TWILIGHT Hits Primary Endpoint, Alters Aspirin Post PCI

Patrice Wendling

September 26, 2019

SAN FRANCISCO — Dropping aspirin after 3 months and continuing ticagrelor monotherapy reduced bleeding without a price to pay in overall ischemic events among patients undergoing percutaneous coronary intervention (PCI) at high risk for either complication, results from the TWILIGHT trial show.

Ticagrelor (Brilinta, AstraZeneca) monotherapy was superior to ticagrelor plus aspirin for the primary endpoint of BARC 2, 3, or 5 bleeding and was noninferior for the key composite ischemic endpoint — providing the strongest support thus far for a strategy of early aspirin discontinuation after PCI.

In presenting the late-breaking results at Transcatheter Cardiovascular Therapeutics (TCT) 2019, Roxana Mehran, MD, Icahn School of Medicine at Mount Sinai, New York City, said reducing aspirin duration after PCI may allow for more prolonged use of potent P2Y12 inhibitors while avoiding aspirin-related bleeding.

"One single trial doesn't change the guidelines but the totality of evidence in the same direction might change clinical practice," she told | Medscape Cardiology. "I think what we show in TWILIGHT is that, in this patient population, a simpler strategy of ticagrelor monotherapy can reduce bleeding significantly — 45% to 50% — and not showing a signal of harm, in such a large population, is really reassuring."

Commenting to | Medscape Cardiology, Davide Capodanno, MD, PhD, University of Catania, Italy, said, "I cannot imagine what will be the recommendation of the task forces but, at the same time, I think this [strategy] should enter clinical practice for some selected patients because there is now the proof in a very well-designed and conducted study [that] it reduces BARC 3 and 5, which is a very bad bleed."

The decision to randomly assign patients at 3 months, he said, is "methodologically very correct and different from the previous studies," and provides a clear picture of the differential effect of dropping aspirin vs remaining on dual antiplatelet therapy (DAPT).

"I think the correct interpretation is that you will do DAPT for 3 months and then you assess the bleeding risk and the thrombotic risk and decide whether to drop aspirin or not, because the bleeding risk is higher than the thrombotic risk," Capodanno said. "It's kind of a call to look at these patients at 3 months."

During a discussion of the results, session moderator Gregg W. Stone, MD, professor of medicine (cardiology), Icahn School of Medicine at Mount Sinai, and director of academic affairs, Mount Sinai Heart, agreed: "I believe you did it right by randomizing at 3 months because we shouldn't be making our decisions at time zero about what we're going to give our patients," he said. "We have to see how they do in the first 3 months, if they're tolerating medications, if they've had recurrent events, if they're able to be compliant on medications."

"I'll also say it doesn’t apply to one-month dual antiplatelet therapy, which has been obviously a real move toward going shorter and shorter, especially when patients are at high-bleeding risk," he said, noting the "less overwhelming" results of the GLOBAL LEADERS trial, in which extended ticagrelor monotherapy with aspirin added for just the first month after PCI failed to reduce bleeding at 2 years and narrowly missed its primary endpoint.

High-risk Population

The TWILIGHT trial, which was published simultaneously in the New England Journal of Medicine, enrolled patients who underwent successful PCI with at least one drug-eluting stent and had completed 3 months of DAPT (ticagrelor 90 mg twice daily and aspirin 81-100 mg daily) without a major bleeding or ischemic event.

Patients also had to have at least one clinical and one angiographic feature that put them at high risk for these events. Clinical features included age at least 65 years, female sex, troponin-positive acute coronary syndrome, established vascular disease, diabetes treated with medication, and chronic kidney disease. Angiographic criteria included multivessel coronary disease, stent length more than 30 mm, a thrombotic target lesion, a bifurcation lesion requiring two stents, an obstructive left main or proximal left anterior descending lesion, and a calcified target lesion requiring atherectomy.

Of the 9006 patients enrolled at 187 sites in 11 countries, 7119 were randomly assigned in a double-blind fashion to receive aspirin or placebo for an additional 12 months plus open-label ticagrelor. The two groups were well balanced, with a mean age of 65 years, 23.8% were female, 36.8% had diabetes, and 64.8% underwent PCI for acute coronary syndrome. Mean total stent length was 40 mm.

At 1 year, the primary endpoint of BARC 2, 3, or 5 bleeding had occurred in 4% of the ticagrelor monotherapy group and in 7.1% of the ticagrelor/aspirin group (hazard ratio [HR], 0.56; 95% confidence interval [CI], 0.45 - 0.68; P < .001). The number needed to treat was 33.

