SEARCH

SEARCH BY CITATION

Keywords:

  • Coumarins;
  • cytochrome P450 2C9;
  • International Normalized Ratio;
  • Pharmacogenetics;
  • Vitamin K epoxide reductase multiprotein complex 1

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References
  11. Appendix
  12. Supporting Information

Summary.  Background:  The required acenocoumarol dose and the risk of underanticoagulation and overanticoagulation are associated with the CYP2C9 and VKORC1 genotypes. However, the duration of the effects of these genes on anticoagulation is not yet known.

Objectives:  In the present study, the effects of these polymorphisms on the risk of underanticoagulation and overanticoagulation over time after the start of acenocoumarol were investigated.

Patients/methods:  In three cohorts, we analyzed the relationship between the CYP2C9 and VKORC1 genotypes and the incidence of subtherapeutic or supratherapeutic International Normalized Ratio (INR) values (< 2 and > 3.5) or severe overanticoagulation (INR > 6) for different time periods after treatment initiation.

Results:  Patients with polymorphisms in CYP2C9 and VKORC1 had a higher risk of overanticoagulation (up to 74%) and a lower risk of underanticoagulation (down to 45%) in the first month of treatment with acenocoumarol, but this effect diminished after 1–6 months.

Conclusions:  Knowledge of the patient’s genotype therefore might assist physicians to adjust doses in the first month(s) of therapy.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References
  11. Appendix
  12. Supporting Information

Coumarin derivatives, such as acenocoumarol, phenprocoumon and warfarin, are widely used oral anticoagulants. These drugs are prescribed for the treatment and prevention of thrombo-embolic events in patients with, for example, venous thromboembolism or atrial fibrillation [1]. Because of the narrow therapeutic window of these drugs, patients need to be monitored frequently by measuring the prothrombin time, expressed as the International Normalized Ratio (INR). A large variability in dose-response exists among coumarin users, which is caused by several factors such as age, concomitant medication and diet, but genetic factors also play an important role [2–6]. Approximately one-third of the variation in coumarin dose requirements can be explained by polymorphisms in the CYP2C9 gene, encoding for the main metabolizing enzyme, cytochrome P450 2C9 (CYP2C9), and the VKORC1 gene, encoding for the target enzyme Vitamin K epoxide reductase multiprotein complex 1 (VKORC1) [6–9].

Carriers of a CYP2C9*2 or *3 or a VKORC1 T-allele require a lower coumarin maintenance dose compared with wild-type patients (CYP2C9*1*1, VKORC1 CC) [7,10]. These patients often receive a supratherapeutic dose at the start of treatment, which may lead to overanticoagulation. Lower dose requirements, an increased risk of overanticoagulation in the first month(s) of therapy and delayed stabilization have been shown in carriers of a variant allele in CYP2C9 and/or VKORC1 in several studies [11–17]. Moreover, the risk of hemorrhagic adverse events increases with an increased INR. Severe overanticoagulated patients (INR > 6) therefore have a considerably increased risk of a bleeding event [18,19]. However, when the INR is below the therapeutic range, coumarin therapy is less effective, with a higher risk of (recurrent) thromboembolic events [20]. When physicians are unaware of the genotype of patients, it is conceivable that patients with wild-type genotypes are more often underanticoagulated than variant allele carriers and the latter group has a higher risk for overanticoagulation than the former group.

This difference in dose requirements led to the hypothesis that CYP2C9 and VKORC1 wild-type patients have an increased risk of subtherapeutic INR values and that carriers of a variant allele have an increased risk of supratherapeutic INR values. Meckley et al. [21] showed an increased risk of overanticoagulation in CYP2C9 variant carriers in the first 6 months and in VKORC1 variant carriers in the first month of warfarin treatment. In several European countries, including the Netherlands, acenocoumarol or phenprocoumon are prescribed more frequently for anticoagulant therapy [22]. Schalekamp et al. [15,16] showed an increased risk of severe overanticoagulation in carriers of a CYP2C9 or VKORC1 polymorphism during the first 6 months of acenocoumarol and phenprocoumon treatment. In these previous studies, the first 6 months were not analyzed separately, but as a whole. Teichert et al. [17] showed an increased risk of severe overanticoagulation for VKORC1 variant alleles after an initial standard dose of acenocoumarol treatment.

Whether carriers of a CYP2C9 or VKORC1 polymorphism only have an increased risk of overanticoagulation in the first month of therapy or whether this effect is also seen after the initiation period is still unknown, as well as the possible risk of underanticoagulation in wild-type patients. The aim of the present study was therefore to investigate the association of the CYP2C9 and VKORC1 polymorphisms with the risk of over- and underanticoagulation after the initiation period of acenocoumarol.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References
  11. Appendix
  12. Supporting Information

Study population

For the present study, we looked at data from three different studies. First, data from the pre-EU-PACT study were used [23]. In this cohort study, patients who were using acenocoumarol in November 2009 with a target International Normalized Ratio (INR) in the lowest intensity category (according to Dutch guidelines INR 2.0–3.5) were included. Schalekamp prospectively followed patients newly starting on acenocoumarol with a target INR in the lowest intensity category (2.0–3.5) for 6 months [15]. In this dataset we therefore could repeat the analyzes for the first half year of treatment. In the Rotterdam study, patients on acenocoumarol were followed for their entire treatment period regardless of their target INR [24]. We selected only the patients with a target INR of 2.0 to 3.5 for the present analyzes.

