Accepted Manuscript

Title: Pharmacogenetic Aspects of TramadolPharmacokinetics and Pharmacodynamics After a Single OralDose

Author: Salumeh Bastami Pernilla Haage Robert KronstrandFredrik C. Kugelberg Anna-Lena Zackrisson SrinivasUppugunduri

PII: S0379-0738(14)00096-6DOI: http://dx.doi.org/doi:10.1016/j.forsciint.2014.03.003Reference: FSI 7534

To appear in: FSI

Received date: 31-10-2013Revised date: 20-2-2014Accepted date: 2-3-2014

Please cite this article as: S. Bastami, P. Haage, R. Kronstrand, F.C. Kugelberg, A.-L.Zackrisson, S. Uppugunduri, Pharmacogenetic Aspects of Tramadol Pharmacokineticsand Pharmacodynamics After a Single Oral Dose, Forensic Science International(2014), http://dx.doi.org/10.1016/j.forsciint.2014.03.003

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 proofbefore it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers thatapply to the journal pertain.

Page 1 of 23

Accep

ted

Man

uscr

ipt

Pharmacogenetic Aspects of Tramadol Pharmacokinetics and Pharmacodynamics After a

Single Oral Dose

Salumeh Bastami*1, Pernilla Haage*1,2, Robert Kronstrand1,2, Fredrik C. Kugelberg1,2, Anna-Lena Zackrisson2, Srinivas Uppugunduri3

* Both authors contributed equally to this work

Affiliations

1 Department of Medical and Health Sciences, Division of Drug Research, Linköping University, Linköping, Sweden

2 National Board of Forensic Medicine, Department of Forensic Genetics and Forensic Toxicology, Linköping, Sweden

3 Department of Clinical and Experimental Medicine, Linköping University, Department of Clinical Chemistry, County Council of Östergötland, Linköping, Sweden.

Corresponding author

Anna-Lena Zackrisson

National Board of Forensic Medicine Department of Forensic Genetics and Forensic Toxicology

Artillerigatan 12 SE 587 58 Linköping, Sweden

E-mail: anna-lena.zackrisson@rmv.se

Phone: +46 13 252153

Fax: +46 13 136005

Page 2 of 23

Accep

ted

Man

uscr

ipt

Pharmacogenetic Aspects of Tramadol Pharmacokinetics and Pharmacodynamics After a

Single Oral Dose

Abstract

The major purpose of this study was to elucidate if genotyping can facilitate interpretations of tramadol (TRA) in

forensic case work, with special regard to the estimation of the time of drug intake and drug related symptoms

(DRS). The association between genetic polymorphisms in CYP2D6, OPRM1 and ABCB1 and pharmacokinetic

and pharmacodynamic properties of TRA was studied. Nineteen healthy volunteers were randomized into two

groups receiving a single dose of either 50 or 100 mg of orally administrated TRA. Blood samples were collected

prior to dosing and up to 72 h after drug intake. The subjects were asked to report DRS during the experimental

day. We found a positive correlation between the metabolic ratio of O-desmethyltramadol (ODT) to TRA and the

time after drug intake for both CYP2D6 intermediate metabolizers and extensive metabolizers. For the only poor

metabolizer with detectable ODT levels the metabolic ratio was almost constant. Significant associations were

found between the area under the concentration-time curve (AUC) and three of the investigated ABCB1 single

nucleotide polymorphisms for TRA, but not for ODT and only in the 50 mg dosage group. There was great

interindividual variation in DRS, some subjects exhibited no symptoms at all whereas one subject both fainted and

vomited after a single therapeutic dose. However, no associations could be found between DRS and investigated

polymorphisms. We conclude that the metabolic ratio of ODT/TRA may be used for estimation of the time of drug

intake, but only when the CYP2D6 genotype is known and taken into consideration. The influence of genetic

polymorphisms in ABCB1 and OPRM1 requires further study.

Key Words Tramadol, Pharmacokinetics, Pharmacodynamics, CYP2D6, ABCB1, OPRM1.

Page 3 of 23

Accep

ted

Man

uscr

ipt

1. Introduction

The use of tramadol (TRA) for treatment of moderate to severe pain has been increasing steadily during the last

decades. Unfortunately, abuse of this drug is also becoming more common. Further, development of addiction to

TRA in association with analgesic treatment within the recommended dose range is another alarming trend. A

history of abuse or use of a drug of abuse seems to be an important risk factor [1]. TRA has also become a more

common cause of death in drug addicts with a similar trend for increase in overdose cases [2, 3].

Analysis of drugs of abuse is a common feature of forensic investigations and correct interpretation of the

measured concentrations is important in both post mortem and human performance toxicology. Accurate

estimation of the time of drug intake and expected drug effects from a certain dose or concentration are also

frequent issues in drug-facilitated crimes.

TRA has a dual mechanism of action, acting as a µ-opioid receptor agonist as well as a serotonin and

norepinephrine reuptake inhibitor. The cytochrome P450 (CYP) enzyme CYP2D6 is involved in the formation of

the active metabolite O-desmethyltramadol (ODT). In comparison to TRA, ODT is a significantly more potent μ-

opioid agonist [4]. The concentration of the parent compound alone is often not sufficient to make an accurate

estimation of the time of drug intake. The ratio between metabolite and parent compound is generally used to

indicate a recent acute intake, e.g. to diagnose suspected acute overdoses. It has earlier been shown, for some

substances other than TRA, that the ratio between metabolite and parent compound also can be helpful in a more

accurate estimation of the time of intake [5, 6]. The amount of ODT formed is largely dependent on the CYP2D6

genotype but also on the time of intake. It is however unclear if the metabolic ratio (MR) of ODT/TRA is a useful

indicator of the time of TRA intake.

There is considerable variation in the interindividual response to the same dose of opioids, including TRA, with

respect to both therapeutic and adverse effects. Development of tolerance due to prolonged use of opioids in

addition to genetic factors could partly explain some of these differences. Several genes and polymorphisms have

been studied in this respect. Reduced or absent metabolite formation with reduced analgesic effect have been

observed after TRA administration in CYP2D6 poor metabolizers (PMs) [7], whereas ultrarapid metabolizers

(UMs), are instead associated with quicker analgesic effects, but with higher risk for adverse effects [8].

