Full GALILEO Results Confirm Harm With Rivaroxaban After TAVR

PHILADELPHIA, PA—Rivaroxaban (Xarelto; Bayer/Janssen) is not the answer to the question of how best to treat patients after a successful TAVR, the full results of the prematurely stopped GALILEO trial affirm.

As seen in a preliminary analysis that brought the study to a halt more than a year ago, the rate of death or thromboembolic events—the primary composite efficacy outcome—was higher with rivaroxaban-based therapy than with antiplatelets (12.7% vs 9.5%; HR 1.35; 95% CI 1.01-1.81). Death alone was significantly increased as well (7.7% vs 4.6%; HR 1.69; 95% CI 1.13-2.53).

Serious bleeding, too, was more frequent in the anticoagulation arm. Although the primary composite safety outcome of VARC-2 major, disabling, or life-threatening bleeding was only borderline significantly increased (5.6% vs 3.8%; HR 1.50; 95% CI 0.95-2.37), there were significant differences in VARC major, TIMI major/minor, ISTH major, and BARC type 2, 3, or 5 bleeding.

The findings were reported here at the American Heart Association (AHA) 2019 Scientific Sessions by George Dangas, MD, PhD (Icahn School of Medicine at Mount Sinai, New York, NY), one of the principal investigators, and published simultaneously online in the New England Journal of Medicine.

“In patients without an indication for oral anticoagulation after TAVR, a 10-mg daily rivaroxaban-based antithrombotic strategy in the GALILEO trial was associated with higher risk of death or thromboembolic events and bleeding compared to an antiplatelet-based strategy,” he concluded. Dangas added that the mechanism underlying the increase in death, which was primarily driven by noncardiovascular mortality, isn’t explained by the excess of other clinical events or bleeding and remains uncertain.

Routine oral anticoagulation . . . post-TAVR in patients who don’t otherwise have an indication for oral anticoagulation cannot be recommended at the present time. Howard Herrmann

The lack of benefit—and indeed, the harm—of a rivaroxaban-based strategy was seen despite the findings of a 231-patient four-dimensional CT substudy—GALILEO-4D, also presented at the AHA meeting and published in NEJM—showing that the direct oral anticoagulant (DOAC) significantly mitigated subclinical leaflet thickening and reduced leaflet motion, which have been associated with worse clinical outcomes in some prior studies, 90 days after TAVR.

Speaking with TCTMD, study investigator Howard Herrmann, MD (University of Pennsylvania, Philadelphia), said the message is clear: “Routine oral anticoagulation, whether with a vitamin K antagonist or a direct oral anticoagulant, post-TAVR in patients who don’t otherwise have an indication for oral anticoagulation cannot be recommended at the present time.”

The study, he added, “emphasizes the risks of even a low-dose anticoagulant in elderly patients with multiple comorbidities.”

Still Searching for an Answer

Current guidelines recommend dual antiplatelet therapy after TAVR. However, the optimal treatment strategy for patients undergoing the procedure remains unclear. Previous observational studies have suggested that oral anticoagulation may lower the risk of subclinical leaflet thrombosis, which has been tied to cerebrovascular events, that develops after TAVR.

GALILEO, an open-label trial conducted at 136 sites in 16 countries in North America and Europe, was designed to test rivaroxaban in this setting. Within a week of successful TAVR, investigators randomized 1,644 patients (mean age 80.6 years; 49.5% women) without an established indication for oral anticoagulation to a rivaroxaban-based or antiplatelet treatment regimen. In the former group, patients received rivaroxaban 10 mg daily plus aspirin 75 to 100 mg daily for 3 months before continuing with rivaroxaban alone. In the antiplatelet group, patients received aspirin plus clopidogrel 75 mg daily for 3 months before continuing with aspirin alone.

The rivaroxaban-based strategy worsened clinical outcomes in addition to bleeding, showing that the field will have to look elsewhere in the search for effective antithrombotic regimens following TAVR.

That search is ongoing, as several trials are exploring use of oral anticoagulation after TAVR in patients who have other indications for such treatment, including POPular-TAVI, ATLANTIS, ENVISAGE-TAVI AF, and AVATAR.

We still don’t have a solid foundation beyond dual antiplatelet therapy that we’re currently using for best treatment and thankfully trials like this help add to our knowledge. Kim Eagle

Commenting for TCTMD, Kim Eagle, MD (University of Michigan, Ann Arbor), said the issue of subclinical valvular thrombosis after TAVR and SAVR has gotten a lot of attention, with observational studies suggesting that it was both relatively common and possibly associated with worsened clinical outcomes. “That really begged for trials like GALILEO to start answering that question,” he said.

The findings of the main GALILEO trial and the CT substudy, which was presented by Ole De Backer, MD, PhD (Rigshospitalet, Copenhagen University Hospital, Denmark), are discordant, Eagle noted, in that rivaroxaban lessened leaflet problems but worsened clinical outcomes.

“We still don’t know for sure how common this is, how important it is,” Eagle said about valvular thrombosis. “We still don’t have a solid foundation beyond dual antiplatelet therapy that we’re currently using for best treatment and thankfully trials like this help add to our knowledge. So far rivaroxaban at 10 milligrams doesn’t look like the answer.”

For now, he said, the standard treatment approach should remain dual antiplatelet therapy.

An Unmet Need

Elaine Hylek, MD (Boston University School of Medicine, MA), who served as a discussant following Dangas’ presentation, said that there was an unmet need providing a strong rationale for the GALILEO trial. The optimal antithrombotic strategy after TAVR is currently unknown, resulting in practice variation and discrepancy among guidelines, she said.

“The risk of stroke . . . remains high early and late post-TAVR despite antiplatelet therapy, which tells us there must be something else going on,” Hylek said.

She indicated that there are a host of unresolved questions after GALILEO revolving around patient population; dosing strategy; the timing of intervention; whether there is any relationship of thromboembolic events and death with treatment interruption or discontinuation or physician response to bleeding; the incidence of subclinical paroxysmal A-fib; and whether there is a link between incident A-fib after TAVR and bleeding related to dose escalation and subsequent treatment cessation.

The risk of stroke . . . remains high early and late post-TAVR despite antiplatelet therapy, which tells us there must be something else going on. Elaine Hylek

Hylek said it’s too early to state what the implications of the trial are, adding that she is eagerly awaiting other trials of anticoagulation in the post-TAVR setting.

Ultimately, she said, “at the end of the day, [in] this very medically complicated, high-risk patient population with a mean age of 80 years, combination therapy may simply not be feasible.”

Herrmann also indicated that there are areas related to oral anticoagulation remaining for possible exploration. “Whether a short-term strategy, 3 months or 6 months, is beneficial could warrant further study, though I have to say I’m a little pessimistic based on the early separation of the curves that was observed in GALILEO,” he said.

“I think at best probably we need to study some selective strategies that involve selection of patients based on echocardiographic criteria, such as an increase in gradient, that then prompt a CT that then might prompt anticoagulation, although even that requires study since we know that many of these increased gradients and decreases in mobility go away spontaneously,” Herrmann said.