The study protocols of the three studies were approved by the Medical Ethics Committee (Leiden University Medical Center, Leiden for pre-EU-PACT, Utrecht Medical Centre, Utrecht for the study of Schalekamp, and Erasmus Medical Center, Rotterdam for the Rotterdam study) and patients provided informed consent before inclusion into the study. All procedures were conducted in accordance with the Helsinki Declaration.

The data of the three studies were combined to increase the power of the analyzes. For the first 6 months, this dataset contains data from all three studies, but for the periods after 6 months only data from Pre-EU-PACT and the Rotterdam study were included.

Data collection

For each participating patient, data on age and gender were obtained from the electronic registry databases of the anticoagulation clinics. INR measurements, prescribed doses and relevant co-medication have been routinely collected and recorded in registry databases at each visit to the anticoagulation clinic in the Netherlands since 1983. Therefore, it was possible to obtain this information for each patient. Patients were genotyped for CYP2C9*2 (rs1799853), CYP2C9*3 (rs1057910) and VKORC1 1173C>T (rs9934438). In the Pre-EU-PACT study and the Rotterdam study, also data on height and weight were available. More details on the three studies can be found elsewhere [15,17,23].

Statistical analyzes

As the target INR for the participants was in the lowest intensity category (INR 2.0–3.5), subtherapeutic INR values were defined as INR < 2 and supratherapeutic INR values as INR > 3.5. As the risk of bleeding events is considerably increased in severe overanticoagulated patients, we also investigated the occurrence of INR values > 6 [18,19]. In several time periods up to one and a half years after treatment initiation, the occurrence of at least one INR < 2, > 3.5 or > 6.0 was studied. When patients reach a stable dose, they require less frequent monitoring than before they were stable. Consequently, the mean number of INR measurements decreases over time. The time periods used in the present study were chosen so as to have a sufficient number of INR measurements in every period. Data from a patient were only included in the analysis of a specific period, if the patient was using the coumarin under study for this entire period. Data on the start of treatment were required for all included patients and follow-up started at the first week of acenocoumarol use.

The difference in risk of at least one INR < 2, > 3.5 or > 6.0 between the different CYP2C9 and VKORC1 genotypes was tested with a chi-square test. If the expected number of observations in a cell was below 5, Fisher’s exact test was used. Patients with missing data were excluded from the analyzes where this data was needed. Because the frequency of homozygous carriers of a variant CYP2C9 allele is low, the CYP2C9 genotype was grouped to increase the group sizes for our analyzes. We grouped the *2 carriers together (*1*2 and *2*2) and the *3 carriers together (*1*3, *2*3 and *3*3). The combined effect of CYP2C9 and VKORC1 was also investigated for the first month of acenocoumarol use, by combining the two genotypes in six groups, with every VKORC1 genotype divided into CYP2C9 wild-type patients and CYP2C9 variant carriers.

In addition to the occurrence of at least one out-of-range INR during the different periods, we also looked at the time within, below and above the therapeutic range. This method is more robust for the difference in number of INR measurements between patients. All analyzes were performed using SPSS 16.0 (SPSS Inc., Chicago, IL, USA.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References
  11. Appendix
  12. Supporting Information

Patient characteristics

In total, 1586 acenocoumarol users from the three studies [15,23,25] were eligible for analyzes in the present study. A flowchart of the selection of patients can be found in the Supplement (Figs S1A–C). From the 471 acenocoumarol users in the pre-EU-PACT cohort [23], 231 in the study of Schalekamp [15] and 2065 in the Rotterdam study [25], 275, 192 and 1119 patients were included in the present study. Patient characteristics and genotypes of all 1586 patients are shown in Table 1. Characteristics per study can be found in the supplement (Tables S1–S3). Data on height and weight were only available in two studies (Pre-EU-PACT and the Rotterdam study). Mean age stratified by genotype varied between 73 and 75 years and 38% to 42% of the participants were male. The most frequent indication for acenocoumarol treatment was atrial fibrillation. The treatment duration at the moment of data collection ranged from 0 to 189 months.

Table 1.   Characteristics of included acenocoumarol users
Acenocoumarol (n = 1586)VKORC1 CYP2C9 
CCCTTTWild-type*2 carriers*3 carriers
  1. INR, International Normalized Ratio. Bold values indicate statistical significance (P < 0.05).

