‘Speaking the Same Language’: New Guidance for Trials of Patients at High Bleeding Risk

An academic research consortium is proposing new recommendations to guide the design of future pharmacological and device studies in patients at high risk for bleeding undergoing PCI, a population that has historically been excluded from randomized controlled trials.

The new consensus document, from the Academic Research Consortium-High Bleeding Risk (ARC-HBR) group, proposes standardized primary and secondary clinical outcomes, including key ischemic/thrombotic and bleeding endpoints, and recommends consistent comparator groups when feasible and appropriate. It also discusses the optimal duration of clinical trials.

Davide Capodanno, MD, PhD (University of Catania, Italy), lead author of the new report, said this is the second initiative of the ARC-HBR group, noting that the panel, which includes researchers from Europe, the United States, and Asia, along with input from regulatory agencies, proposed a consensus HBR definition to standardize the patient population in clinical trials. The ARC-HBR listed 20 major and minor clinical variables, with patients considered high risk for bleeding—defined as a BARC 3 or 5 bleeding risk ≥ 4% and/or a ≥ 1% risk of intracranial hemorrhage at 1 year—if they meet one major or two minor criteria.

“Here, there was a need to define the principles of trial design for patients at high bleeding risk, because now there are multiple clinical trials in this population,” Capodanno told TCTMD. “The first document from ARC-HBR dealt with defining these patients, and the second document is about standardizing the designs of these trials. At this moment, the trials that are out there differ in many aspects, including the type of population, control therapies, study design, primary endpoint, secondary endpoint, and so on.”

Roxana Mehran, MD (Icahn School of Medicine at Mount Sinai, New York, NY), a member of ARC-HBR and senior author of the report, said that by first standardizing the patient population included in HBR trials and now providing recommendations for the design of those clinical trials, it allows the field “to all be speaking the same language.” In their paper, the ARC-HBR group looked at the methodological differences in completed HBR trials and reviewed some of the trials that are underway or in the planning stages.

“The consensus-based document is intended to provide good recommendations to guide future clinical trials in this space,” said Mehran. “These consensus definitions will help have more consistency in the trials going forward.”

Some of the completed randomized trials in HBR patients include LEADERS-FREE, ZEUS-HBR, SENIOR, ONYX ONE, and DEBUT, most of which tested different devices against BMS (ONYX ONE tested a durable-polymer DES against a polymer-free drug-coated stent). Completed drug trials include WOEST, ISAR-TRIPLE, and AUGUSTUS, among others. The MASTER DAPT trial, which is comparing 1 versus 6-12 months of dual antiplatelet therapy (DAPT) in 4,300 patients is ongoing, as is the COBRA-REDUCE trial and several single-arm studies.

The new recommendations were published September 14, 2020, in the Journal of the American College of Cardiology.

Eliminating Confusion

The ARC-HBR group recommends researchers compare device interventions against standard care—current-generation DES—as the control arm. However, they acknowledge that a performance goal or objective performance criteria may be justified in single-arm trials where randomization is not feasible. With medical therapy, the goal is to reduce bleeding without increasing the risk of ischemic events. The ARC-HBR group acknowledges that comparative assessments of antithrombotic strategies can be difficult given the complexity of treatment, noting that the optimal duration of DAPT in HBR is unknown. They state that rapid changes in clinical practice mean there is a need for flexible standard-of-care definitions (and control therapies).   

In an attempt to promote consistency in the design of clinical device and drug trials in HBR patients undergoing PCI, the group recommends using a modified clinical endpoint previously defined by the ARC. For stent trials, they recommend using an ischemic/thrombotic endpoint that includes cardiovascular death, target-vessel MI, and clinically driven target lesion revascularization. For drug therapies, they recommend an endpoint that includes cardiovascular death, any MI, and ischemic stroke, but note that adding peripheral arterial embolism or any revascularization to the composite endpoint may allow drug trials to reduce the size of the study.

To TCTMD, Capodanno said one of the report’s novel aspects is its introduction of ischemic/thrombotic and bleeding events as two co-primary endpoints as opposed to labeling them as efficacy and safety. “Using traditional terms like safety and efficacy can be a little bit misleading, because interventions aimed at bleeding are not really about safety, but rather the efficacy of the intervention to reduce bleeding,” he said. “So we propose this terminology to avoid any ambiguity.”

Mehran said it has been difficult to compare clinical trials given the lack of standardized bleeding and ischemic/thrombotic endpoint definitions, but that research community has been receptive to the idea of consistency. “It’s important if we’re comparing one device to another device that we’re actually talking about the exact same patient,” she said. “This way, the comparison across trials would be apples to apples rather than apples to oranges. On the drug side, it’s also very important because there are so many different approaches.”

Gaps in Knowledge

With regard to how long trials should last, the timing of primary endpoint evaluation is recommended at 12 months at least, or possibly longer depending on the individual trial objectives, according to the ARC-HBR group. Secondary endpoints, which could include definite or probably stent thrombosis (for device trials), BARC 1 or 2 bleeding, hospitalizations related to thromboembolic/ischemic events or bleeding, patient-related quality-of-life metrics, and all-cause mortality, should also be reported at a minimum of 12 months.    

Capodanno hopes these design principles will make it easier to compare clinical trials and to establish standards that result in achieving excellent results—be they positive or negative—so that patient care can be the best it can be. In terms of future steps, Capodanno said a large trial testing the optimal DAPT duration in HBR patients is among the most-pressing issues. “If I had to wish for the next step, it would be a collaboration that would lead to a landmark trial that would be specific for HBR patients,” he said.

For Mehran, there’s a need to include more-diverse patients in future HBR studies. “All of these trials don’t have enough of the underrepresented minorities and women,” she said. HBR patients are really challenging, said Mehran, and “if we don’t have the socioeconomic status, race, and gender issues incorporated into these clinical trials we’re going to be very behind. That, to me, is number one for future trials, for any trial, but specifically in these [HBR] trials.”

Like Capodanno, Mehran said studies testing the best DAPT duration in HBR patients are critical. “Is it 1 month, is it 3 months, or is it 6 months?” she asked. “Is it different in ACS versus patients in stable ischemic heart disease? What happens when we drop aspirin? What do you when a patient requires oral anticoagulants? Do we drop aspirin immediately? Or do we go to a short duration of triple therapy? There are so many unanswered questions.”

Mehran added that a pragmatic trial design, one that enrolls quickly, is needed so that clinicians get answers to these critical questions soon, “not in 10 years when everything will change.”