Indianapolis, IN - The final manuscript for the Multicenter Automatic Defibrillator Implantation Trial II (MADIT-II) trial is expected to be submitted this week, the investigators say, with publication perhaps soon thereafter. In the meantime, heartwire asked experts in this field about the implications the positive findings may have for a greatly expanded pool of potential patients who may now benefit from receiving an implantable cardioverter defibrillator (ICD).
MADIT-II was a prospective randomized trial comparing the use of an ICD to medical therapy in post-MI patients with moderate LV dysfunction (ejection fraction (EF) <30%). The trial, supported solely by Guidant (Indianapolis, IN), included more than 1200 patients from 71 centers in the US and 5 in Europe. On November 20, 2001 the investigators, led by Dr Arthur J Moss (University of Rochester Medical Center, Rochester NY), announced that the trial had been halted by the Data Safety and Monitoring Board (DSMB) for efficacy in the ICD arm: a 30% reduction in total mortality.
That's the truly important part of what this trial shows. No longer will EP testing be necessary - Holter monitoring first, EP testing second, that all becomes just historical fact, not necessary now.
Previously, patients had been required to have a positive EP test with depressed EF before being deemed eligible for ICD therapy. In the MADIT-II trial, no eligibility criteria beyond reduced EF and coronary heart disease (CHD) was required.
"In this patient population - which is a large population - [MADIT-II] shows that defibrillator therapy has a critical role in treating these patients," Dr David Cannom (Good Samaritan Hospital, Los Angeles, CA), a member of the trial's writing committee, told heartwire . "It's as important as any other therapy they're on, and probably more important than any other single therapy in terms of survival."
If approved by the FDA, reduced EF and CHD will become a standard indication for an ICD without any other testing, Cannom said. "Now, that's a big change, and that's the truly important part of what this trial shows. No longer will EP testing be necessary - Holter monitoring first, EP testing second, that all becomes just historical fact, not necessary now."
A spokesman for Guidant said the company is currently preparing their submission to the FDA for consideration of this new indication.Right on the numbers
Although pleased with their results, MADIT-II investigators were not particularly surprised by them, Cannom said. "When we designed the trial, we were predicting a 19%, 2-year mortality on conventional therapy, and predicting roughly a 38% reduction in total mortality with the ICD, and it turned out to be a 30% reduction," he said. "That slight difference accounted for the increased duration of the study, but it didn't surprise us - we were right on the numbers in terms of our prediction of mortality."
The decision to use total mortality as the endpoint rather than sudden death mortality makes the data much more reliable, even though it's a tougher endpoint to achieve, Cannom said. "Most investigators think sudden death is fine but it's very tough to define. It's very difficult to prove a death was sudden rather than non-sudden, so total mortality is just simply unassailable as an endpoint."
Unlike the MADIT trial, during which the treatment of heart failure changed substantially during the trial, MADIT-II was not much affected by emerging therapies, Cannom added.Next step: convince the physicians
There have now been six trials, including MADIT-II, of these devices showing their benefit in both primary and secondary prevention. The challenge going forward, as Cannom sees it, is to convince the physicians who see these patients to use the devices. "The problem in our field today has not been getting reimbursed for the devices, it's been appropriate referral of patients for devices," he said. "The results of these trials have not filtered down to the individuals taking care of these patients at the front lines, the internists and the family practitioners."
The current estimate from surveys conducted by the companies, Cannom said, is that about 1 in 4 patients who are already eligible for these devices, whether in a primary or secondary prevention indication, gets a device, "and that's a small number, it seems to me." Now that the indications may be expanding dramatically, he added, "I suspect there's not going to be a rush to therapy It's going to be very slow, and will depend largely on how much education's come about, what this means for patients."
Cannom's group has spent some time in "strategizing" how best to get this data out, he said. It will be published in a prominent journal, but "after that, the information system has got to be good."Show me the data
Although generally positive, reaction to the announcement was still guarded in the EP community, as they reserve their judgment for the publication of the full data from MADIT-II.
Dr Douglas P Zipes (Krannert Institute of Cardiology, Indiana University School of Medicine, Indianapolis) was very clear that extensive discussion of these findings could not be done in the absence of the full disclosure of the data. "These comments are all based on very preliminary data from a newspaper article, and I eagerly await the final published article," Zipes told heartwire .
