The Ethics of Sham PCI and Clinical Relevance of ORBITA-2

This transcript has been edited for clarity.

John M. Mandrola, MD: Hi, everyone. This is John Mandrola from theheart.org | Medscape Cardiology. I'm at the American Heart Association meeting in Philadelphia, and I'm so excited to present and have a conversation with two saviors of interventional cardiology, primary author Christopher Rajkumar and senior author Rasha Al-Lamee, authors of the ORBITA-2 trial, which came out positive for PCI. First of all, congratulations!

Rasha Al-Lamee, MBBS, MA, PhD: Thank you, John.

Mandrola: You've already done many interviews on this. Before we get into some of the specific things that have been going on, talking online, just give us the basic rundown of what you did and what you found.

ORBITA-2 Key Findings

Christopher A. Rajkumar, MBBS, PhD: ORBITA-2 was a placebo-controlled trial of percutaneous coronary intervention (PCI) for stable angina, just like the first ORBITA trial. Essentially, the key differences in ORBITA-2 compared with ORBITA are that we intentionally diverged from clinical guidelines in ORBITA-2 — I'm sure that will be the focus of some of the questions today — in that we wanted to test the unattenuated efficacy of PCI for angina relief. The only way we could do that is by taking patients off their antianginal medications.

There were some other differences in the trial as well. We incorporated patients with single- and multivessel disease, and evidence of ischemia and evidence of symptoms were both mandatory prior to randomization. The patients were followed up for longer — 12 weeks instead of 6. The big difference is also that we used a novel clinical endpoint called the angina symptom score, which encompassed daily reporting of angina from our patients, how much antianginal restart they needed, and any adverse clinical events like acute coronary syndrome and death.

Mandrola: ORBITA was a placebo-controlled trial of PCI as an add-on to antianginal drugs. This was totally different in that these were symptomatic patients on antianginal drugs, and then a couple of weeks before, you took them off the antianginal drugs to isolate the effect of PCI alone. Rasha, tell us what happened.

Al-Lamee: We found that angioplasty, compared with a placebo procedure, improved that angina symptom score and that patients had a far better health status on our angina symptom score. In fact, that was collected every single day. We saw that impact immediately and it was sustained throughout that 12-week period.

We also saw that patients in the PCI group were three times more likely to be free from angina and more likely to have an improvement in quality of life and an improvement in all of the various categories in terms of exercise time on the treadmill, improvement in CCS class, and also an improvement in the various domains of the SAQ questionnaire.

Mandrola: Thank goodness.

Al-Lamee: Well, I know what it's like to be negative, so it was quite nice for it to be positive this time!

Mandrola: The exercise time on the treadmill, which was the endpoint of ORBITA-1, was also positive by about 60 seconds. Is that a lot or a little?

Al-Lamee: It equates to one full-dose antianginal agent. You're thinking about a procedure with costs and potential risk — and it's not negligible either, that cost or that risk — for one antianginal agent's worth of treadmill exercise time improvement. It's still, as a blinded effect size, far smaller than the unblinded effect size that we saw from the original ACME trial of 96 seconds. Of course, it's far larger than what we saw in ORBITA, which was 16.6 seconds.

Mandrola: Chris, the slide you showed in the presentation of that gnarly lesion in the left anterior descending artery, which no one would walk away from in my neighborhood, that's only equivalent to one antianginal tablet. It's hard to believe.

Rajkumar: An important point is that, of course, that's an average treatment effect.

Mandrola: Okay, fair enough.

Rajkumar: Of course, we've got some huge responders in ORBITA-2 and we've got many patients who saw no response at all. That is a whole other subject of further research.

Al-Lamee: That's a "watch this space" moment, I think.

Is It Ethical?

Mandrola: Your trial has obviously generated many things online. One is the ethics. People ask how is it even ethical to do this. Let's answer that.

Al-Lamee: I've been answering questions on ethics for a long time now. It's been about a decade where people have said that the trials that we're doing are potentially unethical. I'd say it's unethical to do procedures or to give treatments without knowing if they work.

What we wanted to do here was test whether angioplasty worked. I think it's absolutely reasonable to do a placebo-controlled trial as long as the patients understand that they might potentially have a placebo procedure without any potential benefit, but there's an associated risk. Then I guess you're going to come to the ethics of taking off their antianginals.

Mandrola: We want to talk about that, but I also want to emphasize your comments to my atrial fibrillation ablation community, which is seemingly afraid to do a proper sham control.

Al-Lamee: Look, it's time for these trials to become the entry standard, not the gold standard that we try to achieve at the best possible research facilities — the entry standard for a new procedure that might purport to improve a subjective endpoint like quality of life or an established procedure that has risk and cost.

I often say, if you're giving someone homeopathy and you don't think there's any risk associated with it, it doesn't really matter if all of the therapeutic effect is placebo. If you're giving someone a treatment that might have risk and cost, then you need to understand what proportion of the overall therapeutic effect is placebo and what proportion is physical. The only way you can work out that physical effect is to do a placebo-controlled trial.

Mandrola: Thank you. Let's move on to the issue of taking people off their medicines. There's been some conversation that this is like taking someone off their cancer chemotherapy. Either one of you can address that.

Al-Lamee: I'm happy to take that one. There were medications that are there for risk reduction or disease modification, and those medications in ORBITA-2 were continued. Everyone was on their aspirin and their statins. If they were on low-dose statins, they were placed on high-dose statins. As you know, after randomization, they were actually on dual-antiplatelet therapy. We felt that those antianginal agents were there for symptom relief and they have no prognostic benefit.

