Why it is Important to Test Platelet Reactivity

Why do you test platelet reactivity? 

In acute coronary syndrome (ACS) patients who receive stents in the cath lab, studies have demonstrated that if there is an inadequate amount of platelet inhibition despite the administration of dual antiplatelet therapy, many of these patients may go on to experience stent thrombosis or heart attacks, and may even die. Therefore, it makes sense to make sure that these patients are getting the right level of platelet inhibition. 

Based on genetic metabolism methods, we know that up to 25% of patients have significant resistance to clopidogrel (Plavix). Clopidogrel is an inactive drug that has to be activated by the liver. Some people do not have the appropriate enzymes to activate it, and so they are resistant to clopidogrel. Essentially, they are not even getting the drug. Therefore, it seemed reasonable and appropriate to figure out how to appropriately inhibit these patients, and that is why we started measuring platelet inhibition. 

The P2Y12 inhibitor prasugrel (Effient) has the same bottom-line active molecule as clopidogrel, except it is more pre-activated and doesn’t have to go through all the same activating pathways as clopidogrel.  Prasugrel’s incidence of resistance is not as high as with clopidogrel. However, although prasugrel was better in the TRITON trial when it was tested against clopidogrel, there was also a significant amount of bleeding associated with prasugrel. It may well be that for a large number of patients, the trial dose of 10mg was too high, and too much bleeding was likely the result. At least one published study has demonstrated that too much platelet inhibition is associated with an increased risk of bleeding.¹ There appears to be a sweet spot for the best effect, and that is why we test, to try to get our patients to that sweet spot. Baseline characteristics, such as age, gender, and weight, are not adequate enough information to tell us about that sweet spot, and so, just like we check pro-thrombin times in patients on warfarin, we check P2Y12 reactivity units (PRUs) in patients who are on platelet inhibition. The goal is to make sure we have the right drug and the right dose, so there isn’t too much thrombosis or too much bleeding. 

How is platelet function testing implemented into your practice?

Intermountain Healthcare has set up a guideline with the recommendation to consider everyone for testing, but especially patients with ACS or those who are very high-risk patients. Note we did not say, “Don’t test the stable patients.” I personally test all patients, but the guidelines themselves are not so dogmatic. 

The GRAVITAS trial looked at very stable patients found to have ineffective platelet inhibition. Patients were randomized to stay on the regular dose of clopidogrel or receive a double dose. In the post hoc analysis, patients who started out without effective platelet inhibition but who became effectively inhibited did better than the ones who stayed ineffectively inhibited, whether or not these patients were left on the normal dose of clopidogrel or went to the higher dose. The problem with GRAVITAS was that if patients are clopidogrel resistant, it is a sign that they have something wrong with their activation metabolism. It turns out to be extremely difficult, just by giving more clopidogrel, and especially just by doubling the dose of clopidogrel, to increase the level of platelet inhibition in a resistant patient. Now, if the patient is clopidogrel sensitive, and the dose is doubled, then we easily get more inhibition. It is when the patient is clopidogrel resistant to begin with that there is a problem.

We can test for clopidogrel resistance using genetic panels, but the panels don’t account for all of the potential factors that may influence platelet reactivity, and so our approach has been to directly measure platelet inhibition. Rather than looking at genetics and trying to make an inference between genetics and platelet inhibition, we will find out if a patient is resistant by determining whether their PRU is too high. In those circumstances, it has already been demonstrated that it is not particularly helpful to increase the clopidogrel dose. Unfortunately, the two major outcomes trials looking at testing platelet inhibition, GRAVITAS and ARCTIC, used stable patients, and both trials mainly just increased the dose of clopidogrel. Therefore, these trials were limited by their inability to overcome clopidogrel resistance.  At the same time, in both trials, the overall event rates were quite low, which was also limiting, because they picked a patient population less likely to have stent thrombosis. 

