Extended Low-Dose Ticagrelor May Prevent Adverse Events Post-MI, Study Finds

A long-term analysis of PEGASUS-TIMI 54 patients shows benefit to extending therapy but with the trade-off of increased bleeding risk.

Extended Low-Dose Ticagrelor May Prevent Adverse Events Post-MI, Study Finds

Given the high risk of events in patients after an initial MI, new research suggests extending ticagrelor at a low dose for several years may be of benefit despite some increased bleeding.

The value of prolonged ticagrelor (Brilinta; AstraZeneca) post-MI has been debated, but questions have remained regarding how long the drug should be continued and in which patient populations. Additionally, data from the COMPASS trial—presented at the European Society of Cardiology Congress last month—may ultimately lead to an additional option, namely long-term, low-dose rivaroxaban (Xarelto; Bayer/Janssen) on top of aspirin in patients with stable atherosclerotic vascular disease.

For this study, published in the September 12, 2017, issue of the Journal of the American College of Cardiology, Marc Bonaca, MD, MPH (Brigham and Women’s Hospital, Boston, MA), and colleagues looked at more than 21,000 patients from the PEGASUS-TIMI 54 trial who were randomized to one of two doses of ticagrelor—60 or 90 mg twice daily—or placebo after an MI then followed for a median of 33 months. Just over one-quarter of patients (28%) were more than 5 years out from their MI at the conclusion of the trial.

Compared with patients in the placebo arm—who saw their annual risk of cardiovascular death, MI, or stroke increase consistently at roughly 3%—those treated with ticagrelor 60 mg saw a clear reduction in events at year 1, with trends in reductions at years 2 and 3.

Risk of CV Death, MI, or Stroke: Ticagrelor 60 mg vs Placebo



95% CI

Year 1



Year 2



Year 3



When patients were stratified by the gap between their MI and time of randomization, there seemed to be a greater benefit of ticagrelor 60 mg for those within 2 years of their MI (HR 0.77; 95% CI 0.66-0.90) compared with those 2 or more years from their event (HR 0.96; 95% CI 0.79-1.17; P = 0.09 for interaction). In addition, there was a reduction in CV mortality for those within 2 years of their MI (HR 0.68; 95% CI 0.53-0.89) but not for those who were further out (HR 1.12; 95% CI 0.81-1.54; P = 0.019 for interaction).

Timing of the last dose of ticagrelor 60 mg appeared to have no effect on the P2Y12 inhibitor’s efficacy.

The trade-off to using ticagrelor 60 mg was increased TIMI major bleeding compared with the placebo arm, with the greatest hazard seen within the first year of treatment (HR 3.22) relative to year 2 (HR 2.07) and year 3 (HR 1.65). Not surprisingly, bleeding was even higher with ticagrelor 90 mg.

Ticagrelor Not ‘One-Size-Fits-All’

The goal of this study was to see if ticagrelor lends consistent benefit to patients with prior MI over time or if there is a point when the benefit attenuates enough to warrant stopping treatment, Bonaca told TCTMD. The results were expected except for the decreased bleeding hazard observed through 3 years, he commented, adding that this is likely partially due to selection bias (ie, only patients who tolerated ticagrelor remained on it throughout the study). Roughly one-third of patients treated with each ticagrelor dose in PEGASUS-TIMI 54 stopped treatment for various reasons including bleeding.

“The takeaway from this would be to continue on [with ticagrelor if patients] haven't had any bleeding or tolerability issues, and that the plan would be to continue on [it] long-term, provided they continue to tolerate the drug or don't develop a contraindication,” Bonaca said.

Commenting on the study, Mamas Mamas, BMBCh (Keele University, Stoke-on-Trent, England), told TCTMD that he was most struck by the continuing linear risk of events observed in this patient population over 5 years. “Certainly a lot of individuals think that once you’ve had an MI, there's an inflection point at some point where the risk reduces,” he said. “But this doesn't seem to be the case.”