Page 4 of 23

Accep

ted

Man

uscr

ipt

The ABCB1 gene encodes P-glycoprotein (P-gp), which is located in the blood-brain barrier and gut and is

responsible for the cellular efflux of a variety of drugs [9]. The most common single nucleotide polymorphisms

(SNPs) in the coding region are C1236T, G2677T and C3435T [10]. C3435T, the most studied SNP, has been

associated with both increased and decreased expression of P-gp [9]. An altered expression in gut could potentially

change the pharmacokinetics of the drugs being substrates of this transporter. A decreased expression and

functionality of P-gp has been suggested in homozygous for the 3435T allele [11], possibly leading to higher

bioavailability and subsequently a higher concentration of TRA in blood. Although P-gp is of importance for the

pharmacokinetic and pharmacodynamic properties of a number of substances, including opioids [9, 12], its

importance for TRA remains unknown. A clinical study suggested that TRA is a substrate of P-gp [13]. In

contrast, an in vitro study and one in vivo study in rat have demonstrated that TRA and ODT are not P-gp

substrates [14, 15]. A recent study showed no association between the C3435T polymorphism and pain relief in

patients receiving TRA [16]. Taken together, the importance of P-gp for the pharmacokinetic and

pharmacodynamics properties of TRA still remains to be corroborated.

Genetic polymorphisms in OPRM1 have been associated with an altered pain threshold and opioid requirements.

Results from clinical studies have shown that patients’ homozygous for the wild-type of the most common SNP

A118G require less opioids than those with the other allelic variants, AG and GG [17, 18]. It has also been shown

that the 118G allele is associated with a more severe clinical outcome in emergency department patients with acute

drug overdose [19]. There is only limited information regarding the influence of OPRM1 A118G polymorphism

on TRA efficacy [20].

We undertook a human study to elucidate if genotyping can facilitate interpretations of TRA in forensic case work,

with special regard to the estimation of the time of drug intake and drug related symptoms (DRS). Our study had

the following specific aims:

1) Investigate if the metabolic ratio (MR) of ODT/TRA is a useful indicator of the time of TRA intake.

2) Determine the association between polymorphisms in the ABCB1 gene (SNPs G1199A, C1236T, G2677T,

C3435T) and pharmacokinetic parameters of TRA.

3) Study the association between polymorphisms in the CYP2D6, OPRM1 (SNP A118G) or ABCB1 (SNPs

G1199A, C1236T, G2677T, C3435T) gene and DRS.

Page 5 of 23

Accep

ted

Man

uscr

ipt

2. Methods

2.1 Study participants and study design

Twenty healthy volunteers were recruited through advertisements. Nineteen subjects (nine males, ten females)

aged 18 years or older (mean 25.4±4.3) completed the study, whereas one subject did not show up on the

experimental day due to unspecified illness. Participants were informed and examined by a physician before

inclusion in the study. Questions of present or previous drug use were asked. Previous or concurrent use of opioids

or drugs known of interacting with TRA were exclusion criteria. Demographic data such as age, gender, height

and weight were noted. Seven of the female participants were taking oral contraceptives. None of the participants

stated upon question that they were pregnant or breast-feeding. Written informed consent was collected from each

participant. The study was approved by the Regional Ethical Review Board in Linköping (No: 2011/337-31). The

subjects were randomized into two groups receiving a single dose of either 50 or 100 mg of orally administered

TRA (Tramadol HEXAL, Sandoz). During the experimental day, blood samples were obtained from a peripheral

venous catheter, Insyte W in combination with a Mandrin (Becton Dicinson AB), which was inserted in the

forearm. The blood was collected in labeled 7 ml heparinized collection tubes using a Vacutainer Luer-Lok Access

Device (Becton Dicinson AB). Blood samples were collected prior to dosing and at 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4, 5,

6, 7, 8, 10, 24, 48 and 72 h. Samples were stored at –80 ºC pending analysis. The participants were exhorted to eat

breakfast according to their usual routines. Lunch was bought from a nearby restaurant and served around noon.

Fruit, some biscuits, tea, coffee, juice and water were available during the whole day. The subjects were requested

to fill in a form regarding their experience of DRS during the experimental day in conjunction to the last blood

sampling on the first day (10 h). Seven questions about nausea, dizziness, headache, vomiting, dry mouth,

sweating and fatigue were posed. A scale between zero to five were used, where zero was no symptoms at all and

five was worst imaginable symptoms.

2.2 Quantitation of tramadol and O-desmethyltramadol in whole blood

2.2.1 Chemicals and reagents

Acetonitrile (gradient grade), methanol (gradient grade) and formic acid (98 %) were purchased from Merck

(Darmstadt, Germany). Ammonium formate (98 %) and ethanol (95 %) were purchased from Fluka (Basel,

Switzerland) and Kemetyl (Haninge, Sweden) respectively. Water used was first purified with a MilliQ-water

purifying system (Millipore Corporation, Bedford, MA, USA). Reference substances used for making calibrators

and quality controls (QCs), i.e. cis-tramadol and O-desmethyl-cis-tramadol were purchased from Cerilliant

Page 6 of 23

Accep

ted

Man

uscr

ipt

(Austin, TX, USA). Tramadol-13c-d3 and O-desmethyl-cis-tramadol-d6, used for making the internal standard

were also purchased from Cerilliant.

2.2.2 Instrumentation

High performance liquid chromatography was performed on a 1290 Infinity LC instrument (Agilent

Technologies), using a 2.1x100 mm Zorbax Eclipse Plus C18 RRHD column with 1.8 μm particles. A guard filter

with 0.2 μm particles (Waters) was also used. Column temperature was set to 60 ˚C. Mass detection (MS/MS) was

performed on an AT6460 instrument (Agilent Technologies) with an electrospray interface, using positive

ionization. Mobile phase A consisted of 0.05 % formic acid in 10 mM ammonium formate while mobile phase B

consisted of 0.05 % formic acid in methanol. Gradient elution with a total run time of 8 min was used. The total

flow rate was 0.5 ml/min. The software used was Masshunter (Agilent Technologies). Identification criteria were

based on transition ratios. Both TRA and ODT exhibit a very strong base peak of 58.1 and few other fragments,

which have abundance less than 10 % of base peak. In accordance with European recommendations (EUD

2002/657/EC) an acceptance limit of 50 % was used. The following transitions were used for TRA, ODT,

tramadol-13c-d3, and O-desmethyl-cis-tramadol-d6, respectively: 264.2 → 58.1; 246.1, 250.1 → 58.1; 232.1,

268.2 → 58.1 and 256.2 → 64.2. The underlined fragments were used for quantification.