After receiving a stent, a stable patient who takes only aspirin doesn’t always have stent thrombosis. The incidence of stent thrombosis on aspirin alone is only about 8%, without any clopidogrel. While any incidence of stent thrombosis is still too high, 9 out of 10 patients, even without these drugs, are not going to thrombose, and in the stable patients, the rate is going to be even less. No matter what, there will be a lower incidence of stent thrombosis in stable patients. But that still doesn’t mean these patients won’t do better on adequate dual antiplatelet therapy. Even though the relative risk reduction may be similar in both stable and unstable patients, the absolute reduction will be less in the stable patients. That is the reason why TRIGGER-PCI, which used clopidogrel in one arm and prasugrel in the other arm, was stopped early, because of futility. The trial researchers realized that in the chosen patient population, not enough events were going to occur. It didn’t matter whether patients stayed in the clopidogrel arm or whether they switched to the prasugrel arm (which is much more successful at reaching a PRU of <200), there were almost no events in either group. The trial would probably need to be powered for 20,000 patients in order to demonstrate a statistically significant difference in hard efficacy outcomes like death. 

How long have you been testing platelet function?

We have been testing for about three years. We have found a great deal of variability in platelet inhibition in our patient population receiving either clopidogrel or prasugrel and have published our results in an abstract.² We are currently writing up a manuscript incorporating our experience with ticagrelor as well.  More than half (57.2%) of patients were resistant to clopidogrel, with a PRU >200, and 17.3% were over-inhibited (ie PRU<100). This means only about only 25.5% of the patients were in range if given clopidogrel. By range, I mean 100-200 measured PRU. (As to why we picked 100-200 as the therapeutic window, it is still under some debate in the general community. People may quibble about the ends.) In patients receiving prasugrel, it was basically the other way around. A total of 55.7% of patients were over-inhibited, 21.4% were under-inhibited, and only 22.9% were in that 100-200 PRU range. Interestingly, none of the baseline clinical variables of age, sex, hypertension, hyperlipidemia, heart failure, renal failure, prior cardiac history or other non-platelet related medications were predictive of PRU results. 

So, whatever drug you use, there is significant variability in the effect. It is just as important to test people who get prasugrel, because of the risk of bleeds. We found that more than half of patients receiving prasugrel had a PRU of <100, so it is no wonder they have more bleeds. TRITON-TIMI 38 reported significantly more bleeding in patients receiving prasugrel. My hypothesis is that a majority of the excess bleeding came in the patients that had PRUs <100. That data is not available, but that is my guess. 

What we need are appropriate randomized trials, in which patients are titrated into the right dose, and then compare the outcomes of those who reach the right range versus those who don’t. Unfortunately, the two studies thus far have all used mostly clopidogrel. They didn’t actually convert over to prasugrel or ticagrelor, drugs more able to reach the therapeutic range that we would target. We need a study that actually tests what happens when you get patients into range. My sense is that in all of the trials thus far, most patients did not reach into the target therapeutic range. Therefore, it is not surprising that they didn’t get positive results. 

In the meantime, we as clinicians have to do what is best for our patients. We make all sorts of decisions when we don’t have specific trials, and we use the best information we can. At Intermountain, we have concluded that, with the information we now have, it only makes sense to make sure we are getting the therapeutic effect we want out of our antiplatelet therapy. Unfortunately, since we are testing everyone, we can’t do natural history studies. We did propose a study to do it in a randomized fashion, but we did not get funded, and since it is not funded, we have to do what we think is right. 

What are you seeing with your patients?

I have very little bleeding and very little stent thromboses in my patients. But these are low incidence events anyway, so with a single person, it’s hard to say. But I believe I am making a difference. Not as many of my patients come in with bruises. A significant number of my patients on prasugrel, probably half, which is just what was found in the TRIGGER-PCI study, have a PRU <100. Once I measure results at that level, I back off the prasugrel dose to 5mg, which is an FDA-approved dose, but it’s not based on PRU, it’s based on age and weight. Still, it works for PRU. Patients come back after a few weeks on the 5mg dose, and they haven’t had stent thrombosis, their bruising and problems are much less, and they are happier. Platelet function testing in this way is useful; the more I use it, the more I like it.

