This article is more than 5 years old. Information may no longer be current.
CMS panel lukewarm on hospitalizations as outcome to determine coverage of HF technologies
The Medicare Evidence Development and Coverage Advisory Committee voted it had low-to-intermediate confidence that HF hospitalizations are standalone, meaningful primary outcomes in studies, and it had intermediate confidence in the use of HF hospitalizations and events equivalent to hospitalizations such as outpatient IV diuretic therapy.
The committee was asked to advise CMS on which outcomes should be included in studies of technologies for HF treatment. This information will eventually help CMS decide what evidence it prefers to see before agreeing to cover HF technologies.
Regarding the use of surrogate and intermediate endpoints, the panel voted that it had low-to-intermediate confidence on the following: HF with preserved ejection fraction, HF secondary to mitral regurgitation where the focus is the repair or replacement of the mitral valve and HF with reduced ejection fraction.
The panel voted it had intermediate-to-high confidence that quality-of-life measures are an adequate reflection of patient experience, though they had intermediate confidence regarding use of either as a standalone outcome or a composite outcome.
Functionality assessments such as the 6-minute walk test received a vote of intermediate confidence from the panel. Panelists voted that they had low-to-intermediate confidence for its use as a standalone primary outcome vs. intermediate confidence for composite primary outcome.
Unmet needs
Philip B. Adamson, MD, MSc, FAAC, vice president of medical affairs and medical director at Abbott, said identifying the areas of unmet medical needs through these studies will allow physicians to focus on meaningful patient outcomes while improving patient experience.
Benefits such as quality of life and freedom from hospitalization should also be considered as endpoints, John D. Carroll, MD, professor of medicine at the University of Colorado School of Medicine and director of interventional cardiology at University of Colorado Hospital, said to the panel. He said it’s important to measure patient-reported health status, as it will allow physicians to best decide what will best optimize their patient’s survival and health.
“If we want to help patients make informed decisions about new [HF] technologies, then we need their input from the start,” Larry A. Allen, MD, MHS, medical director of advanced heart failure at the University of Colorado, said to the panel. “These treatment decisions not only affect the patient, but also their families and caregivers, and can often be emotional.”
Quality of life
Clinton A. Brawner, PhD, clinical exercise physiologist at the Henry Ford Hospital, who represented the Cardiovascular Research Foundation, told the panel about the benefits of peak VO2 and the 6-Minute Walk Test in clinical studies. “[They’re] closely related to outcomes across the spectrum of [CVD], particularly in HF,” Brawner said.
Adding other measures of quality of life including the Kansas City Cardiomyopathy Questionnaire is clinically significant, according to Nancy Sweitzer, MD, director of the Sarver Heart Center at the University of Arizona. “Failure to include measure of quality of life is failure to comprehensively study an intervention,” Sweitzer said.
“Current and future therapies must focus not only on mortality, but perhaps even more on HF and related outcomes,” Srihari S. Naidu, MD, FACC, FSCAI, director of the hypertrophic cardiomyopathy program and interventional cardiologist at Westchester Medical Center in Valhalla, New York, and associate professor of medicine at New York Medical College, told the panel. He said endpoints such as days alive and out of the hospital or number of hospitalizations should be considered in studies, as they provide value to all involved, including the health care system.
Rita Redberg, MD, MSc, chair of the committee, professor of medicine at the University of California, San Francisco, School of Medicine, and a member of the Cardiology Today Editorial Board, said caution must be exercised before encouraging use of implantable technologies.
“I don’t know of any devices that get pulled off the market when we find out that they actually didn’t work or they’re harmful, and I don’t think right now, post-marketing … [evidence is] robust at all,” she said. “So that means essentially, we’re putting untested or minimally tested and possibly unsafe devices in patients that are now permanently implanted.”– by Darlene Dobkowski
Disclosure: MEDCAC panel members reported the following financial disclosures: Michael J. Fisch, MD, MPH, FACP reports employment with AIM Specialty Health; Eileen Hsich, MD, reports receiving institutional research support from Novartis and St. Jude Medical, and unpaid consulting for Medtronic and St. Jude Medical; and Adi Renbaum, MBA, reports being an industry advisor representative working for variety of medical device companies. Adamson is an employee of Abbott. Allen reports being a member of the advisory committee for the National Quality Forum; consulting for Janssen, Novartis, and ZS Pharma; and serving on boards for Boettcher Foundation and PCPI. Brawner reports laboratory contracts with Actelion, Covance and Heart Metabolics. Naidu reports serving on advisory boards for Abbott Vascular and MyoKardia. Sweitzer reports leading clinical trials sponsored by Merck and Novartis, and consulting for Acorda Therapeutics and Medtronic.
