values. However, Vu et al. (2010)(4), using several different cellulose matrices for DBS, estimated the resulting

MA

volume of blood in the spot in Hct values of 0.20, 0.35 and 0.50 for moxifloxacin determination. For the three Hct values, blood spot areas (ADBS) obtained after spotting volumes in the range of 10 to 60 µL were measured from digital images, using a

D

specific software, and the estimated volume in the spot (Vest) was calculated with

TE

equation Vest = (π . r2 . Vb) / ADBS, being spot radius (r) fixed at 4 mm. This approach was based on the assumption of a linear relation between applied volume of blood and

CE P

DBS area. Additionally, regression lines correlating Vest (y) and Hct (x), normalized to 0.35 (the Hct used for preparation of calibration samples - Hctcal) were obtained, being the intercept denominated as Vstd and the angular coefficient denominated b. A

AC

corrected concentration of the analyte in the DBS was calculated using the equation Ccorrected = Cmeasured . (Vstd / Vest) or using the regression coefficients, Ccorrected = Cmeasured . {Vstd / [Vstd + b (Hct-Hctcal)]}. Using Whatman™ 31 ET CHR paper as DBS matrix, the difference between uncorrected moxifloxacin concentrations between lowest (0.20) and highest (0.5) Hct was about 40%, falling to below 15% after the correction procedure described above. Besides variations in the amount of blood present in a DBS of constant size, the determination of the concentration of a drug in plasma from a DBS measurement also requires knowledge of the its partition between the cellular and the water compartments of blood. The ratio between concentrations of a drug measured in blood and plasma is dependent of the unbound fraction in plasma (fu) and the erythrocyte-to-plasma concentration ratio (ρ), as well as the Hct(5). According to Rowland and Emmons(5), the major concern for estimating plasma concentrations from whole blood

ACCEPTED MANUSCRIPT

5

measurements for drugs with a blood-to-plasma concentration ratio close to the lower limit of 0.55, which indicates almost complete amount in plasma, is the variability in fu. In the other hand, for drugs with larger values of the blood-to-plasma concentration

T

ratio, particularly higher than 2, variability in ρ becomes the critical factor. However,

concern in the use of DBS as an alternative to plasma.

IP

for drugs showing little variability in fu and ρ under clinical conditions, there is little

SC R

An estimated plasma concentration can be calculated using the equation Cplasma = Cblood / [(1-Hct) + Hct . fu . ρ](5). When fu is constant, an alternative approach is based on the knowledge of the fraction of the drug in plasma (fp), which can be established in

NU

an in vitro experiment as previously described(6), and is calculated as fp = (Cblood / Cplasma) . (1-Hct). Once fp is known, the plasma concentration can be calculated using

MA

the simplified formula Cplasma =[Cblood / (1-Hct)] . fp(6)(7)(8). Moreover, as DBS is usually based on blood obtained from finger pricks and is composed from arterial capillary blood and some amount of interstitial fluid, drug

D

concentrations can potentially be different from venous blood. As an example, Ashley et

TE

al.(9) found concentrations of piperaquine about 1.7 times higher in capillary when compared to venous blood. As these differences are dependent of the characteristics of

CE P

particular drugs, case-to-case evaluation is necessary during clinical validation. The use of linear regression correlating plasma and DBS concentrations from clinical data in the validated range of Hct values could potentially account for overall differences in drug

AC

concentrations in both matrices(8). Based on the above, it is obvious that knowledge of hematocrit is critical to the translation of DBS analysis to interpretable plasma levels. The most straightforward approach currently available to determine the Hct of a DBS was proposed by Capiau et al. (2013)(10), and is based on the measurement of the concentration of potassium in a DBS punch. The authors found a linear correlation between potassium concentrations and Hct in the range of 0.19 to 0.63, with acceptable accuracy and precision. Moreover, concentrations of potassium in DBS were stable up to 55 days at room temperature. This approach can be used to calculate plasma concentrations using the described equations or to evaluate if a particular sample has a Hct within the validated range of a certain method. Even considering that a stable relation between concentrations measured in DBS and plasma could be estimated by validation and in vitro experiments, a clinical

ACCEPTED MANUSCRIPT

6

validation study with patient samples is a mandatory step before implementation of a DBS assay in routine. Quality assurance and validation of DBS assays

T

3.

IP

An adequate quality control is essential to ensure meaningful data in TDM testing. Several pre-analytical, analytical and post-analytical variables influence the DBS

SC R

analysis and must be taken into account during the development and validation of a new DBS assay, as discussed below.

NU

3.1 Choice of matrix

MA

During method development, testing for the most appropriate type of matrix can be an efficient strategy to optimize the DBS analysis. The most common matrix for DBS are cellulose-based papers. Papers are mostly differentiated by its composition,

D

thickness and resistance to spreadability of blood. These characteristics may give rise to

TE

differences in extraction recovery, matrix effects, analyte stability and chromatographic, Hct and volume effects (11).

CE P

There are two main types of commercially available paper cards suitable for DBS: chemically untreated and treated papers. The untreated are the most commonly used, particularly the pure cellulose Whatman 903® and Ahlstrom 226®, which are registred by the US Food and Drug Administration (USFDA). Novel substrate materials,

AC

allegedly less subject to Hct effect on spot size and analyte recovery are under research, such as the Agilent Bond Elut DMS® card, with promising results (2). The chemically treated group of matrices consists of cellulose papers treated with different proprietary chemicals intending to lyse cells, inactivate pathogens, and denature enzymes and other proteins. Among them, Whatman FTA®, FTA Elute®, FTA DMPK-A® and FTA DMPK-B® are the most used in this group. Alternatively, untreated paper can be impregnated with chemicals in order to improve stability of some analytes (11). 3.2 Collection procedure: DBS sample collection procedures must follow uniform procedures to minimize the potential effects of pre-analytical errors, such as overlapping or smeared spots. Peck et al. (2009) evaluated the variation of blood volumes and geometries of 422 DBS samples obtained from 138 patients. The study showed non-idealities from blood spot

ACCEPTED MANUSCRIPT

7

collection, including low blood volumes (2 to 72 μL; mean: 25 ± 13 µL), multiple-drop applications, and aberrant sample geometries not consistent with single-drop applications, indicating the need for continuing education of bloodspot collectors (12).

T

Contamination is another concern for DBS sampling, which can lead to inaccurate

IP

determinations. It can result from the use of topic anesthetic creams, disinfectants and handling of the drug prior to collection. In this regard, the European Bioanalytical

SC R

Forum (EBF) proposed the concept of good blood-spotting practices (GBSP) (13,14). An outline for DBS collection procedure is listed below (11,15,16): 1.

Prior to the collection, any contact with the target site of the matrix card must be

2.

NU

avoided.

If the participant’s hands are cold, massaging or warming the collection site before

MA

pricking can stimulate local blood flow. 3.

Clean puncture site with 70% isopropyl alcohol.

4.

Use a sterile, single-use lancet to prick the finger just off the center of the tip of the

Wipe the first blood drop away with a sterile gauze pad to remove the tissue fluid from the sample.

Carefully position the collection paper below the finger and allow the drop to fall of

CE P

6.

TE

5.

D

middle or ring finger (lancets with blades produce better blood flow than needles).

its own weight. Up to five drops of blood (average of 50 µL per drop) are applied to the DBS matrix. For a better blood flow, gently milk the hand starting at the wrist and work down to the base of the finger to produce blood flow, avoiding to squeeze

AC

the finger. The donor’s finger should never touch the DBS matrix. Furthermore, do not place blood on top of blood, what can result in sample concentration. 7.

Allow the spot to dry for 3-4 hours in horizontal position, allowing contact with air on both sides of matrix.

8.

Store/transport samples in plastic bags with desiccant, under room temperature, refrigerated or frozen, depending on analyte stability. Samples must be assessed for quality following collection and prior to assay,

including proper packaging and saturation, appearance and size of spots (15).

3.3 Preparation of calibrators and quality control samples Calibration standards and quality control samples must be prepared at an Hct level similar to the study population. Different approached to prepare quality control samples

ACCEPTED MANUSCRIPT

8

with a target Hct from anticoagulated whole blood were described. One alternative is based on the separation of plasma and red blood cells (RBC) by centrifugation and a proper volume of each are pipetted and pooled again to obtain the target Hct (6–8,17)

T

Alternatively, after centrifugation of blood, the appropriate amount of plasma can be

IP

removed or added to obtain samples with a target Hct (10,18–20). A third, time consuming, approach requires washing erythrocytes with saline and polling cells with

SC R

plasma in proper volumes (4,16,21). Recently, Koster et al. (2015) evaluated the performance of first and second above mentioned approaches. Although the first procedure appears to control both the volume of plasma and RBC, the obtained blood

NU

samples presented a significant bias in the Hct (-7.9% to -16.1%), probably due to the presence of residual plasma in the centrifuged RBC fraction. On contrary, the second

MA

procedure (adding or removing plasma from blood with a known Hct) caused no error in the Hct (2.1% to 3.6%) and was considered the preferred procedure. However, it is recommended that as part of quality assurance the target Hct should be measured after

TE

D

preparation using a hematology analyzer (22).

3.4 Potentials method development challenges of DBS

CE P

There are a number of potential method development challenges related to clinical use of DBS(23). The DBS consortium of EBF identified the major areas of focus to further document the validity of DBS for regulated bioanalysis (14).