Fewer complications would also have economic implications.

Yes, that’s true. In addition, when you switch to a 5mg dose of prasugrel, you take a 10mg pill and cut it in half, and the price also goes in half. Prasugrel is not generic and thus still expensive. Further, we are able to identify a large number of patients that are able to safely stay on clopidogrel, which is now generic and much cheaper for the patients.

How do you talk to patients about testing platelet function?

I tell them that we are going to start them on this medicine and it is our best guess as to what might be an appropriate dose, but not everybody is the same, so I’m going to test it. Luckily, we have a blood test that I can use to tell them whether they are getting too much or too little of this drug, and then adjust accordingly. They love that idea. It’s easy to convince them to do it.

How frequently do you test each patient?

I test patients the morning after their procedure. It is important to make sure that certain drugs are not on board that would compete with the test, especially glycoprotein IIb/IIIa inhibitors like eptifibatide or abciximab. While Accumetrics (manufacturer of the VerifyNow test, which is what we use) says to wait 48 hours for the IIb/IIIa inhibitor to wear off, we found it is not really necessary. It is also hard to wait that long, because patients aren’t in the hospital for that length of time. So we test the next morning, before the patients go home. If we have them on clopidogrel and find that the patient is not responsive, then we switch them to prasugrel. We reload the patient on prasugrel and send them home. We will retest about 2-3 weeks later, after they become stable on the maintenance dose, and then adjust the dose accordingly, if needed. The bleeding complications with prasugrel are more long term rather than short term, so we usually leave the patient on 10mg/day for the first three weeks, and then we come back and test. If the test shows a PRU that is too inhibited, then we will cut the dose to 5mg and bring the patient back 2-3 weeks later to test again, and make sure that they are actually in the appropriate PRU range. Just like with a pro-thrombin time with warfarin, after making a change, you test to make sure the patient is in the right range. PRU is not as variable as warfarin, so while you have to keep checking warfarin every month, we don’t do that with PRU. Once a stable dose is found, then I check once a year thereafter. The reason I check once a year is to detect compliance. Patients often get tired of their blood pressure medicines and they stop taking them. Then they come in and see me, and their blood pressure is high, and I say, “You stopped your medicines, didn’t you?” I can do the same thing with platelet inhibitors. After measuring their PRU, I can tell my patients, “It looks like you stopped taking your medicine.” Then they confess and say, “Yes, it was so expensive, I just couldn’t afford it, and I stopped it.” I tell them, “It is still important, and I’ll help you. We’ll get samples. We’ll figure out some way to get you this medicine.” So testing helps me to monitor compliance as well. 

How did you get your program started at Intermountain Healthcare?

Our chief of cardiology, Dr. Don Lappe, saw the initial data and was very excited. He directed us, me and my partner Dr. Jeffrey Anderson, who is also the chair of the guidelines committee for the American Heart Association, to start working on how to best implement platelet function testing. We worked together to come up with the Intermountain Health guidelines and then proposed it to our physicians, and they bought into it. 

Are you primarily using prasugrel as the alternative to clopidogrel?

We are using ticagrelor (Brilinta) as well, especially in certain groups, including ACS patients who aren’t candidates for prasugrel. These are patients over the age of 75 and patients with a prior history of transient ischemic attack (TIA) or stroke, where there is either an absolute or relative contraindication for the use of prasugrel. Probably once the dose is adjusted, those risks wouldn’t be there. But in the TRITON trial, the dose was not adjusted, they just gave 10mg to everybody, and that was too much for 50% of the patients. But I don’t think the company wants to do another huge TRITON trial to prove that if you adjust the dose you wouldn’t need a warning against TIA, stroke, and patients over the age of 75. 

Could you walk us through a case that demonstrates the value of testing platelet reactivity?