2.2.3 Sample preparation

25 μl of internal standard (tramadol-13c-d3 and O-desmethyl-cis-tramadol-d6, 4.0 μg/ml) and 1 ml of protein

precipitation solvent (0.075 % formic acid in acetonitrile:ethanol; 90:10 v/v) were added to 0.5 g of whole blood.

The samples were then mixed for 10 minutes and centrifuged for another 10 minutes (5000 rpm, 5 ˚C). 100 μl of

the supernatant was finally transferred to a vial for LC-MS-MS analysis. The injection volume was 3 μl. A blank

sample, a low (100 ng/g T, 40 ng/g ODT) and a high (2000 ng/g T and ODT) QC were run within each batch of

samples.

2.2.4 Method characteristics

Calibration models were evaluated by analysis of six replicates at nine levels from 10 to 3000 ng/g blood and were

found to be best fitted by quadratic equations using a 1/X weighting. Calibrators and QCs were made by adding a

volume of standard to 0.5 g of bovine whole blood. Absence of substances capable of interacting with the

quantitative analysis was confirmed by using the described method prior to use. TRA and ODT were added in the

Page 7 of 23

Accep

ted

Man

uscr

ipt

same concentration for each calibrator. Lower limit of quantitation (LLOQ) was assessed by running five

replicates at lower and lower concentrations and defined as the concentration where the imprecision was less than

25 % and the accuracy was 75-125 %. LLOQ was found to be 10 ng/g for ODT, and 20 ng/g for TRA. The method

imprecision and accuracy was evaluated by analysis of triplicates of controls at four levels during eight days

(n=24). The total imprecision for TRA was 2.7 % (at 40 ng/g), 2.5 % (at 200 ng/g), 5.0 % (at 600 ng/g), and 3.5 %

(at 2000 ng/g). The accuracy was between 90.4 and 94.7 % at all levels. The total imprecision for ODT was 2.8 %

(at 40 ng/g), 2.2 % (at 200 ng/g), 4.4 % (at 600 ng/g), and 3.8 % (at 2000 ng/g). The accuracy was between 88.0

and 94.2 % at all levels. The method is currently enrolled in the Nordquant PT scheme and has presented with

good results.

2.3 Genotyping

Genotyping analyses were performed at the Department of Forensic Genetics and Forensic Toxicology, Linköping,

Sweden and at the Department of Clinical Chemistry, County Council of Östergötland, Linköping, Sweden.

Genomic DNA was extracted using NorDiag Arrow (Autogen, Holliston, MA, USA). The extracted DNA was

stored frozen at –20 ºC until analyzed.

Genotyping of CYP2D6 includes three SNPs; CYP2D6*3 (rs35742686), CYP2D6*4 (rs3892097) and

CYP2D6*6 (rs5030655) as well as determination of copy number variation (CNV), which includes identification

of whole gene deletion (CYP2D6*5) and multiple gene copies (CYP2D6xN). These methods have earlier been

described [21, 22] but some minor modifications were made. In brief, the PCR amplification was performed in a

total volume of 10 µl using ~5 ng human genomic DNA, 0.2 µl of 20 µM forward and reverse primers (Invitrogen,

Lidingö, Sweden), 5 µl 2x HotStarTaq Plus Master Mix (Qiagen, Hilden, Germany) and for the CNV reactions 2

µl Q-solution was also needed (Qiagen, Hilden, Germany). Primer sequences are shown in Supplement 1. The

PCR reactions were carried out on a Gene Amp PCR System 9700 (Applied Biosystems) with an initial

denaturation step at 95 °C for 5 min, thereafter 40 cycles of 95 °C for 30 s, 57 °C for 30 s and 72 °C for 30 s,

followed by a final ex tension step at 72 °C for 10 min. Alleles not carrying any of the determined polymorphisms

were classified as *1 (wild-type). The outcomes of the genotype analysis were categorized into four groups;

individuals carrying no active gene (i.e. carrier of only the *3, *4, *5 or *6 alleles, also known as poor

metabolizers, PMs), individuals carrying one active gene (i.e. carrier of *1 in combination with one of the alleles

*3, *4, *5 or *6, also known as intermediate metabolizers, IMs), individuals with two active genes (i.e. carrier of

two *1 alleles, also known as extensive metabolizers, EMs) and individuals carrying more than two active genes

(i.e. carrier of multiple *1 alleles, also known as ultrarapid metabolizers, UMs).

Page 8 of 23

Accep

ted

Man

uscr

ipt

The PCR reactions for OPRM1 A118G, (rs1799971) and ABCB1 G1199A (rs2229109), ABCB1 C1236T

(rs1128503), ABCB1 G2677T/A (rs2032582) and ABCB1 C3435T (rs1045642) were carried out on a Mastercycler

ep (Eppendorf, Hamburg, Germany). For each PCR reaction, 2 µl 10× PCR buffer, 2 mM MgCl2, 0.5 U HotStar

Taq polymerase (Qiagen, Hilden, Germany), 125 µM dNTPs (VWR International, Stockholm, Sweden), 5 pmol of

each forward and reverse primer (Biomers, Ulm, Germany), 20 – 50 ng DNA sample and nuclease free water were

mixed to a final volume of 20 µl. The PCR reaction was initiated with 15 min of enzyme activation at 95 °C

followed by 50 cycles at 95 °C for 30 s, 63 °C for 45 s, and 72 °C for 60 s. The reaction was finished with a 5 min

final extension step at 72 °C followed by 4 °C incubation. Primer sequences are shown in Supplement 1.