Patients in study populationn = 499n = 696n = 211missing: 180n = 938n = 312n = 170missing: 166
Male199 (40%)286 (41%)87 (41%)P = 0.90388 (41%)118 (38%)72 (42%)P = 0.49
Mean age, years (median)74 (75)75 (76)73 (75)P = 0.0675 (76)75 (76)73 (74)P = 0.15
Mean height, cm (median)168 (167)168 (167)168 (168)P = 0.93168 (167)168 (168)169 (168)P = 0.30
Mean weight, kg (median)76 (74)76 (75)76 (76)P = 0.9576 (75)76 (74)77 (78)P = 0.43
Mean treatment duration, months (median)15.5 (5.8)17.4 (5.6)16.6 (5.3)P = 0.6516.6 (5.5)15.2 (5.2)17.3 (6.1)P = 0.38
Amiodaron users16 (3%)13 (2%)3 (1%)P = 0.2120 (2%)7 (2%)6 (4%)P = 0.54
Indication for coumarin therapy
 Atrial fibrillation243 (49%)332 (48%)97 (46%)P = 0.80445 (47%)141 (45%)88 (52%)P = 0.39
 Venous thromboembolism81 (16%)97 (14%)40 (19%)P = 0.18136 (15%)52 (17%)33 (19%)P = 0.22
 Other174 (35%)267 (38%)74 (35%)P = 0.41357 (38%)118 (38%)49 (29%)P = 0.07
Number of INR measurements
 0–1 month (n = 1484)5.225.105.00P = 0.135.145.145.06P = 0.64
 1–3 months (n = 1028)5.034.804.59P = 0.034.884.774.83P = 0.79
 3–6 months (n = 786)5.985.475.74P = 0.0045.775.505.56p = 0.18
 6–9 months (n = 541)5.495.505.35P = 0.865.355.685.71P = 0.10
 9–12 months (n = 489)5.345.395.81P = 0.335.325.595.82P = 0.17
 12–15 months (n = 443)5.304.955.14P = 0.535.085.294.71P = 0.27
 15–18 months (n = 414)5.315.355.37P = 0.895.355.415.19P = 0.95

A VKORC1 CC genotype was seen in 499 patients, a CT genotype in 696 patients and 211 patients had a TT genotype. When the CYP2C9 genotype was grouped as wild type, *2 carriers and *3 carriers, the group sizes were 938, 312 and 170, respectively. All genotypes were in Hardy–Weinberg equilibrium.

Time periods

The following time periods were considered to ensure a sufficient number of INR measurements in every period: 0–1 month (day 1–30), 1–3 months (day 31–90), 3–6 months (day 91–180), 6–9 months (day 181–270), 9–12 months (day 271–360), 12–15 months (day 361–450) and 15–18 months (day 451–540) after treatment initiation.

The average number of INR measurements ranged from 4.6 to 6 for all periods. The INR of carriers of a VKORC1 variant allele was measured less frequently during months 1–3 and 3–6 than the INR of wild-type patients (P = 0.03 and P = 0.004, respectively, see Table 1).

Because the treatment duration at the moment of data collection was different among the patients, the number of patients decreased over time to 414 acenocoumarol users in the last period (15–18 months, Table 1). Only data of patients using acenocoumarol during the entire period were included in the analysis of a specific period. The maximum follow-up in the study of Schalekamp study was 6 months. The distribution of the different genotypes remained similar in the different time periods.

Subtherapeutic INR values

Figure 1 depicts the occurrence of at least one INR < 2 over the different periods in acenocoumarol users with the different VKORC1 and CYP2C9 genotypes. Figure 1A illustrates that during the first month, 73% (95% confidence interval [CI]: 68%–77%) of acenocoumarol users with a VKORC1-CC genotype had at least one INR measurement < 2. This number was significantly lower among patients with a CT genotype (62%, P < 0.001, 95% CI: 58%–66%) and with a TT genotype (45%, P < 0.001, 95% CI: 38%–52%). During months 2–3, the risk of a subtherapeutic INR was 49% (95% CI: 43%–54%) in CC patients, vs. 40% (95% CI: 36%–45%, P = 0.01) and 39% (95% CI: 30%–48%, P = 0.05) in CT and TT patients, respectively. Differences between the VKORC1 genotypes were not significant after the third month of acenocoumarol use. CYP2C9 wild-type patients had a 65% (95% CI: 62%–68%) risk of an INR < 2 in the first month, vs. 64% (95% CI: 58%–69%, P = 0.27) and 54% (95% CI: 46%–62%, P = 0.005) in *2 and *3 carriers, respectively. After the first month, no significant differences were found for CYP2C9 (Fig. 1B).

image

Figure 1.  Percentage of patients (and 95% confidence intervals [CI]) with at least one International Normalized Ratio (INR) < 2 in the different time periods after coumarin initiation. A: VKORC1 genotypes B: CYP2C9 genotypes.

Download figure to PowerPoint

Supratherapeutic INR values

Figure 2 displays the occurrence of at least one INR > 3.5 over the different periods in acenocoumarol users with the different VKORC1 and CYP2C9 genotypes. A significant difference between the VKORC1 genotypes was found up to the third month (Fig. 2A). In the first month, supratherapeutic INR values occurred in 30% (95% CI: 26%–35%) of the wild-type patients, vs. 45% (95% CI: 41%–49%, P < 0.001) and 74% (95% CI: 67%–80%, P < 0.001) in patients with a CT and TT genotype, respectively. This difference was smaller (40% (95% CI: 34%–45%)) for wild-type patients, vs. 43% (95% CI: 38%–47%, P = 0.37) for CT and 62% (95% CI: 53%–70%, P < 0.001) for TT in months 1–3. CYP2C9 wild-type patients had a 41% (95% CI: 38%–44%) risk of an INR > 3.5 in the first month, vs. 50% (95% CI: 44%–56%, P = 0.008) and 51% (95% CI: 43%–59%, P = 0.01) in *2 and *3 carriers, respectively. No significant differences were found between the CYP2C9 genotypes of acenocoumarol users after the first month (Fig. 2B).

image

Figure 2.  Percentage of patients (and 95% confidence intervals [CI]) with at least one International Normalized Ratio INR > 3.5 in the different time periods after coumarin initiation. A: VKORC1 genotypes B: CYP2C9 genotypes.

