The cardiology experience

Meta-analyses have shown an increased rate of poor outcome (stent thrombosis, myocardial infarction, and death) in patients with high persistent platelet reactivity (non-response) treated for symptomatic coronary atherothrombosis.1 However, modification of antiplatelet therapy according to point of care platelet function testing did not reduce the rate of poor outcome (stent thrombosis, myocardial infarction, and death) in the recent ARCTIC (Assessment by a Double Randomization of a Conventional Antiplatelet Strategy versus a Monitoring-guided Strategy for Drug-Eluting Stent Implantation and of Treatment Interruption versus Continuation One Year after Stenting Trial)2 and GRAVITAS (Gauging Responsiveness with a VerifyNow assay—Impact on Thrombosis And Safety)3 trials.

The ARCTIC study prospectively randomized 2440 patients across 38 centers to a monitoring strategy of platelet function assessment with VerifyNow P2Y12 assays and drug adjustment in patients with persistent platelet reactivity, or to a conventional strategy without monitoring and no drug adjustment.2 In the monitoring cohort, persistent platelet reactivity to clopidogrel (≥230 P2Y12 reaction units (PRU) or <15% reduction from baseline) or aspirin (≥550 aspirin reaction units) occurred in 35% and 8%, respectively. This led to administration of an additional load of ≥600 mg bolus of clopidogrel or 60 mg of prasugrel at least 6 h prior to percutaneous coronary intervention (PCI), use of intraprocedural glycoprotein IIb/IIIa inhibitors, and daily maintenance doses of 150 mg of clopidogrel or 10 mg of prasugrel after the procedure, if there was persistent platelet reactivity to clopidogrel, and additional bolus intravenous aspirin for those with persistent platelet reactivity to aspirin. Significant reduction in persistent platelet reactivity (34.5% to 15.6%; p<0.001) was proven by repeating the VerifyNow assays between 2 and 4 weeks after stent implantation.

The primary endpoint of myocardial infarction, stent thrombosis, stroke, urgent revascularization, or death by 1 year occurred in 34.6% of the monitoring group compared with 31.1% of the conventional group (p=0.10). There was no difference in the main secondary endpoint of stent thrombosis or any urgent revascularization (4.9% vs 4.6%; p=0.77). Importantly, the rate of major bleeding events (clinically overt bleeding causing ≥3 g/dl hemoglobin reduction, hemodynamic compromise requiring transfusion or surgical intervention, any retroperitoneal, intracranial bleeding, or fatal bleeding) was not significantly different (2.3% vs 3.3%; p=0.15).

The GRAVITAS trial also failed to show a reduction in the incidence of cardiovascular death, non-fatal myocardial infarction, or stent thrombosis at 6 months in 2214 patients with high on-treatment platelet reactivity (PRU ≥230) measured 12–24 h post PCI and randomized between high dose clopidogrel (600 mg total first day dose followed by 150 mg daily for 6 months) versus standard dose (loading dose of placebo followed by 75 mg of clopidogrel and placebo tablet daily).3

Can we translate the cardiology experience: not really…

When point of care testing is used in NI, it is generally performed prior to elective procedures, mainly endovascular treatment of asymptomatic aneurysms. The clinical goal of point of care testing in NI is to reduce the rate of periprocedural thromboembolic complication by assessing the level of persistent platelet activity, and modifying prophylactic antiplatelet therapy. This is quite different to the endpoints of the large cardiology trials. The most relevant data that could be extrapolated from the cardiology literature are those of stent thrombosis, the frequency of which increases with higher levels of persistent platelet activity.

In a meta-analysis of 3059 patients across six trials using the VerifyNow P2Y12 assay, there was a significant increase in the rate of stent thrombosis in those in the highest quartile of PRU compared with the lowest quartile (3.4% vs 0.4%; p=0.002).1 When a threshold value of 230 PRU was chosen, the rate of stent thrombosis was 3.0% if ≥230 PRU versus 1.0% if <230 (HR 3.11; 95% CI 1.50 to 6.46; p=0.002). However, these data, while interesting, cannot be extrapolated to our general NI use, as stent thrombosis was reported as a delayed outcome, including at 1 year.

What is more relevant for our context is the rate of intraprocedural thromboembolic events or acute stent thrombosis—these are rare, occurring in 3.5% and 0.3% of patients, respectively, undergoing acute PCI.4 There are no randomized controlled data to confirm that modification of antiplatelet therapy according to point of care platelet function testing reduces the rates of intraprocedural thromboembolic events or acute stent thrombosis.

Clopidogrel and neurointervention

Observational studies have shown that clopidogrel premedication reduces the rate of procedure related thromboembolic events in patients undergoing elective coil embolization of intracranial aneurysms.5 ,6 However, clopidogrel is not the best antiplatelet agent for our NI clinical goal because of the marked variability in individual response—per cent platelet inhibition (VerifyNow) in NI patients after routine clopidogrel premedication varies between 0% and 99%.7 ,8 There are a myriad of factors involved in an individual patients response to clopidogrel, including the genetic polymorphism of the CYP2C19 allele. In addition to loss of function variants that result in little platelet inhibition (CYP2C19*2), there are also gain of function variants (CYP2C19*17) that can result in marked platelet inhibition. Marked platelet inhibition in turn increases the risk of hemorrhagic complications.7 ,9

Given the marked variability in the level of platelet inhibition after clopidogrel use, point of care testing has been utilized in the NI community in an attempt to individualize antiplatelet therapy. In spite of a lack of a uniform definition of ‘non-response’, more potent antiplatelet agents (such as prasugrel) are given to patients in an attempt to prevent thromboembolic complications during NI procedures, using regimens borrowed from the cardiology literature. Ticagrelor and cangrelor are also potential alternatives but there are no reports of their use in NI.

