IL1B and VWF Variants Are Associated With Fibrinolytic Early Recanalization in Patients With Ischemic Stroke
Background and Purpose—There is a great interindividual variability among patients with acute ischemic stroke regarding the response to intravenous tissue-type plasminogen activator treatment. The aim of this study was to identify genetic variants associated with recanalization, and thus treatment efficacy, after tissue-type plasminogen activator administration.
Methods—A total of 140 single nucleotide polymorphisms from 97 candidate genes were successfully genotyped by SNPlex in 2 cohorts, accounting for 497 prospectively recruited tissue-type plasminogen activator-treated patients, of whom 33% recanalized during tissue-type plasminogen activator infusion. Functional studies were then performed, including assessment of interleukin 1B mRNA levels and von Willebrand factor, FIII, FVII, FVIII, and FX protein activity.
Results—After replication, the following single nucleotide polymorphisms were associated with early recanalization: rs1143627 in IL1B gene (CC: 53.1% of recanalization, A-carriers: 32.7%; P=0.022; replication cohort: P=0.046), rs16944 in IL1B gene (AA: 50% of recanalization, G-carriers: 32%; P=0.038; replication cohort: P=0.049), and rs1063856 in the vWF gene (GG: 53.8% of recanalization, A-carriers: 31.5%; P=0.006; replication cohort: P=0.046). The functional studies revealed an association between the rs1063856 single nucleotide polymorphisms in vWF and FVIII activity (AA: 115.93%, AG: 156.07%, GG: 83.42%; P=0.005).
Conclusions—Three single nucleotide polymorphisms were associated with tissue-type plasminogen activator efficacy in the Spanish population, and their mechanism of action might be associated with the activity of coagulation factors.
Intravenous tissue-type plasminogen activator (tPA) is the only drug currently approved for acute stroke treatment. The clinical response to intravenous tPA may be poor, because of a lack of efficacy in terms of early recanalization of the occluded vessel in 48% to 65% of patients1 or because of safety concerns such as symptomatic hemorrhagic transformation occurring in 1.7% to 6.4% of cases and leading to death in 6.5% to 12.7% of cases.2 A longer time to recanalization highly correlates with a larger infarcted area and worse neurological outcome, overall leading to poor clinical recovery from stroke.3
The identification of factors predicting tPA efficacy could be useful to improve the management of patients with stroke, especially those not responding well to tPA treatment, because in these patients, the intra-arterial administration of tPA with or without coadjuvant drugs could be more effective than intravenous tPA treatment alone. In previous studies, we already identified some predictors of recanalization efficacy, namely thrombin–antithrombin complex levels and variants in the thrombin activable fibrinolysis inhibitor gene.4,5
The aim of this study was to identify single nucleotide polymorphisms (SNPs) that could be used by physicians as new predictors of early recanalization to individualize and improve acute ischemic stroke treatment.
White patients with an acute ischemic stroke with a documented arterial occlusion by transcranial Doppler (TCD) and who received tPA in a standard dose of 0.9 mg/kg (10% bolus, 90% 1-hour continuous infusion) within the first 4.5 hours after onset of symptoms were consecutively recruited in the emergency department.
The original cohort (Cohort A, n=531) consisted of patients with an acute stroke admitted in 5 Spanish university hospitals between November 2000 and May 2005. We finally included in the study 354 patients with continuous TCD monitoring. These patients were consecutively monitored by continuous TCD in those hospitals that perform a continuous monitoring TCD protocol. This cohort was genotyped in July 2007.
The replication cohort (Cohort B, n=354) consisted of patients with an acute stroke recruited in the same centers between May 2005 and June 2008; 143 of them had continuous TCD determination and were included in the study and genotyped in November 2008.
The clinical data of both cohorts is available in online-only Data Supplement Table I.
Clinical and TCD Protocol
On admission to the emergency department, a detailed history of vascular risk factors and current medication was obtained from each patient. Stroke severity was assessed with the National Institutes of Health Stroke Scale. Functional outcome was assessed at 3 months with the modified Rankin Scale.2
The main end point of the study was recanalization 1 hour after bolus tPA administration, which has been consistently associated with good outcome1,3 and is considered a marker of tPA efficacy. Continuous TCD monitoring during tPA administration, 1 hour, 2 hours, and 6 hours after tPA bolus was implemented to detect recanalization of the occluded vessel.1,5 Partial recanalization on TCD was defined by the Thrombolysis in Brain Ischemia flow grading system when blunted or dampened signals (Thrombolysis in Brain Ischemia 2–3) appeared in a previously demonstrated absent or minimal flow (Thrombolysis in Brain Ischemia 0–1) and complete recanalization if the end-diastolic flow velocity improved to elevated or normal values (Thrombolysis in Brain Ischemia 4–5; stenotic or normal).5 We considered successful recanalization when it was partial or complete. The local ethics committee approved the study and all patients gave informed consent.