I had a patient a while ago, a man who had bypass surgery for left main disease. The surgeon did a funny sort of bypass in which the patient got one big vein. It was a four-vessel bypass and the surgeon hooked it up to 4 different arteries: the left anterior descending, left circumflex, and right coronary artery. With one vein graft, the surgeon bypassed the whole heart. The patient did fine for 10 years, but at year 11, he came in with chest pain with elevated troponins. He was in the coronary care unit and we were planning to cath him the next morning. He was put on aspirin, eptifibatide, and heparin. At the time, I didn’t know what his anatomy was and how his bypass had been done. I then got a call at 6am the next morning, saying the patient was getting really sick, had chest pain, and his blood pressure was dropping. He was going into heart failure. We took him stat into the cath lab and found a 99% stenosis in the proximal saphenous vein graft. It was like severe left main disease with a totally occluded right. We rapidly opened and stented him, and got blood flowing into his whole heart again, and he just perked up like crazy. All of his heart failure went away, he peed out all the extra fluid, and he was feeling great. We thought it was a fabulous result. This was a very acute non-STEMI patient. He was older than the age of 75 and this was before we had ticagrelor, so we gave him a 600 mg load of clopidogrel. We waited the requisite 18-24 hours for eptifibatide to leave his system so we could test him, and his PRU was 325. This patient was totally and completely resistant to clopidogrel. We watched him carefully (remember, this is before ticagrelor), because if that stent was to clot, this patient would die suddenly. He wouldn’t make it to the ER to have an acute rescue angioplasty; he would just die, because there is one stent that is filling his whole heart and it is in a degenerated vein graft, so it is a risky stent. We felt very uncomfortable sending him home with clopidogrel, so we gave him prasugrel. His PRU, after a 60mg load of prasugrel, dropped to 30. We thought that was good and brought him back three weeks later for testing. His PRU had gone down to 25. At that time, he was not so acute, so we cut his prasugrel dose down to 5mg, and brought him back again three weeks later. His PRU came up to 110, in the safe range, but still showing he was well anticoagulated. We have been following him ever since, and he has been doing great. Especially with this patient, I am very convinced, because he was so resistant to clopidogrel and there was such a tremendous need to keep that stent open, that he may well have died if we hadn’t tested his PRU.

Any final thoughts?

Someone needs to do the right trial, which would look at high-risk ACS patients, because then you don’t need 50,000 patients. It would randomize patients to untargeted treatment or targeted treatment, meaning patients are treated until they get into range. There would be two endpoints to the study. The first endpoint would measure success at getting patients into range. In my opinion, you can be very successful, because we have many drugs for getting patients into range. I can get every single one of my patients into range by adjusting doses and switching drugs. Once everyone is in range, the second endpoint would compare appropriately inhibited outcomes to the standard dose of clopidogrel. 

It’s a good point that you are able to get every single one of your patients into range with the drugs we have available today.

All we need is the trial to prove it, so it can be evidence-based. We as physicians have to make decisions based on the best information we have, even if there isn’t a randomized trial to justify it. But we would like the randomized trial, and I acknowledge that I might be wrong. But until I am proven wrong, I think that platelet function testing is the right thing to do. n

New Studies Demonstrate Cost Effectiveness, Clinical Value in a Stent Population and Validation of a Therapeutic Window using the VerifyNow System

Accumetrics, Inc., developer of the VerifyNow system, announced the presentation of a series of important data that solidifies the clinical utility of platelet reactivity testing. Real world outcomes data in high-risk patients receiving stents, a cost-effectiveness analysis, and validation of a therapeutic window continue to demonstrate platelet reactivity as a critical element for improving the quality of care for the millions of patients on antiplatelet therapies worldwide.