Download figure to PowerPoint

Severe overanticoagulation

The risk of severe overanticoagulation (INR > 6) in acenocoumarol users with the different VKORC1 and CYP2C9 genotypes is shown in Fig. 3 over the different periods. In the first month, 3% (95% CI: 2%–5%) of the VKORC1 wild-type patients had an INR > 6, vs. 5% (95% CI: 3%–7%, P = 0.26) and 12% (95% CI: 8%–17%, P < 0.001) in CT and TT patients, respectively. In all, 4% (95% CI: 3%–6%) of the CYP2C9 wild-type patients had an INR > 6 in the first month, vs. 7% (95% CI: 4%–10%, P = 0.07) of the *2 carriers and 9% (95% CI: 5%–14%, P = 0.01) of the *3 carriers. No significant differences were found between the VKORC1 and CYP2C9 genotypes after the first month (Fig. 3A,B).

image

Figure 3.  Percentage of patients (and 95% confidence intervals [CI]) with at least one International Normalized Ratio INR > 6 in the different time periods after coumarin initiation. A: VKORC1 genotypes B: CYP2C9 genotypes.

Download figure to PowerPoint

VKORC1 and CYP2C9 genotypes combined

Figure 4 shows the risk of under- and overanticoagulation in the first month for the different combined genotype groups. The risk of underdosing in the first month was greatest (75%, 95% CI: 69%–79%) in VKORC1 and CYP2C9 wild-type patients. This risk decreased for every variant allele, with the lowest risk (44%, 95% CI: 35%–52%) existing in VKORC1 TT and CYP2C9 wild-type patients. The risk of overdosing increased as the number of variant alleles increased (28% [95% CI: 23%–33%] in VKORC1 and CYP2C9 wild-type patients to 76% [95% CI: 67%–83%] in VKORC1 TT and CYP2C9 wild-type patients). Severe overanticoagulation in the first month of acenocoumarol use was relatively rare in VKORC1 CC patients and in VKORC CT/CYP2C9 wild-type patients (2%–4%) and higher in VKORC1 CT/CYP2C9 variant carriers and VKORC1 TT patients (9%–16%).

image

Figure 4.  Risk of under- or overanticoagulation during the first month of coumarin treatment and combined VKORC1/ CYP2C9 genotypes.

Download figure to PowerPoint

Analyzes per subcohort

The present analyzes of the combined dataset were also performed in the three separate studies. The results of these analyzes were similar to the results in the combined dataset and can be found in the Supplement. Only some small differences in results were seen. In the pre-EU-PACT dataset, the risk of underanticoagulation was higher than in the other two datasets (Figs S2 and S3). In this dataset, a significantly different risk of an INR > 3.5 between the VKORC1 genotypes could be demonstrated up to month 6 (Fig. S4). The risk of severe overanticoagulation was relatively low (mostly below 10%) and the confidence intervals in the datasets of pre-EU-PACT and Schalekamp in these analyzes were large. Therefore only in the Rotterdam dataset could the effect of VKORC1 on the occurrence of an INR > 6 be demonstrated, although this trend was also seen in the other datasets (Fig. S6).

The results of the analyzes on time within, below and above the therapeutic INR range were very similar to the results described above. In the first month, time below the therapeutic INR range was highest in VKORC1 and CYP2C9 wild-type patients (up to 31%) and the time above therapeutic INR range was highest in VKORC1 TT and CYP2C9*3 carriers (up to 30%). For CYP2C9 no difference was found after the first month, but the effect of VKORC1 on time below the therapeutic INR range lasted up to months 1–3 and on time above the therapeutic INR range up to months 3–6.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References
  11. Appendix
  12. Supporting Information

The present study demonstrates that in the first month of acenocoumarol therapy, the risk of underdosing is highest in patients with a VKORC1 wild type. This increased risk of a subtherapeutic INR was also seen in months 2 and 3, but not after the third month of coumarin treatment. In addition, the risk of overdosing was highest in patients with a VKORC1 TT genotype in the first 6 months. For severe overanticoagulation an effect of VKORC1 was only seen in the first month. After the sixth month, no effect of polymorphisms in VKORC1 on the occurrence of out-of-range INRs was found.

The effect of the CYP2C9 genotype on under or overdosing was smaller than the effect of VKORC1. Only when we combined three datasets together was a significant difference in the occurrence of subtherapeutic INR values, supratherapeutic values and severe overanticoagulation found between the wild-type patients and the *2 and *3 carriers. However, this effect was only found in the first month of therapy and not after the initiation period.