DNA was isolated from peripheral blood extracted at admission and stored at −80°C.4 A total of 140 SNPs from 97 candidate genes (online-only Data Supplement Table II) related to coagulation, inflammation, or stroke risk factors were successfully genotyped at the Spanish National Genotyping Centre (CeGen, Barcelona, Spain) using SNPlex technology (Applied Biosystems, Foster City, CA). Internal validation of SNPlex results was performed using Sanger sequencing (complete methodology available in the online-only Data Supplement and online-only Data Supplement Table III).
The blood samples were extracted before tPA administration. The plasma samples were obtained after centrifugation for 15 minutes at 4°C and 15 000 rpm and stored at −80°C. The number of samples analyzed depended on sample availability and technical concerns such as the number of samples that could be added to one plate. von Willebrand factor (vWF) activity was measured in 80 plasma citrate samples by enzyme-linked immunosorbent assay (FVWF200; Axis-Shield Diagnostics Limited, Dundee, UK) in the fully automated Triturus system (Grífols, Barcelona, Spain). Activity of coagulation factors VII, VIII, and X was determined by measuring activated partial thromboplastin time in 125, 94, and 125 baseline citrate plasma samples using the semiautomated coagulometer ST4 (Diagnostica Stago-Roche, Asnières, France) following the manufacturer's user guide. Tissue factor (FIII) baseline activity was measured in 81 plasma citrate samples (Assaypro, St Charles, MO) following the manufacturer's user guide. Gene expression quantification was performed in 41 patients by real-time quantitative polymerase chain reaction analysis (Applied Biosystems, Foster City, CA). Protein measurements were performed by operators blinded to genotype data. All samples were matched by age, sex, and recanalization status at 1 hour. The detailed methodology for real-time quantitative polymerase chain reaction analysis and vWF, FVII, FVIII, FX, and FIII activity determination is available in the online-only Data Supplement.
Sample size calculation was performed using Ene 2.0 software. The 140 SNPs with frequency >0.35 in the no-recanalization group and <0.20 in the recanalization group could be associated with recanalization with a power of 80% and significance of 0.05 in a sample of at least 138 subjects considering 2 groups under a dominant/recessive model or additive model.
Deviation from the Hardy-Weinberg equilibrium was assessed using a χ2 test with one degree of freedom. Statistical significance for categorical variables was assessed by the χ2 or Fisher exact test. For continuous variables, analysis of variance and t test or Kruskal-Wallis and Mann-Whitney U test were used depending on the normality of the distribution, as indicated by Kolmogorov-Smirnoff test and PP or QQ plots. ORs and 95% CIs for the effect of each SNP on recanalization after tPA infusion were estimated using a forward stepwise logistic regression model adjusted by vascular risk factors and associated clinical variables. All statistical analysis was performed using the SPSS statistical package, Version 15.0. False discovery rate was used as multivariable test correction.
Clinical and Radiological Predictors of Recanalization
The univariate analysis revealed an inverse association between early recanalization and atherothrombotic etiology, absence of atrial fibrillation, and absence of proximal occlusion on baseline TCD (Tables 1 and 2). As expected, recanalization at 1 hour was associated with a lower median National Institutes of Health Stroke Scale at 24 hours after symptoms onset (Rec1 hour: 4, No Rec1 hour: 11; P=1.73−8), 48 hours (Rec1 hour: 4, No Rec1 hour: 10; P=4.98−7), and at discharge (Rec1 hour: 2, No Rec1 hour: 8; P=2.41−5) and better functional outcome at 3 months (Rec1 hour: 2, No Rec1 hour: 3; P=7.87−4).
Successful recanalization at 1 hour was obtained in 33.6% of Cohort A and in 29.4% of Cohort B. Three variants in Cohort A were associated with recanalization at 1 hour and were further validated in the replication cohort (Table 3; online-only Data Supplement Tables IV and V): rs1063856 (Cohort A: 53.8% of GG recanalized at 1 hour versus 31.5% of A-carriers, P=0.006; Cohort B: 50.0% of GG recanalized at 1 hour versus 26.9% of A-carriers, P=0.046) in the vWF gene (coagulation pathway) and rs1143627 (Cohort A: 53.1% of CC versus 32.7% of T-carriers, P=0.022; Cohort B: 53.3% of CC versus 27.4% of T-carriers; P=0.039) and rs16944 (Cohort A: 50% of TT versus 32% of C-carriers, P=0.038; Cohort B: 50.0% of TT versus 26.2% of C-carriers, P=0.049) in the interleukin (IL)1B gene (inflammation pathway). When we analyzed the results of these 3 SNPs combining the 2 cohorts, the values were rs1063856 (P=0.001), rs1143627 (P=0.002), and rs16944 (P=0.005). When we performed a multivariable test correction of these probability values using false discovery rate and considering only the significant results (Table 3), the adjusted q-values remained significant (rs1063856, q-value=0.007; rs1143627, q-value=0.006, and rs16944, q-value=0.01). However, when we used false discovery rate or Bonferroni correction with the probability values obtained in the discovery cohort, none of the probability values remained significant. Similarly when we used false discovery rate or Bonferroni correction in the replication cohort, none of the probability values remained significant.