Pyrosequencing was carried out on a Q96 MD according to the manufacturer’s recommendations (Qiagen,

Hilden, Germany). Briefly, to 10 µl PCR products a mix containing 2 µl steptavidine-coated sepharose beads

(Amersham Biosciences, Piscataway, NJ, USA), 40 µl 1x binding buffer (10 mM Tris-HCl, 2M NaCl, 1mM

EDTA, 0.1% Tween 20, pH 7.6) and 28 µl water was added. The samples were shaken on a thermomixer

(Eppendorf, Hamburg, Germany) at 1400 rpm for 5 min. Biotinylated single-stranded DNA (ssDNA) was prepared

on a Vacuum Prep Workstation according to the manufacturer’s instructions (Qiagen, Hilden, Germany). The

ssDNA was hybridised to 15 pmol sequencing primer in 12 µl 1x annealing buffer (200 mM Tris-acetate and 50

mM MgAc2, pH 7.6) at 80 ºC for 2 min using a block thermostat (Grant Instrument, Cambridge, UK). A SNP

Reagent containing substrate, enzyme and dNTP mixtures (Qiagen, Hilden, Germany) were added to a reagent

cartridge, and the pyrosequencing reaction was carried out according to the dispensation orders shown in

Supplement 1.

2.4 Data analysis

The area under the concentration-time curve (AUC0-10) was calculated using the linear trapezoidal method. For

AUC0-∞ the area was extrapolated to infinity using the logarithmic trapezoidal method. Since the extrapolated area

exceeded 20% of the total AUC0-∞ we opted to always use AUC0-10 for subsequent group comparisons. AUC

therefore refers to the AUC0-10 values throughout this article. Peak blood concentrations (Cmax) and corresponding

times (tmax) of TRA and ODT were read directly from the data. Metabolic ratios of ODT/TRA are in the present

study abbreviated MR according to the following: MR = CODT/CTRA, AUC MR = AUCODT/AUCTRA, Cmax MR =

Cmax ODT/Cmax TRA.

Statistical analysis was performed by using the SPSS statistical program (Version 19.0 for Windows; IBM SPSS).

Non-parametric analysis method was used to compare the pharmacokinetics parameters in subjects taking 50 and

100 mg TRA. Mann-Whitney test was also used to compare allelic variation in genes between groups. The

Page 9 of 23

Accep

ted

Man

uscr

ipt

hypothetical effect of allelic variation on AUC MR and Cmax MR was estimated by regression analysis.

Significance level was set at P<0.05.

3. Results and Discussion

The pharmacokinetic parameters, i.e. mean values of Cmax, tmax and AUC0-∞, observed for TRA and ODT (Table 1)

in this study were in agreement with earlier published data [23]. Concentration-time profiles of TRA and ODT in

whole blood are shown for both dosage groups in Fig. 1. As expected, Cmax and AUC0-10 for TRA was almost two

times higher in the subjects given 100 mg compared to those given 50 mg (P<0.001; Fig. 1 and Table 1). A similar

trend was observed for ODT but the differences were not statistically significant, possibly due to the high

interindividual variation between the subjects in each group. Similar median values were obtained for tmax for both

TRA and ODT in respective dosage group (Table 1).

3.1 Estimation of the time of drug intake using the metabolic ratio of O-desmethyltramadol to tramadol

The subjects were grouped on the basis of the CYP2D6 genotyping analysis as follows: eight subjects were

classified as EMs, nine subjects as IMs and two subjects as PMs, lacking all CYP2D6 activity. A significant

association was found for the AUC MR and CYP2D6 genotype, (P< 0.01). The EMs and IMs showed a mean

AUC MR of 0.41±0.1 and 0.24±0.1, respectively, while the only PM individual with detectable levels of ODT

showed a ratio of 0.09. A similar level of significance (P=0.005) was observed for Cmax MR, with the highest mean

value for EMs (0.33±0.1) followed by IMs (0.20 ± 0.1) and the PM individual (0.07). Our results are in accordance

with those of Levo et al [24], who demonstrated that the ratio of TRA to ODT, in a forensic autopsy material, is

well correlated to the different genotypes of CYP2D6, the more functional alleles the lower ratio. To our

knowledge, MR in relation to time after TRA intake has not been documented earlier. We found a positive

correlation between the mean MRs and the time after drug intake for both IMs (R2 = 0.96) and EMs (R2 = 0.97).

For the only PM individual with detectable ODT levels the MR was almost constant during the 10 studied hours

(Fig. 2). The linear increase in ratio for both EMs and IMs may be used to estimate the time of intake when the

genotype is known. For example, a ratio of 0.3 may represent an intake about 1 hour prior to sampling in an EM

whereas it may represent a time of intake 10 hours earlier in an IM. In the present study, the PM always showed a

low ratio indicating a recent intake. In these interpretations it is however important to consider interindividual

variation within the genotype groups, as well as concomitant medication. Drugs capable of inhibiting the CYP2D6

enzyme [25] will change the MR but not the genotype. On the basis of our data it is possible to conclude that an

estimation of the time of drug intake is not valid without genotyping.

Page 10 of 23

Accep

ted

Man

uscr

ipt

No significant associations were found between CYP2D6 genotype and the pharmacokinetic parameters Cmax, tmax

or AUC, neither for TRA or ODT (Table 2a and b).

3.2 A putative association between ABCB1 polymorphisms and tramadol pharmacokinetics

In the 50 mg dosage group (Table 2a), there were significant associations between AUC and the three ABCB1

SNPs C1236T, G2677T/A and C3435T (P≤0.05), with the highest AUC value for homozygous of the variant

allele. The correlation held true for the AUC of TRA but not for ODT. There were no statistically significant

associations between the same SNPs and Cmax for TRA or ODT. All individuals in the 50 mg dosage group were

homozygous for the wild-type allele of the SNP G1199A.

In the 100 mg dosage group (Table 2b), there were no significant associations between the four SNPs G1199A,

C1236T, G2677T/A and C3435T in the ABCB1 gene and the pharmacokinetic parameters Cmax and AUC, neither

for TRA nor ODT. Only one subject was heterozygous for the SNP G1199A.

Although not statistically significant, Slanar et al [13] showed that both the average Cmax and AUC0-24 of TRA

alone increased with the number of 3435T alleles in healthy volunteers. The same trend was however not shown

for G2677T/A. These contradictions and also the fact that associations in our study were found only in the 50 mg

dosage group and not in the 100 mg dosage group suggests that further studies are needed to elucidate the potential

importance of ABCB1 for the pharmacokinetics of TRA.