An increased risk of overanticoagulation among warfarin users with a CYP2C9 or VKORC1 polymorphism during the initiation period was also found in the previous study of Limdi et al. [26]. They found that patients with a variant allele had a higher risk of an INR above 4 during the first 30 days. In the present study, we found a larger influence of VKORC1 than of CYP2C9. This difference was also seen in the previous study of Schwarz et al., [27] who demonstrated that the initial variability in INR response to warfarin was more strongly associated with VKORC1 than with CYP2C9. However, none of these studies investigated the effect of genetic variation on the risk of overanticoagulation after the first month. We found no effect of CYP2C9 after the first month, but for VKORC1 we demonstrated an increased risk of overanticoagulation among variant carriers up to 6 months after treatment initiation.

The aforementioned studies focused on the risk of (severe) overanticoagulation, but not on the risk of underanticoagulation. We found an increased risk of a subtherapeutic INR in CYP2C9 wild-type patients during the first month and in VKORC1 wild-type patients during the first 3 months. If physicians fear to prescribe high doses, because of uncertainty whether the patient is sensitive or not, frequent underdosing of VKORC1 and CYP2C9 wild-type patients is to be expected. For these patients, the standard dose is often not high enough. The dose will then be adjusted after a couple of INR measurements, mainly during the first weeks of coumarin therapy. Our results thus correspond with these expectations. Patients with venous thromboembolism are often also treated with a low-molecular-weight heparin (LMWH) during the first days, until an INR > 2 is obtained. For these patients the risk of a subtherapeutic INR is compensated by this LMWH during this period. This is not the case for patients with other indications, as for example atrial fibrillation.

Because the Pre-EU-PACT study selected patients using acenocoumarol in November 2009, patients in this cohort did not all have the same length of follow-up. At the time of data collection, some patients were already using acenocoumarol for years, whereas others had just started using acenocoumarol. This is a limitation to the present study, not only because of the lower numbers of patients in the later periods, but also because we might have missed very unstable patients. These patients often stop using acenocoumarol early and therefore might be underrepresented in this cohort. However, we do not believe this had a large influence on our results because the results in the pre-EU-PACT study were very similar to the results in the other studies and the follow-up time was not different among the different genotypes (Table 1). We also performed a survival analysis using the prospective data of Schalekamp, and we found no differences in loss to follow-up between the genotypes. Of the 192 patients included from this cohort, four patients (2%) stopped within the first month.

Patients in the three Dutch cohorts were treated with a therapeutic INR range of 2.0–3.5. This is standard care in the Netherlands, but differs from other countries where normally a range of 2.0–3.0 is used. In the present study, we therefore defined supratherapeutic INR values as INR > 3.5. As we used this as a marker of instability and we obtained similar results in our analysis on time above therapeutic range, we believe our results are also relevant for other countries.

A difference in risk of out-of-range INRs was found between the three studies. The data have been collected in different clinics, and as clinics perform differently, this can explain this variation. However, the effect of the different genotypes and the trend over time remains similar across the different clinics/datasets. Because data from different clinics were used, we believe our study population reflects the Dutch population well.

The likelihood of a patient having an out-of-range INR value depends on how often the INR is checked. This could influence the results when we use occurrence of at least one INR below or above a certain value. In the Netherlands, patients are monitored frequently (on average 21 times per year), especially in the first year. However, we also studied the effect of the genotypes on the percentage time within, below and above the therapeutic INR range, and these analyzes yielded similar results. Using this metric (% of time), the results are relatively robust for the frequency of INR monitoring.

Although we did not find an effect of being a carrier of CYP2C9 or VKORC1 polymorphisms after the sixth month of therapy, for VKORC1 we did find an effect on subtherapeutic and supratherapeutic INR values after the first month. This could mean that knowledge of the patient’s genotype could help to determine dose adjustments for acenocoumarol users. This might be useful if the patient has an out-of-range INR. In this case, carriers of a variant allele could be treated with smaller dose adjustments than wild-type patients. This would not only be useful for the sensitive patients to prevent supratherapeutic INR values and thereby decrease the risk of bleeding, but also for wild-type patients. In the present study, we have shown an increased risk of underdosing in wild-type patients, which exposes them to an increased risk of thromboembolic events. Oake et al. [28] investigated the risk of adverse events in different INR ranges and demonstrated that although the risk of bleeding or thromboembolic events was lowest with an INR between two and three, an INR just above three was safer (less events) than an INR below 2. The results from the present study suggest that sensitive patients could be treated with smaller dose adjustments, thereby decreasing their bleeding risk, and that wild-type patients could be treated with larger dose adjustments to decrease the time below therapeutic INR range, thereby decreasing their risk of thromboembolic events. This knowledge is especially useful in the first months of therapy, as in the months thereafter the physician more often uses the previous INRs and doses of a patient to determine the magnitude of dose adjustments.

