Taggart Rebuts EXCEL Response, Insisting Concerns Over MI and Mortality Remain
Taggart, who withdrew his name from the NEJM paper, says the EXCEL investigators’ response does not answer all his questions.
CORRECTED—The leading detractor of the EXCEL trial is not backing down from his criticisms of the trial’s leadership following a failure to publish myocardial infarction rates using the prespecified Universal Definition of MI, among other key issues.
On Tuesday, David Taggart, MD, PhD (University of Oxford, England), the chairman of the EXCEL surgical committee during the design and recruitment phase of the trial, issued a 14-point rebuttal to the EXCEL investigators’ response from last week, expanding on many of the arguments he’d already raised to TCTMD.
“My profound concerns remain the same and, in my opinion, the very long rebuttal response by the EXCEL investigators does not adequately respond to the core issues,” Taggart states in the rebuttal he sent TCTMD via email.
The NEJM told TCTMD last week that the concerns and criticisms raised by Taggart and the BBC do not meet any of their criteria for retraction.
Spotlight on MI Definitions
The EXCEL controversy exploded following a BBC Newsnight report suggesting that the rate of MI adjudicated using the Universal Definition was 80% higher in patients with left main coronary artery disease treated with PCI compared with CABG surgery. The rate of MI using this definition, which was leaked to the BBC, was never reported by EXCEL investigators despite it being listed in the study protocol as a secondary endpoint.
Eleven EXCEL investigators “on behalf of the EXCEL Leadership” followed up with a 3,500-word response to the BBC’s “misleading narrative,” stating they unequivocally did not conceal the MI data or downplay a signal of harm with PCI. The Universal Definition of MI preferentially uses cardiac troponin for biomarker assessment; its ascertainment was optional in EXCEL and infrequently performed, said the EXCEL investigators. “Thus, reporting procedural MI rates according to Universal Definition was not possible.”
In his email to TCTMD, Taggart stressed that the Universal Definition specifically states that if troponin assays are not available, the best alternative is CK-MB.
“If CK-MB data was available, and it must have been as it was leaked to the BBC, but never presented [by] the investigators and the protocol was committed to publishing this data, why did this not happen?” asks Taggart. In his statement, he once again raises the specter of data concealment, questioning whether the EXCEL investigators examined the CK-MB data and if they did, did they suppress the results? He also questions whether or not the MI rate adjudicated by the Universal Definition would have altered the interpretation of the primary endpoint at 3 and 5 years.
For their part, the EXCEL investigators have stated they are committed to publishing further studies reporting the rates and implications of MI according to the protocol definition, the Universal Definition using CK-MB data, and other definitions, if possible, such as the SCAI and ARC-2 definitions. In the email sent to TCTMD with the EXCEL response, co-principal investigator Gregg Stone, MD (Icahn School of Medicine at Mount Sinai, New York, NY), said that investigators will not be responding to further requests for comment.
Framing All-Cause Mortality
Taggart, in his rebuttal, says he is also not satisfied with the explanation from the EXCEL investigators as to how they framed the increased risk of all-cause mortality with PCI seen at 5 years. While the EXCEL group claims the excess mortality risk isn’t biologically plausible, noting that cardiovascular mortality was nearly identical in the PCI and CABG study group, Taggart maintains that it is and as such should have received greater scrutiny and prominence when the 5-year results were presented at TCT 2019 and published in the NEJM.
The BBC Newsnight report alleged that trial investigators downplayed concerns from the data and safety monitoring board (DSMB), something the EXCEL leadership adamantly denies. While the DSMB met frequently to review the unblinded data, they said, the board recommended the study continue as planned without changes.
This, for Taggart, is problematic. He says that he himself was never made aware of any concerns from the DSMB and that these should also have been discussed with the trial’s steering committee, of which he was a member. He points out that Abbott stopped the trial early, cutting enrollment from the planned 2,600 to 1,906 patients. “Was this decision influenced because the investigators and company were aware of mortality concerns raised by the DSMB or simply coincidental?” asks Taggart. “Did the DSMB also have access to or comment on the protocol-specified Universal Definition CK-MB data?”
Finally, Taggart notes that the EXCEL response letter is signed by 11 of the 33 EXCEL authors, including the four principal investigators—Stone, Patrick Serruys, MD (Erasmus Medical Center, Rotterdam, the Netherlands), Joseph Sabik III, MD (University Hospitals Cleveland Medical Center, OH), and A. Pieter Kappetein, MD (formerly Erasmus Medical Center, the Netherlands, and now Medtronic). What’s unclear, Taggart says, is whether the other study authors also stand by the formal EXCEL response.
Immediately following their rebuttal to the BBC, Stone said the EXCEL investigators will not engage in further debate about the trial “in the press or on Twitter.” He has since clarified his remarks to say they eagerly invite conversations about the treatment of patients with left main disease or any other topic. “Start the discussion and support your opinion with facts,” he wrote. “Let’s promote evidence-based understanding and collaboration.”
Correction: In a previous version of this story, it was mistakenly reported that Taggart called for a retraction of the EXCEL paper when in fact he only referenced the fact that others are calling for retraction, including Medscape blogger and electrophysiologist John Mandrola, MD.
Note: Stone is Co-Director of Medical Research and Education at the Cardiovascular Research Foundation (CRF), the publisher of TCTMD. Several additional EXCEL co-authors are faculty members of CRF.