Experts Ponder Expanding Transcatheter Valves Indication to Lower-Risk Patients

Reed Miller

February 25, 2010

February 25, 2010 (Washington, DC) — As surgeons and interventionalists await the results of the pivotal Placement of Aortic Transcatheter Valve Trial (PARTNER) of the Sapien (Edwards Lifesciences) transcatheter aortic-replacement valve in patients with high surgical risk, some have already begun looking forward to the day when the therapy will supplant surgery for lower-risk patients.

Here at CRT 2010, the annual Cardiovascular Research Technologies meeting held this week, a group of physicians and industry and FDA scientists discussed the possibility of transcatheter aortic-valve implantation (TAVI) becoming the first-line choice for aortic replacement and the clinical trials that will be needed to convince physicians and regulators that the technology is appropriate for that wider population.

PARTNER is comparing TAVI with open valve surgery in patients whose predicted risk of operative mortality is >15% and comparing TAVI with optimal medical management in patients whose probability of death or serious, irreversible morbidity exceeds 50%. But even without seeing the results of PARTNER, the panelists at the CRT meeting agreed that minimally invasive valve-repair systems such as Sapien will eventually be used in patients with lower surgical risk.

Surgery has to be considered as a morbidity event in itself. It's perceived as traumatic because it is traumatic.

"The general [consensus] in the surgical community is that [transcatheter valve technology] is a game changer," cardiac surgeon Dr Michael Mack (Medical City Dallas Hospital, TX) said. He said that there are some surgeons who are resistant to the shift from open surgery to a less invasive approach for valve replacement, but most of his colleagues think transcatheter valves will be to valve surgery what PCI has been to CABG surgery. Dr Eberhard Grube (Helios Heart Center, Siegburg, Germany) recalled that "we had a similar discussion about PCI and surgery--it was furious and opinionated, but today we don't discuss this anymore, because today PCI is an excellent treatment [alternative to] surgery."

"The way this has rolled out as a collaborative approach [between surgeons and interventionalists] has been an education for that. I think that surgeons are much more open-minded than they were when [PCI was starting]," Mack said.

Grube said that, based on what is already known about the materials and design of percutaneous valves, there is no reason to think that they will not eventually be as durable as surgically implanted valves. And, as more long-term data accumulate and the technology improves, intravascular aortic valves should supplant invasive surgical replacement in a wider patient population, who will have a better quality of life and faster recovery with a less invasive option. "Surgery has to be considered as a morbidity event in itself. It's perceived as traumatic because it is traumatic."

Matt Hillebrenner, an engineer in FDA's Center for Devices and Radiological Health, pointed out that as the device is tested in less sick patients, the importance of mortality as an end point will diminish, and therefore perhaps a composite end point that includes quality of life or other comorbidities will be more appropriate.

Are More RCTs Needed, or Can Doctors Just Figure It Out?

The panelists agreed that the patients' desire for less invasive procedures will put pressure on physicians to offer TAVI to a wider patient population than just the high-risk patients represented in PARTNER. Dr Howard Herrmann (University of Pennsylvania, Philadelphia) said, "There is resistance to sternotomy. If you tell them [the incision is] coming down the middle of their chest, they don't want it. But as physicians we need data to be able to advise the patients when they're making stupid choices," he said. Of course, some options--like a 50-year-old choosing a replacement valve that will last only 10 years would be so bad that the surgeon wouldn't offer that to the patient--but in other cases, the patient may choose a less invasive procedure that may carry slightly more long-term risk, he explained. "We do need data from randomized trials to be their advocate and allow them to make the choice that they want. It may turn out that the data show that one option isn't as good as the other. They can still make that choice, but they have to have that information."

As physicians we need data to be able to advise the patients when they're making stupid choices.

Most of the panel agreed that TAVI devices should be tested in randomized clinical trials in the lower-risk population. Dr William H Maisel (Beth Israel Deaconess Medical Center, Boston, MA), a former chair of FDA's Circulatory System Devices Advisory Panel, said it is "striking that there's unanimity for randomized controlled trials of these valves. I don't think that was the case even a few years ago."

However, cardiologist Dr Peter Block (Emory University, Atlanta, GA) suggested that once a device is approved as safe and effective for the high-risk population, additional randomized trials may not be necessary for less risky patients. "Why can't we as physicians use our judgment? . . . Granted, we will probably hurt some people, but we will also help some people whom otherwise we could not help because the randomized trial would not allow us to do so [because the inclusion criteria for randomized trials are strict]."

Block agreed that patients in their 50s and 60s who need a new valve have a much different pathophysiology than the much older patients currently being enrolled in PARTNER, so that population would definitely require a new randomized trial. But he pointed out that clinical data already show that Sapien is durable for at least seven years with no indications of falling apart, so he suggested that it could be safely implanted in, for example, a 70- or 75-year-old patient with a 4% operative risk and 20-year life expectancy, a patient who would currently be excluded from the PARTNER trial. "We don't have to prove what we think we already know."

Balance of Postmarket and Premarket Data Collection

Other panelists disagreed. "I'm not sure that we're smart enough to develop [the technology] just by experience without rigorous study," Dr Michael Domanski (National Institutes of Health, Bethesda, MD) argued.

Maisel thinks Block's aggressive approach is "half right."

"It always amazes me how good the cardiologists and surgeons participating in these trials are at walking into a room and figuring out what their patient's mortality is," he said. "Some of the risk scores we use . . . don't accurately pick up on what a physician can see at the bedside."

However, Maisel points out that, in the past, devices have been approved for one population and "physicians have strayed off course into off-label use of devices--not that all off-label use is bad--but here's a place where we could really take a left-hand turn and have the device implanted off-label into thousands or tens of thousands of patients. I'd be concerned about going head first into an off-label territory."

He cited the sudden drop-off in drug-eluting-stent sales that followed the discovery that they carried a higher risk of stent thrombosis as an example of the market overreacting to a safety concern that could have been contained better if it had been detected in the premarket studies instead of in off-label patients after the device was approved. A similar problem could arise with TAVI devices.

Maisel agreed, however, that not every potential population should be subject to a randomized trial and that well-designed and carefully analyzed postmarket registries can provide a lot of important information about off-label indications.

We're not going to be able to not do randomized trials.

Mack agreed that some balance between reliance on randomized trials and postmarket registries will be the key to moving the technology forward into new patient populations. "The reality is that we're not going to be able to not do randomized trials. But, regardless, enhanced postmarket surveillance is going to be absolutely necessary." He cited the Interagency Registry for Mechanical Assisted Circulatory Support (INTERMACS), a government-sponsored registry for patients with ventricular assist devices, as the type of comprehensive registry that ought to be constructed for transcatheter-valve implants. He said that the Society of Thoracic Surgeons and American College of Cardiology are already working on developing that registry.

Meanwhile, an Edwards spokesperson, also participating in the panel discussion, acknowledged that valve makers are already gearing up to sponsor randomized clinical trials using hard end points, to satisfy the FDA, in order to identify a broader population of patients who would benefit.