In summary, the novel finding of the present study is that acenocoumarol users with the CYP2C9 and VKORC1 wild type have an increased risk of underanticoagulation in the first period of therapy. This suggests that pre-treatment genotyping could not only be useful in preventing overanticoagulation in the limited group of carriers of a CYP2C9 and VKORC1 polymorphism, but also to prevent underanticoagulation in the larger group of patients without a CYP2C9 and VKORC1 polymorphism. It has been suggested that pre-treatment genotyping could identify patients requiring a lower or higher coumarin dose, and thereby, reduce the risk of over anticoagulation or underanticoagulation in variant carriers and wild-type patients. Currently, the effectiveness and cost-effectiveness of a genotype-guided dosing regimen is being investigated in clinical trials [29–31]. If the genotype of a patient is known, this might help to prevent subtherapeutic or supratherapeutic INRs in the first months of therapy and thereby reduce the risk of adverse events. The trade-off between the health gained through this risk reduction and the extra costs of genotyping should be investigated in a cost-effectiveness analysis. As the costs of a genetic test are still decreasing, we believe that genotyping could be an attractive option in the future [32].

Addendum

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References
  11. Appendix
  12. Supporting Information

Contributions of the authors of this manuscript were as follows. T.I Verhoef was involved in the conception and design of the study, acquisition of data, analysis and interpretation of data, drafting of the manuscript and statistical analysis. W.K. Redekop was involved in conception and design of the study, analysis and interpretation of data, critical revision of the manuscript and supervision. M.M. Buikema was involved in analysis and interpretation of data, drafting of the manuscript and statistical analysis. T. Schalekamp was involved in conception and design of the study, acquisition of data and critical revision of the manuscript. F.J.M. van der Meer was involved in conception and design of the study, acquisition of data and critical revision of the manuscript. S. le Cessie was involved in analysis and interpretation of data and critical revision of the manuscript. J.A.M. Wessels was involved in acquisition of data and critical revision of the manuscript. R.M.F. van Schie was involved in conception and design of the study, acquisition of data and critical revision of the manuscript. A. de Boer was involved in conception and design of the study, analysis and interpretation of data, critical revision of the manuscript and supervision. M. Teichert was involved in acquisition of data, and critical revision of the manuscript. L.E. Visser was involved in conception and design of the study, acquisition of data and critical revision of the manuscript. A.H. Maitland-van der Zee was involved in conception and design of the study, analysis and interpretation of data, critical revision of the manuscript, obtaining funding and supervision.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References
  11. Appendix
  12. Supporting Information

We would like to thank the Anticoagulation Clinic Medial, E. van Meegen, A. Hofman, P. Buhre, S. van der Meer and J. Berbee for their support during the data collection period.

Disclosure of Conflict of Interest

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References
  11. Appendix
  12. Supporting Information

