SORT OUT IX Investigators Reverse Conclusion on Primary End Point

Neil Osterweil

December 03, 2018

Investigators in the SORT OUT IX trial have announced that, contrary to what they reported at Transcatheter Cardiovascular Therapeutics (TCT) 2018, the polymer-free BioFreedom stent (Biosensors International) was not, at the end of the day, noninferior to the bioresorbable polymer-coated sirolimus-eluting Orsiro stent (Biotronik) for the primary end point of target lesion failure at 1 year.

The reversal of fortune for the biolimus A9-coated BioFreedom device came after investigators discovered a statistical error while preparing the study for publication.

The revised calculation showed that the value for noninferiority was .123 rather than the .010 originally reported, meaning that the trial did not meet its primary end point of noninferiority for the BioFreedom device for a composite of cardiac death, myocardial infarction (MI) related to the index lesion, and target lesion revascularization.

The end point occurred in 5.2% of patients treated with BioFreedom and 4.0% of patients treated with Orsiro.

The error was brought to wide attention in a brief press release from Biotronik, maker of the comparator device.

Biosensors' chief medical officer, Hans-Peter Stoll, MD, who heard about the error in a phone call from coprincipal investigator Lisette Okkels Jensen, PhD, Odense University Hospital, Denmark, said that Jensen was "devastated when she found out that this mistake had occurred."


As originally reported by | Medscape Cardiology, SORT OUT IX was a randomized, multicenter study of 3151 patients with chronic stable coronary artery disease or acute coronary syndromes. There were no restrictions on the number of treated lesions, treated vessels, or lesion length.

Other data from that trial have not changed: cardiac deaths occurred in 1.2% of patients in the BioFreedom group and 2.1% in the Orsiro group. The rate ratio of 0.57 was not statistically significant (= .055).

As previously reported, MI rates at 1 year were similar between the groups, as were rates of definite stent thrombosis and definite/probable stent thrombosis. There were, however, significantly more target lesion revascularizations required in the polymer-free stent group.

Stoll pointed out that although SORT OUT IX was an "all-comers" trial, the biolimus-eluting stent was specifically designed for use in patients at high risk for bleeding for whom dual antiplatelet therapy (DAPT) beyond 1 month would be contraindicated.

In the LEADERS FREE II trial, also reported at TCT 2018, the BioFreedom stent was associated with better efficacy and safety outcomes at 1 year than a bare metal stent in a high-bleeding-risk population.

"We believe that for these patients, the BioFreedom is a unique solution and they benefit from it," he said.

Stoll also noted that in SORT OUT IX, patients with stable angina received 6 months of DAPT and patients with acute coronary syndromes received 12 months of DAPT, so the device "didn't have the chance to play to its strengths."

Report in Haste, Repent at Leisure

Apart from being a cause for embarrassment for investigators and for Biosensors International, the error highlights the blurring of the line between clinical science and marketing hype, and suggests that in the rush to publicize results, science may get short shrift.

"I think it points to a challenge with the current emphasis on late-breaking trials, when sometimes they're a little too late-breaking," said David J. Cohen, MD, MSc, University of Missouri–Kansas City, in an interview with | Medscape Cardiology.

Cohen was an invited discussant for a briefing at TCT 2018 during which Jensen presented the results.

"I actually asked her at the press conference whether she was sure it was noninferior because it was clear to me that it was not," he said.

Despite the data correction, Cohen said his opinions about the BioFreedom stent haven't changed, noting that it may be a niche device but iit s a suitable choice for patients with high bleeding risk for whom a longer course of DAPT could pose unwarranted risks.

Giulio Stefanini, MD, Humanitas University, Milan, an invited discussant on the original SORT OUT IX presentation, told | Medscape Cardiology that although he had not heard about the revised results, they did not change his opinion about the device either.

"I have to say that I'm partly surprised. However, we always need to think of the P value not as a threshold but rather as a gray zone," he said in an interview. "The truth is that the .01— which is now a .12 — is truly a matter of a slight change in probability. As a scientific community, we accept as significant the threshold of .05; however, if you talk to methodologists and statisticians, they can very well debate whether it's acceptable to use that threshold or not.

"What I'm saying is that if it was not noninferior as they claimed during TCT, still the claim of noninferiority was not too strong," he added.

He noted that because the stent is designed to release biolimus over only 4 weeks, its efficacy in terms of restenosis rates is somewhat compromised, compared with polymer-based drug-eluting stents, as reflected by the device narrowly missing its target lesion failure end point.

Stefanini added, however, that SORT OUT IX results have helped to allay concerns about the safety of the BioFreedom device because of the higher risk for stent thrombosis seen in LEADERS FREE.

"What I think the SORT OUT IX tells us is that in all-comer patients, actually, the stent thrombosis rates are comparable to other drug-eluting stents, so in that sense, I think it provides a favorable message for BioFreedom," he said.

SORT OUT IX was sponsored by Odense University Hospital ; no study funding source was disclosed. Jensen disclosed grant/research support from Biotronik, Biosensors, and St. Jude Medical. Stoll is an employee of Biosensors International. Cohen disclosed grant/research support from Abbott Vascular, Boston Scientific, Edwards Lifesciences, Medtronic, Corvia Medical and Svelte, Inc., and fees from Edwards and Medtronic. Stefanini disclosed grant/research support from Boston Scientific, and fees from that company, as well as from B. Braun and Biosensors.

Transcatheter Cardiovascular Therapeutics (TCT) 2018. Originally presented September 22, 2018.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.