Keywords
direct oral anticoagulants - dabigatran - rivaroxaban - apixaban - serum - methods
Serum samples are taken from all patients in acute and nonacute clinical situations.
On adequate handling, clinical chemical analyses are performed on automated clinical
chemistry analyzers. Blood coagulation parameters need to be collected into plastic
or siliconized glass tubes containing an anticoagulant such as sodium citrate to inhibit
blood coagulation in vitro. Limitations of blood drawing for coagulation parameters
include incomplete filling of the tube resulting in an incorrect ratio of anticoagulant
to blood leading to incorrect coagulation results and activation of blood clotting
during and after blood sampling due to incorrect handling. Correct handling and centrifugation
at given temperatures, analysis within a given time frame, and specific caution for
freezing and thawing of samples are also required.[1]
Serum samples of patients are often used to determine the concentration of drugs during
therapy.[2] Drug levels or other clinical chemical parameters may be requested by clinicians
hours or days after original blood collection as an add on test. Therefore, serum
samples are frequently stored routinely for several days. This becomes important for
patients with drug overdose or intoxication or for forensic purposes.[3]
[4] Heparin, low-molecular-weight heparin, fondaparinux, and danaparoid need antithrombin
or heparin-cofactor II for accelerating the inhibition of specific coagulation enzymes.[5] A specific determination of hirudins requires activation through meizothrombin.[6] Antithrombin and meizothrombin are consumed during blood clotting as would take
place during preparation of serum. Therefore, antithrombin and meizothrombin-dependent
anticoagulant drugs may encounter limitations when analyzed from serum samples.
DOACs bind specifically either to thrombin or to factor Xa. These coagulation proteases
are also consumed during blood clotting that occurs during preparation of serum samples.
DOAC concentrations are determined by liquid chromatography techniques after extensive
purification from plasma proteins.[7]
[8]
[9] These methods require specialized techniques and expert operators, which are rarely
available in routine laboratories. UV (ultraviolet)-detection photometry analysis
provides an easy, less time consuming, and sensitive analysis for pharmaceuticals.
Here, we describe a simple and rapid photometric method for the determination of DOAC
in serum. We have compared data obtained by spiking DOAC to human serum and plasma
samples as well as from patients on treatment.
Methods
Patients
These studies were approved by the local ethics committee and participants gave written
informed consent before blood sampling. Serum and plasma samples were taken from patients
under treatment with rivaroxaban 10-mg once daily before and between days 4 to 6 following
primary elective total knee and hip replacement surgery 12 hours after intake of medicine,
from patients with atrial fibrillation on treatment with dabigatran 110- or 150-mg
twice daily (2 or 12 hours after intake of medicine) or treated with apixaban 5-mg
twice daily 12 hours after intake of the drug and for control studies from patients
not taking any anticoagulant.
Preparation of Samples
Serum Samples
Blood was collected from control persons without intake of an anticoagulant and from
patients on treatment with one of the DOACs into kaolin containing plastic tubes to
generate and obtain serum. Blood was allowed to clot at room temperature for 30 minutes.
Samples were centrifuged at 1,800 g and room temperature for 10 minutes and the supernatant
was shock frozen in aliquots and stored at –72°C until analyzed.
Plasma Samples
Blood for platelet poor plasma (PPP) samples were taken during the same venipuncture
as for those to obtain serum, collected into plastic tubes containing 3.8% sodium
citrate (1/9, v/v, citrate/blood), centrifuged within 30 minutes at 1,800 g and room
temperature for 10 minutes and several aliquots of the supernatant were shock frozen
and kept at −85°C until analyzed.
Origin and Quality of DOACs
Dabigatran and rivaroxaban were purified from commercially available Pradaxa (Boehringer,
Ingelheim, Germany) and Xarelto (Bayer HealthCare, Wuppertal, Germany; optical density
at 405 nm) and their purity was characterized by analytical methods as described.[10]
[11] Apixaban was provided by BMS (Plainsboro, NJ).