This project is funded by the European Community’s Seventh Framework Programme under grant agreement HEALTH-F2-2009-223062. The division of Pharmacoepidemiology & Clinical Pharmacology employing authors T.I. Verhoef, T. Schalekamp, R.M.F. van Schie, A. de Boer and A.-H. Maitland-van der Zee, has received unrestricted funding for pharmacoepidemiological research from GlaxoSmithKline, Novo Nordisk, the private–public funded Top Institute Pharma, which includes cofunding from universities, government and industry), the Dutch Medicines Evaluation Board and the Dutch Ministry of Health. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed. No writing assistance was utilized in the production of this manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References
  11. Appendix
  12. Supporting Information
  • 1
    Ansell J, Hirsh J, Hylek E, Jacobson A, Crowther M, Palareti G. Pharmacology and management of the vitamin K antagonists: American college of chest physicians evidence-based clinical practice guidelines (8th edition). Chest 2008; 133(Suppl. 6): 160S98S.
  • 2
    Custodio das Dores SM, Booth SL, Martini LA, de Carvalho Gouvea VH, Padovani CR, de Abreu Maffei FH, Campana AO, Rupp de Paiva SA. Relationship between diet and anticoagulant response to warfarin: a factor analysis. Eur J Nutr 2007; 46: 14754.
  • 3
    Harder S, Thurmann P. Clinically important drug interactions with anticoagulants. an update. Clin Pharmacokinet 1996; 30: 41644.
  • 4
    Torn M, Bollen WL, van der Meer FJ, van der Wall EE, Rosendaal FR. Risks of oral anticoagulant therapy with increasing age. Arch Intern Med 2005; 165: 152732.
  • 5
    Schalekamp T, de Boer A. Pharmacogenetics of oral anticoagulant therapy. Curr Pharm Des 2010; 16: 187203.
  • 6
    Bodin L, Verstuyft C, Tregouet DA, Robert A, Dubert L, Funck-Brentano C, Jaillon P, Beaune P, Laurent-Puig P, Becquemont L, Loriot MA. Cytochrome P450 2C9 (CYP2C9) and vitamin K epoxide reductase (VKORC1) genotypes as determinants of acenocoumarol sensitivity. Blood 2005; 106: 13540.
  • 7
    Wadelius M, Chen LY, Lindh JD, Eriksson N, Ghori MJ, Bumpstead S, Holm L, McGinnis R, Rane A, Deloukas P. The largest prospective warfarin-treated cohort supports genetic forecasting. Blood 2009; 113: 78492.
  • 8
    Teichert M, Eijgelsheim M, Uitterlinden AG, Buhre PN, Hofman A, De Smet PA, Visser LE, Stricker BH. Dependency of phenprocoumon dosage on polymorphisms in the VKORC1, CYP2C9, and CYP4F2 genes. Pharmacogenet Genomics 2011; 21: 2634.
  • 9
    Teichert M, Eijgelsheim M, Rivadeneira F, Uitterlinden AG, van Schaik RH, Hofman A, De Smet PA, van Gelder T, Visser LE, Stricker BH. A genome-wide association study of acenocoumarol maintenance dosage. Hum Mol Genet 2009; 18: 375868.
  • 10
    Gage BF, Eby C, Milligan PE, Banet GA, Duncan JR, McLeod HL. Use of pharmacogenetics and clinical factors to predict the maintenance dose of warfarin. Thromb Haemost 2004; 91: 8794.
  • 11
    Lindh JD, Lundgren S, Holm L, Alfredsson L, Rane A. Several-fold increase in risk of overanticoagulation by CYP2C9 mutations. Clin Pharmacol Ther 2005; 78: 54050.
  • 12
    Aithal GP, Day CP, Kesteven PJ, Daly AK. Association of polymorphisms in the cytochrome P450 CYP2C9 with warfarin dose requirement and risk of bleeding complications. Lancet 1999; 353: 7179.
  • 13
    D’Andrea G, D’Ambrosio RL, Di Perna P, Chetta M, Santacroce R, Brancaccio V, Grandone E, Margaglione M. A polymorphism in the VKORC1 gene is associated with an interindividual variability in the dose-anticoagulant effect of warfarin. Blood 2005; 105: 6459.
  • 14
    Reitsma PH, van der Heijden JF, Groot AP, Rosendaal FR, Buller HR. A C1173T dimorphism in the VKORC1 gene determines coumarin sensitivity and bleeding risk. PLoS Med 2005; 2: e312.
  • 15
    Schalekamp T, Brasse BP, Roijers JF, Chahid Y, van Geest-Daalderop JH, de Vries-Goldschmeding H, van Wijk EM, Egberts AC, de Boer A. VKORC1 and CYP2C9 genotypes and acenocoumarol anticoagulation status: interaction between both genotypes affects overanticoagulation. Clin Pharmacol Ther 2006; 80: 1322.
  • 16
    Schalekamp T, Brasse BP, Roijers JF, van Meegen E, van der Meer FJ, van Wijk EM, Egberts AC, de Boer A. VKORC1 and CYP2C9 genotypes and phenprocoumon anticoagulation status: interaction between both genotypes affects dose requirement. Clin Pharmacol Ther 2007; 81: 18593.
  • 17
    Teichert M, van Schaik RH, Hofman A, Uitterlinden AG, de Smet PA, Stricker BH, Visser LE. Genotypes associated with reduced activity of VKORC1 and CYP2C9 and their modification of acenocoumarol anticoagulation during the initial treatment period. Clin Pharmacol Ther 2009; 85: 37986.
  • 18
    van der Meer FJ, Rosendaal FR, Vandenbroucke JP, Briet E. Assessment of a bleeding risk index in two cohorts of patients treated with oral anticoagulants. Thromb Haemost 1996; 76: 126.
  • 19
    Hylek EM, Chang YC, Skates SJ, Hughes RA, Singer DE. Prospective study of the outcomes of ambulatory patients with excessive warfarin anticoagulation. Arch Intern Med 2000; 160: 16127.
  • 20
    Merli GJ, Tzanis G. Warfarin: what are the clinical implications of an out-of-range-therapeutic international normalized ratio? J Thromb Thrombolysis 2009; 27: 2939.
  • 21
    Meckley LM, Wittkowsky AK, Rieder MJ, Rettie AE, Veenstra DL. An analysis of the relative effects of VKORC1 and CYP2C9 variants on anticoagulation related outcomes in warfarin-treated patients. Thromb Haemost 2008; 100: 22939.
  • 22
    Pengo V, Pegoraro C, Cucchini U, Iliceto S. Worldwide management of oral anticoagulant therapy: the ISAM study. J Thromb Thrombolysis 2006; 21: 737.
  • 23
    van Schie RM, Wessels JA, le Cessie S, de Boer A, Schalekamp T, van der Meer FJ, Verhoef TI, van Meegen E, Rosendaal FR, Maitland-van der Zee AH, for the EU-PACT Study Group. Loading and maintenance dose algorithms for phenprocoumon and acenocoumarol using patient characteristics and pharmacogenetic data. Eur Heart J 2011; 32: 190917.
  • 24
    Hofman A, Breteler MM, van Duijn CM, Krestin GP, Pols HA, Stricker BH, Tiemeier H, Uitterlinden AG, Vingerling JR, Witteman JC. The rotterdam study: objectives and design update. Eur J Epidemiol 2007; 22: 81929.
  • 25
    Hofman A, van Duijn CM, Franco OH, Ikram MA, Janssen HL, Klaver CC, Kuipers EJ, Nijsten TE, Stricker BH, Tiemeier H, Uitterlinden AG, Vernooij MW, Witteman JC. The rotterdam study: 2012 objectives and design update. Eur J Epidemiol 2012; 8: 65786.
  • 26
    Limdi NA, Wiener H, Goldstein JA, Acton RT, Beasley TM. Influence of CYP2C9 and VKORC1 on warfarin response during initiation of therapy. Blood Cells Mol Dis 2009; 43: 11928.
  • 27
    Schwarz UI, Ritchie MD, Bradford Y, Li C, Dudek SM, Frye-Anderson A, Kim RB, Roden DM, Stein CM. Genetic determinants of response to warfarin during initial anticoagulation. N Engl J Med 2008; 358: 9991008.
  • 28
    Oake N, Jennings A, Forster AJ, Fergusson D, Doucette S, van Walraven C. Anticoagulation intensity and outcomes among patients prescribed oral anticoagulant therapy: a systematic review and meta-analysis. CMAJ 2008; 179: 23544.
  • 29
    van Schie RM, Wadelius MI, Kamali F, Daly AK, Manolopoulos VG, de Boer A, Barallon R, Verhoef TI, Kirchheiner J, Haschke-Becher E, Briz M, Rosendaal FR, Redekop WK, Pirmohamed M, van der Zee AH. Genotype-guided dosing of coumarin derivatives: the european pharmacogenetics of anticoagulant therapy (EU-PACT) trial design. Pharmacogenomics 2009; 10: 168795.
  • 30
    Kangelaris KN, Bent S, Nussbaum RL, Garcia DA, Tice JA. Genetic testing before anticoagulation? A systematic review of pharmacogenetic dosing of warfarin. J Gen Intern Med 2009; 24: 65664.
  • 31
    French B, Joo J, Geller NL, Kimmel SE, Rosenberg Y, Anderson JL, Gage BF, Johnson JA, Ellenberg JH, COAG (Clarification of Optimal Anticoagulation through Genetics) Investigators. Statistical design of personalized medicine interventions: the clarification of optimal anticoagulation through genetics (COAG) trial. Trials 2010; 11: 108108.
  • 32
    Verhoef TI, Redekop WK, Darba J, Geitona M, Hughes DA, Siebert U, de Boer A, Maitland-van der Zee AH, Barallon R, Briz M, Daly A, Haschke-Becher E, Kamali F, Kirchheiner J, Manolopoulos VG, Pirmohamed M, Rosendaal FR, van Schie RM, Wadelius M, EU-PACT Group. A systematic review of cost-effectiveness analyses of pharmacogenetic-guided dosing in treatment with coumarin derivatives. Pharmacogenomics 2010; 11: 9891002.

