FAME II: High-Risk, Stable CAD Patients Randomized to FFR-Guided PCI with OMT vs. OMT Alone
An interim analysis of the FAME II trial found a highly statistically significant reduction in the need for hospital readmission and urgent revascularization when fractional flow reserve (FFR) was used to direct treatment. As a result, the FAME II independent Data Safety Monitoring Board recommended stopping patient enrollment. CLD talks to Dr. Fearon about FFR, the original FAME trial, and what it all means.
How does FAME II differ from the original FAME trial?
The original FAME trial, FAME I, was a multi-center international study comparing two strategies for deciding which lesions to stent in patients who have multi-vessel coronary disease. Patients were randomized to either a traditional angiography-guided strategy or an FFR-guided strategy. All patients underwent coronary angiography and if the patient had two or three vessel coronary disease and had consented to participate, the operator identified which lesions, based on the angiogram and clinical information, required PCI. If the patient was randomized to an angiography-guided approach, then the operator stented the identified lesions. If the patient was randomized to an FFR-guided approach, then FFR was measured across each lesion, and only if the FFR was significant would the operator go ahead and stent. Greater than 90% of patients received stents in FAME I, but half of the patients, those in the FFR-guided arm, received significantly fewer stents compared to the angio-guided arms. On average, there were about three stents per patient in the angio-guided arm and only two in the FFR-guided arm. Outcomes showed that the FFR-guided patients had lower rates of major adverse events: death, myocardial infarction (MI), and need for repeat revascularization. FAME I included all patients with multi-vessel disease, meaning patients had two or three vessels involved, and it could be stable or unstable patients, so patients with acute coronary syndrome were also included. If the patient had had an ST-elevation MI, the culprit lesion was fixed first and they could be enrolled a week later if they still had multi-vessel disease.
FAME II includes only stable coronary artery disease (CAD) patients, and the trial includes both single and multi-vessel disease patients. In a sense, the idea behind FAME II was to redo the COURAGE study. COURAGE showed no difference in death and MI when comparing optimal medical therapy alone to percutaneous coronary intervention (PCI) plus optimal medical therapy in patients with stable CAD. One of the criticisms of COURAGE is that the patient population was lower risk; they didn’t have a large amount of ischemia, based on their non-invasive tests, and importantly, their procedure was guided by the angiogram alone. If the patient was randomized to the PCI arm in COURAGE, operators did the angiogram and used their own judgment as to whether lesions were significant and required stenting. As a result, there may have been an excess number of stents placed and lesions that didn’t warrant therapy may have been treated. The hypothesis behind FAME II is that if an FFR-guided strategy was used, only high-risk patients with significant ischemia would be included and operators would only stent lesions that actually needed it. In FAME I, FFR-guided PCI had lower death and MI rates, and better outcomes than angio-guided PCI. We took the next logical step, positing that a comparison of FFR-guided PCI to medical therapy might show better outcomes, unlike COURAGE. So, in FAME II, patients who had stable coronary disease were referred for a cath and underwent an angiogram. FFR had to be measured and only if the patient had a significant lesion, based on the angiogram and an abnormal FFR, could the patient be included in the trial. If they fit those criteria, then the patient was randomized to either stenting or to medical therapy, and again, the idea was that this would include only higher-risk patients who had significant ischemia, and would maximize the benefit of stenting while minimizing the risks, because no unnecessary stenting would be performed.
It was a challenging study in which to enroll patients, because we had to consent lot of patients up front, not knowing for sure what they would have. If the patient agreed to participate, then the angiogram was performed, and if they had a lesion that was 50% or greater in at least one major epicardial vessel, then they could be included in the trial, and FFR would be measured. If the FFR was below .80, meaning it was a significant lesion, then the patient could be randomized. The randomization would occur right then and there on the table. If the patient was randomized to PCI, the operator would then go on and stent. If the patient was randomized to medical therapy, the procedure stopped, and the patient was placed on optimal medical therapy and followed.
A large percentage of patients had what appeared to be significant lesions, either one or more, on the angiogram, and yet when FFR was measured, it wasn’t significant. Although they had a narrowing, it wasn’t causing significant ischemia, and was unlikely to be responsible for their symptoms. These patients were not randomized to a treatment group, but were included in a second cohort or registry that received standard of care, and we are following those patients as well. In a sense, we weeded out patients who didn’t need stenting because they didn’t have significant ischemia. FFR is very useful for identifying lesions that are higher risk and causing ischemia. Using an FFR-guided approach, we honed in on just the patients who would benefit from stenting, and avoided treating patients who didn’t need stents.
