FAME II Stopped Enrollment Early, More Events in the Medical Arm. What Does It Mean for PCI?

Morton Kern, MD
Clinical Editor
Chief Cardiology, Long Beach Veterans Administration Hospital;
Associate Chief Cardiology, University California Irvine;
Professor of Medicine, UCI
Orange, California

Morton Kern, MD
Clinical Editor
Chief Cardiology, Long Beach Veterans Administration Hospital;
Associate Chief Cardiology, University California Irvine;
Professor of Medicine, UCI
Orange, California

Since the FAME II study made such headlines recently, I thought I would add my two cents and respond to some of the comments, the good and bad, that surrounded the press release from St. Jude at the end of January 2012. 

But first, my disclaimer: I use, teach, and write about FFR and coronary physiology. I speak for the two companies that make pressure-sensor guide wires and I participate directly or as a committee member on studies that use FFR. Specifically, at this time I am a member of the critical events committee for the adjudication of events that are reported to the Data Safety Monitoring Board of the FAME II study. I am not a participating center or involved in any other way. 

What’s the big news? Reported on theheart.org January 18, 2012, according to the sponsor’s (St Jude Medical) press release of the same date, the independent data and safety monitoring board (DSMB) recommended St Jude Medical’s FAME II clinical trial stop enrollment following a positive interim analysis. The interim analysis of the FAME II study — which compares fractional flow reserve (FFR)-guided stenting with optimal medical treatment (OMT) to OMT alone — showed a clear benefit for the FFR-guided approach. The analysis revealed a statistically significant reduction in the need for hospital readmission and urgent revascularization when FFR-guided assessment was used to direct treatment in coronary artery disease (CAD) patients. As a result, the DSMB recommended that patient enrollment be stopped, but the follow-up continue. “The DSMB considers it unethical to continue to randomize patients to OMT alone,” St Jude Medical stated in their press release.

FAME II was designed to look at percutaneous coronary intervention (PCI) compared to optimal medical therapy as was done in the COURAGE trial, but differed in the requirement to have a measure of ischemia — FFR — to guide the approach to intervention, rather than rely on the operators’ appreciation of CAD by angiography alone. In addition, since most CAD narrowings were characterized as ischemic by FFR, the efficacy of medical therapy for ischemia could be tested directly in this fashion, something COURAGE did not do. 

FAME II randomized 1,219 patients with stable CAD in 28 centers in Europe, the U.S., and Canada; patients who are already participating will continue to be followed according to the trial protocol, but no new patients will be enrolled. Currently, there is no difference in the rates of death or myocardial infarction (MI) between the two study arms.

In this issue of CLD, Dr. William Fearon, one of the U.S. principal investigators, addresses some key questions regarding the FAME II study.  This enlightening discussion will put the FAME family of studies into focus, remind us how important ischemia guided-PCI is, and why we should approach patients we think need stents with some objective evidence that stenting will help and be more than just a cosmetic addition to the angiogram. 

Since the discussion about FAME II will likely continue until its formal presentation, I thought it might be interesting to address some of the comments that accompanied the news of the early termination of FAME II enrollment on theheart.org. The comments reflected a broad range of opinion, bias, hopes, and ignorance regarding the issues of ischemia-guided PCI. All comments were posted in the public record on the Internet. Portions of those comments are reproduced below with my replies.

Theheart.org headline: “FFR shows benefit in FAME II; enrollment halted.”

Comment 1: “[This study had] Industry sponsorship and bias?”

Response: With the exception of NIH or other government-sponsored trials, almost all such large studies have industry support. The bias does not come from the sponsors, since it is mandatory in good research that the principal investigators, data safety board, and monitors are separated from the company. The sponsor does not have access to the information. It is true that the design of the study may get comments from the sponsors and possibly modifications before the study is funded, but once funded and submitted to institutional review board (IRB) reviews, the sponsors no longer have participation. The fact that a study is industry-sponsored does not mean there is industry bias.

Comment 2 (part 1): “I despise this type of study. First, as a disclaimer, I think that the use of FFR-guided stenting plus OMT is probably superior to OMT alone. That said, enrollment should not have been stopped in this study! You get a group of patients with angina. You randomize them to getting a stent (which we know quickly resolves angina, but has little if any effect on MI or death) vs. medical management that treats angina more slowly than stenting. We then tell the patients to let us know if they have any chest symptoms. Those not stented will still have their symptoms, present to the ER where they get stented. We call this an adverse outcome even though it was completely predictable at the onset of the study.

“This study has demonstrated a fact that we already know, e.g., stenting relieves angina quickly. I was hoping to find out whether or not stenting based on FFR measurement was better than OMT alone with respect to MI or coronary death.”

Response: This individual wants a study to show that FFR-guided stenting is better than OMT for death or MI. The FAME I study showed that FFR-guided was better than angio-guided PCI for reducing death and MI. Because of the event rates, mostly urgent revascularization for poorly controlled angina (despite OMT), it is unethical to continue in hopes of seeing more MIs or deaths. At the end of a two-year follow-up, this may indeed be the case. What is also shown, although only as a presumption prior to this study, was that symptomatic ischemia treated only medically over time remains symptomatic and in some cases, progresses to infarction. FFR identifies those patients with ischemic CAD beyond merely the angiographic presence of CAD who might benefit from PCI or be suitable candidates to continue medicines for atherosclerosis. 

Comment 2 (Part 2) (continuing): “I accuse the sponsors of this study of setting up criteria so that the study would be stopped with the initial predictable endpoint of decreased angina was apparent. They were afraid to wait to a later endpoint that may have shown no benefit with MI or coronary death, or possibly a worsening with MI or coronary death in the stented group.”