AC

3.4.1 Application of IS

Usual application of internal standards (IS) in liquid samples is based on the addition of small volume of IS solution directly to the samples or in the extraction solvent. However, since DBS is a solid matrix, compensating satisfactorily for sample preparation and matrix effects can be challenging. Thus, a variety of possibilities for applying IS in DBS samples has been considered, including the application of IS on the paper before DBS sampling; the addition to blood sample before application to paper; the incorporation of IS in extraction solvent and the direct application of IS to the DBS before extraction (2,11). Abu-Rabie et al. (2011) compared three different methods of IS application: addition of IS to control blood prior to DBS sample preparation, incorporation of IS into extraction solvent, and the use of a novel spray technology to apply the IS solution to DBS samples. The study indicated no significant difference in accuracy and precision

ACCEPTED MANUSCRIPT

9

data using these three techniques (24). Despite of the widespread use of IS into extraction solvent in DBS methods (4,6–8,17,20,21,25), it has been recognized that this procedure does not cover all aspects of analytical compensation intended by an IS,

T

particularly for variations in extraction of the analyte from the DBS punch.

IP

3.4.2 Sample stability

One of the potential benefits inherent to the DBS technique is the possibility to

SC R

collect, ship and store dry samples under ambient conditions. This increase in compound stability can largely be attributed to removal of esterase activity upon drying or even enhanced photostability(26). However, given the wide temperature fluctuations

NU

that can be faced during sample transport and the potential long-term storage of samples under ambient conditions, it is important to assess the stability of every compound to be

MA

analyzed in DBS samples. Currently, data on the stability of drugs and metabolites on DBS samples under different temperatures and storage conditions are limited. Unless data on the interest compound is available, full validation experiments must be

D

performed, including short-term stability in room temperature, long-term stability in a

TE

wide range of temperature (including extreme temperatures) and post-extraction stability.

CE P

3.4.3 Sample dilution procedure The principles of dilution used for fresh blood also applies for DBS, and samples that exceed the upper limit of quantification (ULOQ) must be measured after prior

AC

validated dilution (11). However, due to its solid characteristic, the DBS sample cannot be directly diluted as liquid matrices. One acceptable approach is to dilute the DBS after the extraction of the punched disk with an extract of a blank DBS sample (13). The dilution with blank DBS extract ensures that the matrix composition and the ratio of punch sizes and volumes of extraction solutions remains the same, independently of samples being diluted or not. In addition, the dilution can be processed after the concentration is measured above the linear range. In the case of remnant volume of tested sample, the extract is added to the diluent. 3.4.4 Effect of the hematocrit According to the DBS consortium of EBF, changes in the Hct remain the single most important parameter defining compound behavior and DBS assay performance (14). Hct affects spot formation, homogeneity and size, drying time, recovery of the

ACCEPTED MANUSCRIPT

10

analyte, as well as robustness and reproducibility of the assays (13,23). In addition, these effects are compound-dependent, demanding its evaluation during method validation.

T

Hct and DBS size: DBS size on cellulose-based paper decrease proportionally

IP

with the increase of Hct (viscosity) in the blood (27). Dennif & Spooner (2010) investigated the effect of hematocrit on assay bias and found a decrease of blood spots

SC R

areas of ~15% on Whatman 903® and FTA® papers when increasing the Hct from 0.3 to 0.7 (28). Consequently, a calibration curve performed in blood with intermediate Hct may result in an underestimation of concentration in low Hct and overestimation in high

NU

Hct samples. As described previously, sample volume on the DBS can be corrected for the particular Hct (4). Other validation parameters related to the spot size are drying

MA

time and temperature, related to and relative humidity (RH) (13), which may need to be examined as part of validation.

DBS homogeneity: Another important DBS sampling challenge is the

D

inhomogeneity of the sample (coffee-stain effect; volcano effect), which can be

TE

influenced by the Hct, filter paper quality, humidity and drying conditions. Depending on the extent of the inhomogeneity, punches on different spot locations could result in

CE P

significantly different concentrations. O’Mara et al. (2011) found a notable bias (>15%) on analyte concentration while investigating the impact of punch location on the quantification of acetaminophen and four proprietary compounds in 20 µL DBS from

AC

different matrices (Ahlstrom 226®, Whatman FTA®, FTA Elute®, and FTA DPMK-C®) (29). Concentrations measured from FTA Elute® cards were higher in the center of the spot, and the opposite was found for the other three r cards. The influence of the relative humidity (RH) during drying of DBS has been evaluated by Lenk et al (2015). The study showed that the coffee-stain effect, which represents an over-proportional collection of solid particles at the edge of the spot after drying, caused inhomogeneities in DBS and that the RH and the card position during drying can affect the analyte distribution (30). In order to minimize the inhomogeneity of DBS samples, EBF recommends taking a punch big enough to be a representative sample and/or punches always from the same location of the spots. However, for cases when is necessary to analyze multiple punches (e.g. to increase sensitivity, reanalysis or additional metabolite quantification), it is recommended to use a replicate spot over punching from the same spot, if this

ACCEPTED MANUSCRIPT

11

replicate spot is available (13). In addition, testing for accuracy of different sites of the DBS punch must be performed and documented. Hct and recovery: Hct may have an impact on the extraction efficiency and,

T

consequently, on the recovery of the analyte. It has been indicated that concentrations

IP

may be underestimated in DBS with high Hct when compared to normal or low Hct, as

SC R

a result of a lower recovery (23). Previous studies demonstrated a significant effect of Hct and analyte concentration on extraction recovery(31,32). The DBS consortium of EBF evaluated the effect of Hct on the recoveries of 12 analytes. DBS were prepared at three Hct levels (0.20, 0.45 and 0.75) on four different DBS cards. For some analytes,

NU

recovery varied along with Hct and card type (e.g. compound 1 on Ahlstrom 226® card had a recovery at Hct 0.70 of 49% and of 105% at Hct 0.20 - while on Bond Elut® card

MA

the recovery was 106% for both Hct values). For other analytes, on the other hand, recovery was constant whatever the conditions (31). Similarly, Koster et al. (2013) found constant recoveries for tacrolimus (99%; CV 2.3%) and cyclosporin A (87%; CV

D

4.3%) at varying Hct, while sirolimus and everolimus presented a significance influence

TE

on the extraction recovery (ER) by the concentration as well as the Hct value. The lowest extraction recoveries were observed at the highest concentrations in combination

CE P

with the lowest Hct values (sirolimus: ER of 93% at 3 mg L-1 and a Hct 0.45 and ER of 69% at 50 mg L-1 and Hct 0.25; everolimus ER of 87% at 3 mg L-1 and Hct of 0.35 and ER of 49% at 50 mg L-1 L and Hct of 0.25)(32). Although the impact of diverging Hct

AC

values on recovery may be alleviated by optimizing the extraction (solvent and paper choice), this validation parameter still need to be investigated using blood samples covering a broad range of Hct values. Avoiding/minimizing overall Hct effect: Different strategies have been proposed to manage the Hct issue in DBS analysis. The whole-spot analysis of a volumetric DBS can overcome the Hct effect (except for the physiological plasma/cell drug distribution). However, the obtainance of accurate volumetric blood spots is less attractive for patient self-sampling, once requires some particular skills (2). Although eliminating Hct effects is unlike in DBS testing, the assay needs to have acceptable biases over the Hct range of the studied population. Hct varies with age, gender, health status, and slightly with ethnicity. Reference ranges are typically 0.40-0.50 for adult men and 0.35-0.45 for adult woman. De Kesel et al. (2013) found 95% of routine Hct measurements between 0.23– 0.48 in a hospital population (23). As stated previously, it is mandatory to prepare

ACCEPTED MANUSCRIPT

12

calibration standards and quality control samples at an Hct level that corresponds to the patients samples, with proper corrections to sample volume and analyte partition. 3.5 Methods of analysis

T

The small amount of blood (~5-50 µL) and the complexity of DBS samples

IP

represent a significant analytical challenge. Due to its high sensitivity and selectivity,

SC R

mass spectrometry (MS) methods coupled to chromatographic separation has been the predominant analytical tool for DBS analysis. The use of MS-based analyzers, particularly GC-MS and LC-MS/MS, enables detection of many drugs and metabolites present in blood at very low levels (7,17,33,34,32). To date, triple quadrupole mass

NU

spectrometers remain the gold standard for quantification and were used in most DBS studies (35).

