Believing SCOT-HEART: New Analysis Probes Plausible Mechanisms of Benefit

 

(UPDATED) Faced with a barrage of complaints that using CT angiography (CTA) to triage stable chest pain patients could not possibly produce a more than 40% drop in death/MI over 5 years, SCOT-HEART investigators have released a new paper trying to drill down to the mechanisms of benefit.

“Various people have sort of said, or tweeted, ‘How could this have happened? No trials have shown this before, what’s the mechanism, is it really plausible, is it believable?’ And this was an attempt to address those sorts of 'Wow, how can this be true?' responses,” senior SCOT-HEART investigator David Newby, MD, PhD (University of Edinburgh, Scotland), told TCTMD.

The main SCOT-HEART results, presented at the European Society of Cardiology Congress 2018 and published simultaneously in the New England Journal of Medicine, showed a 41% reduction in the endpoint of coronary heart disease death or nonfatal MI at 5 years when coronary CT angiography was used instead of standard care, predominantly stress testing, to guide the management of patients presenting with chest pain. But skepticism over these dramatic results has dogged the trial, prompting the current analyses.

This was an attempt to address those sorts of ‘Wow, how can this be true?' responses. David Newby

“We have presented a multifaceted analysis that consistently and robustly demonstrates the plausibility of a reduction in long-term coronary events consequent on investigating patients with stable chest pain using coronary CTA,” write Philip D. Adamson, MD, PhD (Christchurch Heart Institute, University of Otago, New Zealand), and colleagues in the paper, published online today ahead of print in the Journal of the American College of Cardiology. “If we are to improve the prevention of future myocardial infarction, coronary CTA would appear to be the most effective and indeed the only proven investigative approach in patients with stable chest pain.”

Adamson previously proposed some of these mechanistic insights at the Society of Cardiovascular Computed Tomography meeting in July 2019, as reported by TCTMD.

Symptoms, Risk, and Actions

In the paper, which investigators note is comprised of post hoc analyses, Adamson, Newby, and colleagues look first at reductions in subsequent events across the range of symptom and risk categories in the trial. Here, they point out, patterns were consistent across categories, but were most pronounced among the highest-risk patients. For example, among patients presenting with nonanginal chest pain, just 50% of these were found to have nonobstructive or obstructive CAD on coronary CTA. By comparison, among patients with “possible” angina, rates of CAD were—as might be expected—much higher at 66.9%.

Next, they looked at temporal patterns of benefit and found these to differ between groups based on their initial presentation. Among patients with prior coronary heart disease, the benefits of a CTA strategy were “more immediately apparent” with event curves separating very early, a finding likely explained by the higher number of patients sent for revascularization in the first year in the CT group. By contrast, in patients with nonanginal chest pain or possible angina, the investigators saw a delay in the separation of event curves, whereby a CT-based strategy was associated with fewer subsequent events than usual care over time. These delays were approximately 1 year among the nonanginal group and several months in the possible-angina group, both of which are plausible, Newby explained, given the time that typically elapsed between the ordering of the CT test, the communication of the results, and the time needed for any new or uptitrated prescriptions to take effect.

Revascularization rates were higher among the CT-guided patients than in the standard-care group in the first year (246 vs 208; HR 1.21, 95% CI 1.01-1.46). This pattern was reversed between years 1 and 4, however, such that more patients in the standard-care group underwent revascularization than in the CTA arm (59 versus 33; HR 0.59, 95% CI 0.38-0.90). In all, 18 of these procedures were prompted by an MI in the standard-care group compared with eight in the CTA group, although the difference was not statistically significant.

SCOT-HEART proponents have always argued that better use of preventive therapies in the CTA group is the other most likely reason for the improved survival seen with CTA-guided care. Delving into these numbers, SCOT-HEART investigators found that just 5% of patients in the standard-care arm had their treatment tweaked within 6 weeks of their clinic consultation, versus 23% in the CTA group. While use of antiplatelet therapy actually fell in the standard-care group over the first year (48% to 41%), it rose slightly in the CTA group (49% to 52%). Overall statin use increased significantly in both groups, from 43% to 50% in the standard-care group and from 44% to 59% in the coronary CTA group. But for both drugs, usage remained higher over the 5 years of the trial in the CTA-guided group than in the standard-care group.

And while critics of the trial have pointed out that a 10% difference in statin usage couldn’t possibly explain the subsequent reduction in hard events, SCOT-HEART investigators, in response, did additional analyses to demonstrate that prescription of these drugs appeared to be more appropriately targeted among the CTA patients.

Newby estimated that approximately one in three patients in the standard-care group were overtreated, while roughly the same proportion were undertreated in SCOT-HEART. New analyses reveal that despite having similar 10-year risk scores, patients with documented CAD on CTA were significantly more likely to be taking preventive medication than were patients with no evidence of CAD on CTA, and this was “most apparent in participants with the lower cardiovascular risk scores,” Adamson et al write.

“What CT did is it cancelled treatment in some and started it in others, so that the 10% difference is not [the point],” Newby said. “It's that the people who had disease got the treatment.”

This underscores the fact that cardiovascular risk scores “are not great,” Newby continued, despite the fact that they are heavily relied upon for guiding care.

