麻豆传媒

SAN FRANCISCO -- Cutting aspirin from the ticagrelor (Brilinta)-based dual antiplatelet therapy (DAPT) regimen after 3 months was safer for high-risk stent recipients, the TWILIGHT trial showed.

Continuing ticagrelor with placebo in place of aspirin after that point reduced clinically-relevant bleeding -- adjudicated as Bleeding Academic Research Consortium (BARC) type 2, 3, or 5 events -- by a relative 44% compared with both agents at 1 year (4.0% vs 7.1%, P<0.001). The number needed to treat was 33.

On the key secondary endpoint of composite all-cause mortality, nonfatal MI, or nonfatal stroke, both groups came in at 3.9% in the per-protocol analysis, which met criteria for noninferiority, Roxana Mehran, MD, of Mount Sinai Hospital in New York City reported at a late-breaking clinical trial session here at the Transcatheter Cardiovascular Therapeutics meeting.

"Although previously considered relatively benign, post-PCI bleeding has been shown to be associated with a substantial and durable risk of death, approximating or even exceeding that associated with myocardial infarction," her group wrote in a paper simultaneously released in the .

"This is probably the strongest trial to date that suggests that we can de-escalate our dual antiplatelet therapy fairly early on, after 3 months, and actually have improved outcomes," commented Hadley Wilson, MD, of Sanger Heart & Vascular Institute in Charlotte, North Carolina.

While it's no surprise that one antiplatelet agent led to less bleeding than two, it adds to the evidence for peeling back DAPT, agreed Robert Harrington, MD, of Stanford University in California and president of the American Heart Association.

Even though somewhat underpowered for ischemic events due to low rates, it was sufficient to rule out a clinically relevant difference in a "very clinically relevant population," commented Michael Rinaldi, MD, also of Sanger Heart & Vascular Institute and an investigator on the 2014 DAPT trial that showed a benefit to 30 months of thienopyridine plus aspirin after stenting.

A 51-patient thrombogenicity substudy presented at the same session by Usman Baber, MD, also of Mount Sinai, showed no significant difference in ex-vivo platelet-dependent thrombus area, nor in platelet reactivity to thrombin or adenosine diphosphate (ADP), between the time on DAPT vs ticagrelor alone.

That finding "corroborates the clinical observations of no incremental ischemic risk upon aspirin withdrawal seen in TWILIGHT," Baber told reporters at a press conference.

The difference in reactivity to collagen and arachidonic acid versus thrombin and ADP after going off aspirin "really does help tell the story that ticagrelor plus aspirin is mechanistically different from ticagrelor alone," and suggests the mechanism provided by the P2Y12 inhibitor alone is sufficient, Harrington said. "I don't want to overread this, but at least it's nice that it's consistent."

Confirmatory data clarifying whether this is a class effect across the P2Y12 receptor inhibitors, across the , or for certain drugs more than others, would likely be necessary before changing the guidelines, Wilson told 麻豆传媒.

The new findings contrast with those of the open-label GLOBAL LEADERS trial, which showed 1 month of DAPT followed by 23 months of ticagrelor alone after a biodegradable-polymer stent wasn't better than a standard regimen for bleeding risk in all-comers.

While benefit in one trial and no harm in the other may be sufficient to shift practice, cost is likely to be a big factor, Rinaldi suggested. Ticagrelor is not only a lot more expensive than generic clopidogrel and prasugrel, but it's not as tolerable due to dyspnea and twice-daily dosing, he added.

"It's going to be harder to put everybody on ticagrelor, and a lot of people will resist that because the important thing is compliance," Rinaldi pointed out. If patients stop ticagrelor monotherapy due to expense or adverse effects, they're left with no antiplatelet at all.

However, Harrington noted that the drug may well be cost-effective given the savings associated with bleeding events that were prevented.

Determining whether to extrapolate the findings to clopidogrel (of which some patients are poor metabolizers) or prasugrel is going to be a "tough calculus," he predicted. "This gives us some freedom to get rid of aspirin in high bleeding risk patients, but we'll also have to figure out, is ticagrelor feasible?"

"I think we're going to see a number of trials beyond this with shorter durations of DAPT," Wilson said. "We may see more trials now trying to whittle that 3-month period back down to a 1-month, but we're going to need to see that it won't be at the expense of ischemia in these patients."

TWILIGHT included 9,006 patients for whom clinicians had already decided on ticagrelor plus aspirin as their DAPT regimen after drug-eluting stent placement. The 7,119 without a major bleeding event, stroke, MI, or coronary revascularization by 3 months after discharge were randomly assigned to open-label ticagrelor (90 mg twice daily) and double-blind treatment with enteric-coated aspirin (81 to 100 mg daily) or matching placebo for an additional 12 months.

High-risk was defined by anatomic criteria (multivessel or left main disease, calcified or thrombotic lesions, total stent length >30 mm, or certain bifurcation lesions requiring at least two stents) along with at least one clinical criteria: age ≥65, female, troponin-positive acute coronary syndrome, preexisting vascular disease, diabetes, or chronic kidney disease.

Ticagrelor monotherapy reduced the incidence of major bleeding (BARC type 3 or 5) by a relative 51% compared with continued DAPT (1.0% vs 2.0%).

For other secondary endpoints, safety was statistically similar with placebo versus aspirin:

• All-cause mortality (1.0% and 1.3%, respectively)
• MI (2.7% in both groups)
• Definite or probable stent thrombosis (0.4% and 0.6%)
• Ischemic stroke (16 vs eight cases, 0.5% and 0.2%)

Harrington questioned the lack of consistency on stroke risk, but Mehran countered that this may have been the play of chance for these carefully adjudicated events in a population selected for no prior stroke. She pointed out that mortality, on the other hand, was numerically lower with ticagrelor monotherapy.

Results were consistent across predefined subgroups.

Ticagrelor adherence at 1 year post-randomization was similar at 86%-87% in the two arms.

Comment