When the 3 variants were analyzed regarding recanalization at later time points, a significant association was observed 2 hours after tPA administration for rs1143627 (60.5% of CC Rec2 hours versus 43% of T-carriers Rec2 hours, P=0.03) and a trend was observed for rs1063856 (P=0.076) and rs16944 (P=0.088).
Considering the additive model and recanalization at 1 hour, no association observed in Cohort A was replicated in Cohort B (Table 4).
After logistic regression including early recanalization and atherothrombotic etiology, atrial fibrillation, and proximal occlusion, 2 SNPs, early recanalization and atherothrombotic and proximal occlusion, were the only independent variables associated with successful recanalization at 1 hour: rs1063856GG (OR, 3.3; 95% CI, 1.8–6.3; P=2.3E−4), rs1143627CC (OR, 2.8; 95% CI, 1.4–5.3; P=0.03), no early recanalization and atherothrombotic (OR, 3.8; 95% CI, 2.02–7.3; 3.7E−5), and proximal occlusion (OR, 2.6; 95% CI, 2.6–4.2; 6.6E−5). When we performed the multivariable test correction, the probability value of rs1063856 remained significant even when conservative Bonferroni correction was used taking into account the 140 SNPs analyzed (adjusted probability value=0.03).
No association was observed between rs1063856 and vWF activity at baseline (P=0.293) nor between vWF activity and recanalization (P=0.54). However, an association was detected between rs1063856 and FVIII activity in 95 patients analyzed (AA: 115.93%, AG: 156.07%, GG: 83.42%; P=0.005; Figure 1); in addition, after Bonferroni multivariable test correction, this probability value remained significant (P=0.01). However, FVIII activity at baseline was not associated with recanalization at 1 hour (P=0.46) nor with short-term evolution (P>0.05) but correlated with modified Rankin Scale functional outcome (Spearman coefficient 0.263, P=0.009).
No difference was observed between rs1143627 and rs16944 genotypes and IL1B mRNA expression (P=0.29; Figure 2). As previously mentioned,6 we looked for an association between IL1B variants and FIII activity; however, no association was observed between rs1143627 (P=0.57) or rs16944 (P=0.54) and FIII activity.
We also checked the association between IL1B variants and FVII and FX coagulation factors, which form a complex with FIII in the extrinsic coagulation pathway. No association was observed between rs1143627 (P=0.62) or rs16944 (P=0.4) and FVII activity or FX activity (P=0.81 and P=0.89, respectively).
This genetic association study uncovered a role for VWF and IL1B polymorphisms in early arterial recanalization after tPA treatment. These 3 SNPs did not reach significant probability values after multivariable test correction in the discovery Cohort A. However, we observed a nominal significance for these 3 SNPs in the replication cohort, and a clear functional effect was observed for the rs1063856 (vWF gene) variant. In addition, rs1063856 SNP (vWF gene) was significant after logistic regression and Bonferroni correction.
IL-1β is a member of the interleukin 1 cytokine family. This cytokine is an important mediator of the inflammatory response and is involved in a variety of cellular activities, including cell proliferation or apoptosis. IL-1β is regulated at the transcriptional level, depending on haplotypes. However, this regulation is highly complex and the influence observed for SNPs strongly changes depending on multiple factors including the experimental approach used.7 In stroke, plasmatic IL-1β is increased during the acute phase, both in plasma8 and cerebrospinal fluid,9 although it remains in the picomolar range.
In our study, rs1143627 CC and rs16944 TT variants of IL1B were associated with arterial recanalization 1 hour after tPA administration in patients with stroke. The rs1143627 C/T SNP is located at position −-31 in the putative TATA box of the IL1B gene. The rs16944 C/T SNP is located at position −511. The haplotype, composed of the T allele at −511 and the C allele at −31 is significantly associated with a 2- to 3-fold increase of IL1B protein secretion.10 Thus, carriers of this haplotype may produce increased IL-1β protein and an enhanced inflammatory response that, by unknown mechanisms, is associated with high recanalization rates in patients with ischemic stroke after tPA treatment.