To TCTMD, Newby noted that lifestyle changes may have also been more common among patients found to have coronary disease. “Unfortunately, trial design didn't allow us to capture this, so we didn't ask them about smoking cessation, whether they walked more, whether they had a better diet, but given . . . that adherence to therapy was better if you knew you had coronary disease, it's not unreasonable to suspect that they also might have lead healthier lives. Once you've been told you have coronary disease and you might have a heart attack, you tend to get your act together.”

Convincing the Unconvinced

Asked whether he felt these analyses will persuade those who’ve struggled with the SCOT-HEART results, Newby hedged his bets. “I think there's a lot of resistance,” he said. “Nuclear has been the bedrock of imaging and assuming [a test is] positive, you get an angiogram and you get revascularized, and I think there's a lot of buy-in with that. CT is reducing caths, and it’s minimizing the [value] of functional tests in which a lot of people have put a lot of faith over the years. It's just getting your head around the change.”

A reluctance to let go of the revenue stream also afforded by functional tests also poses “a slight reimbursement issue,” Newby added. He alluded to a fact frequently raised by CT proponents, namely that there has never been a trial to show that a noninvasive functional test has an impact on outcomes. “We have arguably two or three trials that have shown that CT does improve outcomes,” Newby said. “If we are truly evidence based, why are we not embracing this?”

Bottom line: this is an intriguing result, but what we really need is another trial that comes along, in another area but similar to this, and that also shows a benefit. Leslee Shaw

An accompanying editorial by Leslee Shaw, PhD (Weill Cornell Medical College, New York, NY), and colleagues echoes that point. The diagnostic pathway based on the demonstration of ischemia is “long intrenched in cardiovascular medicine,” despite the lack of evidence supporting an impact on outcomes, they write. Although there are “lingering” issues left unresolved in SCOT-HEART, the editorialists note, “the question that now remains is whether our traditional diagnostic pathway should evolve alongside the ever-changing contemporary evidence on the value of noninvasive imaging for suspected CAD.”

Speaking with TCTMD, Shaw pointed to a number of leaps of faith within the new SCOT-HEART paper. Chief among these, she said, is the theory that early revascularization may have had an impact on outcomes since no trial, with the exception of FAME, have demonstrated a benefit of revascularization for death/MI in patients with stable ischemic heart disease.

That said, Shaw continued, the analysis “does open some interesting thoughts and ideas,” particularly as to the timing of benefit across the different levels of baseline symptoms and risk as well as the importance of appropriately targeting preventive medications.

Sanjay Kaul, MD (Cedars-Sinai Medical Center, Los Angeles, CA), has previously set out his concerns about SCOT-HEART in JAMA Cardiology. Contacted this week by TCTMD for his views on this latest paper, Kaul pointed out that the three “plausible mechanisms” put forward by the authors to explain the more favorable outcomes with CTA are fundamentally “not verifiable.”

First, he said in an email, “it is difficult to confirm that nonobstructive plaques were indeed the ‘culprit’ lesions. And the difference in prognostically significant CAD was too small (64 vs 72) to account for the large treatment effect. Second, as I and others have previously argued, the difference in antiplatelet and statin therapy is insufficient to explain the large treatment effect. Third, no compelling evidence exists to support reduction in risk of MI with revascularization in the stable angina/CAD cohort”—the point also made by Shaw.

Last but not least, he continued, “it is difficult to explain why a similarly increased frequency of preventative therapies and coronary revascularization in the CTA arm failed to improve outcomes in PROMISE trial. So, I am not sure whether, with this exercise in speculation and modeling, the authors have really moved the needle to satisfy the critics.”

That said, some of the criticism SCOT-HEART has faced is unwarranted, Kaul noted. “Critics have argued that lack of a clear and plausible mechanism underlying the primary outcome benefit is a major limitation of the SCOT-HEART trial. This is a rather unfair criticism, because outcome trials are not designed to unravel the potential mechanisms of benefit.”

Shaw made a similar point, noting that efforts like this one by the SCOT-HEART investigators, also highlights some of the problems of pragmatic trials. Trials like these are not designed for follow-up care to be “protocolized and case-managed” like, for example, the ISCHEMIA trial coming out next month, she said.

Famously, SCOT-HEART was delivered at a cost of less than a million UK pounds, however, whereas the long-running National Institutes of Health-sponsored ISCHEMIA trial has run up a bill in the range of 100 million US dollars.

“Bottom line: this is an intriguing result, but what we really need is another trial that comes along, in another area but similar to this, and that also shows a benefit,” Shaw stressed.

That said, she added, “in today's environment it would be hard to replicate SCOT-HEART, [because] there's just so much evidence about the importance of nonobstructive atherosclerotic plaque.”

This is where Newby believes the field has moved on from ischemia as a means of guiding care in the diagnosis of stable chest pain. Ischemia, he reminded TCTMD, “is just picking up severely stenotic disease, and that's very important but not as important as how much plaque you've got.” ECGs and nuclear tests of ischemia evolved to get around the safety hazards of early invasive coronary imaging, but those days are long gone,” he said. “If we’d had a CT scan 40 years ago, I don't think ischemia tests would have happened to the level that they have today.”