Ultimately, what we were doing was akin to stopping someone's paracetamol when they're having headaches or stopping their co-dydramol when they've got knee arthritis. We were not stopping their chemotherapy agents because they had cancer. Now, of course, if they were on an antianginal agent that had other properties, such as rate control for atrial fibrillation or heart failure, those continued, and they were counted as part of the angina symptom score.

Importantly — and this was important to the ethics — if patients had symptoms that they deemed to be too much, they called the team, they had 24/7 access to us, and they were then able to be restarted on antianginals or uptitrated, and that was counted in the score.

When we sat down with the results of ORBITA and designed ORBITA-2, we did it with a patient focus group, and we asked them what they wanted to understand and what was important to them. I think many physicians will understand this. Many patients told us that, for them, taking more tablets was worse than having a little bit more pain. That's why the trial was designed the way it was, with more antianginal medication scoring higher than higher frequencies of angina.

Clinical Relevance?

Mandrola: It's a scientifically great study. It shows pretty clearly that PCI has a placebo-resistant effect. How does it apply clinically? I've heard some people say this doesn't answer a clinical question and shouldn't be used in the clinic. Either of you, how do you feel about this?

Rajkumar: We didn't recruit 500 patients across two placebo-controlled trials just to generate data. Of course it's there to apply to clinical guidelines. Real-world evidence, which I showed yesterday, shows that actually the average patient having an elective cardiac catheterization is not on three doses of antianginal medications as in the first ORBITA trial.

In fact, about half were either taking none at all or one. As Rasha touched upon, many patients struggle to take high-dose antianginal medications for what is essentially an indefinite period of time. We think this certainly can be, and should be, practice-changing in that we should be able to have a more patient-centered approach and offer patients more of a choice, because now we have an evidence base for that choice of an upfront antianginal procedure, such as PCI, or upfront antianginal medications.

Al-Lamee: It is also important for people to remember which patients were in this trial. They were patients who are symptomatic with angina. They had severe anatomical stenoses and they had evidence of ischemia. What we wouldn't want is a throwaway message that now takes patients with wishy-washy symptoms, nonsevere stenoses, and no evidence of ischemia for having angioplasty. People need to think about the criteria of the patients in this trial. They were pretty symptomatic with, as you saw in some of those angiograms, some pretty severe stenoses.

Mandrola: I think many patients would just say, "Fix it." Of course, doctors in our health system are incentivized to fix it. I can imagine that few patients would decide not to fix.

Al-Lamee: Perhaps it does reflect different patients in different geographies. Our patients are quite up for that conversation, actually. We do find that there are certainly patients who don't want that risky procedure. I think what's also possible — we know this from COURAGE and ISCHEMIA now — is to take your time.

Actually, one of the things that the guidelines have told us is that if you take multiple antianginal medications and it doesn't work, consider a procedure on top. ORBITA showed us that the incremental benefit of angioplasty is very small, so we're probably subselecting a bunch of patients who have the least to gain.

Perhaps actually the strategy — and I'm just going to put that out there that it is, of course, not tested — but maybe when you give your patient their antianginal and they start to feel better, perhaps that's the patient with cardiac pain. Maybe the conversation needs to be, "Are you happy to continue on this or do you want us to consider reducing your burden and fix your stenosis?"

Mandrola: When I look at the graphs that you show for the anginal scores and for the angina class, patients who had PCI can still have residual pain, and patients who didn't have PCI can have resolution. What's happening there?

Rajkumar: One thing that we learned from this is that, going back to the medical school textbooks, we learn how to take a history. We think we know how to diagnose angina. Actually, the link between coronary stenosis, ischemia, and angina is much less linear than we thought it was. Our trial wasn't designed to look into the cause of those residual symptoms. It will be, I suspect, a mixture of perhaps some patients with some microvascular disease, albeit the ischemia was reduced significantly; some patients with some noncardiac symptoms; and, of course, some people who continue to have symptoms from their epicardial coronary arteries. PCI isn't perfect. That's a clear message.

What's Next: ORBITA-3, -4…

Mandrola: Last question. What's next? Are you just going to take it easy and do PCI?

Al-Lamee: Our team's restless. We've already got a plan. As of yesterday, I had a plan for ORBITA-3. By the time we had dinner last night, we've now got a plan for ORBITA-4 as well. We've got all sorts of ideas about what's next.

This question about residual symptoms is clearly important, but also what everyone's going to see before we get to ORBITA-3 and ORBITA-4 are many secondary analyses from the data we have in ORBITA-2, thinking about the predictors of response. It's time we stop thinking about mean responses and consider which patient feels absolutely 100% the next day and which patient doesn't feel that. How can we predict that before we put in our stent?

The other thing we're going to do is combine the data from ORBITA-1 and ORBITA-2.

Mandrola: Okay.

Al-Lamee: This will be a patient-level meta-analysis, where we'll be able to look at the introduction of antianginals and what they did — with the consideration, of course, that there are some endpoints that match up and others that don't. In a combined dataset, what can we understand from that data?

What's also really important to consider is that, when you've got your patient in the clinic and you're considering whether to take them to the cath lab, many physicians will probably have an idea already of which patients might have benefit. The bit of ORBITA-2 that we're really excited about is that we did video diaries of many of the patients telling us about their symptoms. We're going to analyze whether just the description of symptoms, the way a patient tells us, is a predictor of response.

Mandrola: A qualitative study. That will be awesome.

Al-Lamee: I keep saying the videos are going to be the gold dust because the videos are what we have in clinic, right? Them telling us their history and then how we can work out which ones we can fix.

Mandrola: Excellent. Thank you so much for coming in and settling some of these questions for us. I really appreciate it.

Al-Lamee: Thank you, John. It was a pleasure.

Follow Medscape on Facebook, X (formerly known as Twitter), Instagram, and YouTube

Comments

Commenting is limited to medical professionals. To comment please Log-in.

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.