3.3 Self reported score of drug related symptoms unrelated to polymorphisms in CYP2D6, OPRM1 or

ABCB1

Subjects given 100 mg TRA reported higher scores in the DRS form compared to subjects given 50 mg, a mean

score of 7.3±8.7 and 3.1±2.6, respectively (Table 3). Fatigue was the most common symptom in both groups while

nausea and dizziness occurred more frequently and severely in the 100 mg dosage group. Great interindividual

variation was found in DRS score, especially in the group administered 100 mg. The correlations between

polymorphisms in CYP2D6, ABCB1 and OPRM1 and DRS score are shown in Fig. 3a and b. CYP2D6 UMs have

been associated with a higher risk for adverse effects when administered TRA [8] and CYP2D6 IMs have been

reported to have a lower risk compared to EM individuals [26]. One could therefore hypothesize that the more

functional CYP2D6 alleles, the higher risk for DRS. As a corollary, one could also hypothesize that PMs would

have less DRS compared to EMs. This could not be shown in the present study (Fig 3). There were two subjects

who were CYP2D6 PMs in our study, one in each dosage group. The one with non-detectable levels of ODT,

Page 11 of 23

Accep

ted

Man

uscr

ipt

being in the 50 mg dosage group, reported a score of four while the other one reported a score of zero. The

subjects with the highest DRS scores (13, 16, 25) were all in the 100 mg dosage group and having the IM

genotype. Not much is known about the potential association between the genes OPRM1 and ABCB1 and DRS

following TRA intake. Kim et al [26] found that homozygous for OPRM1 118G had considerably lower risk of

side effects like nausea and vomiting, following treatment with paracetamol and TRA, than wild-types. There were

no subjects with the GG variant in our study, and only two with the AG variant, both in the 100 mg dosage group.

There were major interindividual differences between these two heterozygous subjects with one reporting a score

of three (subject 13) and the other a score of 25 (subject 16). Subject 16 become unconscious for about 15 seconds

at 1 h and 15 minutes after drug intake, and vomited around 4 h after TRA intake and also later in the evening. The

subject had already passed the tmax for both TRA and ODT (2 h and 30 minutes) by the time of vomiting and was

therefore not excluded from the study. Subject 13, on the other hand, was generally in good condition throughout

the study. Another interesting contradiction between these individuals is the fact that in spite of being administered

the same dose and both being CYP2D6 IMs, subject 13 had an AUCODT value about three times higher than

subject 16, 656 ng h/g compared to 214 ng h/g. Both individuals had the same genotypes for the other genes

studied, with the exception of ABCB1 C3435T. Subject 16 was heterozygous for this SNP while subject 13 had

allele variant TT. The CYP2D6 PM alleles included in our routine genotype analysis are by far the most common

in a Caucasian population, covering 93-97% of the poor metabolisers [27], [28]. It is however possible that the

CYP2D6*1 allele of subject 16 is not a wild-type allele but instead a rare PM allele and that could possibly explain

the much lower AUCODT value for this individual. Demographics is also a possible contributing factor to the

variation in concentration of ODT, but in this particular case we do not find it very likely. Both subjects were

females, 22 and 24 years of age. Their heights were 162 and 164 cm and their weights 60 kg and 55 kg,

respectively. No association was found between DRS score and the ABCB1 SNPs G1199A, C1236T, G2677T/A

and C3435T (Fig 3). It is however difficult to draw general conclusions regarding the relationship between

investigated SNPs and DRS of TRA in this study. Ideally, a larger study population should have been used, one

also representing all SNPs variants. Further, the self reported questionnaire used an arbitrary scale, which is an

additional limitation. We find it reasonable to conclude that a single therapeutic dose of TRA can certainly affect a

subject´s well-being and that genetics could be an underlying cause. It could be speculated that other factors, not

included in this study, could contribute to the variability in DRS. As the (+)-enantiomer of ODT exerts most of the

opioid effects [4] it could be postulated that the ratios between (+)- and (-)-ODT could potentially explain some of

the observed differences between subject 13 and 16. TRA is also metabolized to NDT, which though lacking any

Page 12 of 23

Accep

ted

Man

uscr

ipt

opioid effects [4] could possibly also shed some light over the differences in AUCODT. As mentioned earlier TRA

exerts its effects also through the serotonin and norepinephrine system and it has been speculated that the

neurotoxicity of TRA in cases of abuse and/or overdose is related to those systems rather than the opioid effects

[29]. These factors will be taken into consideration for future studies.

3.4 Sensitivity of the quantitative method for detection of tramadol and O-desmethyltramadol

TRA could be detected up to 24 h in seven of nine subjects given 100 mg TRA and in three of ten subjects given

50 mg. ODT was detectable in five of nine subjects 24 h after intake of 100 mg. In all subjects both TRA and ODT

could be detected up to 10 h after drug intake, however with one exception since one individual never presented

with a positive ODT sample. A higher sensitivity of the quantitative method, yielding a longer detection time,

would have made the estimation of the time of drug intake possible also beyond 10 h. This is however dependent

on a linear association between the MR and the time of drug intake. Many forensic cases involving TRA where the

estimation of time of drug intake is absolutely critical concern cases of suspicion of driving under the influence of

the drug, especially when the suspect claims an intake a certain time period before the driving. A longer detection

time would obviously also be beneficial in forensic cases where there is a delay between perpetration of crime and

specimen collection contributing to potentially false negative values.

4. Conclusions

Estimation of the time of drug intake using the MR of ODT/TRA is not valid without CYP2D6 genotyping. A

putative association was found between ABCB1 polymorphisms (C1236T, G2677T/A, C3435T) and TRA

pharmacokinetics. There were large interindividual variations in the self reported score of DRS, which seemed to

be unrelated to polymorphisms in CYP2D6, OPRM1 and ABCB1.

Conflict of interest

The authors declare that they have no conflict of interest.