In an oral presentation at the 2013 American College of Cardiology (ACC) Scientific Sessions in San Francisco, CA, Craig I. Coleman, PharmD, Associate Professor of Pharmacy at the University of Connecticut School of Pharmacy, discussed data from his cost-effectiveness analysis, which demonstrated that using a platelet reactivity driven antiplatelet therapy treatment strategy was more cost-effective than using a universal approach of generic clopidogrel in all PCI patients. This follows his recent presentation at the 2013 annual Cardiovascular Research Technologies meeting in Washington, D.C., where his presentation demonstrated that a platelet reactivity driven antiplatelet strategy is also more cost-effective than a universal strategy of giving the more potent inhibitor, ticagrelor, to all PCI patients. In both analyses, the VerifyNow System was used to support the cost model.

“Antiplatelet therapy has undergone a rapid evolution following the approval of two potent and more costly agents, and the recent generic availability of clopidogrel,” said Dr. Coleman. “The results of our model may have significant implications on which antiplatelet agents clinicians use in PCI patients, by attempting to focus on reducing healthcare costs and hospital readmission while improving the quality of patient care.”

Data from University Hospitals of Cleveland also presented at the ACC meeting demonstrated that managing dual antiplatelet therapy (DAPT) and high on-treatment platelet reactivity using a VerifyNow P2Y12 platelet reactivity testing strategy and treatment algorithms in ACS and other high-risk PCI patients led to a reduction in 30-day event rates to one which is equal to that of the facility’s stable, elective PCI population.

“At the Harrington Heart and Vascular Institute/University Hospitals of Cleveland, we have been ahead of the curve regarding the use of platelet reactivity testing in high-risk patients, having utilized a comprehensive platelet function testing management program for the last two and a half years,” said Dr. Tom Lassar, Associate Director of the Adult Cardiac Catheterization Laboratory and Interventional Cardiology at University Hospitals, Case Western Reserve in Cleveland, OH. “Incorporating platelet reactivity measurements in our decision making provides us with invaluable information to make treatment management decisions in our higher risk PCI patients. The data reinforces our conviction that a platelet function testing program is improving the quality of care for our PCI patients. We look forward to expanding our platelet reactivity program to other facilities in the University Hospitals System.” 

The association between high platelet reactivity and risk for subsequent ischemic events has been well characterized by numerous studies. Additional analysis of ADAPT-DES, discussed at an ACC industry expert theater, has also validated the association between very low levels of platelet reactivity and an increased risk of bleeding complications.  

“The concept of a therapeutic window for platelet reactivity has been previously postulated,” stated Ajay J. Kirtane, MD, SM, Chief Academic Officer of the Center for Interventional Vascular Therapy at Columbia University Medical Center/New York-Presbyterian Hospital in New York City. “In over 8500 patients, we were able to demonstrate that patients in the lowest quintile of response [PRU of 0-94] receive virtually no additional protection from ischemic events, but had a significantly higher incidence of bleeding. Recognition of this finding could potentially allow clinicians to minimize the risk of bleeding complications for patients who are on potent antiplatelet therapies.” 

The continued focus on antiplatelet therapies, including a late-breaking clinical trial with phase III results from the first IV P2Y12 inhibitor to potentially come to market, coupled with numerous platelet reactivity presentations, further illustrates the need for greater clarity on optimal use of antiplatelet agents in patients with cardiovascular disease. 

Disclosure: Dr. Muhlestein reports he is on the speaker’s bureau for Daiichi Sankyo and Eli Lilly and Company. 

J. Brent Muhlestein, MD, can be contacted at brent.muhlestein@imail.org.

References

1. Campo G, Parrinello G, Ferraresi P, Lunghi B, Tebaldi M, Miccoli M, Marchesini J, Bernardi F, Ferrari R, Valgimigli M. Prospective evaluation of on-clopidogrel platelet reactivity over time in patients treated with percutaneous coronary intervention:  relationship with gene polymorphisms and clinical outcome. J Am Coll Cardiol. 2011; 57: 2474–2483.
2. Bair TL, May HT, Horne BD, Bennett ST, Whisenant BK, Anderson KL, Lappe DL, Muhlestein JB.  Variability of platelet inhibition as determined by the VerifyNow assay among patients receiving clopidogrel or prasugrel: results from a real world registry. Circulation. 2012; 126: A17054.