MA

From the analytical point of view, DBS sample preparation is, in most of the cases, a simple procedure. Typically a pre-defined diameter disc is punched from DBS, extracted with organic solvent (or a mixture of solvents), followed by direct injection of

D

diluted extract into the analytical system or by injection of a concentrated extract, after

TE

organic phase evaporation (6–8,34,36–38). In some cases, a derivatization step may be necessary, especially when using CG-MS (39). In the last few years, the need for

CE P

automating the sample preparation procedure to make it compatible with highthroughput requirements has aroused the interest on the development of several instrumental solutions (e.g. on-line solvent extraction, direct flow-extraction of the spot)

AC

(11,40). Many specialized laboratories have been embracing these technologies, making it easier to introduce DBS samples in high throughput TDM analysis. 3.6 Validation parameters applied for DBS-based LC-MS/MS and GC-MS methods The extensive use of DBS in regulated bioanalysis and its introduction in the TDM field raised the concern with the particular validation requirements of DBS methods. Although overall acceptance criteria for method validation are equivalent to other biological matrices, additional validation evaluation are necessary (13,14), such as thermal stability, effects of Hct on spot volume and extraction recovery, punch location and plasma/whole blood drug distribution. The development of strategies for estimating equivalent concentrations with the usual analytical matrices is a critical element for the application of DBS in therapeutic drug monitoring. Thus, cross-validation of DBS tests with plasma assays (or venous

ACCEPTED MANUSCRIPT

13

blood for some substances) is mandatory before introduction of a DBS assay into a clinical laboratory routine (6). A summary of proposed experiments for full validation of new DBS analytical

T

methods intended for TDM use is presented in table 1. Recommended experiments,

SC R

previously published papers and consortium statements,

IP

evaluation of data and criteria of acceptance are proposed by the authors, based on

Table 1

Reported applications of DBS sampling in therapeutic drug monitoring

NU

4.

MA

Numerous applications of DBS sampling associated to mass spectrometric analytical methods in the context of TDM were reported in literature. Table 2 summarizes an overview these methods, coupled to both gas and liquid

D

chromatographic separation systems, reported from 2005 to 2015. The search for the

TE

articles was performed in Pubmed using the following keywords: DBS, mass

CE P

spectroscopy, and therapeutic drug monitoring.

Table 2

AC

DBS sampling was reported for a large diversity of therapeutic classes of drugs, such as: ACE inhibitors (51); analgesic (48,76); antibiotics (4,21,25,60,63,74) ; antiepileptic (6,37,53,67,68,75,80); antidepressants (17,39,81); anthelmintic (70); antimalarial (65); antifungal (61,69); antiretroviral (18,34,49,59,64,73,82); diuretic (52); histamine

H2-receptor

antagonist

(72);

immunosuppressant

(32,54–57,41,77);

chemotherapeutic (7,8,36,62); statin (51); β-blocker (50,51,71); µ-opioid agonist (33). Several different matrices were employed, with a significant predominance of the cellulosic matrix Whatman 903® paper, used in 60% of the studies presented in table 1. This preference could be attributed to the wide availability of this matrix, its previous clearance for regulatory agencies, in the context of newborn screening, as well as its highly reproducible behavior (1). Cross-validation tests among different DBS matrices papers were performed by some authors. Lawson et al. (50) tested two different cellulose-based papers (Whatman 903® and Ahlstrom 226®) and a non-cellulose based

ACCEPTED MANUSCRIPT

14

matrix (Agilent DMS®). Despite the small spot diameter on the DMS matrix, similar quantitavie results were obtained in all three matrices, but with a better precision was with Whatman 903® (50). Vu et al. (4) reported same observations among volume spot.

T

Another two researchers reported no significant differences between the different

IP

evaluated matrices (Ahlstrom 226®, Whatman 31 ET CHR® and Whatman DMPK-C®) (32,69).

SC R

There is a predominance of LC-MS/MS methods for the analysis of DBS samples (80% of reported applications), mainly due to the possibility to inject extracts containing significant amounts of water and its intrinsic high sensibility, considering the

NU

small sample amount available in a DBS.

The “classical” analytical validation, according to generally accepted guidelines

MA

for bioanalytical methods, was adequate in all publications (83). On the other hand, important aspects specific to DBS analysis are lacking in many reports. The evaluation of Hct influence in the accuracy and precision of the analytical measurements was

D

performed in 56% of the studies. Additionally, studies aimed to estimate plasma

TE

concentrations from DBS measurements had made use of Hct as a correction factor (21%), but just a few of them had taken into consideration the partition of the analyte

CE P

between blood cells and plasma, what had a significant potential impact on the estimated plasma concentration. Alternatively, five studies evaluated the effect of drug partitioning between whole blood and plasma and applied a correction factor to DBS concentrations to estimate the equivalent values in the normally utilized matrices (6–

AC

8,37). This approach demonstrated to be adequate for estimation of the drug concentration by improving the correlation between DBS and plasma drug concentration. Another critical point in DBS analytical method validation is long-term stability, particularly at room and higher temperatures. Most of the analytes (72%) in reported applications were stable these conditions, which favors to the use of DBS either at home and remote site conditions. However, some authors attempted only for short-term stability (48)(18,33,36,65) and/or in lower temperatures (18,33,34,52,64,72). Clinical application, with analysis of paired samples of venous and capillary blood ranging from 1 to 167 patients, was performed in only 38% of the studies presented in table 2. This comparison is mandatory before clinical application, once variation from the in vitro behavior had been reported (6). Clinical validation studies used either Passing-Bablok, Demming or simple linear regressions, as well as Bland and Altmann plots and parametric correlation analysis to evaluate the comparability of

ACCEPTED MANUSCRIPT

15

estimated plasma concentrations from DBS testing and actual plasma levels. The majority of the studies found that both data rendered clinical equivalent information.

Conclusion

IP

T

5.

The interest on DBS sampling for TDM purposes is increasing continuously.

SC R

Despite several potential advantages, such as self-sampling and facilitated logistics, several particular aspects must be carefully investigated before clinical implementation of DBS assays. The major variable to be taken into consideration for translating DBS

NU

concentrations to reference plasma levels is the blood sample hematocrit, which also is reflected in the validation experiments of new DBS assays. Mass spectrometry

MA

measurement systems, preferentially coupled to liquid chromatography, have been the mainstay of DBS assays for TDM. These methods present the required sensitivity for measurement of drug concentrations of small DBS samples, as well as the required

D

specificity for clinical applications. Currently, just a minority of the published drug

TE

assays using DBS includes comprehensive clinical validation, what can be considered as the major pitfall to its implementation in the routine of TDM. Clinical validation is

References

Edelbroek, P.M., van der Heijden, J. and Stolk L. Dried blood spot methods in

AC

1.

CE P

mandatory before DBS assays could be implemented in routine TDM testing.

therapeutic drug monitoring: methods, assays, and pitfalls. Ther Drug Monit. 2009;31(3):327–36. 2.

Wilhelm AJ, den Burger JCG, Swart EL. Therapeutic Drug Monitoring by Dried Blood Spot: Progress to Date and Future Directions. Clin Pharmacokinet. 2014;53(11):961–73.

3.

Keevil BG. The analysis of dried blood spot samples using liquid chromatography tandem mass spectrometry. Clin Biochem. 2011;44(1):110–8.

4.

Vu DH, Koster RA, Alffenaar JWC, Brouwers JRBJ, Uges DRA. Determination of moxifloxacin in dried blood spots using LC-MS/MS and the impact of the hematocrit and blood volume. J Chromatogr B Anal Technol Biomed Life Sci. 2011;879(15-16):1063–70.

ACCEPTED MANUSCRIPT 5.

16

Rowland M, Emmons GT. Use of dried blood spots in drug development: pharmacokinetic considerations. AAPS J. 2010;12(3):290–3.

6.

Rhoden L, Antunes MV, Hidalgo P, Silva CA, Linden R. Simple procedure for

Antunes MV, Raymundo S, de Oliveira V, Staudt DE, Gössling G, Peteffi GP, et

SC R

7.

IP

spectrometry. J Pharm Biomed Anal. 2014;96:207–12.

T

determination of valproic acid in dried blood spots by gas chromatography-mass

al. Ultra-high performance liquid chromatography tandem mass spectrometric method for the determination of tamoxifen, N-desmethyltamoxifen, 4-

NU

hydroxytamoxifen and endoxifen in dried blood spots. Talanta. 2015;132:775– 84.

Antunes MV, Raymundo S, Sandrine, Wagner C, Mattevi VS, Vieira N, et al.

MA

8.

DBS sampling in imatinib therapeutic drug monitoring : from method

Ashley EA, Stepniewska K, Lindegardh N, Annerberg A, Tarning J, McGready

TE

9.

D

development to clinical application. Bioanalysis. 2015;7(16):2105–17.

R, et al. Comparison of plasma, venous and capillary blood levels of piperaquine

CE P

in patients with uncomplicated falciparum malaria. Eur J Clin Pharmacol. 2010;66(7):705–12. 10.

Capiau S, Stove V V., Lambert WE, Stove CP. Prediction of the hematocrit of

AC

dried blood spots via potassium measurement on a routine clinical chemistry analyzer. Anal Chem. 2013;85(1):404–10. 11.

Wagner M, Tonoli D, Varesio E, Hopfgartner G. The use of mass spectrometry to analyze dried blood spots. Mass Spectrom Rev. 2014;9999:1–78.

12.

Peck HR, Timko DM, Landmark JD, Stickle DF. A survey of apparent blood volumes and sample geometries among filter paper bloodspot samples submitted for lead screening. Clin Chim Acta. 2009;400(1-2):103–6.

13.

Timmerman P, White S, Globig S, Lüdtke S, Brunet L, Smeraglia J. EBF recommendation on the validation of bioanalytical methods for dried blood spots. Bioanalysis. 2011;3(14):1567–75.

14.