Page 13 of 23

Accep

ted

Man

uscr

ipt

Figure 1

References [1] M. Tjaderborn, A.K. Jonsson, J. Ahlner, S. Hagg, Tramadol dependence: a survey of spontaneously reported cases in Sweden, Pharmacoepidemiol Drug Saf. 18 (2009) 1192-8. [2] S. Shadnia, K. Soltaninejad, K. Heydari, G. Sasanian, M. Abdollahi, Tramadol intoxication: a review of 114 cases, Hum Exp Toxicol. 27 (2008) 201-5. [3] K. De Decker, J. Cordonnier, W. Jacobs, V. Coucke, P. Schepens, P.G. Jorens, Fatal intoxication due to tramadol alone: case report and review of the literature, Forensic Sci Int. 175 (2008) 79-82. [4] C. Gillen, M. Haurand, D.J. Kobelt, S. Wnendt, Affinity, potency and efficacy of tramadol and its metabolites at the cloned human mu-opioid receptor, Naunyn Schmiedebergs Arch Pharmacol. 362 (2000) 116-21. [5] M.A. Huestis, A. Barnes, M.L. Smith, Estimating the time of last cannabis use from plasma delta9-tetrahydrocannabinol and 11-nor-9-carboxy-delta9-tetrahydrocannabinol concentrations, Clin Chem. 51 (2005) 2289-95. [6] R. Kronstrand, I. Nystrom, M. Andersson, L. Gunnarsson, S. Hagg, M. Josefsson, et al., Urinary detection times and metabolite/parent compound ratios after a single dose of buprenorphine, J Anal Toxicol. 32 (2008) 586-93. [7] U.M. Stamer, K. Lehnen, F. Hothker, B. Bayerer, S. Wolf, A. Hoeft, et al., Impact of CYP2D6 genotype on postoperative tramadol analgesia, Pain. 105 (2003) 231-8. [8] J. Kirchheiner, J.T. Keulen, S. Bauer, I. Roots, J. Brockmoller, Effects of the CYP2D6 gene duplication on the pharmacokinetics and pharmacodynamics of tramadol, J Clin Psychopharmacol. 28 (2008) 78-83. [9] K. Linnet, T.B. Ejsing, A review on the impact of P-glycoprotein on the penetration of drugs into the brain. Focus on psychotropic drugs, Eur Neuropsychopharmacol. 18 (2008) 157-69. [10] O. Levran, K. O'Hara, E. Peles, D. Li, S. Barral, B. Ray, et al., ABCB1 (MDR1) genetic variants are associated with methadone doses required for effective treatment of heroin dependence, Hum Mol Genet. 17 (2008) 2219-27. [11] D. Campa, A. Gioia, A. Tomei, P. Poli, R. Barale, Association of ABCB1/MDR1 and OPRM1 gene polymorphisms with morphine pain relief, Clin Pharmacol Ther. 83 (2008) 559-66. [12] L. Karlsson, U. Schmitt, M. Josefsson, B. Carlsson, J. Ahlner, F. Bengtsson, et al., Blood-brain barrier penetration of the enantiomers of venlafaxine and its metabolites in mice lacking P-glycoprotein, Eur Neuropsychopharmacol. 20 (2010) 632-40. [13] O. Slanar, M. Nobilis, J. Kvetina, O. Matouskova, J.R. Idle, F. Perlik, Pharmacokinetics of tramadol is affected by MDR1 polymorphism C3435T, Eur J Clin Pharmacol. 63 (2007) 419-21. [14] M. Kanaan, Y. Daali, P. Dayer, J. Desmeules, Uptake/efflux transport of tramadol enantiomers and O-desmethyl-tramadol: focus on P-glycoprotein, Basic Clin Pharmacol Toxicol. 105 (2009) 199-206. [15] B. Sheikholeslami, M. Hamidi, B. Sheikholeslami, H. Lavasani, M. Sharifzadeh, M.R. Rouini, Lack of evidence for involvement of P-glycoprotein in brain uptake of the centrally acting analgesic, tramadol in the rat, J Pharm Pharm Sci. 15 (2012) 606-15. [16] O. Slanar, P. Dupal, O. Matouskova, H. Vondrackova, P. Pafko, F. Perlik, Tramadol efficacy in patients with postoperative pain in relation to CYP2D6 and MDR1 polymorphisms, Bratisl Lek Listy. 113 (2012) 152-5. [17] W.Y. Chou, L.C. Yang, H.F. Lu, J.Y. Ko, C.H. Wang, S.H. Lin, et al., Association of mu-opioid receptor gene polymorphism (A118G) with variations in morphine consumption for analgesia after total knee arthroplasty, Acta Anaesthesiol Scand. 50 (2006) 787-92. [18] P. Klepstad, T.T. Rakvag, S. Kaasa, M. Holthe, O. Dale, P.C. Borchgrevink, et al., The 118 A > G polymorphism in the human mu-opioid receptor gene may increase morphine requirements in patients with pain caused by malignant disease, Acta Anaesthesiol Scand. 48 (2004) 1232-9. [19] A.F. Manini, M.M. Jacobs, D. Vlahov, Y.L. Hurd, Opioid receptor polymorphism A118G associated with clinical severity in a drug overdose population, J Med Toxicol. 9 (2013) 148-54.