In our study, the 2 IL1B variants were not associated with IL1B mRNA expression nor with FIII, FVII, and FX activity, in contrary to what was previously reported.6 It is probable that the measurement of these molecules during the acute phase of stroke might influence the results of the functional studies as previously indicated.11
The glycoprotein encoded by the vWF gene is an antihemophilic factor carrier and also a platelet-vessel wall mediator in the blood coagulation system. vWF is critical in the hemostasis process. Mutations in this gene result in von Willebrand disease, characterized by episodic mucocutaneous bleeding.12 After activation, vWF promotes platelet aggregation and thrombus formation by the GPIb receptor.12 Alternatively, vWF directly interacts with collagen IV, releasing FVIII and leading to thrombus formation.12
vWF plasmatic levels show great variability among individuals. rs1063856-GG was first associated with higher vWF levels and myocardial infarction risk in patients with Type I diabetes.13 This association has been recently extended to the general population, in which rs1063856 also influenced FVIII levels.14 We found the highest arterial revascularization for rs1063856 “GG” individuals, although the “G” allele had been associated with higher protein levels for FVIII in the general population. In addition, this SNP resisted Bonferroni correction after logistic regression analysis. Our results did not reveal any association between GG genotype and vWF levels (P>0.05). A significant association between the GG genotype and lower FVIII activity might be one of the reasons for the increased recanalization rates observed in these patients. FVIII levels are higher in plasma samples during the acute phase of stroke.15 The altered FVIII activity during the acute phase of stroke might influence the role of the rs1063856 SNP. In our study, we hypothesized that lower FVIII activity presented by GG patients could partly explain the increased recanalization rates observed in these patients.
To summarize, 3 SNPs found in IL1B and VWF genes are associated with recanalization rates after tPA. These 3 SNPs did not reach significant probability values after multivariable test correction in the discovery Cohort A. However, these SNPs also reached nominal significance in the replication cohort, and a clear functional effect was observed for the rs1063856 (vWF gene) variant. The mechanisms by which these SNPs modulate recanalization could be related to homeostasis modulation by modification of coagulation factors activity. However, replication in an independent sample will be crucial to confirm these findings.
We believe that the replication and functional studies support the hypothesis that the 3 SNPs are not false-positives. Nevertheless, the significant probability values observed in Cohort A do not withhold the Bonferroni multivariable test correction. Moreover, the functional analysis are restricted to sample availability; thus, the sample size of these studies is limited. In addition, our cohort is one of the largest cohorts with genetic and recanalization data in ischemic tPA-treated patients. However, at present, genetic studies usually analyze cohorts composed by >500 subjects. For all these reasons, further replication work is needed to confirm these results and to expand them to other ethnic backgrounds (blacks, Asians, etc). We did our analysis in Cohorts A and B with all genetic models. However, based on the results of the replication cohort, we thought that the dominant/recessive model was more relevant than additive model due to none of the SNPs of the additive model were replicated; nevertheless, we do not have other statistical arguments to justify the use of dominant/recessive model instead of additive model.
Sources of Funding
This study was funded by a grant of the Spanish government (PI10/01212.). Dr Delgado is supported by the Miguel Servet programme (CP09/136) from the Spanish Ministry of Health (Instituto de Salud Carlos III). The research leading to these results has received funding from the European Union's Seventh Framework Programme (FP7/2007-2013) under grant agreements #201024 and #202213 (European Stroke Network). This study was funded in part by the Ministry of Health, Instituto de Salud Carlos III. Red HERACLES RD06/0009, FEDER.
We are deeply grateful to all participants, neurologists, and nurses who participated in this study. The Neurovascular Research Laboratory takes part in the International Stroke Genetics Consortium ISGC, the Spanish Stroke Genetics Consortium (www.genestroke.com), and in the Cooperative Neurovascular Research RENEVAS (RD06/0026/0010).
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.112.657007/-/DC1.
- Received March 12, 2012.
- Revision received June 5, 2012.
- Accepted June 20, 2012.
- © 2012 American Heart Association, Inc.
- Demchuk AM,
- Burgin WS,
- Christou I,
- Felberg RA,
- Barber PA,
- Hill MD,
- et al
- Christou I,
- Alexandrov AV,
- Burgin WS,
- Wojner AW,
- Felberg RA,
- Malkoff M,
- et al
- Fernandez-Cadenas I,
- Alvarez-Sabin J,
- Ribo M,
- Rubiera M,
- Mendioroz M,
- Molina CA,
- et al
- Iacoviello L,
- Di Castelnuovo A,
- Gattone M,
- Pezzini A,
- Assanelli D,
- Lorenzet R,
- et al
- Chen H,
- Wilkins LM,
- Aziz N,
- Cannings C,
- Wyllie DH,
- Bingle C,
- et al
- Schönbeck U,
- Mach F,
- Libby P
- Kiefer TL,
- Becker RC
- Lacquemant C,
- Gaucher C,
- Delorme C,
- Chatellier G,
- Gallois Y,
- Rodier M,
- et al
- Smith NL,
- Chen MH,
- Dehghan A,
- Strachan DP,
- Basu S,
- Soranzo N,
- et al