Spiking of Serum and Plasma Samples with DOAC
Serial dilutions of dabigatran, rivaroxaban, and apixaban were added to 6 (rivaroxaban
and apixaban) or 4 (dabigatran) individual serum and plasma samples from persons not
taking an anticoagulant. Samples were analyzed within 2 hours for the content of the
DOAC by the methods described later. The mean of the six determinations was calculated
for serum and plasma samples of each DOAC. The concentration of each sample was calculated
from the optical density at 405 nm (OD) concentration curve for serum and plasma samples
of each DOAC using SAS release 9.3 program (SAS Institute Inc., Cary, NC).
Analysis of Dabigatran
Serum Samples
Dabigatran was determined from serum samples by a colorimetric method[12] (international patent application No PCT/EP2012/002540). The assays were run on
a microtiter plate reader with duplicate samples (Multiskan FC, Thermo Fisher Scientific,
Langenselbold, Germany) connected to the software program SkanIt 3.1 (Thermo Fisher
Scientific, Germany). Initially, the assay was developed by using different amounts
of the reagents and incubation periods (data not shown). The final assay procedure
was as initially described also for plasma samples.[13] Ten µL serum and 100 µL tris buffer (0.05 mol Tris, 0.075 mol EDTA, 0.175 mol NaCl,
37 KIU aprotinin/mL, pH 8.4) were incubated for 5 minutes followed by addition of
100 µL human thrombin (Sigma Aldrich, Deisenhofen, Germany, 0.74 NIH units/mL tris
buffer), for 60 seconds at 37°C. Fifty µL chromogenic substrate S2238 (1.59 mmol/L
in distilled water, Chromogenix, Essen, Germany) were added and incubated for 5 minutes.
The reaction was stopped by adding 25 µL 20% acetic acid. For all samples, a background
OD at 405 nm was used by pipetting the reagents from backward, that is, first acetic
acid, to ensure a similar final matrix but with no activity. Absorption of samples
was read at 405 nm and the concentration of dabigatran of the samples was calculated
from a concentration/optical density curve using the software program and prepared
from pooled human serum samples (obtained from 20 volunteers) spiked with dabigatran
ranging from 25 to 500 ng/mL.
Plasma Samples
The assay was performed with the same solution and incubation steps as for serum samples
using the software program SkanIt 3.1 for Multiskan FC.
Analysis of Rivaroxaban and Apixaban
Serum Samples
Rivaroxaban and apixaban were determined by a colorimetric method[14] in the presence of human factor Xa and a substrate (patent No GB1019674.9). The
samples and reagents were pipetted into microtiter plates and after a special time
period the results of the reaction were measured in a special reader (Multiskan FC,
Thermo Fisher Scientific, Langenselbold, Germany) connected to the software program
SkanIt 3.1 (Thermo Fisher Scientific, Germany).
The optimal assay conditions were developed using serial dilutions of the reagents
and incubation times (data not shown). The final assay was conducted as follows: 25
µL serum (diluted 1:15 with tris-buffer) and 25 µL factor Xa (7.1 nkat/mL) were incubated
for 2 minutes followed by addition of 50 µL chromogenic substrate S2222 (3.37 mmol/L
distilled water, Chromogenix, Essen, Germany). After 20 minutes of incubation, the
reaction was stopped by adding 25 µL of 20% acetic acid. For all samples, a background
OD at 405 nm was used by pipetting the reagents from backward, that is, first acetic
acid. Absorption of samples was read at 405 nm and the concentration of rivaroxaban/apixaban
of the samples was calculated from a concentration/optical density curve prepared
from pooled human serum samples (obtained from 20 volunteers) spiked with rivaroxaban/apixaban
using the software program.
Plasma Samples
The antifactor Xa activity was performed with the same assay conditions as described
for serum samples to determine rivaroxaban/apixaban and using the software program
SkanIt 3.1 for Multiskan FC.
Statistical Analysis
Data are given as mean and standard deviation (SD). Box plots were used for the figures
of concentration of DOAC in serum and plasma samples. Differences between groups were
calculated using the SAS release 9.3 software program. SAS mixed procedure was used
to analyze differences between control and patients groups. The level of significance
was determined by the npar one-way procedure. The level of significance was set at
p < 0.01.