However, Zipes did say that he thinks these results need to be confirmed by the findings of a second ongoing primary prevention trial, Sudden Cardiac Death - Heart Failure (SCD-HeFT). "If confirmed, then they have the potential of radically altering how we treat patients with heart failure," Zipes said. "They also have major implications for the healthcare budget, since these devices are expensive and suddenly, the market will increase pretty considerably."
Dr Eric Prystowsky (St Vincent Hospital, Indianapolis, IN) made much the same points. "The problem at present is that there is no published manuscript on MADIT-II, and therefore no way to give an educated opinion on its merits," he told heartwire . "That said, I am upbeat about the results, and if the FDA gives a new indication for ICD use based on this study, I think many EPs will adopt them. However, it is also appropriate to wait for the manuscript before making a decision."
Although it may be reasonable, Prystowsky added, to wait in addition for confirmation from SCD-HeFT, Prystowsky said he would make that decision after seeing the MADIT-II results.
Cannom, however, pointed out that while the two trials have similarities, there are substantial differences between SCD-HeFT and MADIT-II. Investigators in SCD-HeFT have included patients with both ischemic heart failure and dilated cardiomyopathy in proportions of roughly 50-50, and use an EF cutoff of 35% rather than 30%. In addition, they have enrolled 2500 patients, but have randomized them to 3 arms: the device, amiodarone, and placebo.
"It dilutes the population and adds a therapy that I'm not sure is effective - amiodarone," Cannom said. "This was designed before the trials that showed that amiodarone didn't help in this population were published, so there's a chance that whole arm will be useless in showing survival." The DSMB for SCD-HeFT did review their data in October 2001, he added, and apparently found no reason not to continue.Volkswagen or Rolls Royce?
One of the clear issues from the potential use of these devices in a much wider population is that of cost. "This is very expensive technology and suddenly the number of patients who might benefit from an ICD increases pretty dramatically," Zipes said. "That has the potential of rupturing the health care budget."
Zipes raised again the idea he'd first put forward in an editorial appearing in the March 13, 2001 issue of Circulation. If MADIT-II and SCD-HeFT were both positive, he pointed out, device manufacturers should consider creating a selection of ICDs from which to choose. "For some patients, a sophisticated ICD that could be used to treat comorbidities and monitor a variety of physiological functions may be indicated, whereas in others, an inexpensive ICD in the $10 000 to $15 000 range might be preferable," he wrote. "The latter could be achieved if the manufacturers made an ICD with limited function capabilities - a Volkswagen instead of a Rolls Royce."
The lower-cost ICD could have more restricted but adequate detection and storage capabilities, he proposed, and a battery capacity to deliver only 8 to 10 shocks. If the patient used up that many shocks, the device could be replaced with a more advanced model.
There is no paper, no approval, no guideline that would state these patients should get an ICD, and importantly, that one would get reimbursed for implanting it, so this is all very premature.
However, Zipes added, this discussion in relation to the MADIT-II announcement is still theoretical. "There is no paper, no approval, no guideline that would state these patients should get an ICD, and importantly, that one would get reimbursed for implanting it, so this is all very premature."
Others agreed with the potential merit of a lower-cost ICD, using much the same analogy (although the comparisons varied perhaps based on personal preference: Lexus and Jaguar, rather than Rolls Royce), with the basic point that the devices for this larger population could be simpler, and therefore cheaper. However, Cannom said his first concern is still the education of the physicians about use of the devices. "I worry much less about payment frankly," he said. " Yes, it will help if the device is cheaper, no doubt about it. And I think there will be cheaper devices based on this - this has been a huge concern for the defibrillator companies, to halve the price of the devices." However, he added, "Yes, it's going to cost more, but it's better than a lot of other things we do where the data is not so good."
The spokesman for Guidant, however, pointed out that both their company and the other manufacturers already offer ICDs with a range of functions and costs. Depending on the model, and taking into account what hospitals choose to charge for them, costs can differ by about $15 000 from the bottom to the top of the range, he said, beginning from about $15 000.Cost-effective perhaps, but still a cost
Coincidentally, a paper looking at cost-effectiveness projections from a group at Stanford University, focusing on the use of ICDs in the population studied in MADIT-II, was published just last month in the Annals of Internal Medicine.