Response: The study was constructed by the multinational investigators from FAME I in light of the controversy surrounding the outcome of the COURAGE trial. While conventional wisdom might suggest the result could be anticipated, without a randomized trial, we would still be speculating as to best management, to the detriment of appropriate PCI strategy. I do not believe the investigators were afraid of adverse events in the stented group showing a negative outcome.

Comment 3: “[The commentators above] are a little quick to criticize a trial they haven’t even read. That, to me, indicates bias. Not every cardiovascular endpoint needs to be defined by death or MI. Hip or knee replacement doesn’t save lives, but I bet the [commentators above] would certainly want one if they were hurting.

Also, I teach my kids not to hate. It’s OK to like certain things more or less than others, but it’s not OK to “despise”. Lesson to be learned, even for adults...”

Response: I concur both on endpoints and about being a hater.

Comment 4: “The FAME II study has not been terminated per se. Enrollment was stopped short because of ethical concerns over not giving the best care to those patients with proven ischemia — a significant number of the OMT-only patients came back to the hospital for “urgent revascularization.” But the multi-year follow-up and analysis of the data will proceed as planned. […] Give the investigators a little time and I think many of these questions will be answered.”

Response: I agree. The study must unfold in the normal manner before we can say what the results will mean. This was true for FAME I and will be so for FAME II. The follow-up over the next two years is where the real impact lies.

Comment 5: “The endpoints used by the DSMB to stop the trial were soft. This is going to be a major criticism of this study. Think about it: the patient who is randomized to OMT is probably pissed about not being randomized to the PCI arm, gets a bout of angina (expected), then freaks out because they’re in a trial and are told to report every anginal event, goes to the ER (hospitalization), and the interventional doc on call breaks protocol and stents the lesion, thereby creating the urgent TVR event, without allowing the PI to further intensify medical therapy.”

Response: Death, MI and revascularization for unstable angina are hard endpoints. Revascularization electively or non-urgently is a softer endpoint that is based on physician-patient interaction. Nonetheless, in the treatment of the FAME II patients, excess events could have gone either way, but safety considerations mandated stopping enrollment. 

Other criticisms heard

Comment 6 (Part 1): “The downside of FFR is that you have to perform a lesser intervention (angiography) in order to determine the validity of the greater intervention (PCI). Everyone in the trial underwent angiography and FFR prior to randomization, and only those with ischemic lesions (a qualifying FFR <0.80) were randomized. So it’s fair to conclude that if someone undergoes angiography, it would make sense to do FFR, and if an ischemic lesion is [present], implant a stent.”

Response: The trial was about how patients with CAD undergoing angiography and then possible intervention might be best treated. We concur here.

Comment 6 (Part 2): “BUT [FAME II] doesn’t say anything about which patients actually should undergo angiography in the first place, and the use (or overuse) of angiography is probably the most important unresolved problem in cardiology today. So we’re back to the floodgate problem. If the floodgates (angiography) are open, then FAME II makes a lot of sense and is widely applicable. But if the floodgates are closed or only opened selectively, the trial really doesn’t help very much.”

Response: Patients with symptoms and risk factors should be screened and treated according to good clinical practice and undergo angiography for good clinical indications. Unfortunately, many patients may slip through the routine practice, more so in centers which have self-referred reimbursement linkage to the testing and imaging modalities, and undergo angiography without clear indications. On finding CAD without evidence of ischemia beforehand, FFR makes excellent sense and the FAME studies prove this.

Comment 7: “Did FAME II ensure the intensity of OMT? The intensity of OMT in FAME II will need to be scrutinized very carefully once the results are published to ensure that it is comparable to COURAGE and BARI-2D. Additionally, if the follow-up is short by virtue of early cessation (two years or less), there will also be some ambiguity, since in COURAGE, those patients in the PCI arm who had complete anatomical revascularization at baseline versus those who had incomplete revascularization showed no difference in death/MI over the first two years, but subsequently over the next 2.5 years, the patients with complete anatomical revascularization tended to do better (NS, underpowered) with respect to death/MI (Mancini et al. Circulation: Cardiovasc Qual Outcomes 2009; 2: 320–327). It’s always a double-edged sword when trials are stopped prematurely, especially for endpoints that don’t reflect death/MI.”

Response: Without seeing the data, the intensity of OMT is unknown at the moment. Presuming it was vigorous, the fact that enrollment was stopped, but follow-up continued, should provide the important late events that give the study even more weight.

Comment 8: “[…] This [study] doesn’t demonstrate the benefit of FFR. It demonstrates the benefit of PCI (with defined ischemia).”

Response: True, but without FFR, PCI would have been performed on many lesions that were not ischemia-producing and would have missed some that were. The study demonstrates the benefit both of FFR and PCI.

My bottom line

FAME II was designed to address the limitations of COURAGE and will likely parallel some of the results of the Ischemia Study when concluded. It is interesting that such basic concepts as the need to know ischemia before stenting bring out such an emotional response in some practitioners. While no study is perfect, FAME II will add to FAME I, COURAGE, and BARI 2D, and provide a stronger basis for the best treatment of multivessel CAD patients with ischemia (abnormal FFR) with PCI.

Disclosure: Dr. Kern reports that he is a speaker for Volcano Therapeutics and St. Jude Medical, and is a consultant for Merit Medical and InfraReDx, Inc.


  1. Husten L. FAME II: Additional thoughts about FFR in the real world. Forbes. January 18, 2012. Available online at http://www.forbes.com/sites/larryhusten/2012/01/18/fame-ii-additional-thoughts-about-ffr-in-the-real-world/?partner=yahootix. Accessed February 8, 2012.
  2. Nainggolan L. FFR shows benefit in FAME II; enrollment halted. Theheart.org. January 18, 2012. Available online at http://www.theheart.org/article/1342001.do. Accessed February 8, 2012.