Results
Spiking of DOAC to Serum and Plasma Samples
Spiking of serial dilutions of the DOAC showed no differences for serum and plasma
concentrations for dabigatran (r = 0.9997), rivaroxaban (r = 0.9991), and apixaban (r = 0.9998), respectively ([Fig. 1A–C]).
Fig. 1 Comparative concentrations of direct oral anticoagulants (DOACs) for serum and plasma
samples of healthy volunteers (n = 6, mean, SD) spiked with different concentrations of dabigatran (A), rivaroxaban
(B), and apixaban (C), as measured by test procedure described in text and OD at 405
nm.
Patients on Treatment with Dabigatran
Patients not on treatment with an anticoagulant displayed serum and plasma concentrations
of 16.6 ± 18.7 ng/mL and 8.4 ± 7.7 ng/mL, respectively (p = 0.0001) ([Fig. 2A]). In patients on therapy, the concentrations of dabigatran in serum (n = 219) and plasma (n = 363) were 75.9 ± 70.2 ng/mL and 112.9 ± 85.8 ng/mL, respectively (p = 0.0223) ([Fig. 2B]). The p values for differences of serum samples of control versus patients was 0.0085 and
for plasma samples less than 0.0001. The correlation of serum and plasma values was
r = 0.165.
Fig. 2 Box plots of concentrations of dabigatran in serum and plasma samples of control
persons not treated with an anticoagulant (A), and on treatment with dabigatran (B).
Patients on Treatment with Rivaroxaban
Control patients without intake of an anticoagulant had serum (n = 143) and plasma (n = 144) concentrations for rivaroxaban of 18.0 ± 17.6 ng/mL and 14.5 ± 12.4 ng/mL
respectively (p = 0.36) ([Fig. 3A]). In patients on therapy, the concentrations of rivaroxaban in serum (n = 132) and plasma (n = 135) were 81.7 ± 40.3 ng/mL and 66.5 ± 40.7 respectively (p = 0.0003) ([Fig. 3B]). The p values for differences of serum and plasma samples of controls versus patients were
both less than 0.0001. Values of serum and plasma samples displayed a correlation
of 0.7251.
Fig. 3 Box plots of concentrations of rivaroxaban in serum and plasma samples of control
persons not treated with an anticoagulant (A) and of patients on therapy with rivaroxaban
(B).
Patients on Treatment with Apixaban
Control patients without intake of an anticoagulant had serum (n = 48) and plasma (n = 48) concentrations for apixaban of 24.4 ± 18.1 ng/mL and 16.8 ± 23.2 ng/mL, respectively
(p = 0.0031) ([Fig. 4A]). In patients on therapy, the concentrations of apixaban in serum (n = 58) and plasma (n = 69) were 263.6 ± 141.7 ng/mL and 190.7 ± 82.4 respectively (p = 0.0003) ([Fig. 4B]). The p values for differences of serum and plasma samples of controls versus patients were
both less than 0.0001. The correlation of serum and plasma values in patients was
r = 0.709.
Fig. 4 Box plots of concentrations of apixaban in serum and plasma samples of control persons
not treated with an anticoagulant (A) and of patients treated with apixaban (B).
Discussion
The results of the present investigations demonstrate that dabigatran, rivaroxaban,
and apixaban can be determined from serum samples after spiking samples with the DAOC.
The correlations of serum and plasma samples were very high and in the same range
as reported for other drugs.[15] The correlation of values using serum and plasma samples of patients on treatment
was lower but also in the range reported for other drugs.[16] Dabigatran showed lower relation between serum and plasma levels using samples of
patients on treatment compared with rivaroxaban or apixaban. Instability of drugs
in serum has been reported for other compounds.[17] The reason of a higher variability of serum compared with plasma concentrations
of patients on treatment with the thrombin inhibitor dabigatran may be manifold and
remain to be investigated.
