BBC Makes New EXCEL Allegations, Says Investigators Ignored Safety Concerns
The latest revelations say the DSMB and the ESC had mortality concerns that weren’t shared, but the ESC disputes this.
A new report from BBC Newsnight doubles down on allegations that the EXCEL investigators ignored concerns raised by the data safety and monitoring board (DSMB) about an increased risk of death among patients with left main CAD treated with PCI.
Safety concerns were first raised in December 2019 following a similar report from Newsnight, one that also drew attention to missing MI data and added more fuel to a contentious disagreement between the study’s principal investigators and David Taggart, MD, PhD (University of Oxford, England). Taggart, originally the chairman of the EXCEL surgical committee during the design and recruitment phase, broke from the trial following a dispute over the MI data and the characterization of the 5-year results, specifically the risk of all-cause mortality.
In their latest report, BBC reporters state that when the main study results were published in the New England Journal of Medicine in 2016, investigators were aware of a potential risk of harm with PCI versus surgery at 5 years but did not mention this in their 3-year report.
Commenting on the allegations to TCTMD, a spokesperson for the trial sponsor, Abbott, said that the 3-year EXCEL data “reflects the original follow-up period and endpoints the study was powered to assess.” EXCEL, which included 1,905 patients with left main CAD of low-to-intermediate anatomical complexity, was powered for noninferiority testing of the primary composite endpoint of all-cause mortality, stroke, or MI.
In EXCEL, the risk of all-cause mortality at 3 years was not statistically different between the PCI- and CABG-treated patients (HR 1.34; 95% CI 0.94-1.91), but the curves were separating such that the 5-year results, which were published last year in the New England Journal of Medicine, did show that PCI was associated with more deaths from any cause when compared with CABG (13.0% vs 9.9%; OR 1.38; 95% CI 1.03-1.85).
Several EXCEL investigators, including principal investigator Gregg Stone, MD (Icahn School of Medicine at Mount Sinai, New York, NY), have previously told TCTMD that the study showed there was no heightened risk of cardiovascular mortality with PCI compared with CABG (6.8% vs 5.5%, respectively; OR 1.26; 95% CI 0.85-1.85) and that the majority of excess deaths at 5 years in the PCI arm were from noncardiovascular causes.
Emails from the DSMB
For the BBC article, reporters Deborah Cohen and Ed Brown reviewed emails from the DSMB, including from chairperson Lars Wallentin, MD (Uppsala University, Sweden), who in 2017 stated to EXCEL investigators that “it might be very concerning if in the future, suspicions were raised that already available information on mortality was withheld from the cardiology and thoracic surgery community.”
The BBC says Wallentin was worried that new guidelines on myocardial revascularization were being drafted and that the 3-year EXCEL results were being used to inform recommendations for the treatment of left main CAD even though there was emerging longer-term data about higher all-cause mortality with PCI.
Wallentin, contacted by TCTMD, said he was out of the office and currently unavailable for comment.
In 2018, the European Society of Cardiology (ESC)/European Association for Cardio-Thoracic Surgery (EACTS) published their guidelines and awarded PCI a class I (level of evidence A) recommendation for left main CAD in patients with a low SYNTAX score. PCI was also given a class IIa recommendation, meaning it should be considered, in patients with an intermediate SYNTAX score. CABG has a class I (level of evidence A) indication in all patients with left main CAD regardless of SYNTAX score.
According to Newsnight, there was even disagreement when the 2018 revascularization guidelines were drafted, and that an external reviewer was brought in by the ESC to examine the evidence. That review ultimately concluded stents were worse than surgery for patients with left main CAD, but the BBC alleges this information was not shared with all members of the guideline writing committee.
Contacted by TCTMD, a spokesperson for the ESC said that, in fact, the guideline authors did request a review of the evidence for left main CAD. The external review concluded there were uncertainties as to the equivalence of PCI and CABG, and that both procedures should not receive the same class of recommendation.
“This analysis was made available to all task force members and the review coordinators,” the ESC affirmed to TCTMD.
While the guidelines were being drafted, new evidence came to light from a pooled patient-level meta-analysis by Stuart J. Head, MD (Erasmus University Medical Center, Rotterdam, the Netherlands). Published in the Lancet in 2018, that analysis showed PCI was associated with a higher risk of all-cause mortality at 5 years compared with surgery, but there was no survival difference seen in patients with left main CAD (10.7% with PCI vs 10.5% with CABG; HR 1.07; 95% CI 0.87-1.33).
“Careful consideration of all reviewers’ comments, the external methodologist’s analysis, and the new evidence from Head et al led to the final, published recommendations after the third round of review,” said the ESC. “In the final document, PCI and CABG did not receive the same class of recommendation but PCI was downgraded for intermediate and complex left main disease. The document was approved by the task force, the review coordinators, and both ESC and EACTS.”
The ESC said the guidelines were reviewed by close to 100 experts. The task force, chaired by a cardiologist and cardiovascular surgeon, included noninvasive cardiologists, invasive cardiologists appointed by the ESC, and cardiovascular surgeons appointed by EACTS.
Pulled Support for ESC/EACTS Guidelines
Since the controversy surrounding EXCEL erupted in November 2019, EACTS has pulled its support from the left main CAD section of the European guidelines given the missing MI data and allegations that the EXCEL investigators downplayed the DSMB’s concerns. The current recommendations are unsafe for patients with left main CAD, according to EACTS, and they have called on Abbott, the study sponsor, to make individual patient-level data available to an independent research group for reanalysis. They also suggest that EACTS, as well as the ESC, be involved in the process.
Abbott told TCTMD they have “contracted an independent biostatistician to conduct analysis of the EXCEL data to include alternative definitions of MI.” No other information on the reanalysis is available at this time.
The Cardiovascular Research Foundation (CRF) has clarified that its review is looking specifically at the standard operating procedures of its Clinical Trial Center (CTC) as followed during the course of the trial. The CTC, a division of CRF, was engaged by Abbott to provide services, including organizing the independent clinical endpoint committee (CEC), for the EXCEL trial. The CRF review does not include a reanalysis of any of the EXCEL data.
Stone, Patrick Serruys, MD (Imperial College London, England), and Joseph Sabik III, MD (University Hospitals Cleveland Medical Center, OH), the trial’s principal investigators along with Pieter Kappetein, MD (Medtronic; formerly Erasmus Medical Center, the Netherlands), did not respond to requests from TCTMD for their reaction to the BBC story.
Note: Stone is Co-Director of Medical Research and Education at CRF. Several additional EXCEL co-authors are faculty members of CRF. CRF is the publisher of TCTMD.