Measuring Platelet Reactivity: What Do the Latest Data Mean for Clinical Practice?

Thirty-day ADAPT-DES (Assessment of Dual AntiPlatelet Therapy with Drug-Eluting Stents) data were presented at TCT in November. Can you describe the study?

ADAPT-DES is a large, multi-center registry looking at the relation between on-treatment platelet function and stent thrombosis after drug-eluting stents were placed. It determined platelet function after the patient was treated with dual antiplatelet therapy, so after clopidogrel and aspirin. VerifyNow (Accumetrics) was used to measure platelet function ≥6 hours after a 600 mg clopidogrel load or ≥12 hours after a 300 mg load.  In patients treated with 75 mg clopidogrel qd, platelet function was measured at ≥5 days. The timing of the measurements were intended to reflect platelet function at steady state.

ADAPT-DES included a very large number of patients: 8,575 patients were enrolled at 11 sites around the world. Investigators not only looked at clopidogrel response, meaning platelet reactivity influenced by clopidogrel, but also at platelet reactivity influenced by aspirin. The VerifyNow assay can assess both an aspirin and a clopidogrel effect.

What did investigators find?

Patients were followed for 30 days. Overall, the stent thrombosis rates were low. The bulk of the stent thrombosis occurrences  were within the first twelve days. Several events did occur after that time period, but the vast majority occurred on day 12 or earlier. The definite and probable stent thrombosis rate was .46%. Quite low, but this is what is being seen now in the era of the new-generation stents (remember, however, that this is 30-day data and not one year.) The VerifyNow readout, called a P2Y12 reaction unit (PRU) level, was used to assess platelet reactivity during clopidogrel therapy. The PRU level was markedly higher in patients who had a stent thrombosis than in those who did not. P-value was significant at the .0001 level.

It is important to note that stent thrombosis in the highest two quintiles of platelet reactivity — the two highest quintiles for the PRU readout — was .8%. Stent thrombosis in the two lowest quintiles was about .2%. As we move from the lowest two quintiles to the top two quintiles, there is about a four-fold increase in the risk for stent thrombosis. In fact, a multivariate analysis found that 50% of the stent thromboses were  attributable to high platelet reactivity.

How should we implement these findings into clinical practice? 

Overall, there is a strong relation between platelet reactivity and stent thrombosis; however, in the individual patient, it is not very predictive. The prevalence is so low that the positive predictive value is not going to be great. This is the problem. If you have high platelet reactivity, the positive predictive value of the test is not very high. However, high platelet reactivity means you are in the group of patients with the highest risk for stent thrombosis. Yet while stent thrombosis is highly attributable to high platelet reactivity, in the individual patient, the positive predictive value is not very high. Testing each individual would result in the placement of many patients in the high-risk category although they would never end up having a stent thrombosis. Such an approach could lead to over-treatment in these patients with more potent antiplatelet strategies, and it would be costly. That is one way to look at it. The other way to look at it would be to say, well, I would like to test my patients to be sure that I can virtually eliminate the risk for stent thrombosis, because stent thrombosis is the most feared event to happen to a stented patient. It is catastrophic.

We still need to see further data on individual platelet function in the patients with stent thrombosis versus those without stent thrombosis. There may be a lower level of platelet reactivity below which we do not see stent thrombosis, or its occurrence is very, very low. We may be able to find an achievable level that is not associated with an increased risk for bleeding, raising the concept of whether there is a true therapeutic window for P2Y12 inhibitors. It is intriguing to think that if a patient’s PRUs are below a certain level, they are immune to stent thrombosis. It is a very exciting area of research, but we still need to do more work.

What would finding such a window mean for antiplatelet therapy?

You would titrate your antiplatelet drug to that therapeutic level of platelet reactivity. Which would entail, potentially, testing platelet function more than once, because you wouldn’t want to drive the platelet reactivity too low, placing patients in the range for bleeding risk. If you chose to use clopidogrel, you would test your patient, and if they were above the risk level, then you would either give them more  daily clopidogrel or put them on a new drug.

The American College of Cardiology/American Heart Association/Society for Cardiovascular Angiography & Interventions acute coronary syndrome (ACS) guidelines have platelet reactivity testing listed as a IIb recommendation for testing in patients with ACS, if the results of testing will lead to a change in treatment. This is not a very strong recommendation, but guideline writers finally acknowledged the overwhelming body of data that links high platelet reactivity to stent thrombosis. The European Society of Cardiology guidelines also make a similar recommendation for the ACS population. In the most recent percutaneous coronary intervention guidelines, routine analysis is a class III recommendation, with no benefit. However, if the patient is felt to be at high risk, then it can be considered. High risk would be the ACS patient, or patients with complex anatomy, such as  bifurcation stenting, treatment with long stents or small-caliber stents, multivessel stents, poor left ventricular function, etc.

What about the use of prasugrel or ticagrelor?

These drugs almost uniformly eliminate high platelet reactivity. If, from the beginning, you are going to use prasugrel or ticagrelor, there is no proven utility for platelet function testing, unless you wanted to be certain that the platelet function has not been driven too low, and the patient has an excess risk for bleeding.

How are you incorporating platelet function testing into your practice?

We started off doing registry studies, so at first, it was a research tool. Now we are using it according to the guidelines. In high-risk patients, if we treat them with clopidogrel, we will measure their platelet function before they leave the hospital. If their platelet function is high, and we believe they are at a high risk for thrombosis, they will get an alternative therapy. Up to this point, we have only been using prasugrel. Now that ticagrelor is available, we will be using that agent as well.

Has platelet reactivity as it relates to stent thrombosis been studied before?

Yes, the relation of platelet physiology to risk of stent thrombosis has been studied since the early 2000’s. and more recently, in registries of thousands of patients. ADAPT-DES is just a much, much larger study. We knew about the relationship and had evidence that platelet reactivity was higher in those with stent thrombosis than those without it and published our findings in the Journal of the American College of Cardiology in 2005.¹ This was the first report of the relation between platelet reactivity to adenosine diphosphate and stent thrombosis. The major question now is whether modification of the platelet reactivity actually lowers the stent thrombosis rate, although this question was not addressed in ADAPT-DES. 

Any final thoughts?

I would like to highlight the very important role of the platelet in stent thrombosis. The fact that the patient with high platelet reactivity is at an increased risk for stent thrombosis rests on a rock-solid biologic rationale. In fact, the development of all antiplatelet drugs is fundamentally based on an assessment of platelet function. How did we come to decide to study prasugrel or ticagrelor? These decisions were initially based on an assessment of ex vivo platelet function. Prasugrel and ticagrelor had better antiplatelet effects than clopidogrel. It makes sense that the patient who has a poor antiplatelet response is going to do more poorly than someone with a good antiplatelet response, just like it is logical that prasugrel, which is associated with a better antiplatelet response, is associated with better outcomes.

Dr. Gurbel can be contacted at pgurbel@lifebridgehealth.org.

Reference

1. Gurbel PA, Bliden KP, Samara W, Yoho JA, Hayes K, Fissha MZ, Tantry US. Clopidogrel effect on platelet reactivity in patients with stent thrombosis: results of the CREST Study. J Am Coll Cardiol 2005 Nov 15;46(10):1827-1832.