Is there an alternative?

The ideal drug for our NI clinical goal is one that provides rapid, consistent, and predictable platelet inhibition. A consistent dose–response effect would make point of care testing unnecessary. Defined doses could then be tailored to the clinical context, depending on whether we are trying to prevent or treat thromboembolic complications. There is already a potential alternative that is being explored in the NI literature. Prasugrel results in more potent, consistent, and predictable platelet inhibition compared with clopidogrel, irrespective of age, weight, and dose.10 ,11 The key to achieving our NI goal depends on defining the correct dose of prasugrel to achieve enough platelet inhibition to prevent thromboembolic events without increasing the risk of hemorrhage compared with the current standard of aspirin and/or clopidogrel.

In the TRITON TIMI 38 trial (Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition with Prasugrel–Thrombolysis in Myocardial Infarction 38), 13 608patients undergoing PCI were randomized to prasugrel (60 mg loading dose/10 mg daily maintenance dose) or clopidogrel (300 mg loading dose/75 mg daily maintenance dose).12 Major bleeding occurred in 2.4% of the prasugrel cohort and in 1.8% of the clopidogrel cohort (p=0.03). Intracranial hemorrhage occurred in 0.3% of both cohorts; fatal hemorrhage was more frequent in the prasugrel cohort (0.4% vs 0.1%; p=0.002). Three subgroups were identified with higher rates for hemorrhage: age ≥75 years, weight <60 kg, and prior history of stroke or transient ischemic attack (TIA).

Subsequently, the TRILOGY ACS trial (Targeted Platelet Inhibition to Clarify the Optimal Strategy to Medically Manage Acute Coronary Syndromes) (n=9326) modified the prasugrel regimen: patients with a history of stroke or TIA were excluded; a 30 mg loading dose was administered for all patients and a reduced 5 mg maintenance dose for those aged ≥75 years or with weight <60 kg.13 There were no longer any differences in the rates of minor or major (2.1% vs 1.7%; p=0.11), intracranial (0.3% vs 0.4%; p=0.42), or fatal (0.2% both groups; p=0.68) hemorrhagic complications between prasugrel and clopidogrel.

There are only small case series in the NI literature about the use of prasugrel for elective NI procedures, mainly elective aneurysm treatment.14–16 Akbari et al reported on their strategy of prasugrel (60 mg load/10 mg maintenance) for 25 patients with <40% P2Y12 receptor inhibition (VerifyNow) after premedication with clopidogrel, and compared hemorrhagic complication rates with 51 patients treated with aspirin and standard clopidogrel therapy with >40% P2Y12 receptor inhibition.15 Excluding the patient with a procedural basilar artery perforation, major hemorrhagic complications (requiring blood products/surgical intervention) occurred in five patients treated with aspirin/prasugrel compared with one patient with aspirin/clopidogrel. One intracranial hemorrhage occurred in both arms; one thromboembolic complication (TIA) was reported in the aspirin/prasugrel arm. Stetler et al reported their safety data on a more conservative strategy of prasugrel (40 mg load/5–10 mg maintenance for a mean of 3 months) for 16 patients with <20% P2Y12 receptor inhibition (VerifyNow) after premedication with clopidogrel.14 There were no major bleeding events (requiring blood products/surgical intervention) or intracranial hemorrhage. A single retroperitoneal hematoma occurred in a patient with a high femoral puncture site that was managed conservatively. No thromboembolic complications occurred. Notably, the reduced 40 mg prasugrel load in 13 patients produced therapeutic inhibition in all patients; almost 80% of non-responders had ≥50% P2Y12 receptor inhibition after the single dose of prasugrel compared with 0–9% inhibition after 1 week of premedication with clopidogrel.

We must remember that the level of platelet inhibition achieved by a 60 mg loading of prasugrel after just 30 min is similar to the maximal inhibition achieved by a 600 mg loading dose of clopidogrel.17 By 6 h, significantly greater (51% VerifyNow P2Y12 assay; 95% CI 45.5 to 57.4; p<0.0001) platelet inhibition is achieved by prasugrel (89.5 ± 10.5% inhibition) compared with 600 mg of clopidogrel (38.4 ± 26.1% inhibition). A 10 mg prasugrel maintenance dose also achieves significantly greater levels of platelet inhibition than 150 mg of clopidogrel daily.17

So where to now?

There is no clear answer. Clopidogrel is effective in preventing thromboembolic complications for most patients. However, there is marked individual variability in platelet response, and in some patients undergoing NI procedures marked platelet inhibition can result in increased rates of hemorrhagic complications.7

So far efforts have focused on making our NI procedures safer by individualization of clopidogrel therapy using point of care testing. However, this is a drug that does not produce a consistent dose–response effect, which we attempt to characterize using a test for which there are no consistent definitions for interpretation. Moreover, we modify therapy and add more potent antiplatelet agents in regimens borrowed from the cardiology literature, which used clinical endpoints significantly different to our clinical goals. This strategy for routine clinical NI practice seems unjustified.

Periprocedural antiplatelet therapy is a cornerstone to make our elective procedures safer. Maybe the ideal level of prophylactic platelet inhibition for our procedures is somewhere between 50% and 80%. The consistency of dose–response makes prasugrel an attractive choice but the 60 mg prasugrel load and 10 mg maintenance dose appear too potent for our NI clinical goal. A multicenter dose escalation trial or registry using prasugrel at 20, 30, and 40 mg loading doses and 2.5–5 mg maintenance doses would provide helpful safety and efficacy data; once an effective strategy with acceptable safety profile is established, point of care testing would be unnecessary. As a NI community, we can take charge of this issue, and work collaboratively to achieve better safety outcomes for our patients.