Perspective
Back to Top
It was great that MEDCAC had the forethought to look at this as an issue. Historically, we’ve had most clinical trials that really only valued mortality as the endpoint that is beyond compare, and it set a very high bar for all types of therapies which either require large patient numbers or some really strong benefits on mortality.
What we’ve realized and they have realized is that there are more technologies coming up, which have used other types of surrogate endpoints, and there has been, I assume, perhaps criticism or discussion about which endpoints are reasonable because we are approving these devices and there’s a sense that they are meaningful products, but we need some kind of consensus going forward to standardize the approval process.
It’s very important for MEDCAC to take this on and to understand that these endpoints are necessary and reasonable to use in clinical trials. I was very happy about that.
In terms of the vote, after hearing everything, they’re faced with a decision about how much evidence do we have for each of these things, be they some of the more subjective endpoints vs. the ones that are more anatomic-based, for example. What we saw from their voting was that they were not very confident about some of the surrogate endpoints, but certainly the ones that were more patient-centric, they’re very confident about. That is reassuring, that some of these endpoints, and especially those that represent the patient experience, will be in clinical trials to help approve the novel devices.
A large part of the impetus for discussing this is based on interventional treatments that have come up, which often cost more than other types of treatments. So I do believe that once they’ve nailed down exactly which endpoints they feel are important, this will have consequences in terms of how trials are designed, to make sure the endpoints that have the most confidence are included in up and coming HF trials.
The implications are that at least it gives industry some idea of which ones Medicare will look at for reimbursement, so they don’t have to make these decisions in a vacuum as they construct their clinical trials.
The question is, will it take more time or not? Will the inclusion of different endpoints increase the time to approval or make things easier? There was minimal discussion about the duration of follow-up that’s required. There was some discussion about registries and some discussion about the horizon on how long you’d like to see improvements, but this is going to be very device-specific. A device that makes a more immediate benefit, you’re going to see these improvements in patient outcomes a lot sooner. Many of our trials are 1- to 3-year trials, so I would imagine that within that time period, we’d want to see these more patient-centric variables parallel the clinical outcomes, so I don’t think that, a priori, this requires a longer duration of trial or a longer duration for approval. That would be determined by the device-specific outcomes that are impacted by the device; for example, reduction in mitral regurgitation for a device that works on the mitral valve.
The reason I believe that these novel and surrogate endpoints are very important is because they have significant meaning for everybody that is involved in clinical care … but most importantly patients and physicians. Our overriding goal is to keep patients feeling well and out of the hospital. And I think that is also a patient preference as well, so I think that the added benefits really are that it’s very meaningful to patients and they will come to see clinical trials as much more relevant and more patient-centric as opposed to variables that are oftentimes not very patient-centric.
For example, if you have a device that reduces mitral regurgitation, that’s fine, but a patient then has to make the leap that it will improve left ventricular volumes and cardiac output, which will perhaps decrease their diuretic requirement, which perhaps will decrease their rate of refractory HF and hospitalizations.
If you tell them that with this particular device or this particular technology, they will have only an X% chance of being in the hospital the next year as opposed to this other technology (or no therapy), that has significant meaning for patients that goes above and beyond what the device actually does anatomically, the latter of which is important for doctors, but certainly less important for patients.
It’s a much better way of balancing the patient expectations of device therapies with the physician expectation, where we’re driven to make the connection between an anatomical or a physiological improvement and ultimate improvements in a clinical state, whereas patients don’t always make that assumption. For them, a decrease in hospitalization may be a much harder endpoint than we see it, and similar to the patient outcomes, the quality of life aspects or the functional status aspect, those may be the harder outcomes that they have more interest in rather than the more anatomy- or physiology-based ones.
The panel voted low and intermediate more frequently than I would have. I have more confidence on several of these questions than they seem to have. I probably would have been in the three to four range for the majority and a little bit over four on some of these questions. The evidence in my mind is higher for some of these, and they should be used more often going forward. There’s enough evidence to move in that direction. If you compare them to other metrics, of course they don’t stand as tall because mortality is a very black and white outcome. Other ones are still novel metrics that we’re trying to justify prospectively, so that’s why I think they gave them lower scores, but I would have given them somewhat higher scores. Now that being said, it’s all relative and what we’re going to have to wait to see is how does that translate into whether they use some of these and which ones they use going forward in different trials.
[HF] is a huge epidemic in this country and in the world. We need to juggle the patient experience and expectation, the physician experience and expectations and of course cost as well as how rapidly devices and technology can make it to the market. It is a tough job balancing all those things, but I would like to hope that CMS will be forward-thinking so that we can continue to improve outcomes in this space.