Timmerman P, White S, Cobb Z, de Vries R, Thomas E, van Baar B. Update of

ACCEPTED MANUSCRIPT

17

the EBF recommendation for the use of DBS in regulated bioanalysis integrating the conclusions from the EBF DBS-microsampling consortium. Bioanalysis. 2013;5(17):2129–36. Ostler MW, Porter JH, Buxton OM. Dried blood spot collection of health

T

15.

IP

biomarkers to maximize participation in population studies. J Vis Exp.

16.

SC R

2014;(83):e50973.

Mcdade T. Development and validation of assay protocols for use with dried blood spot samples. Am J Hum Biol. 2013;9:1–9.

Berm EJJ, Paardekooper J, Brummel-Mulder E, Hak E, Wilffert B, Maring JG. A

NU

17.

simple dried blood spot method for therapeutic drug monitoring of the tricyclic

MA

antidepressants amitriptyline, nortriptyline, imipramine, clomipramine, and their active metabolites using LC-MS/MS. Talanta. 2015;134:165–72. Amara AB, Else LJ, Tjia J, Olagunju A, Puls RL, Khoo S, et al. A validated

D

18.

TE

method for quantification of efavirenz in dried blood spots using highperformance liquid chromatography-mass spectrometry. Ther Drug Monit.

19.

CE P

2015;37(2):1.

Enderle Y, Meid AD, Friedrich J, Grünig E, Wilkens H, Haefeli WE, et al. Dried Blood Spot Technique for the Monitoring of Ambrisentan, Bosentan, Sildenafil,

AC

and Tadalafil in Patients with Pulmonary Arterial Hypertension. Anal Chem. 2015;87(24):12112–20. 20.

Verweij-Van Wissen CPWGM, De Graaff-Teulen MJA, De Kanter CTMM, Aarnoutse RE, Burger DM. Determination of the HCV protease inhibitor telaprevir in plasma and dried blood spot by liquid chromatography-tandem mass spectrometry. Ther Drug Monit. 2015;37(5):626–33.

21.

Vu DH, Koster RA, Bolhuis MS, Greijdanus B, Altena R V., Nguyen DH, et al. Simultaneous determination of rifampicin, clarithromycin and their metabolites in dried blood spots using LC-MS/MS. Talanta. 2014;121:9–17.

22.

Koster RA, Alffenaar JW, Botma R, Greijdanus B, Touw DJ, Uges DR, et al. What is the right blood hematocrit preparation procedure for standards and quality

control

samples

for

dried

blood

spot

analysis?

Bioanalysis.

ACCEPTED MANUSCRIPT

18

2015;7(3):345–51. 23.

De Kesel PM, Sadones N, Capiau S, Lambert WE, Stove CP. Hemato-critical issues in quantitative analysis of dried blood spots: challenges and solutions.

IP

24.

T

Bioanalysis. 2013;5(16):2023–41.

Abu-Rabie P, Denniff P, Spooner N, Brynjolffssen J, Galluzzo P, Sanders G.

SC R

Method of applying internal standard to dried matrix spot samples for use in quantitative bioanalysis. Anal Chem. 2011 Nov 15;83(22):8779–86. 25.

Vu DH, Bolhuis MS, Koster RA, Greijdanus B, De Lange WCM, Van Altena R,

NU

et al. Dried blood spot analysis for therapeutic drug monitoring of linezolid in patients with multidrug-resistant tuberculosis. Antimicrob Agents Chemother.

26.

MA

2012;56(11):5758–63.

Bowen C, Volpatti J, Cades J. Evaluation of glucuronide metabolite stability in

Chao TC, Trybala A, Starov V, Das DB. Influence of haematocrit level on the

TE

27.

D

dried blood spots. Bioanalysis. 2012;4(23):2823–32.

kinetics of blood spreading on thin porous medium during dried blood spot

28.

CE P

sampling. Colloids Surfaces A Physicochem Eng Asp. 2014;451(1):38–47. Denniff P, Spooner N. The effect of hematocrit on assay bias when using DBS

29.

AC

samples for the quantitative bioanalysis of drugs. Bioanalysis. 2010;2:1385–95. M O, Hudson-Curtis B, Olson B, Yueh Y, Dunn J, Spooner N. The effect of hematocrit and punch location on assay bias during quantitative bioanalysis of dried blood spot samples. Bioanalysis. 2011;3(20):2335–47. 30.

Lenk G, Hansson J, Beck O, Roxhed N. The effect of drying on the homogeneity of DBS. Bioanalysis. 2015;7:1977–85.

31.

de Vries R, Barfield M, de Merbel VN, Schmid B, Siethoff C, Ortiz J, et al. The effect of hematocrit on bioanalysis of DBS: results from the EBF DBS-sampling consortium. Bioanalysis. 2013;5(17):2147–60.

32.

Koster RA, Alffenaar JWC, Greijdanus B, Uges DRA. Fast LC-MS/MS analysis of tacrolimus, sirolimus, everolimus and cyclosporin A in dried blood spots and the influence of the hematocrit and immunosuppressant concentration on

ACCEPTED MANUSCRIPT

19

recovery. Talanta. 2013;115:47–54. 33.

Clavijo CF, Thomas JJ, Cromie M, Schniedewind B, Hoffman KL, Christians U, et al. A low blood volume LC-MS/MS assay for the quantification of fentanyl

T

and its major metabolites norfentanyl and despropionyl fentanyl in children. J

Castillo-Mancilla JR, Zheng J-H, Rower JE, Meditz A, Gardner EM, Predhomme

SC R

34.

IP

Sep Sci. 2011;34(24):3568–77.

J, et al. Tenofovir, emtricitabine, and tenofovir diphosphate in dried blood spots for determining recent and cumulative drug exposure. AIDS Res Hum

35.

NU

Retroviruses 2013;29(2):384–90.

Déglon J, Leuthold LA, Thomas A. Potential missing steps for a wide use of

36.

MA

dried matrix spots in biomedical analysis. 2015;7:2375–85. Ansari M, Uppugunduri CRS, Déglon J, Théorêt Y, Versace F, Gumy-Pause F, et

D

al. A simplified method for busulfan monitoring using dried blood spot in

TE

combination with liquid chromatography/tandem mass spectrometry. Rapid Commun Mass Spectrom. 2012;26(12):1437–46. la Marca G, Malvagia S, Filippi L, Fiorini P, Innocenti M, Luceri F, et al. Rapid

CE P

37.

assay of topiramate in dried blood spots by a new liquid chromatography-tandem mass spectrometric method. J Pharm Biomed Anal. 2008;48(5):1392–6. Nirogi R, Ajjala DR, Kandikere V, Aleti R, Srikakolapu S, Vurimindi H. Dried

AC

38.

blood spot analysis of an iron chelator--deferasirox and its potential application to therapeutic drug monitoring. J Chromatogr B Analyt Technol Biomed Life Sci 2012;907:65–73. 39.

Déglon J, Lauer E, Thomas A, Mangin P, Staub C. Use of the dried blood spot sampling process coupled with fast gas chromatography and negative-ion chemical ionization tandem mass spectrometry: Application to fluoxetine, norfluoxetine, reboxetine, and paroxetine analysis. Anal Bioanal Chem. 2010;396(7):2523–32.

40.

Demirev P a. Dried blood spots: Analysis and applications. Anal Chem. 2013;85(2):779–89.

ACCEPTED MANUSCRIPT 41.

20

Sadilkova K, Busby B, Dickerson JA, Rutledge JC, Jack RM. Clinical validation and implementation of a multiplexed immunosuppressant assay in dried blood spots by LC-MS/MS. Clin Chim Acta. 2013;421:152–6. Koster RA, Botma R, Greijdanus B, Uges DRA, Kosterink JGW, Alffenaar J-

T

42.

IP

WC, et al. The influence of the dried blood spot drying time on the recoveries of

43.

SC R

six im-munosuppressants. J Appl Bioanal. 2015;1(4): 116-122. Peters AFT, Hartung M, Saar H, Schmitt MHG. Requirements for the validation of analytical methods. Toxichem Krimtech. 2009;76(3): 185-208. Spooner N, Lad R, Barfield M. Dried blood spots as a sample collection

NU

44.

technique for the determination of pharmacokinetics in clinical studies:

Chem. 2009;81(4):1557–63.

FDA F and DA. Guidance for industry. Bioanalytical method validation. 2001.

D

45.

MA

Considerations for the validation of a quantitative bioanalytical method. Anal

from:

TE

Available

http://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/

46.

CE P

guidances/ucm368107.pdf

Peters FT, Drummer OH, Musshoff F. Validation of new methods. Forensic Sci Int. 2007;165(2-3):216–24. European Medicines Agency. Guideline in Bioanalytical Method Validation.

AC

47.

Available

at

http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2 011/08/WC500109686.pdf. Access 03/28/2016. 48.

Taylor RR, Hoffman KL, Schniedewind B, Clavijo CF, Galinkin JL, Christians U. Comparison of the quantification of acetaminophen in plasma, cerebrospinal fluid and dried blood spots using highperformance liquid chromatographytandem mass spectrometry. J Pharm Biomed Anal. 2013;83:1–9.

49.

ter Heine R, Rosing H, van Gorp ECM, Mulder JW, van der Steeg WA, Beijnen JH, et al. Quantification of protease inhibitors and non-nucleoside reverse transcriptase inhibitors in dried blood spots by liquid chromatography-triple quadrupole mass spectrometry. J Chromatogr B Anal Technol Biomed Life Sci.

