New ORBITA Insights: Invasive Physiology Tests Track Closely With Ischemia, but Not Symptom Improvement

PARIS, France—Invasive physiology tests used in the ORBITA trial did not predict which patients would benefit from PCI more than placebo in terms of the trial’s primary endpoint of improvement in exercise time, according to new data presented today.

But in a finding sure to provoke fresh debate over the importance of ischemia reduction with stenting, fractional flow reserve (FFR) and instantaneous wave-free ratio (iFR) values tracked closely with improvements on stress echocardiography: the lower the baseline FFR or iFR value, the greater the magnitude of ischemia benefit following PCI as compared with a sham procedure.

In an additional analysis—one that was not part of the original trial design—more patients reported being angina free at follow-up after PCI than following a sham procedure, although neither FFR nor iFR modified this effect. Importantly, no relationship was seen between prerandomization functional test results and change in angina frequency at 6 weeks.

“Many people were critical of ORBITA because they felt that excluding FFR-negative and iFR-negative patients would have improved the clinical power of the study,” principal investigator Rasha Al-Lamee, MBBS (Imperial College London, England), told TCTMD. “In fact, if you expect that, what you would expect is that exercise time and symptoms might vary along the FFR spectrum, and what we've shown in our study, albeit with some limitations, is that [there wasn’t] a strong signal that patients who have very highly ischemic FFR or iFR values gained more exercise time increment with stenting. Now that was a surprise, for sure, but having said that you can entirely predict the ischemia reduction that you see on stress echo with stenting—it strongly correlates with the baseline FFR and iFR.”

It's not to say that stenting doesn't work, because stenting certainly improves the blood supply to these patients’ hearts and in fact predictably so. Rasha Al-Lamee

The overall ORBITA trial was neutral, Al-Lamee stressed, which limits the ability to do any secondary analyses. Still, the new data provide some compelling support for the role of angioplasty in this setting, even if the link between ischemia relief and symptoms is not illuminated, something people may find “hard to get their head around.”

“It's not to say that stenting doesn't work, because stenting certainly improves the blood supply to these patients’ hearts and in fact predictably so,” Al-Lamee explained. “The worse their blood supply is, the better stenting is for them. This brings us back to that question of whether symptom and exercise time are as tightly linked to relief of stenosis as we expected.”

Al-Lamee presented the new ORBITA data during a late-breaking clinical trial session here at EuroPCR 2018. Results were published simultaneously in Circulation.

ORBITA: A Widening Sphere

As previously reported by TCTMD, ORBITA was the first sham-controlled trial to compare PCI and optimal medical therapy in patients with stable coronary artery disease and at least one stenosis of 70% or greater suitable for PCI. At 6 weeks, change in exercise time—the primary endpoint in ORBITA—improved significantly from baseline in patients in both trial arms, and while the change was greater in the PCI group, it was not significantly greater than that seen in the sham-treated patients. Measures of symptom improvement and quality of life were also no different between groups, although ischemia, as assessed by dobutamine stress echocardiography, was significantly reduced with PCI.

The current analysis, according to Al-Lamee, was actually the key reason for conducting ORBITA in the first place.

“The main purpose of ORBITA was not to prove whether or not angioplasty worked, because we assumed for sure we'd see a benefit in exercise time and symptoms with angioplasty. So, despite what’s been written and what’s been said, in fact the trial was designed to be positive for angioplasty,” she said. “We are interventional cardiologists, we thought we knew that what we do definitely works in every subject.”

As such, ORBITA’s original intent had been to look at how well baseline measures of ischemia—FFR, iFR, and dobutamine stress echo—can be used to predict the benefit of PCI. The current analysis looks only at the predictive ability of the invasive tests; a future report is planned for the echo data, Al-Lamee said.

In all, invasive physiology data were available for 103 PCI-treated patients and 93 sham-procedure patients. Mean FFR was 0.69 and mean iFR was 0.76, and these values were nearly identical between the PCI and sham groups. Nearly all the patients in the trial (97%) had one or more positive tests for ischemia, although just 75% of patients had FFR tests of 0.80 or less and 69% of patients had iFR results of less than or equal to 0.89.