Appendix

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References
  11. Appendix
  12. Supporting Information

The members of the EU-PACT group are:

R. Barallon, LGC Limited, Middlesex, UK.

A. de Boer, Utrecht University, Utrecht, the Netherlands.

A. Daly, Newcastle University, Newcastle upon Tyne, UK.

E. Haschke-Becher, Elisabethinen Hospital Linz, Linz, Austria.

F. Kamali, Newcastle University, Newcastle upon Tyne, UK.

A.-H. Maitland, Utrecht University, Utrecht, the Netherlands.

K. Redekop, Erasmus University, Rotterdam, the Netherlands.

J. Stingl, Humboldt University of Berlin, Berlin, Germany.

V.G. Manolopoulos, Democritus University of Thrace, Alexandroupolis, Greece.

M. Pirmohamed, University of Liverpool, Liverpool, UK.

F.R. Rosendaal, Leiden University Medical Center, Leiden, the Netherlands.

M. Wadelius, Uppsala University, Uppsala, Sweden.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References
  11. Appendix
  12. Supporting Information

Figure S1. (A) Selection of patients from Pre-EU-PACT. (B) Selection of patients from the study of Schalekamp. (C) Selection of patients from the Rotterdam study.

Figure S2. Percentage of patients (and 95% confidence intervals) with at least one International Normalized Ratio (INR) < 2 in the different time periods after coumarin initiation – VKORC1 genotypes. (A) Pre-EU-PACT, (B) Schalekamp, (C) Rotterdam.

Figure S3. Percentage of patients (and 95% confidence) with at least one International Normalized Ratio (INR) < 2 in the different time periods after coumarin initiation – CYP2C9 genotypes. (A) Pre-EU-PACT, (B) Schalekamp, (C) Rotter- dam.

Figure S4. Percentage of patients (and 95% confidence intervals) with at least one International Normalized Ratio (INR) > 3.5 in the different time periods after coumarin initiation – VKORC1 genotypes. (A) Pre-EU-PACT, (B) Schalekamp, (C) Rotterdam.

Figure S5. Percentage of patients (and 95% confidence intervals) with at least one International Normalized Ratio (INR) > 3.5 in the different time periods after coumarin initiation – CYP2C9 genotypes. (A) Pre-EU-PACT, (B) Schalekamp, (C) Rotterdam.

Figure S6. Percentage of patients (and 95% confidence intervals) with at least one International Normalized Ratio (INR) > 6 in the different time periods after coumarin initiation – VKORC1 genotypes. (A) Pre-EU-PACT, (B) Schalekamp, (C) Rotterdam.

Figure S7. Percentage of patients (and 95% confidence intervals) with at least one INR > 6 in the different time periods after coumarin initiation – CYP2C9 genotypes. (A) Pre-EU-PACT, (B) Schalekamp, (C) Rotterdam.

Table S1. Characteristics of included acenocoumarol users in Pre-EU-PACT.

Table S2. Characteristics of included acenocoumarol users in Schalekamp.

FilenameFormatSizeDescription
JTH_4633_sm_FigS1-S7-TableS1-S3.pdf119KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.