Page 14 of 23

Accep

ted

Man

uscr

ipt

Figure 1

[20] Y.C. Liu, W.S. Wang, Human mu-opioid receptor gene A118G polymorphism predicts the efficacy of tramadol/acetaminophen combination tablets (ultracet) in oxaliplatin-induced painful neuropathy, Cancer. 118 (2012) 1718-25. [21] A.L. Zackrisson, P. Holmgren, A.B. Gladh, J. Ahlner, B. Lindblom, Fatal intoxication cases: cytochrome P450 2D6 and 2C19 genotype distributions, Eur J Clin Pharmacol. 60 (2004) 547-52. [22] E. Soderback, A.L. Zackrisson, B. Lindblom, A. Alderborn, Determination of CYP2D6 gene copy number by pyrosequencing, Clin Chem. 51 (2005) 522-31. [23] S. Grond, A. Sablotzki, Clinical pharmacology of tramadol, Clin Pharmacokinet. 43 (2004) 879-923. [24] A. Levo, A. Koski, I. Ojanpera, E. Vuori, A. Sajantila, Post-mortem SNP analysis of CYP2D6 gene reveals correlation between genotype and opioid drug (tramadol) metabolite ratios in blood, Forensic Sci Int. 135 (2003) 9-15. [25] S. Rendic, Summary of information on human CYP enzymes: human P450 metabolism data, Drug Metab Rev. 34 (2002) 83-448. [26] E. Kim, C.B. Choi, C. Kang, S.C. Bae, G. Ultracet Study, Adverse events in analgesic treatment with tramadol associated with CYP2D6 extensive-metaboliser and OPRM1 high-expression variants, Ann Rheum Dis. 69 (2010) 1889-90. [27] D. Marez, M. Legrand, N. Sabbagh, J.M. Lo Guidice, C. Spire, J.J. Lafitte, et al., Polymorphism of the cytochrome P450 CYP2D6 gene in a European population: characterization of 48 mutations and 53 alleles, their frequencies and evolution, Pharmacogenetics. 7 (1997) 193-202. [28] C. Sachse, J. Brockmoller, S. Bauer, I. Roots, Cytochrome P450 2D6 variants in a Caucasian population: allele frequencies and phenotypic consequences, Am J Hum Genet. 60 (1997) 284-95. [29] R.A. Sansone, L.A. Sansone, Tramadol: seizures, serotonin syndrome, and coadministered antidepressants, Psychiatry (Edgmont). 6 (2009) 17-21.

Page 15 of 23

Accep

ted

Man

uscr

ipt

Figure 1

Figure 1. Mean blood concentrations ± SEM for tramadol (filled circles) and O-desmethyltramadol (open circles) versus time curves after a single oral administration of 50

mg tramadol (n=10) in (a) and 100 mg (n=9) in (b).

Figure 2. Correlation between the mean blood concentration ± SEM for ODT/TRA and time after drug intake for extensive metabolisers, EMs (n=8, filled diamonds),

intermediate metabolisers, IMs (n=9, open diamonds) and a poor metaboliser, PM (n=1, filled triangles).

Figure 3. The self-reported score of drug related symptoms (DRS) in correlation to different single nucleotide polymorphisms in the OPRM1 and ABCB1 gene. The subjects

are grouped by CYP2D6 genotype: extensive metabolisers, EMs (cross), intermediate metabolisers, IMs (open ovals) and poor metabolisers, PMs (line). Subjects receiving 50

mg tramadol (n=10) is shown in (a) and 100 mg (n=9) in (b).

Page 16 of 23

Accep

ted

Man

uscr

ipt

Figure 1

0

50

100

150

200

0 2 4 6 8 10 12

Time (h)

Con

cent

ratio

n (n

g/g

bloo

d)

0

100

200

300

400

0 2 4 6 8 10 12

Time (h)C

once

ntra

tion

(ng/

g bl

ood)

a b

Page 17 of 23

Accep

ted

Man

uscr

ipt

Figure 2

y = 0.0312x + 0.2646R2 = 0.97

y = 0.0215x + 0.1347R2 = 0.96

y = 0.0031x + 0.0709R2 = 0.92

0,0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0 2 4 6 8 10 12

Time (h)

OD

T/TR

A co

ncen

tratio

n ra

tio

EMs

IMs

PM

Page 18 of 23

Accep

ted

Man

uscr

ipt

Figure 3

a

b

Page 19 of 23

Accep

ted

Man

uscr

ipt

Figure 3

Table 1 Pharmacokinetic parameters of tramadol (TRA) and its metabolite O-desmethyltramadol (ODT) after a

single oral administration of 50 and 100 mg tramadol, respectively.

50 mg TRA (n=10) 100 mg TRA (n=9) Pharmacokinetic variables Median Mean Range Median Mean Range

TRA

Cmax (ng/g) 155 156 121-214 334 335 274-399

tmax (min) 149 132 90-181 178 156 87-208

AUC0-10 (ng h/g) 871 900 668-1258 2027 2015 1628-2321

AUC0-∞ (ng h/g) 1066 1229 767-2045 3310 3123 1817-3965

Last detectable time point (min) 602 855 597-1504 1455 1267 596-1485

ODT

Cmax (ng/g) 47 44 0-62 69 68 27-126

tmax (min) 180 170 90-240 182 190 93-300

AUC0-10 (ng h/g) 318 301 0-426 503 479 199-896

AUC0-∞ (ng h/g) 411 410 0-620 834 819 266-1451

Last detectable time point (min) 599 600 597-606 1425 1129 600-1480

Cmax maximum blood concentration; tmax time to reach Cmax; AUC0-10, AUC0-∞ area under the blood concentration-time curve (AUC) from 0 to 10 h, and from 0 to infinity, respectively.

Page 20 of 23

Accep

ted

Man

uscr

ipt

Figure 3

Table 3 Self-reported scores on drug related symptoms during the experimental day. A scale between zero to

five, where zero was no symptoms at all and five was worst imaginable symptoms was used. The subjects were

grouped after dose received; (a) 50 mg tramadol (n=10) and (b) 100 mg tramadol (n=9).

a Subject no.

Symptom #1 #2 #3 #4 #5 #6 #7 #8 #9 #10 Score/symptom Mean SD

Nausea 0 0 0 0 0 0 0 0 0 2 2 0.2 0.6

Dizziness 0 0 0 0 0 1 1 1 0 1 4 0.4 0.5

Headache 0 1 0 0 0 1 1 0 0 0 3 0.3 0.5

Vomiting 0 0 0 0 0 0 0 0 0 0 0 0.0 0.0

Dry mouth 1 0 0 0 1 1 0 2 0 0 5 0.5 0.7

Sweating 0 1 0 0 1 0 0 2 0 1 5 0.5 0.7

Fatigue 1 0 1 0 2 1 3 3 0 1 12 1.2 1.1

Score/subject 2 2 1 0 4 4 5 8 0 5 31 3.1 2.6

b Subject no.

