CABANA Results Largely Generalizable to Real-World A-fib Patients

The CABANA trial, which evaluated the effect of catheter ablation versus medical therapy on hard clinical outcomes in patients with A-fib, failed to meet its primary endpoint, sparking ongoing debates about what can be gleaned about the usefulness of ablation. But, whatever one thinks of the trial, the results are likely to be generalizable to patients seen in everyday practice, a large, observational analysis suggests.

Indeed, in data pulled from a time period overlapping the enrollment period for CABANA, 73.8% of patients who either underwent ablation or were treated with medical therapy (including antiarrhythmic or rate control drugs) met trial eligibility criteria, Peter Noseworthy, MD (Mayo Clinic, Rochester, MN), and colleagues report in a study published in the April 21, 2019, issue of the European Heart Journal.

“So the CABANA population does not appear to be a highly selected population in terms of basic demographics and patient characteristics,” Noseworthy said. “Whatever your take is from CABANA, it is likely to be applicable to the patients we see in everyday practice.”

It’s in the group of CABANA-like patients that ablation was associated with the largest reduction in the risk of a composite of all-cause mortality, stroke, major bleeding, and cardiac arrest—a composite similar to the primary endpoint in the trial—in the real-world data set (HR 0.70; 95% CI 0.63-0.77). That hazard ratio is similar to the one seen in the “treatment-received” analysis from CABANA (HR 0.67), which was performed to account for the high numbers of patients who either crossed over from the drug group to undergo ablation (27.5%) or did not undergo ablation as assigned (9.2%).

Still, Noseworthy told TCTMD, the consistency of the findings from the two types of analyses—which both have limitations—should not be interpreted as being supportive of a large, beneficial impact of ablation on hard outcomes.

“My suspicion is that there’s probably a small benefit in terms of hard clinical outcomes with ablation, but if you’re interested in reducing stroke, the key is anticoagulation with good adherence when appropriate, and if you’re interested in reducing mortality, it’s all of the other cardiovascular and lifestyle interventions that are known to reduce mortality,” he said.

“Catheter ablation probably plays a very small role in that—if at all—and I think its main role is for symptom control, even in light of what we’ve seen,” he continued. “It would be a very inefficient way to try to reduce stroke to perform catheter ablation on asymptomatic patients and expose them to the up-front risks of an ablation procedure. That would be hasty and it would not be a correct interpretation of our data.”

Commenting for TCTMD, Rahul Doshi, MD (University of Southern California, Los Angeles), said he feels that most in the electrophysiology (EP) community believe CABANA reflects real-world practice and that this analysis bears that out.

And the results, he added, are compelling in that they suggest a beneficial impact of ablation on hard clinical outcomes. Noting the limitations of both the observational analysis and the secondary CABANA analyses, however, he acknowledged that there is not yet best-level evidence to support that effect.

“I think the entire EP and cardiology community are, of course, waiting for the randomized, prospective trial that is appropriately designed, followed, and powered to demonstrate a benefit in a true intention-to-treat analysis—ie, the best level of data,” Doshi said. Considering the difficulty in completing CABANA, it might be challenging to get another trial up and running, but over time with the accumulation of more and more analyses like this one, it might provide the impetus needed to make it happen, he added.

Assessing Generalizability

Noseworthy explained that the current analysis was completed in parallel with the CABANA trial without knowledge of the trial’s results. The idea was to look at whether the trial results—whatever they were—could be applied to a broader swath of centers beyond the experienced ones participating in CABANA.

The investigators pulled information from the OptumLabs Data Warehouse, which contains administrative data from US patients covered by private insurance or Medicare Advantage, on 183,760 patients with A-fib who were treated with ablation (6.5%) or medical therapy (93.5%) between August 2009 and April 2016, when CABANA was enrolling.

Through a mean follow-up of over 2 years, ablation was associated with a lower rate of the composite endpoint (4.51 vs 6.07 per 100 person-years; HR 0.75; 95% CI 0.70-0.81), with a stronger association seen in those who would have met CABANA inclusion criteria.

Among the 22.4% of patients who met at least one of the trial’s exclusion criteria, ablation was still associated with a reduction in the composite endpoint, but the relationship was weaker (HR 0.85; 95% CI 0.75-0.95). And in the 3.8% of patients who did not meet inclusion criteria (but also didn’t meet any of the exclusion criteria), the relationship was not significant (HR 0.67; 95% CI 0.29-1.56).

That latter group of patients had a low risk of events overall, Noseworthy pointed out. “That’s in some ways reassuring that these patients are generally healthy and are unlikely to suffer the consequences of A-fib, but it means that we shouldn’t take CABANA to motivate preemptive catheter ablation in asymptomatic patients who are young for the purpose of reducing stroke or mortality,” he said. “I think that’s clear.”

‘Apples and Oranges’

In an accompanying editorial, A. John Camm, MD (St. George’s, University of London, England), says “there is an unwritten implication that the results of the present study reinforce the superiority of left atrial ablation over other treatments as demonstrated in the as-treated analysis in CABANA, and to some extent legitimizes the as-treated analysis of the outcomes. This conclusion should be rejected, although in any guideline process the results of well-conducted observational studies can certainly add strength to any recommendations.”

Camm points out that characteristics of the study populations and event rates differed between this observational analysis and the CABANA trial.

“This does not detract from the main conclusion of the authors that the CABANA results may be generalizable, but reduces our temptation to assume that the study necessarily validates the CABANA as-treated outcome analysis,” Camm says. “Apples and oranges are both edible fruit, but they certainly do not taste the same.”

If you have a hammer, everything looks like a nail. Marc Moon

Acknowledging for the differences between the studies and the limitations of observational data sets and also pointing out that as-treated or per protocol analyses from randomized trials have shortcomings or their own, Noseworthy said, “You always have to interpret these results—either per protocol or [from] an observational, adjusted analysis—with that grain of salt.”

What’s clear “is that the patients enrolled in CABANA looked like patients we encounter in everyday practice,” Noseworthy added. “The vast majority of patients with atrial fibrillation have at least one risk factor and would be eligible for the trial.”

Marc Moon, MD (Washington University School of Medicine in St. Louis, MO), commented to TCTMD that what can be taken away from this study is that both treatment  approaches—ablation and medical therapy—are acceptable. Physicians who perform ablations might see benefit with that approach, whereas general cardiologists might see benefit with medications, he added.

“If you have a hammer, everything looks like a nail,” said Moon, a member of the Surgeon’s Council for the American College of Cardiology. “I think that’s why we always have to take multiple studies into account when we try to make recommendations, and the recommendations actually have to be made on a per-patient basis.”

What both CABANA and the observational data indicate is that “the medicines that we use are pretty darn good and you don’t have to feel you’re getting slighted therapy if you don’t undergo an ablation, because it doesn’t always work and it’s a modestly invasive procedure,” Moon said.