The benefit was consistent across all subgroups, including the important subgroup with acute coronary syndrome, Mehran said.

Further, similar reductions were observed across four additional prespecified bleeding endpoints:

  • BARC 3 or 5 bleeding: 1% vs 2% (HR, 0.49; 95% CI, 0.33 - 0.74)

  • TIMI major: 0.5% vs 1% (HR, 0.50; 95% CI, 0.28 - 0.90)

  • GUSTO moderate/severe: 0.7% vs 1.4% (HR, 0.53; 95% CI, 0.33 - 0.85)

  • ISTH major: 1.1% vs 2.1% (HR, 0.54; 95% CI, 0.37-0.80)

The key composite ischemic endpoint of all-cause death, myocardial infarction, or stroke occurred in 3.9% of patients in the two study groups in the per protocol cohort (-0.06% difference; HR, 0.99; 95% CI, 0.78 - 1.25; P noninferiority < .001).

There was a numeric but nonsignificant increase in ischemic strokes in the ticagrelor monotherapy group (16 vs 8 events; HR, 2.00; 95% CI, 0.86 - 4.67) and a nonsignificant decrease in mortality (34 vs 45; HR, 0.75; 95% CI, 0.48 - 1.18).

"There's no question that we are not powered for those rare events," Mehran said during a media briefing. "It should be noted that all strokes were adjudicated by board-certified neurologists, so we probably captured more than you would have imagined."

Thrombogenicity Substudy

In a late-breaking clinical science session, Usman Baber, MD, also with the Icahn School of Medicine, presented results of a thrombogenicity substudy embedded within TWILIGHT among 18 patients in the ticagrelor group and 24 patients in the ticagrelor/aspirin group who underwent whole blood testing at randomization and at a second visit 1 to 6 months later.

The mean difference in ex-vivo platelet-dependent thrombus area between the two study groups was 218.2 µm2, indicating no significant difference in thrombus size, as measured using the validated Badimon perfusion chamber (P = .22).

Changes in thrombus area from baseline to follow-up were also similar in the ticagrelor monotherapy (3130.7 µm2 vs 3152.4 µm2P = .87) and ticagrelor/aspirin (3741.5 µmvs 3903.8 µm2P = .16) groups, despite overall higher levels of thrombus in the latter group, likely due to older age and higher rates of diabetes, he said.

Platelet reactivity to collagen (40.2 U vs 35.9 U; P = .03) and arachidonic acid (19.4 U vs 11.6 U; P = .02) was increased in the absence of aspirin, while aggregation to adenosine diphosphate (13 U vs 16.8 U; P = .47) and thrombin receptor activating peptide (77 U vs 79.1 U; P = .81) was unchanged with or without aspirin.

"These findings suggest that aspirin withdrawal does not modulate ex-vivo blood thrombogenicity in the presence of strong P2Y12 blockade with ticagrelor and corroborates the clinical observations of no incremental risk upon aspirin withdrawal seen in TWILIGHT," Baber concluded.

Speaking to | Medscape Cardiology, Robert Harrington, MD, Stanford University, California, said while "there is no difference in the composite ischemic endpoint, stroke is numerically higher and cannot be ignored, even with the mechanistic data."

Nevertheless, he said, "based on the overall evidence, I think it's very reasonable to down-titrate to a single antiplatelet agent and using just the ADP [adenosine diphosphate receptor] inhibitor seems very rational given everything we know."

The study was funded by AstraZeneca. Mehran reports grants from AstraZeneca, Bayer, Beth Israel Deaconess, Bristol-Myers Squibb, CSL, DSI, Medtronic, Novartis, Orbus Neich, Osprey Medical, PLC/Renal Guard, and Abbott Vascular; and personal fees from Abbott Vascular, Boston Scientific, Medscape/Web MD, Siemens Medical Solutions, Philips/Volcano/Spectranetics, Roviant Sciences, Sanofi, Bracco Group, Janssen, Watermark Research Funding, Medintelligence (Janssen), ACC, Bayer, and AstraZeneca. Baber reports consulting fees/honoraria from AstraZeneca and Boston Scientific. Harrington has disclosed no relevant financial relationships.

Transcatheter Cardiovascular Therapeutics (TCT) 2019: Presented
September 26, 2019.

N Engl J Med. Published online September 26, 2019. Abstract

Follow Patrice Wendling on Twitter: @pwendl. For more from | Medscape Cardiology, join us on Twitter and Facebook


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.