Symptom #11 #12 #13 #14 #15 #16 #17 #18 #19 Score/symptom Mean SD

Nausea 4 0 0 0 2 5 1 0 0 12 1.3 1.9

Dizziness 1 0 2 0 3 5 0 1 0 12 1.3 1.7

Headache 0 0 0 0 2 3 0 0 0 5 0.6 1.1

Vomiting 0 0 0 0 0 4 0 0 0 4 0.4 1.3

Dry mouth 3 0 0 1 3 0 0 2 0 9 1.0 1.3

Sweating 4 0 0 0 0 4 0 0 0 8 0.9 1.8

Fatigue 4 0 1 3 3 4 1 0 0 16 1.8 1.7

Score/subject 16 0 3 4 13 25 2 3 0 66 7.3 8.7

Page 21 of 23

Accep

ted

Man

uscr

ipt

Figure 3

Table 2 Pharmacokinetic parameters stratified by genotype in subjects given (a) 50 mg TRA and (b) subjects given 100 mg.

n number of subjects of each genotype/SNP; Cmax maximum blood concentration; tmax time to reach Cmax;

AUC area under the blood concentration-time curve (AUC) from 0 to 10 h; - only one value, no range can

be given; nd not detected; a ODT was only detected in one of the two subjects, no range can be given.

TRA 50 mg ODT 50 mg

Cmax (ng/g) tmax (min) AUC (ng h/g) Cmax (ng/g) tmax (min)

Median Mean Range Median Mean Range Median Mean Range Median Mean Range Median Mean Range Med

32 139 121-161 148 134 90-152 693 760 668-939 48 48 33-62 149 155 90-210 334

82 175 122-214 150 150 120-181 1059 1024 722-1258 38 40 33-49 182 196 179-240 290

70 170 - 92 92 - 1106 1 106 - nd nd nd nd nd nd nd

20 0.32 0.41 0.21 0.14 0.33

55 156 121-214 149 136 90-181 871 900 668-1258 47 44 33-62 180 173 90-240 318

55 156 121-214 149 136 90-181 871 900 668-1258 47 44 33-62 180 173 90-240 318

32 133 121-149 140 130 90-149 687 715 668-819 48 48 33-62 149 149 90-210 307

64 162 122-198 150 143 120-152 931 960 722-1258 38 40 33-50 181 195 179-240 290

92 192 170-214 137 137 92-181 1151 1151 1106-1195 49 49 a 181 181 a 323

08 0.49 0.05 0.54 0.34 0.86

32 137 121-166 140 131 90-149 708 751 668-922 45 44 33-62 164 160 90-210 299

80 180 161-198 152 152 151-152 1099 1099 939-1258 42 42 33-50 210 210 180-240 300

92 192 170-214 137 137 92-181 1151 1151 1106-1195 49 49 a 181 181 323

06 0.23 0.04 0.73 0.42 0.87

44 144 122-166 135 135 120-149 822 822 722-922 38 38 33-43 181 181 179-182 273

32 133 121-149 140 130 90-149 687 715 668-819 48 48 33-62 149 149 90-210 307

84 186 161-214 152 144 92-181 1151 1125 939-1258 49 44 33-50 181 200 180-240 323

06 0.32 0.03 0.49 0.31 0.67

Page 22 of 23

Accep

ted

Man

uscr

ipt

Figure 3

b TRA 100 mg

Gene Genotype/ Cmax (ng/g) Tmax (min) AUC (ng h/g) Cmax (ng/g)

SNPs (n) Median Mean Range Median Mean Range Median Mean Range Median Mean Range Median

CYP2D6 EM (3) 322 322 310-334 181 179 147-208 1709 1788 1628-2027 86 98 83-126 210

IM (5) 336 337 271-399 178 162 120-181 2116 2093 1862-2321 56 58 27-98 182

PM (1) 366 366 - 119 119 - 2305 2035 - 27 27 - 93

P value 0.62       0.22    0.15       0.12       0.50

OPRM1 AA (7) 334 331 271-377 180 162 119-208 2010 1980 1628-2321 69 69 27-126 182

A118G AG (2) 350 350 301-399 164 164 150-178 2136 2136 2116-2156 63 63 27-98 165

P value 1 0.89 0.5 0.89 0.67

ABCB1 GG (8) 329 335 271-399 164 157 119-181 2063 2013 1628-2321 63 66 27-126 181

G1199A GA (1) 334 334 208 208 2027 2027 83 83 241

P value 1       0.22    1       0.89       0.44

ABCB1 CC (3) 334 337 310-366 147 158 119-208 2027 1987 1628-2305 83 79 27-126 120

C1236T CT (3) 377 359 301-399 178 169 150-180 2156 2198 2116-2321 69 65 27-98 180

TT (3) 322 310 271-336 181 161 120-181 1862 1860 1709-2010 56 60 39-86 210

P value 0.49       0.84    0.15       0.93       0.43

ABCB1 GG (3) 334 340 310-377 180 178 147-208 2027 1992 1628-2321 83 93 69-126 180

G2677T/A GT (2) 350 350 301-399 164 164 150-178 2136 2136 2116-2156 63 63 27-98 165

TT (3) 322 310 271-336 181 161 120-181 1862 1860 1709-2010 56 60 39-86 210

GA (1) 366 366 119 119 2305 2305 27 93

P value 0.72       0.43    0.33       0.42       0.46

ABCB1 CC (3) 366 359 334-377 180 169 119-208 2305 2218 2027-2321 69 60 27-83 180

C3435T CT (3) 310 316 301-336 147 139 120-150 2010 1918 1628-2116 39 64 27-126 120

TT (3) 322 331 271-399 181 180 178-181 1862 1909 1709-2156 86 80 56-98 239

P value 0.39       0.25    0.19       0.58       0.15

Page 23 of 23

Accep

ted

Man

uscr

ipt

Figure 3

Acknowledgement

We would like to thank the study volunteers for their participation and Malin Forsman for her help with the

recruitment of participants. We are grateful to Ingela Jacobsson, Johan Ahlner, Ahmed Omran, Eva Tärning,

Maria Dolores Chermá and Karin Björnström-Karlsson for taking care of participants during the experimental day

and to Ewa Lönn Karlsson and Abdimajid Osman for their help with genotyping analysis. This work was

financially supported by the National Board of Forensic Medicine in Sweden, the Swedish Academy of

Pharmaceutical Sciences (SAPS), the National Board of Health and Welfare in Sweden and the County Council of

Östergötland.