ACCEPTED MANUSCRIPT

21

2008;867(2):205–12. 50.

Lawson G, Cocks E, Tanna S. Quantitative determination of atenolol in dried blood spot samples by LC-HRMS: A potential method for assessing medication

51.

IP

T

adherence. J Chromatogr B Anal Technol Biomed Life Sci. 2012;897:72–9. Lawson G, Cocks E, Tanna S. Bisoprolol, ramipril and simvastatin determination

SC R

in dried blood spot samples using LC-HRMS for assessing medication adherence. J Pharm Biomed Anal. 2013;81-82:99–107. 52.

Suyagh MF, Kole PL, Millership J, Collier P, Halliday H, McElnay JC.

NU

Development and validation of a dried blood spot-LC-APCI-MS assay for estimation of canrenone in paediatric samples. J Chromatogr B Anal Technol

53.

MA

Biomed Life Sci. 2010;878(9-10):769–76.

Shokry E, Villanelli F, Malvagia S, Rosati A, Forni G, Funghini S, et al.

D

Therapeutic drug monitoring of carbamazepine and its metabolite in children

TE

from dried blood spots using liquid chromatography and tandem mass spectrometry. J Pharm Biomed Anal. 2015;109:164–70. Wilhelm AJ, den Burger JCG, Vos RM, Chahbouni A, Sinjewel A. Analysis of

CE P

54.

cyclosporin A in dried blood spots using liquid chromatography tandem mass spectrometry. J Chromatogr B Anal Technol Biomed Life Sci. 2009;877(14-

55.

AC

15):1595–8.

Wilhelm AJ, Klijn A, den Burger JCG, Visser OJ, Veldkamp AI, Janssen JJWM, et al. Clinical Validation of Dried Blood Spot Sampling in Therapeutic Drug Monitoring of Ciclosporin A in Allogeneic Stem Cell Transplant Recipients. Ther Drug Monit. 2013;35(1):92–5.

56.

Hinchliffe E, Adaway JE, Keevil BG. Simultaneous measurement of cyclosporin A and tacrolimus from dried blood spots by ultra high performance liquid chromatography tandem mass spectrometry. J Chromatogr B Anal Technol Biomed Life Sci. 2012;883-884:102–7.

57.

Hinchliffe E, Adaway J, Fildes J, Rowan A, Keevil BG. Therapeutic drug monitoring of ciclosporin A and tacrolimus in heart lung transplant patients using dried blood spots. Ann Clin Biochem. 2014;51(Pt 1):106–9.

ACCEPTED MANUSCRIPT 58.

22

Patel P, Tanna S, Mulla H, Kairamkonda V, Pandya H, Lawson G. Dexamethasone quantification in dried blood spot samples using LC-MS: The potential for application to neonatal pharmacokinetic studies. J Chromatogr B

Zheng J-H, Guida L, Rower C, Castillo-Mancilla JR, Meditz A, Klein B, et al.

IP

59.

T

Anal Technol Biomed Life Sci. 2010;878(31):3277–82.

SC R

Quantitation of Tenofovir and Emtricitabine in Dried Blood Spots (DBS) with LC-MS/MS. J Pharm Biomed Anal. 2014;88(1):1–23. 60.

la Marca G, Giocaliere E, Villanelli F, Malvagia S, Funghini S, Ombrone D, et

NU

al. Development of an UPLC-MS/MS method for the determination of antibiotic ertapenem on dried blood spots. J Pharm Biomed Anal. 2012;61:108–13. Van Der Elst KCM, Span LFR, Van Hateren K, Vermeulen KM, Van Der Werf

MA

61.

TS, Greijdanus B, et al. Dried blood spot analysis suitable for therapeutic drug monitoring of voriconazole, fluconazole, and posaconazole. Antimicrob Agents

D

Chemother. 2013;57(10):4999–5004. Kralj E, Trontelj J, Pajič T, Kristl A. Simultaneous measurement of imatinib,

TE

62.

CE P

nilotinib and dasatinib in dried blood spot by ultra high performance liquid chromatography tandem mass spectrometry. J Chromatogr B Anal Technol Biomed Life Sci. 2012;903:150–6. la Marca G, Villanelli F, Malvagia S, Ombrone D, Funghini S, De Gaudio M, et

AC

63.

al. Rapid and sensitive LC-MS/MS method for the analysis of antibiotic linezolid on dried blood spot. J Pharm Biomed Anal 2012;67-68:86–91. 64.

Meesters RJW, Van Kampen JJA, Reedijk ML, Scheuer RD, Dekker LJM, Burger DM, et al. Ultrafast and high-throughput mass spectrometric assay for therapeutic drug monitoring of antiretroviral drugs in pediatric HIV-1 infection applying dried blood spots. Anal Bioanal Chem. 2010;398(1):319–28.

65.

Geditz MCK, Lindner W, Lämmerhofer M, Heinkele G, Kerb R, Ramharter M, et al. Simultaneous quantification of mefloquine (+)- and (-)-enantiomers and the carboxy metabolite in dried blood spots by liquid chromatography/tandem mass spectrometry. J Chromatogr B Anal Technol Biomed Life Sci. 2014;968:32–9.

66.

Hooff GP, Meesters RJW, Van Kampen JJA, Van Huizen NA, Koch B, Al

ACCEPTED MANUSCRIPT

23

Hadithy AFY, et al. Dried blood spot UHPLC-MS/MS analysis of oseltamivir and oseltamivircarboxylate- a validated assay for the clinic. Anal Bioanal Chem. 2011;400(10):3473–9. la Marca G, Malvagia S, Filippi L, Luceri F, Moneti G, Guerrini R. A new rapid

T

67.

IP

micromethod for the assay of phenobarbital from dried blood spots by LC-

68.

SC R

tandem mass spectrometry. Epilepsia. 2009;50(12):2658–62.

Villanelli F, Giocaliere E, Malvagia S, Rosati A, Forni G, Funghini S, et al. Dried blood

spot

assay for

the

quantification

of

phenytoin

using

Liquid

69.

NU

Chromatography-Mass Spectrometry. Clin Chim Acta. 2015;440:31–5. Reddy TM, Tama CI, Hayes RN. A dried blood spots technique based LC-

MA

MS/MS method for the analysis of posaconazole in human whole blood samples. J Chromatogr B Anal Technol Biomed Life Sci. 2011;879(30):3626–38. Meister I, Leonidova A, Kovac J, Duthaler U, Keiser J, Huwyler J. Development

D

70.

TE

and validation of an enantioselective LC-MS/MS method for the analysis of the anthelmintic drug praziquantel and its main metabolite in human plasma, blood

71.

CE P

and dried blood spots. J Pharm Biomed Anal. 2016;118:81–8. Della Bona ML, Malvagia S, Villanelli F, Giocaliere E, Ombrone D, Funghini S, et al. A rapid liquid chromatography tandem mass spectrometry-based method

AC

for measuring propranolol on dried blood spots. J Pharm Biomed Anal. 2013;7879:34–8. 72.

Yakkundi S, Millership J, Collier P, Shields MD, McElnay J. Development and validation of a dried blood spot LC-MS/MS assay to quantify ranitidine in paediatric samples. J Pharm Biomed Anal. 2011;56(5):1057–63.

73.

Jimmerson LC, Zheng J-H, Bushman LR, MacBrayne CE, Anderson PL, Kiser JJ. Development and validation of a dried blood spot assay for the quantification of ribavirin using liquid chromatography coupled to mass spectrometry. JounaL Chromatogr B Anal Technol Biomed Life Sci. 2014;944:18–24.

74.

Parsons TL, Marzinke MA, Hoang T, Bliven-Sizemore E, Weiner M, Mac Kenzie WR, et al. Quantification of rifapentine, a potent antituberculosis drug, from dried blood spot samples using liquid chromatographic-tandem mass

ACCEPTED MANUSCRIPT

24

spectrometric analysis. Antimicrob Agents Chemother. 2014;58(11):6747–57. 75.

la Marca G, Malvagia S, Filippi L, Innocenti M, Rosati A, Falchi M, et al. Rapid assay of rufinamide in dried blood spots by a new liquid chromatography-tandem

76.

IP

T

mass spectrometric method. J Pharm Biomed Anal. 2011;54(1):192–7. Siebenhaar M, Küllmer K, de Barros Fernandes NM, Hüllen V, Hopf C.

SC R

Personalized monitoring of therapeutic salicylic acid in dried blood spots using a three-layer setup and desorption electrospray ionization mass spectrometry. Anal Bioanal Chem. 2015;407(23):7229–38.

Hoogtanders K, van der Heijden J, Christiaans M, Edelbroek P, van Hooff JP,

NU

77.

Stolk LML. Therapeutic drug monitoring of tacrolimus with the dried blood spot

78.

MA

method. J Pharm Biomed Anal. 2007;44(3 SPEC. ISS.):658–64. Koop DR, Bleyle LA, Munar M, Cherala G, Al-Uzri A. Analysis of tacrolimus

D

and creatinine from a single dried blood spot using liquid chromatography

2013;926:54–61.

Li Q, Cao D, Huang Y, Xu H, Yu C, Li Z. Development and validation of a

CE P

79.

