FAME 3 published and guideline updates: PCI vs CABG in multivessel disease debate continues.

By Dr Oliver Jones

Key points:

In FAME 3, PCI failed to meet noninferiority versus CABG for composite MACE in three-vessel CAD patients without severe LVSD
In contrast, the new 2021 ACC/AHA/SCAI 2021 Guideline for Coronary Artery Revascularization downgraded the recommendation for CABG in multivessel CAD from strong to weak – the same recommendation as PCI
US, European, and other international surgical societies have strongly rejected the latest 2021 ACC/AHA/SCAI guideline and recommended clinicians rely on other existing guidelines

The 13th January 2022 issue of the New England Journal of Medicine (NEJM) included publication of the one-year results of the FAME 3 trial, alongside an accompanying editorial from Frederick Welt MD: CABG versus PCI — End of the Debate?

First presented at TCT 2021 in November, FAME 3 is an investigator-initiated trial to assess non-inferiority of Fractional Flow Reserve (FFR)-guided percutaneous coronary intervention (PCI) with second-generation coronary artery drug-eluting stents (DES) against coronary artery bypass grafting (CABG) in three-vessel coronary artery disease (CAD).

CABG has previously outperformed PCI in multi-vessel CAD in SYNTAX, FREEDOM, and BEST trials – FAME 3 sought to test the latest PCI technologies against this standard. The hypothesis was that the combination of FFR and second-generation DES might improve PCI and make it comparable in efficacy to CABG. The key finding was that PCI failed to meet non-inferiority for the composite efficacy outcome of major adverse cardiac events (MACE), which included death, myocardial infarction, stroke, or repeat revascularisation.

Approximately 1500 patients were randomised to FFR-guided PCI or CABG following an angiogram demonstrating at least 50% stenosis in each of the three major epicardial arteries (not including left main disease), and subsequent assessment by a heart team that both PCI and CABG were reasonable. Average age was 65 years, 82% were male, 28% had diabetes, 82% had ejection fraction (EF) >50%, and 39% presented as acute coronary syndromes – patients with EF <30% or recent ST-elevation myocardial infarction were excluded. The composite primary endpoint occurred in 10.6% in the PCI arm and 6.9% in the CABG arm, giving a hazard ratio of 1.5, with a 95% confidence interval of 1.1-2.2, which exceeded the prespecified threshold of 1.65. The P-value for noninferiority was 0.35.

These results are particularly remarkable as they compare 1-year outcomes: in previous studies, CABG has demonstrated higher up-front risk but accrued benefits in the longer-term.

As expected, safety endpoints, including major bleeding, arrhythmia, acute kidney injury, and rehospitalisation within 30 days, were all significantly higher in the CABG arm. Additionally, a prespecified subgroup analysis found that in patients with a low SYNTAX score (<23), PCI was numerically superior to CABG – although it is important to note that the study design means this subgroup analysis is hypothesis-generating only.

Arms race

These results were surprising to some. Second-generation drug-eluting stents (DES) were previously demonstrated to outperform older technologies in the SPIRIT meta-analysis. Equally, FAME 1 showed a lower rate of MACE for FFR-guided PCI compared to angiographically-guided PCI – though the benefit of FFR is more controversial, and some of the difference in FFR-guided PCI may be seen because higher SYNTAX score patients can be reclassified to lower risk groups.

This apparent paradox points to a second important finding in FAME 3: CABG is not a static target. In fact, despite comparable patient demographics to SYNTAX in 2009, rates of MACE at one year were lower for both CABG and PCI in FAME 3. In SYNTAX, MACE was 17.8% for PCI and 12.4% for CABG, whereas in FAME 3 outcomes were 40-50% lower in both arms. This likely reflects the significant scientific and clinical advances made in both PCI and CABG.

Guidelines move in the opposite direction: Double-downgrade for CABG

In the interval since TCT 2021, the American College of Cardiology (ACC), American Heart Association (AHA) and Society for Cardiovascular Angiography and Interventions (SCAI) jointly released the 2021 Guideline for Coronary Artery Revascularization.

In stark contrast to these findings, their own previous 2011 guidelines, and 2018 Guidelines on Myocardial Revascularization from the European Society of Cardiology (ESC) and European Association for Cardio-Thoracic Surgery (EACTS), the new ACC/AHA/SCAI guideline downgraded CABG from a class I (strong) recommendation in the previous guidelines to IIb (weak) in patients with stable multivessel CAD and normal left ventricular EF. This downgrade also means that CABG and PCI are given the same class of recommendation (COR) in the guidelines. In severe left ventricular systolic dysfunction (LVSD) and multivessel CAD, the COR for CABG remains I, and in mild-moderate LVSD the COR is now IIa.

The principal evidence cited for this decision is the ISCHEMIA trial, which found that an initial invasive strategy did not provide reduction in MACE when compared to an initial conservative strategy in stable CAD and moderate to severe ischemia. The guideline also references a “more detailed review of earlier studies,” and appears to acknowledge controversy by mentioning “several hours of deliberation.”

“Extreme Concern”

The reaction from surgical societies was swift and explicit.

In a preprint published in the annals of thoracic surgery, the American Association for Thoracic Surgery (AATS) and the Society of Thoracic Surgeons (STS) withdrew their support for the updated ACA/AHA/SCAI guidelines, citing both the downgrading of CABG in the treatment of multivessel CAD and a “lack of recognition of superior long-term benefits of CABG vs PCI in decreasing repeat re-interventions and post-procedural myocardial infarctions.”

Focussing specifically on the interpretation of ISCHEMIA, AATS/STS highlighted that ISCHEMIA grouped PCI and CABG into a single invasive arm and was neither designed nor powered to assess CABG alone – no CABG subgroup analysis of ISCHEMIA has been published. They pointed out that only 20% of patients in the initial invasive arm received CABG, less than the number who received OMT (21%).

EACTS also published a statement on their website outlining their “extreme concern” that the downgrading of CABG “may lead to avoidable loss of life.” They recommended clinicians rely on the ESC/EACTS guideline exclusively, and that AHA/ACC/SCAI review the evidence.

Both AATS/STS and EACTS also called in to question the design of the guideline committee and argued that surgical societies were not adequately represented. In their statement, EACTS described it as “astonishing that no surgical association was involved, co-authored or endorsed these guidelines,” and “unfathomable [to] include only one side of the Heart Team.”

Next steps

There has so far been no published response from the guideline writers. Further debate is inevitable, especially as PCI and CABG – and medical therapy – continue to evolve rapidly. PCI with intravascular imaging and trials in low SYNTAX score patients are both likely avenues for future comparison. Meanwhile, upcoming trials in 2022 will provide valuable information about the comparative roles of PCI and CABG in LVSD.

The challenge for clinicians remains translating the data to individual patients. FAME 3 was widely received as favouring CABG in the groups studied. While three- and five-year follow-up of the FAME 3 participants is only expected to strengthen the case for CABG, most acknowledged complexity and agreed that decisions should be patient-specific, and guided by a multi-disciplinary Heart Team. This was summed up by Professor A J Kirtane, Professor of Medicine at Columbia University Irving Medical Center: “for 3VD gotta take patient off the table.”