FUTURE of Fractional Flow Reserve–Guided PCI Still Seen as Bright After Halted Trial

November 28, 2016

NEW ORLEANS, LA — Entry to a large randomized clinical-outcomes trial was stopped early after its data and safety monitoring board (DSMB) saw a significant jump in early mortality in patients undergoing coronary angiography with revascularization guided by fractional flow reserve (FFR) readings[1].

At the interim data check covering the trial's first 836 patients, the risk for death by any cause was more than doubled (P<0.02) in the group following the FFR-guided strategy compared with the control group, whose revascularization decisions were based on standard angiography.

With a target enrollment of over 1700 patients, the Functional Testing Underlying Coronary Revascularization (FUTURE) trial was then stopped at 933 patients. Of the 797 patients followed for 12 months, 1-year mortality remained numerically increased but was no longer significant. Nor at that time was there a significant difference in the clinical composite primary end point, of which mortality was only a part.

Speculations on Why Mortality Was Increased

The signal of raised mortality in a study of an established strategy like FFR-supplemented angiography was attributed to either undetermined non-FFR differences between the groups or, more likely, the play of chance by expert observers speaking to the media here at the American Heart Association 2016 Scientific Sessions where the trial was presented.

Most of the deaths in the trial were cardiovascular in nature, and most occurred within the first 1 or 2 months after randomization, according to Dr Gilles Rioufol (Hospices Civils de Lyon, France), who was slated to present FUTURE at the meeting. So the mortality signal was presumably not related to FFR itself, he told heartwire from Medscape. "That's reassuring for FFR. It's not a device-related problem. If there is a problem with the strategy, it's more linked to the strategy and not the device."

Also, the trial's jump in mortality in the FFR group shouldn't be taken as a "negative signal in terms of safety," according to Rioufol, because mortality was only a component of the primary end point—which was all-cause mortality, MI, repeat revascularization, or stroke—and that primary end point was not increased at 1 year.

Risk of Early "Overinterpretation"

"This is a reminder of how difficult it is to work [on the DSMB] and not overinterpret early signals that may give you huge swings in outcomes, which are not necessarily confirmed when you have more data available at the end of the trial," according to Dr Philippe Gabriel Steg (Bichat Hospital, Paris, France), who wasn't involved in the study.

"Not only would it have been unethical to continue the trial in order to demonstrate harm, but I think it would have been unlikely that harm would have been demonstrated," Dr Herbert D Aronow (Rhode Island and The Miriam Hospitals, Providence), assigned discussant for the FUTURE presentation, said at the briefing.

Prevalence of Revascularization Decisions, FFR Guidance vs No FFR Guidance

Revascularization decision FFR guidance, % Angiography-only guidance, %
PCI + OMT 71 78*
CABG + OMT 12 12
OMT only 17 9
*P=0.002 vs FFR guidance group
OMT=optimal medical therapy

"If there really were a mortality excess due to FFR, it would have had to have been due to the FFR procedure itself or to the way in which the information we get from FFR changed treatment decisions," he said.

But complications of the FFR procedure were few, Aronow noted. And FFR guidance was followed by fewer patients getting PCI and more getting only optimal medical therapy (OMT) compared with the control group, he noted. If that difference were behind the mortality increase, "one would have had to argue that performing more PCI would have led to a lower mortality rate. In fact, we know that in this study cohort there's no evidence to suggest mortality is reduced by performing PCI."

Therefore, he said, FFR-based changes in treatment decision didn't lead to the excess mortality, according to Aronow. "In all likelihood, I believe this was chance," he said. "I think in the end this will remain a head scratcher."

Groups Comparable at Baseline

Short of the planned enrollment of 1728 patients, FUTURE ultimately entered 941 patients across 31 centers in France. Of the cohort, 83% were men and 21% had stable angina, 46% an acute coronary syndrome, and the rest other presenting conditions.

The patients were described as an "all-comers" population with stable or stabilized angina and multivessel disease including the left anterior descending coronary artery at the time of angiography.

The interim analysis included 465 patients randomized to FFR guidance and 469 patients to the standard-strategy control group. After the procedure, they were stratified by treatment decision: PCI plus OMT, CABG plus OMT, or OMT only. The two randomization groups at baseline were similar with respect to Canadian Cardiovascular Society class, left ventricular ejection fraction, prevalence of ACS, diabetes, number of involved coronary arteries, and SYNTAX lesion-complexity score.

When the DSMB recommended a halt to the trial, it had noted a hazard ratio (HR) of 2.39 (P<0.02) for mortality in the FFR group vs controls; 72% of the deaths were cardiovascular in nature. After entry to the trial was halted, a cohort of 797 patients followed to 1 year showed an HR of 1.09 (95% CI 0.76–1.56, P=0.63) in the FFR-guided group for the full primary end point that also included MI, stroke, and revascularization.

One-Year Outcomes for 797 Patients in FUTURE, by Randomization Strategy

End points FFR guidance (n=399), % Angiography only guidance (n=398), % HR (95% CI)
Primary end point* 15.1 13.2 1.09 (0.76–1.56)
Death from any cause 3.9 1.8 1.98 (0.85–4.60)
MI 6.5 5.3 1.23 (0.71–2.11)
Stroke 0.4 0.9 0.48 (0.09–2.62)
Repeat revascularization 7.6 7.6 0.97 (0.60–1.58)
*All-cause mortality, MI, stroke, or repeat revascularization

"I think most will feel that FFR is safe," Aronow said. "As to the question of whether it reduces clinical events, in this broader population, I don't think we know. But it doesn't have to reduce clinical events to be worthwhile. It could reduce resources and have no difference in clinical events, and that would be important." And FFR was associated with less resource use, he noted, in the form of less PCI by 7 percentage points relative to the control group.

"So I believe that, moving forward, employ an FFR-guided strategy with formulating our treatment decisions in patients with coronary disease who undergo coronary angiography."

FUTURE was sponsored by Hospices Civils de Lyon. Rioufol discloses receiving research support from St Jude Medical and Volcano; honoraria from St Jude Medical and AstraZeneca; and serving as consultant or on advisory boards for St Jude Medical and Boston Scientific. Aronow had no disclosures. Steg reports grants, personal fees, and nonfinancial support from Servier; personal fees from Amarin, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol Myers-Squibb, Daiichi-Sankyo, GlaxoSmithKline, Lilly, Merck Sharp & Dohme, Novartis, Pfizer, Roche, Medtronic, Servier, Janssen, CSL Behring, and Regeneron; grants and personal fees from Sanofi; and personal fees and nonfinancial support from The Medicines Company.

Follow Steve Stiles on Twitter: @SteveStiles2. For more from theheart.org, follow us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.