TE

tandem mass spectrometry. J Chromatogr B Anal Technol Biomed Life Sci.

sensitive LC-MS/MS method for determination of tacrolimus on dried blood spots. Biomed Chromatogr. 2013;27(3):327–34. Popov TV, Maričič LC, Prosen H, Vončina DB. Development and validation of

AC

80.

dried blood spots technique for quantitative determination of topiramate using liquid chromatography-tandem mass spectrometry. Biomed Chromatogr. 2013;27(8):1054–61. 81.

Berm EJJ, Brummel-Mulder E, Paardekooper J, Hak E, Wilffert B, Maring JG. Determination of venlafaxine and O-desmethylvenlafaxine in dried blood spots for TDM purposes, using LC-MS/MS. Anal Bioanal Chem. 2014;406(910):2349–53.

82.

Verweij-Van Wissen CPWGM, De Graaff-Teulen MJA, De Kanter CTMM, Aarnoutse RE, Burger DM. Determination of the HCV Protease Inhibitor Telaprevir in Plasma and Dried Blood Spot by Liquid Chromatography–Tandem Mass Spectrometry. Ther Drug Monit. 2015;37(5):626–33.

ACCEPTED MANUSCRIPT 83.

25

United States Food and Drug Administration. Guidance for Industry: Bioanalytical

method

validation.

Available

at

http://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/

IP

T

guidances/ucm368107.pdf. Access 03/28/2016.

SC R

LIST OF TABLES

NU

Table 1. Proposed experiments for validation of DBS assays for TDM

MA

Table 2. Overview of reported applications of DBS sampling associated to mass

AC

CE P

TE

D

spectrometry for TDM purposes (2005-2015)

ACCEPTED MANUSCRIPT

T

26

CR

IP

Table 1. Proposed experiments for validation of DBS assays for TDM Proposed experimental design

Evaluation and acceptance criteria

Blood-plasma partitioning at different Hct levels

-Prepare whole control blood samples at two levels (QCL and QCH), in at least 3 different Hct levels. -Analysis in triplicate of QCL and QCH, measuring both whole blood and, after centrifugation, plasma concentration.

DBS Homogeneity

- Prepare DBS blood samples at two levels (QCL and QCH), in at least 3 different Hct levels. -Analysis in triplicate of 2 punches sites from the same spot (center and near perimeter) for each QC sample and Hct level. - Run a calibration curve prepared in fixed Hct level, correcting concentrations according to Hct.

Influence of Hct on DBS measurement accuracy

US

Parameter

Related reference (6)

Considerations: Hct levels should be prepared according to study population (e.g. 0.25 to 0.60). The effect of different types of paper on DBS spreading can be evaluated as well. Evaluation: Calculate the center/perimeter concentration ratio (C/P). Acceptance: A C/P ratio of 1 is expected for homogeneous spot, C/P < 1 indicate a preferential distribution of the analyte to the perimeter of the spot, rather the center (volcano effect).

(31)

- Prepare DBS blood samples at two levels (QCL and QCH), in at Considerations: Hct levels should be prepared according to least 3 different Hct levels. study population (e.g. 0.25 to 0.60). -Run a calibration curve prepared in fixed Hct level, do not correct Evaluation: The percent deviation of the mean calculated

(7)

AC

CE P

TE D

MA N

Considerations: Hct levels should be prepared according to study population (e.g. 0.25 to 0.60). Blood hemolysis is avoided by drying the standards in the test tubes before adding the blood. Samples are incubate at 37◦C for 1 h, with occasional gentle agitation Evaluation: Blood-to-plasma concentration ratio low, close to 0.55, indicates almost complete amount in plasma. The determination of the fraction in plasma (fp) can be used to estimate plasma concentrations from whole blood measurements.

ACCEPTED MANUSCRIPT 27

from the nominal value for each Hct level (bias). Acceptance: Bias ± 15%

IP

T

concentrations according to Hct.

- Prepare DBS blood samples at two levels (QCL and QCH), Evaluation: The percent deviation of the mean calculated prepared at fixed Hct level. from the nominal value for each spotted volume (bias). -Spot at least 3 different blood volumes in triplicate on paper Acceptance: Bias ± 15% -Analyze in triplicate same size punch taken from the center of each DBS -Run a calibration curve prepared at fixed Hct and blood volume.

(41)

Recovery (RE)

- Prepare spiked DBS samples at two levels (QCL and QCH), prepared in at least 3 different Hct levels. Extract and analyze in triplicate. - Prepare blank DBS samples, prepared in at least 3 different Hct levels. Extract and analyze in triplicate. -Perform whole punch analysis of fixed volume.

(32)

AC

CE P

TE D

MA N

US

CR

Influence of blood spot size

Matrix effect (ME)

-Considerations: Hct levels should be prepared according to study population (e.g. 0.25 to 0.60).The effect of drying time on recovery can be included this validation assay: QCH samples can be processed and analyzed in a period of 48 h (e.g. 3, 24 and 48h) (42). -Evaluation: Calculate RE for each Hct and QC sample:

-Acceptance: Extent of recovery should be consistent for the different Hct at QCL and QCH .

-Prepare net standard solutions at QCH and QCL concentrations, -Evaluation: analyzed in quintuplicate. - Prepare blank DBS samples, using 5 different blank blood donors. Extract and analyze in quintuplicate. -Acceptance: ±25 % for average ME for five different blank matrix samples. A RSD of 25% is acceptable when

(43)

ACCEPTED MANUSCRIPT 28

T

deuterated internal standards are used, and in other cases, 15% are acceptable (20% near the limit of quantification). - Prepare 6 DBS samples with a concentration 2x the ULOQ. Extract and analyse. -Prepare diluent with a pool of 10 sources of blank DBS extracted as test samples. -Dilute 10-fold each QC sample with the diluent. -Run a calibration curve with a fixed Hct level

Evaluation: Results corrected for the dilution factor. The percent deviation of the mean calculated from the nominal value for each spotted volume (bias). Acceptance: Bias ± 15%

Cross-validation between papers (if it is necessary to change DBS matrix)

-Prepare a 8 point calibration curve, prepared at fixed Hct, for each -Evaluation: 95% confidence intervals (CI) of the intercepts paper. Extract and analyse. and slopes of each calibration curve to determine differences between the papers. -Acceptance: Similar results when the 95% CI between papers are overlapped.

(32)

Long-term stability

-Prepare DBS QCL and QCH prepared at a fixed Hct level. -After drying, store DBS samples at -20, 4 and 25°C for 1 month -Analyse in triplicate each QC sample at day “zero” and every “x” wanted days. -Extreme temperature variations: LQC and HQC storage at 37, 45 and 60°C, analyzed for 5 days, in triplicate. -Run daily calibration curve prepared at fixed Hct.

-Considerations: DBS samples must be placed into zip-lock storage bags along with a desiccant pouch, to avoid humidity. To mimic potential transportation times and potential conditions, extreme temperature variations should be evaluated. -Evaluation: Comparison of all measured concentrations to the mean of back-calculated values for QC samples from the first day of long-term stability testing. -Acceptance: Results within ±15% of nominal concentrations.

(41)

Short-term stability

-Prepare DBS QCL and QCH prepared at a fixed Hct level. -Considerations: Duration of short-term stability is based on - QCL and QCH kept at room temperature from 4 to 24 hours, the expected duration that samples will be maintained at room temperature in the intended study. analysis in triplicate. -Evaluation: Comparison of all measured concentrations to - Run daily calibration curve prepared at fixed Hct. the mean of back-calculated values for QC samples from the first test.

(45)

(44)

AC

CE P

TE D

MA N

US

CR

IP

Dilution of samples above the ULOQ

ACCEPTED MANUSCRIPT 29

within

±15%

of

nominal

T

-Acceptance: Results concentrations.

- Prepare at least 6 replicates of QCL and QCH samples. -Evaluation: Regression analysis plotting absolute peak - Pool extracts obtained at each concentration and repeat injection areas corresponding to compound at each concentration vs. injection time. at certain time intervals (e.g. every 1 h, during 12 h). Acceptance: Negative slope significantly different from 0 indicates instability. A 10% increase or decrease in the measured peak areas is acceptable.

(7,46)

Selectivity

- Prepare at least 6 DBS blank samples (blank without IS). Extract -Evaluation: Checking for interfering signals and analyse. -Acceptance: Absence of interfering signals at the LLOQ -Co-medications that may potentially interfere should be taken into account.

(45,47)

Calibration model

- Prepare DBS calibration samples at 6 to 8 levels, prepared at fixed -Considerations: Concentrations must cover the expected Hct. Extract and analyze in quintuplicate. range, including LLOQ. For endogenous compounds, the intercept of the calibration can be used to correct for the natural presence of the analyte in reference samples. -Evaluation: Analysis of behavior of variances across calibration range (F-test), evaluation of weighting factor in case of heteroscedastic data, evaluation of linear or nonlinear model, and statistical test of model fit. -Acceptance: 15% deviation (20% for LLOQ) of standards from nominal concentration. At least four out of six nonzero standards should meet the above criteria, including the LLOQ and the calibration standard at the highest concentration.