But for observers hoping to see that functional tests would be able to pinpoint the patients who would benefit from PCI, ORBITA investigators found no such thing. The effect of PCI on the improvement in exercise time, after adjustment for prerandomization between-arm differences, showed no interaction with FFR or iFR. Moreover, as previously reported, PCI did not improve angina frequency score more than a sham procedure and here, too, there was no detectable evidence of an interaction with FFR or iFR.

Of note, however, in a new finding that Al-Lamee stressed was not prespecified in the original study, investigators looked at the more “black and white” finding of patient-reported freedom from angina and found that significantly more patients reported have no angina after PCI than after a sham procedure (49.5% vs 31.5%; OR 2.47; 95% CI 1.30-4.72). Notably, neither FFR nor iFR modified that effect.

Importantly, FFR and iFR were evaluated as continuous variables, as prespecified. Physicians, Al-Lamee noted, would prefer a cut point for FFR/iFR, below which intervening on a lesion would unequivocally make sense. In ORBITA, however, “we have not found a cut point at which point we can tell that all patients do better, or all patients do worse. What we have seen is that the more negative the FFR or iFR value, the less improvement you will have with angioplasty in terms of ischemia relief on stress echo, but that is a gradation along the spectrum and things don't suddenly improve at any particular cut point,” she said.

Commenting on the paper for TCTMD, Robert Yeh, MD (Beth Israel Deaconess Medical Center, Boston, MA), called the analysis “meticulous,” adding that it should help people interpret the original study, even if it comes as a disappointment to FFR/iFR enthusiasts.

“All along the continuum of FFR/iFR values, they found there were probably some patients who didn't derive any significant benefit over what they would have observed with placebo alone. And I think that on the one hand that might be surprising to those who are really strong believers in invasive physiology as a surrogate for symptomaticity, and I think that it basically proves that FFR/iFR is not a good surrogate for symptomaticity,” Yeh said. “But what it is a good surrogate for is the physiologic stenosis. We know that is true, and that was found here again.”

The essential interaction demonstrated by Al-Lamee et al here, Yeh continued, is that the lower the FFR/iFR value, the greater the improvement on the dobutamine stress test.

“FFR does measure what we think it does: if you're a ‘true-ist’ and you're very meticulous about what we think we're measuring, it's true that FFR does measure physiological significance of stenosis and that's confirmed here again,” he said. “What's really complicated is that symptoms are this really hard to describe entity and the attribution of symptoms to physiology is challenging. Symptoms are subjective.”

It's true that FFR does measure physiological significance of stenosis and that's confirmed here again. What's really complicated is that symptoms are this really hard to describe entity and the attribution of symptoms to physiology is challenging. Robert Yeh

Also commenting on the findings for TCTMD, Allen Jeremias, MD (St. Francis Hospital, Roslyn, NY), pointed out that exercise time is to some degree fraught as an endpoint, given that patients who are uncertain whether they had PCI or a sham procedure might have been reluctant to exercise to maximal capacity. This might partially explain why severity of the lesion on functional tests didn’t track with exercise time. Sample size might also be an issue, he noted, particularly since 25-30% of patients did not have flow-limiting lesions on invasive tests at baseline. “Also, I think that we don’t really understand what is causing chest pain,” Jeremias said. “In the vast majority of studies where patients were treated with stents, they still have residual chest pain after 1 year in the range of 20 to 30%, so there might be other factors involved.”

Al-Lamee, Yeh, and Jeremias all emphasized the same point, namely that a key contribution of this new ORBITA analysis is to question the link between ischemia and symptoms.

“I do think there is something about the way the brain interprets ischemia which probably varies from one person to the next, and it does so in a way that is not a 1:1 correlation,” Yeh observed. “And if ever that was proven, that was proven here, that more ischemia as measured by FFR does not correlate with improvements in symptoms, and what that probably means is that ischemia itself does not correlate in a 1:1 manner with symptoms.”