Methods for determination of drugs in serum samples are required if plasma samples
are not available (3). Serum samples are preferred for determination of drugs because
of the absence of many proteins that relates mainly to coagulation proteins. Purification
of drugs from serum samples is therefore easier to perform.[18] DOACs can also be the determined from serum samples using sophisticated liquid chromatography
techniques (7, 8, 9). With these analyses, we demonstrated that DOACs can be determined
also by adding an excess of thrombin or factor Xa followed by addition of an enzyme-specific
substrate that releases quantitatively a chromophor from the substrate negatively
related to the presence of the DOAC (12, 14). These assays can be performed in laboratories
equipped with coagulation platforms to determine anticoagulants.
Clinical trials with DOACs were performed without determination of the anticoagulant
effect of dabigatran,[19] rivaroxaban,[20] or apixaban.[21] Pharmacokinetic analyses were computed using a population-based model for dabigatran[22]
[23] and rivaroxaban.[24]
[25]
[26] Markers for activation of blood coagulation were reported from a clinical study
using apixaban.[27] Few reports were published for small patient groups treated with dabigatran[28]
[29]
[30] or rivaroxaban.[29]
[ 31]
[ 32]
[ 33] The reported values for subjects not on DOACs are in the same range as reported
in our patients using plasma and serum samples. For apixaban, similar ranges of steady
state plasma concentrations were reported for volunteers as in our patients.[34]
There were some differences between the results of the three DOACs when determined
in serum and in plasma samples from patients on treatment. For dabigatran, control
values were higher in serum than in plasma samples. In patients, however, dabigatran
concentrations were lower in serum than in plasma samples. However, differences were
significant between persons not on dabigatran and patients on treatment in both cases.
Therefore, patients on treatment can be separated from control values using serum
samples as well as plasma samples. One reason for the lower values in serum samples
of patients on treatment may that dabigatran is consumed during coagulation of blood
after withdrawal to obtain serum. This may explain in part also the low correlation
of serum and plasma values in patients on therapy with dabigatran. The mechanism of
fibrin formation in serum samples in the presence of dabigatran remains to be investigated.
Values for controls without treatment of rivaroxaban were not different but somewhat
higher for apixaban in serum compared with plasma samples. In patients on treatment,
serum values were higher for rivaroxaban and apixaban when determined from serum samples
compared with plasma samples. The mechanisms of the interaction of these inhibitors
on fibrin formation and the difference to dabigatran remain to be determined. Importantly,
serum samples are significantly higher in patients on therapy compared with controls
and serum and plasma concentrations show a good correlation.
The amount of anticoagulant present in patient may be necessary to be determined in
deterioration of renal function, in the elderly or oldest population, for surgery
or in acute clinical situations, during recurrent events or during bleeding complications,
or to check compliance.[35]
[36] The concentration of DOACs can be accurately determined by high-pressure liquid
chromatography methods. These techniques are, however, not suitable for clinical routine
use. Thrombin- and factor Xa-dependent coagulation or chromogenic substrate assays
are developed to determine the concentration or activity of dabigatran, rivaroxaban,
and apixaban.
The limitations of this investigation are that the determination DOAC in serum samples
remains to be validated in clinical studies. However, a validation of the plasma concentrations
of DOACs also has to be undertaken in relation to the incidences of thrombotic or
bleeding events. The time of blood sampling after the intake of the last dose of the
anticoagulant plays an important role and remains to be standardized in general. The
strength of the investigations is that the feasibility of using serum samples for
the determination of DOACs was shown using a simple and reliable method that can be
applied to a coagulation platform connected with a photometer.
In conclusion, the determination of dabigatran, rivaroxaban, and apixaban from serum
samples of patients may offer an additional tool in specific clinical situations such
as lack of withdrawal of plasma samples or facilitated freezing compared with plasma.
The suitability of the determination of DOACs from serum samples by other anti-factor
Xa chromogenic substrate assays remains to be proven. If this results in positive
findings, adaption of the assays to specific coagulation platforms may reduce the
variability of the results.[37]