(45,46)

Sensitivity

- Prepare DBS samples at LLOQ, prepared at fixed Hct. Extract and -Consideration: Experiment preferably performed together analyze in triplicate. with accuracy/precision assays. -Evaluation: Accuracy evaluated by calculating the percent deviation of the mean calculated from the nominal value -Run a calibration curve prepared at fixed Hct. (bias). Precision analyzed by CV% calculation using

(45)

AC

CE P

TE D

MA N

US

CR

IP

Processed sample stability

ACCEPTED MANUSCRIPT 30

IP

T

ANOVA approach. -Acceptance: Accuracy: bias ± 20%. Within-run and between-run precision: CV ≤ 20%. - Prepare DBS samples at three levels (QCL, QCM and QCH), -Evaluation: Accuracy evaluated by calculating the percent prepared at fixed Hct. Extract and analyze in triplicate. deviation of the mean calculated from the nominal value (bias). Precision analyzed by CV% calculation using ANOVA approach. -Run a calibration curve prepared at fixed Hct. -Acceptance: Accuracy: bias ± 15%. Within-run and between-run precision: CV ≤ 15%.

(45)

Clinical validation

Analysis of a minimum of 50 clinical samples paired DBS, venous blood and plasma samples

(8)

CE P

TE D

MA N

US

CR

Accuracy/ Precision

-Considerations: Estimation of plasma concentrations from DBS measurements adjusted by the fraction of drug in plasma (fp) and adjusted by individual Hct and/or a fixed Hct as an average expected for the study population. -Evaluation: Bland Altman plots to describe agreement of methods and Passing Bablok or Deming regression for detecting constant or proportional difference.

AC

DBS= dried blood spot; QC= quality control; QCL= low quality control sample at low concentration; QCM= quality control sample at medium concentration; QCH= quality control sample at high concentration; LLOQ= lower limit of quantification; ULOQ: upper limit of quantification; CV= coefficient variation; IS=internal standard.

ACCEPTED MANUSCRIPT 31

3.04-20000 ng/mL

903

LCMS/MS

Ambrisentan

2.5-1000 ng/mL

DMPK-C

LCMS/MS

Bosentan

2.5-4000 ng/mL

Sildenafil

5-2500 ng/mL

Tadalafil Amitriptyline

10-1000 ng/mL 20-500 µg/L

Nortriptyline

20-500 µg/L

Imipramine

20-500 µg/L

Clomipramine Atazanir

20-500 µg/L 0.1-20 µg/Ml

Darunavir

0.05-10 µg/mL

Lopinavir

0.1-20 µg/mL

Ritonavir

0.05-10 µg/mL

Efavirenz

0.1-20 µg/mL

Nevirapine Atenolol

0.1-20 µg/mL 25-1500 ng/L

903 DMS

No

-

Lower Hct at LQC: Blood viscosity

No

Correlation DBS x plasma Well correlated r2=0.980 # SRA, BA Well correlated r2>0.880 # DR, PC

Drug stability on DBS

Clinical Validation

24 h at RT 1 week at 4 and 20 °C 1 month at -80 °C 147 days at RT 14 days at 37 and -20 °C

Yes n= 167

Yes n= 27 n= 55

Capillary blood collection Yes

Ref.

Yes

(19)

Yes

(17)

Yes

(49)

No

(50)

(48)

n= 53

LCMS/MS

n.e.

-

Well correlated r2>0,900 Except to clomipramine (r2=0,730) # SRA n.e.

LC-TOFMS

n.e.

-

n.e.

CE P

903

LCMS/MS

AC

DMPK-C

Concentration correction for Hct -

IP

Acetominophen

Evaluation of Hct Influence n.e.

CR

Analytical Technique

US

Matrix

MA N

Calibration Range

TE D

Drug

T

Table 2. Overview of reported applications of DBS sampling associated to mass spectrometry for TDM purposes (2005-2015)

3 months at RT

2 months at RT

n= 29 Yes n= 6 n= 12 n= 0

7 days at 30 °C

n= 12 Yes (Only for Darunavir and Ritonavir) n= 1

Yes n= 2

ACCEPTED MANUSCRIPT 32

Ahl-226 903

Simvastatin Bussulfan

100-2000 ng/mL

Canrenone

n.e.

-

903

LCMS/MS

n.e.

-

25-1000 ng/mL

903

LC-APCIMS

n.e

Carbamazepine

1-40 mg/L

903

LCMS/MS

No

Carbamazepineepoxide Cyclosporin A

0.25-20 mg/L 903

LCMS/MS

3 months at RT

Yes n= 15

No

(51)

n.d.

6 h at RT 24 h at 4 °C 1 month at -20°C 1 month at -20 °C

No

No

(36)

Yes n= 37

No

(52)

1 month at -20, 4, RT and 37 °C

Yes n= 12

Yes

(53)

17 days at RT 45 days at 4 °C

No

-

(54)

Yes n= 37

Yes

(55)

Yes n= 150

No

(56)

Yes n=91 Yes n= 79

Yes

(57)

No

(41)

IP n.e.

Cplasma= CDBS/(100− Hct%)

Well correlated r2>0.920 # SRA n.e.

MA N

TE D No

-

-

Cyclosporin A

75-1500 µg/L

Tacrolimus

2.5-50 µg/L

30-100 ng/mL

903

903

AC

CE P

25-1440 µg/L

LCMS/MS

LCMS/MS

n.e.

No

CR

Ramipril

Cyclosporin A

n.e.

T

LC-TOFMS

US

0.1-100 ng/mL

Bisoprolol

-

-

Well correlated r2=0.990 # DR BA Well correlated r2>0,900 # PB, BA

Well correlated r2>0.950

14 days at RT

30 days -20, 4 and 25 °C 5 days at 60 °C

ACCEPTED MANUSCRIPT 33

#

1.2-40 ng/mL

SRA

Tacrolimus

1.2-40 ng/mL

Cyclosprin A

20-2000 µg/L

Everolimus

1-50 µg/L

Sirolimus

1-50 µg/L

Tacrolimus

1-50 µg/L

Dexamethasone

15-800 ng/mL

903

LC-MS

Efavirenz

25-5000 ng/mL

903

LC-MS

Emtricitabine

2.5-5000 ng/mL

903

Tenofovir

2.5-1000 ng/mL

Emtricitabine

2.5-5000 ng/mL

Tenofovir

2.5-1000 ng/mL

Ertapenen

0.5-100 mg/L

Lower Hct at HQC Blood viscosity

No

US

LCMS/MS

TE D

MA N

31ETCHR DMPK-C

Well correlated* r2>0.870 # SRA, PB

-

n.e.

Hct ≤ 27.9 at HQC: Blood viscosity

No

LCMS/MS

No

-

903

LCMS/MS

No

-

903

LCMS/MS

n.e.

n.d.

Well correlated r=0.976 # DR, BA Well correlated r2>0.960 # SRA Well correlated r2>0,960 # SRA Well correlated** r2=0.985 # SRA

AC

CE P

No

CR

IP

T

Sirolimus

30 days -20, 4 and 25 °C 5 days at 60 °C 30 days -20, 4 and 25 °C 24 h at 60 °C

n= 68

28 days at 22 and 37 °C 13 days at 22 °C 2 days at 37 °C 7 days at 22 and 37 °C 28 days at 22 and 37 °C

Yes n= 57 n= 55

7 days at RT 28 days at 4 °C 24 h at 40 °C 1 year at -40°C

n= 115

No

(32)

Yes n= 2 Yes n= 46

No

(58)

No

(18)

11 months at -20 and -80 °C 18 months at RT

Yes n= 30

No

(59)

7 months at 4, -20 and -80 °C

Yes n= 3

No

(34)

1 month at -20 °C

No

-

(60)

n= 36 n= 50

ACCEPTED MANUSCRIPT 34

Pozaconazole

0.1-10 µg/mL

Voriconazole

0.1-10 µg/mL

Fluoxetine

1-500 ng/mL

Norfluoxetine

1-500 ng/mL

Paroxitine

20-500 ng/mL

Reboxitine Imatinib

DMPK-C

LCMS/MS

No

-

n.e.

T

0.1-100 ng/mL 0.5-100 µg/mL

-

3 days at RT 144 days at -20 and -80 °C

Yes n= n.d.

No

(33)

12 days at RT, 37, 50 and -80 °C

Yes n= 10 n= 8

Yes

(61)

IP

Despropionyl fentanyl Fluconazole

n.e.

CR

0.1-100 ng/mL

LCMS/MS

US

Norfentanyl

903

CG-NICIMS-MS

20-500 ng/mL 50-4000 ng/mL

903

LCMS/MS

Imatinib

50-5000 µg/L

DMS

LCMS/MS

Nilotinib

50-5000 µg/L

n.e.

AC

Well correlated r2>0.950 # PB, BA

n= 10

-

n.e.

30 days at -20, 4, 20 and 40 °C

Yes n= 1 (Fluoxetine and Norfluoxetine)

Yes

(39)

No

EPC = (DBSconc/ [1Hct%/100]) x fp

36 days at -20, 25 and 40 °C

Yes n= 50

Yes

(8)

No

Cplasma= CDBS/(100− Hct%)

Well correlated r>0.960 # PB, BA Well correlated r2=0.977 # SRA, BA

28 days at 25 °C 3 days at -20 °C 3 days at 40 °C 28 days at 25 °C 3 days at -20 °C 3 days at 40 °C 28 days at 25 °C 3 days at -20 °C 1 month at RT, 4, 37 and -20°C

Yes n= 18

No

(62)

n= 3 Yes n=9

Not clear

(63)

2 months at RT

Yes

Yes

(25)

CE P

CF12

MA N

0.1-100 ng/mL

TE D

Fentanyl

Dasatinib Linezolid

2.5-500 µg/L 1-100 mg/L

903

LCMS/MS

n.e.