After COURAGE, perhaps a lot of stable angina patients avoided having a cath and went right to optimal medical therapy. Now, it seems more will be having caths.
Yes, I agree, and one of the side effects of COURAGE was just what you said. People would say, well, this patient is stable, I can just treat them medically. But the problem with that is COURAGE was not actually set up that way. In COURAGE, first the cath was done, and then patients were included, based on what was found. There were a number of patients who were excluded for various reasons, such as left main disease or high-grade multi-vessel disease that was better suited for bypass. So if the baseline cath is not done first in order to figure out what is going on, many patients may be under-treated. The pendulum may have swung a little too far in one direction. FAME II should bring patients back to the cath lab and we can make more educated decisions using coronary physiology to guide us.
FAME II was stopped early by the DSMB. Can you tell us more?
I am privy to the decision process, but this was based on a recommendation from the data safety monitoring board, and we don’t have all the data in our possession yet. We are still waiting on events to be adjudicated, so everything is preliminary. As the press release showed, there was a significantly higher major adverse event rate in the patients randomized to optimal medical therapy alone, and that was driven primarily by the need for hospitalization and urgent revascularization, which was part of the composite endpoint. The primary endpoint of FAME II is death, MI, and urgent revascularization requiring hospitalization. Because the major adverse event rate showed such a significant difference, the data safety monitoring board felt that it was highly unlikely that there would be any change in this event rate by enrolling more patients and it wasn’t appropriate to continue to enroll.
What can we conclude so far?
FAME II provides more data supporting the paradigm of what we’ve been calling “functional” angioplasty, meaning that we stent ischemia-producing lesions and we treat with optimal medical therapy the non-ischemia-producing ones. It further reinforces the fact that although the angiogram is the “gold standard” for diagnosing coronary artery disease, it does have limitations and if we want to optimally treat our patients, we need adjunctive techniques like FFR to tell us which lesions are the ones causing ischemia and will benefit most from stenting, and which lesions we can avoid and treat safely with medical therapy.
Many interventionalists anticipate future requirements for documentation showing a PCI is appropriate to perform.
Certainly, there is more and more scrutiny, and we want to do what is right for the patient. It is incumbent upon us to document ischemia, and whether that is with a non-invasive test or with FFR, I doubt requirements will be specific enough to necessitate it be done one way or the other. The nice thing about FFR is that it gives a vessel-specific and lesion-specific index for determining which areas are most responsible, whereas non-invasive tests just tell us a particular region is the area that is causing the problem, but not which vessel is the culprit.
What’s happening next for FAME II?
We are collecting all the data and analyzing the event rates, and will present the results. We haven’t figured out when and where as of yet, but that’s actively going on now and hopefully will be expedited.
We are planning future studies as well, including one that will compare FFR-guided PCI to bypass surgery in patients who have three-vessel disease. In SYNTAX, the event rates were higher in the PCI arm, primarily due to increased need for revascularization, but again, PCI was guided by the angiogram, and it’s possible that excess or unnecessary stents were placed that increased the adverse event rate in the PCI arm. If we use an FFR-guided approach in the PCI arm, we might have fewer stents with equivalent, if not better, outcomes, like we saw in FAME, and the event rate may be similar, if not better, than it is with coronary bypass surgery.
Any plans for an economic analysis?
I was involved in an economic analysis for FAME I. It had unique findings with respect to FFR-guided PCI compared to angio-guided PCI. Not only did FFR use improve outcomes, but it also saved money. One concern was that, well, you have to use a wire and adenosine. Both cost money, so is it cost-effective to use FFR-guided PCI? The economic analysis showed that compared to angio-guided PCI, the FFR-guided approach saved roughly $2,000 per patient. Compared to other strategies, such as a defibrillator or dialysis, these hopefully improve outcomes, but cost a significant amount of money. FFR not only improved outcomes, but it saved money, so it was a win-win situation. We will be doing an economic analysis for FAME II, as well. We don’t have any of that data yet, but it will be interesting to see how that pans out.
It is an exciting time in interventional cardiology. FAME II is good news for interventional cardiologists, and it should hopefully improve the way that we treat our patients.
Disclosure: Dr. Fearon reports receiving institutional research grant support from St. Jude Medical.
Dr. Fearon can be contacted at wfearon@stanford.edu.