Al-Lamee made a similar point: “If you want to improve the blood supply to your patients' hearts, stenting certainly does that. We saw that in the first paper in the Lancet, and we see it in this paper as well. If you want to use FFR and iFR to predict the benefit that you're seeing, we have a very strong correlation which we can use. Again, if you want to be able to tell patients how much relief of symptoms they can get, you can tell them that 20% more patients are free of angina following angioplasty—not 100%, but 20%. And then, more than that we haven't been able to show, much as we weren't able to discern much more with our original primary manuscript.”

But, she continued, “many people from the interventional community will hope that you would be able to use FFR/iFR to predict symptom benefit and exercise time benefit,” and that was not established in ORBITA. “I think that’s probably because the link between that anatomical stenosis and symptom and exercise time is not as simple as we thought,” she said.

One thing the paper does not look at, she continued, is prognosis. “It may be that when you improve peoples’ ischemia—and we've shown that you do that with stenting—it changes their prognosis, and perhaps when we get the ISCHEMIA trial or more data becomes available, we may be able to see where this fits in terms of the long-term implications for angioplasty,” Al-Lamee said. “For the moment, this is just another piece to the puzzle, really, and probably once again doesn't entirely impact what we do. But I think it does impact what we should say to our patients and what we can expect to achieve from angioplasty, because it's the first time that there's been a blinded comparison using FFR and iFR and stress echo and [demonstrating] ischemia relief.”

An oft-heard argument is that patients “don’t care about ischemia,” Al-Lamee noted, adding that she doesn’t know “where that comes from.”

“When I tell my patients that I can improve the blood supply to their heart, that is important to a lot of patients, and intuitively, you would believe that that would be a good thing,” she said. “Now I can tell them that one in five more patients I treat with stenting will be symptom free, and for some patients that's very important. But it does challenge expectations and it allows you to think about who you would be most likely to help with a stent, and clearly it is in those with the most significant lesions that we should be considering stenting for, because you will have such a great deal of ischemia reduction. Maybe in others that's a bit more borderline, and maybe they are better off on their tablets.”

Clearly it is in those with the most significant lesions that we should be considering stenting for, because you will have such a great deal of ischemia reduction. Maybe in others that's a bit more borderline, and maybe they are better off on their tablets. Rasha Al-Lamee

Yeh predicted the new ORBITA data may face some flak for including the post-hoc, freedom-from-angina analysis. But in his mind, the important, hypothesis-generating finding in some ways more closely mirrors the conversations that doctors have with patients: do you still have angina? Yes or no?

“I think this study goes a long way to reassuring community physicians and interventional cardiologists that, under the right circumstances, offering PCI can still be the right therapy in the stable, elective setting,” Yeh said. As for whether the analysis represents a downgrade for FFR/iFR, Yeh insisted that invasive functional testing “still tells you something about the physiology. Now whether you should go in and intervene on that becomes an entirely different question, and I think that question is dictated less so by the physiology that you find on FFR and more so by how severe are the symptoms and what have you tried in terms of medications thus far.”

A Second Look at Stats

In an accompanying editorial, Ajay Kirtane, MD (NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY), points to a second important contribution of the current paper: a do-over of the statistical modelling used for the primary endpoint analysis in the original paper, to account for the impact of prerandomization changes from baseline. These changes were significantly higher in the PCI group than the sham group. The new numbers largely mirror those originally presented last year, Kirtane notes, but suggest an even greater improvement in exercise time with PCI versus sham, “further raising the question of whether the trial would have been ‘positive’ had it enrolled more patients or had it been conducted among patients on less intensive antianginal therapy.”

Kirtane as well as other experts interviewed by TCTMD spoke to the heated debate that has swept into ORBITA’s wake. Several made the observation that the trialists have borne the brunt of criticism that is better directed at the trial’s interpretation than its core findings.

The link between FFR/iFR and stress echo in this analysis, the fact that functional results did not track perfectly with symptoms, as well as the observation that fewer PCI-treated than sham-treated patients reported experiencing angina after 6 weeks are all important new insights, Jeremias said. “At the end of the day, all of these indicate that there is benefit to PCI. The question is, of course, how much and who benefits the most? I think the debate should be around that: who are the patients who benefit the most rather than whether they benefit at all.”

Photo Credit: Allen Jeremias

Note: Kirtane and Jeremias are faculty members of the Cardiovascular Research Foundation, the publisher of TCTMD.