Cplasma= CDBS/(100− Hct%)

Linezolid

0.05-40 mg/L

31ETCHR

LC-

No

-

Well correlated r2>0.900 # SRA, BA Well

n= 2

ACCEPTED MANUSCRIPT 35

FTA-PK

LCMALDIMS

n.e.

-

IP

0.25-50 µmol/L

Ritonavir LCMS/MS

n.e.

31ETCHR 903 No3

LCMS/MS

Yes Blood viscosity

Cplasma= CDBS x (Vstd/Vstd + b x (Hct%-35)

Oseltamivir and

5-1500 ng/mL

2992

LCMS/MS

n.e.

-

Oseltamivir carboxylate Phenobarbital

20-1500 ng/mL 1-100 mg/L

903

LCMS/MS

Yes – drug blood distribution

Cplasma= CDBS/(100− Hct%)

Phenytoin

0-100 mg/L

903

LCMS/MS

n.e.

Not clear

Posaconazole

5-5000 ng/mL

Alh-226 DMPK-C

LCMS/MS

No (Hct 2541%)

Praziquantel (enantiometers) Trans-4- hydroxy praziquantel

0.01-2.5 µg/mL

DMPK-C

LCMS/MS

Hct > 60% Blood viscosity, spot homogeneity No

0.1-25 µg/mL

MA N

13-2600 nM 0.05-6 mg/L

TE D

Carboxymefloquine Moxifloxacin

-

US

903

CE P

5-2000 nM

AC

Mefloquine (enantiometers)

CR

Lopinavir

-

correlated # PB, BA Well correlated r2>0.750 # SRA, BA

T

MS/MS

Well correlated # n.d. Well correlated r2=0.996 # SRA, PB n.e.

Well correlated r2>0.995 # SRA, BA Well correlated r2=0.982 # SRA n.e.

Well correlated # n.d.

and 37 °C 1 week at 50 °C 24 h at 4 °C 20 days at -20 °C

n= 8 Yes n= 19

No

(64)

Yes n= n.d.

No

(65)

4 weeks at RT, 50 and -80 °C

Yes n= 6

Yes

(4)

7 days at RT and 4 °C 24 h at 40 and -20 °C 33 days at RT, 4 and -20 °C

Yes n= 3

Yes

(66)

Yes n= 50

No

(67)

1 month at RT, 37, 4 and -20 °C

Yes n= 17

No

(68)

13 days at RT

No

-

(69)

6 months at RT

Yes n= 1

Yes

(70)

24 h at 4 °C 20 days at -20 °C 4 h at RT

ACCEPTED MANUSCRIPT 36

2.5-200 µg/L

903

LCMS/MS

Yes: Blood viscosity

Cplasma= CDBS×100/ (100−Hct%)

Ranitidine

10-500 ng/mL

903

n.e.

-

Ribavirin

0.05-10 µg/mL

903

LCMS/MS LCMS/MS

n.e.

Rifapentine

50-80000 ng/mL

903

LCMS/MS

Rifampicin

0.15-30 mg/L

31ETCHR

LCMS/MS

Yes Hct < 20% Hct > 60% Blood viscosity Hct = 20 % Drug blood distribution Lower and higher Hct – Blood viscosity

Desacetyl rifampicin

0.15-10 mg/L

Clarithromycin

0.05-10 mg/L

14hydroxyclarithromycin

0.05-10 mg/L

Rufinamide

0.48-47.60 mg/L

903

LCMS/MS

n.e

Cplasma= CDBS/(100− Hct%)

Well correlated r2>0.900 # SRA

Salicylic acid

10-200 mg/L

Three layer

DESI-MS

n.e.

-

n.e.

IP CR

US

MA N

TE D

Cplasma= CDBS/(100− Hct%) Cplasma= CDBS x (Vstd/Vstd + b x (Hct%-35)

Well correlated r2=0,984 # SRA Well correlated # PB, BA Well correlated r2>0.950 Except to DAc-RIF (r2=0.685) # SRA, DR

AC

CE P

Well correlated R=0,936 # SRA n.e.

T

Propranolol

1 month at -20, 4 °C and RT

Yes n= 7

No

(71)

24h at 20 °C 6 months at -20 °C 140 days at RT and -20 °C

Yes n= 36 Yes n= 28

No

(72)

No

(73)

11 weeks at RT

Yes n= 26

No

(74)

60 days at RT 10 days at 37 °C 3 days at 50 °C

Yes n= 27

Yes

(21)

60 days at RT 10 days at 37 °C 3 days at 50 °C

n= 27

60 days at RT 30 days at 37 °C 15 days at 50 °C

n= 10

Yes n= 14

No

(75)

Yes n= 1

Yes

(76)

n= 10 60 days at RT 30 days at 37 °C 15 days at 50 °C 4 weeks at RT, 4 and -2- °C

n.e.

ACCEPTED MANUSCRIPT

Tacrolimus

1-50 ng/mL

Tacrolimus

IP

1-30 µg/L

n.e.

-

DMPK-A

LCMS/MS

n.e.

1-80 ng/mL

903

LCMS/MS

No

Tamoxifen

7.5-300 ng/mL

903

LCMS/MS

N-desmethyltamoxifen

15-600 ng/mL

Endoxifen

1-40 ng/mL

4-hydroxytamoxifen

0.5-50 ng/mL

Telaprevir

0.1-10 mg/mL

903

Topiramate

0.5-50 mg/L

903

Topiramate

10-2000 ng/mL

DMPK-C

CR

LCMS/MS

US

Tacrolimus

setup (589/3 903 Cellulose) CF12

T

37

TE D

MA N

-

EPC = [DBSconc/(100Hct%) x CF]

CE P

Hct 45-50%: Blood viscosity

-

Well correlated # BA, PB Well correlated* r2=0.742 # SRA, BA Not well correlated

Well correlated r>0.800 # PB, BA

7 days at RT, 37 and -20 °C 1 day at 70 °C 31 days at 4 °C 4 weeks at RT

Yes n= 24

Yes

(77)

Yes n= 21

Yes

(78)

10 days at RT and 4 °C 35 days at -20°C 24 h at 50 °C 20 days at -20, 25 and 45 °C

Yes n= 50

No

(79)

Yes n=91

Yes

(7)

20 days at -20 and 25°C 3 months at 4 and 20 °C 1 week at RT

No

-

(20)

Yes n= 21

No

(37)

194 days at RT

No

-

(80)

20 days at -20, 25 and 45 °C

AC

20 days at -20 and 25°C

LCMS/MS LCMS/MS

No

-

n.e.

Yes – drug blood distribution

Well correlated # BA

LCMS/MS

No

Cplasma= CDBS x CF CF=2.2 (newborns) or 1.79 (adults) -

n.e.

ACCEPTED MANUSCRIPT 38

903

CG-MS

Lower Hct – Blood viscosity

No

Well 21 days at 45 °C Yes Yes correlated n= 17 r=0.994 # SRA Venlafaxine and 20-1000 µg/mL DMPK-C LCLower Hct: No Well n.e. No OMS/MS Blood correlated desmethylvenlafaxine viscosity r2>0,980 # PC, PB ® ® ® 903: Whatman 903 ; AHL-226: Ahlstrom 226 ; DMPK-A: Whatman FTA DMPK-A ; DMPK-C: Whatman FTA DMPK-C®; 31ETCHR: Whatman 31 ET CHR®; FTA-PK:

CR

US

MA N

Whatman FTA®; No3: Whatman No3®; DMS: Agilent Bond Elut DMS®; 2992: Whatman Schleicher & Schuell 2992®; CF12: Whatman, Grade CF 12®. DBS= dried blood spot; n.d.= not described; n.e.= not evaluated; EPC= Estimated plasma concentration; CF= Correction factor; fp= fraction of the drug in the plasma; Vstd= estimated volume in the spot (Vest) at standardized Hct (35%); b is the regression coefficient between Vest and Hct; RT= Room temperature; LC-MS/MS= liquid chromatography-tandem mass spectrometry; LC-TOF-MS= liquid chromatography-time-of-flight-mass spectrometry; CG/MS= Gas chromatography coupled to mass spectrometry; MALDI-MS= matrix

regression,

PC=

Pearson´s

CE P

Deming

AC

DR=

TE D

assisted laser desorption/ionization time-of-flight mass spectrometry. *correlation with whole blood; **in vitro assay correlation; #statistical analysis: SRA= Simple regression analysis,

(6)

T

5-250 µg/mL

IP

Valproic acid

correlation,

BA=

Bland-Altman,

PB=

Passing-Bablock.

(81)

ACCEPTED MANUSCRIPT

39

Highlights

Dried blood spots are promising new samples for therapeutic drug monitoring.

-

DBS assays require highly sensitive analytical methods, mainly mass

T

-

IP

spectrometric.

Particular validation assays for DBS methods are described and discussed.

-

Applications of DBS sampling for TDM in the last 10 years are reviewed.

AC

CE P

TE

D

MA

NU

SC R

-