While the data show some value to continuing ticagrelor therapy in the longer term, it comes “obviously at the expense of increased major bleeding complications,” Mamas observed. “What this really tells us is that prolonged ticagrelor isn’t a one-size-fits-all because here we really are balancing the risk of ischemia versus major bleeding complications.”

What is really needed is some sort of risk stratification tool similar to the DAPT Score for PCI patients that can help physicians make informed decisions for those post-MI, he said. “It’s very clear that there will be patients that will benefit, but there are also patients who are going to get harm from this approach.”

There’s also the new COMPASS trial to consider, said Mamas, particularly in the context of individualized patient risk/benefit decisions. “Here we have a situation where one trial is saying, ‘Use prolonged ticagrelor in these patients and you'll have better outcomes.’ [And] the COMPASS trial, albeit a very heterogeneous cohort at high risk of future cardiovascular events, which post-MI patients are, shows that adding rivaroxaban also reduces the risk of future ischemic events,” he observed. “So you’ve got a situation where, what evidence do you take?”

Bonaca responded that “COMPASS is very interesting, but the lack of a P2Y12 inhibitor [in the study] makes it a bit challenging in the coronary space. How do you switch and when? What if they need a PCI? Do you switch back?”

Perhaps a head-to-head study of the two drugs in a post-MI population would be the next logical step, Mamas added.

Relative vs Absolute Risk

In an accompanying editorial, John Bittl, MD (Munroe Regional Medical Group, Ocala, FL), and David Maron, MD (Stanford University School of Medicine, CA), agree that there is a benefit to extending low-dose ticagrelor in patients who can tolerate it over time, but they point to unanswered questions. For example, “are the 10% to 21% reductions in ischemic events worth the 1.7- to 3.2-fold increases in major bleeding?” they ask. “Trialists often use relative risks to summarize study results, but clinicians prefer to use absolute event rates to measure and compare various therapies.”

To help clinicians measure events during each year of therapy, Bittl and Maron calculated an annual number needed to treat (NNT) of 168 patients to prevent one cardiovascular death, MI, or stroke, which “might seem like a lot of effort,” they say. By way of comparison, treating high blood pressure to prevent one event has an annual NNT of 14.

“It’s important for investigators and for reviewers to use the same NNT measurement,” Bittl told TCTMD. “Since we’re treating an overall population rather than subgroups, I think when we analyze a study it might be a little clearer just to use the overall population like we did. On the other hand, in practice when you're trying to assess a patient's risk for an ischemic event—say somebody closer to their MI who might be at higher risk—then it might make sense under those circumstances to base a NNT on a subgroup analysis, but that brings another set of factors that might be criticized.”

Bonaca said that he and the editorialists are not on two different sides of an argument. “These decisions are complicated, but the benefit is clearly there,” he said. “Both the paper and editorial say the same thing. The benefits are long-term whether or not you believe the magnitude is compelling.”

Future research should focus on continued methods “to help clinicians decide for whom to treat longer with ticagrelor,” Bonaca concluded. A study of ticagrelor use in Europe will likely be published before the end of the year, he said, and ongoing work on novel predictors like biomarkers should also advance the field. “There may be ways of coupling biomarkers with clinical characteristics to help clinicians pick out the populations that are going to get the greatest benefit,” he said.

  • The TIMI Study Group has received research grant support from Accumetrics, Amgen, AstraZeneca, Beckman Coulter, Bristol-Myers Squibb, CV Therapeutics, Daiichi-Sankyo, Eli Lilly, GlaxoSmithKline, Integrated Therapeutics, MedImmune, Merck, Nanosphere, Novartis Pharmaceuticals, Nuvelo, Ortho-Clinical Diagnostics, Pfizer, Roche Diagnostics, Sanofi, Sanofi-Synthélabo, Siemens Medical Solutions, and Singulex.
  • Bonaca reports serving as a consultant for AstraZeneca, Merck, Aralez, and Bayer and receiving grant support from AstraZeneca.
  • Mamas reports receiving an educational grant from Daiichi Sankyo and consulting fees from AstraZeneca unrelated to ticagrelor.
  • Bittl and Maron report no relevant conflicts of interest.

We Recommend