After previous trials, including the Multicenter Automatic Defibrillator Implantation Trial (MADIT) and Antiarrhythmics Versus Implantable Defibrillators (AVID) trial, had shown the ICD was very effective in high-risk patients who had already had an arrhythmic event, Dr Gillian Sanders (Stanford University) and colleagues used a model to explore how effective the prophylactic use of an ICD would have to be in order to be considered cost-effective in the lower-risk population being studied in SCD-HeFT and MADIT-II.
Using survival, cardiac death, and inpatient costs estimated from the Myocardial Infarction Triage and Intervention (MITI) registry, the researchers stratified the patients into 3 cohorts based on their EF: under 30%, 30-40%, and over 40%. "The MADIT-II trial was looking at these patients with an EF of less than 30%, and that is the population that in our study was showing it was most cost-effective to be implanting an ICD," Sanders told heartwire in an interview.
It would be a big up-front cost, but from a societal perspective over time, that cost is certainly made up by the lives you're saving.
When Sanders et al carried out this study, she said, they thought the actual results would not be available for years. "We were excited to see the MADIT-II results since it certainly supports the assumptions we made in our paper."
Although it may be cost-effective, implanting an ICD still represents a substantial cost, Sanders acknowledged. "It's certainly not cost-saving, it's a cost, but it's a cost that's considered . . . cost-effective," she said, depending on the available resources, and the decisions that are made about how to use them.
In their paper, the researchers made an initial estimate, given the number of people who have heart attacks every year, the number who survive them, and the number of those likely to be in the low EF group, of the cost of giving an ICD to all eligible patients, Sanders said, and found it to be about a "billion dollars."
"But that, in our [estimate], saved over 2000 lives in the first year, and continued to save lives," Sanders said.
The assumptions of efficacy the researchers used to make that calculation may have been more conservative than the benefit now being shown in the MADIT-II results, Sanders added. "It would be a big up-front cost, but from a societal perspective over time, that cost is certainly made up by the lives you're saving."Industry analysts satisfied
Industry analysts, at least, seem persuaded by the results of MADIT-II in their estimates of the potential for not only Guidant, but for other ICD makers such as Medtronic (Minneapolis, MN) and St Jude Medical (St Paul, MN), to benefit from this announcement.
In her December 3, 2001 report entitled "MADIT-II: Just what the doctor ordered," Sheryl Zimmer, an analyst for Deutsche Banc Alex Brown (DBAB) points out that although many had speculated and hoped that the trial would end early, it still came as a surprise, given it was not expected to finish until the summer of 2003. That it did end early, she writes, "is a significant positive for the worldwide ICD market, which had been suffering from moderating growth over the past several quarters. Moreover, while solely sponsored by Guidant, the study will serve to help all three cardiac rhythm management companies."
The most important implication of MADIT-II from the industry standpoint is that it should revive growth in the ICD market, Zimmer points out. That the market has been showing slow growth, she writes, seemed to stem from a combination of "complacency among the companies and physicians, a lack of recent high-profile studies supporting the case for ICDs, and the physician distraction with time-consuming new technologies (ie heart failure)." MADIT-II, she adds, should add an incremental 275 000 patients to the 300 000 currently indicated for an ICD in the US, "essentially doubling the market opportunity."
"While we are not changing our estimates at the moment, we think this enhances visibility and creates potential upside for the companies leveraged to this area, namely GDT (Guidant), MDT (Medtronic) and STJ (St Jude Medical)," Zimmer writes. Accordingly, DBAB give a "strong buy" rating to Guidant and Medtronic, and a "buy" rating to St Jude.
Goldman, Sachs & Co Investment Research also upgraded Guidant on the basis of the news from MADIT-II from "Market Performer" to "Market Outperformer."
"While a boon for all defib[rillator] players, including MDT and STJ, we believe that this event, coupled with strong momentum in GDT's coronary stent business, could send GDT shares into the low $50s range in the near-term, prompting an upgrade to Market Outperformer," a report from Goldman Sachs notes.
Heartwire from Medscape © 2001
Cite this: MADIT-II: Implications unclear as the potential pool of ICD-eligible patients may double - Medscape - Dec 17, 2001.