Clinical background and unmet need
Drug-eluting stents (DESs) remain the default treatment for most de novo coronary lesions in percutaneous coronary intervention (PCI). They have dramatically reduced restenosis compared with bare-metal stents, but they do not eliminate long-term adverse events. Stents leave behind a permanent metallic scaffold, which can contribute to late stent thrombosis, in-stent restenosis, impaired vasomotion, and the need for repeat revascularization. These limitations have sustained interest in “leave nothing behind” strategies, particularly for lesions that may not require permanent scaffolding.
Drug-coated or drug-eluting balloons are attractive in this context because they can deliver antiproliferative therapy without leaving an implant. The challenge has been that conventional balloon-based drug delivery has often shown more variable performance in coronary arteries than in peripheral arteries, especially for de novo coronary disease. The current trial evaluated a novel sirolimus-eluting balloon (SEB) that uses biodegradable polymer microreservoir technology to release sirolimus over 90 days. The central question was whether an SEB-based strategy with provisional stenting could match systematic DES implantation for clinical outcomes at 1 year.
Study design and noninferiority framework
This was a multicenter, open-label, randomized, noninferiority trial conducted at 62 sites. Adults with de novo coronary lesions in vessels 2 to 5 mm in diameter were randomized 1:1 before PCI to either an SEB strategy with provisional DES or systematic DES implantation. In the SEB arm, bailout stenting was permitted if the angiographic result was unsatisfactory or complications occurred.
The primary endpoint was target vessel failure, a composite of cardiac death, target vessel-related myocardial infarction, and clinically driven target vessel revascularization at 1 year. The main analysis followed the intention-to-treat principle and included all randomized subjects who underwent attempted or completed revascularization. A per-protocol analysis served as a sensitivity analysis. Noninferiority was tested using an absolute margin equal to 50% of the combined event rate at a one-sided significance level of 0.025.
From a methodological standpoint, the trial design is clinically relevant because it mirrors real-world decision-making: can an operator begin with a stent-sparing approach and reserve stenting only for bailout situations, without sacrificing safety or efficacy?
Key efficacy and safety results
Between August 27, 2021, and July 29, 2024, 3323 participants were randomized and treated. Among 1661 participants assigned to the SEB strategy, bailout stenting was required in 343 patients, or 20.7%. This figure is important: it shows that in roughly one in five cases, the balloon-only approach was not sufficient and a stent had to be added to achieve an acceptable procedural result.
At 1 year, target vessel failure occurred in 88 participants in the SEB group (5.3%) and 73 participants in the systematic DES group (4.4%). The absolute risk difference was 0.91% with a 95% confidence interval from -0.55% to 2.38%. The one-sided P value for noninferiority was 0.02, and the prespecified noninferiority margin was 2.44%. On this basis, the intention-to-treat analysis met the criterion for noninferiority.
However, the individual components of the composite outcome help refine interpretation. Clinically driven target vessel revascularization occurred more often with the SEB strategy than with systematic DES implantation: 3.3% versus 2.1%, corresponding to a risk difference of 1.22% with a 95% confidence interval of 0.11% to 2.33%. This suggests that although the overall composite endpoint was noninferior, the balloon-based strategy may carry a modestly higher likelihood of repeat treatment of the target vessel within the first year.
Safety outcomes were reassuring. Events including lesion thrombosis were low and appeared similar between groups. This is an important finding because one of the principal concerns with a stent-sparing strategy is whether it might trade lower device burden for higher acute or late vessel complications. In this study, at least over 1 year, that concern was not borne out in a major way.
The per-protocol analysis, which included 3194 participants (96%), did not confirm noninferiority because the upper boundary of the 95% confidence interval reached 2.63. Nevertheless, the effect size and direction were similar to the intention-to-treat findings. This discrepancy is not unusual in noninferiority trials, where protocol deviations, crossover, and bailout procedures can meaningfully influence the estimated treatment difference. Still, it does introduce interpretive caution.
How should clinicians interpret the noninferiority result?
The most defensible interpretation is that the SEB-plus-provisional-stenting strategy performed close to systematic DES implantation for the primary 1-year composite endpoint, but it did not clearly outperform it. In practical terms, the SEB strategy may be acceptable in selected patients and lesions, especially if the goal is to minimize permanent metal implantation. Yet the higher rate of clinically driven repeat revascularization suggests that the strategy may not be fully equivalent in all settings, even if the overall composite narrowly satisfies the noninferiority criterion in intention-to-treat analysis.
Noninferiority trials deserve careful reading because statistical success does not always translate into clinical interchangeability. Here, the 1-year data support feasibility and broadly comparable short-term outcomes, but they do not establish superiority or unequivocal equivalence. The upcoming 5-year assessment will be particularly important, because the main theoretical advantage of leaving less metal behind is often expected to emerge over longer follow-up through lower late thrombosis, fewer very late restenosis events, or better vessel healing.
Expert commentary and limitations
Several strengths make this trial noteworthy. It was large, multicenter, randomized, and directly compared a novel balloon-based strategy against current standard-of-care DES implantation. The sample size and event rates provide meaningful statistical power, and the use of clinically relevant endpoints increases translational value.
There are also important limitations. First, the trial was open-label, which is unavoidable in many interventional studies but can influence downstream decisions such as revascularization. Second, the primary endpoint included clinically driven revascularization, a component that may be more susceptible to operator judgment than hard endpoints such as death or myocardial infarction. Third, 20.7% bailout stenting in the SEB arm indicates that the intended balloon-only strategy could not be maintained in a substantial minority of cases. Fourth, the per-protocol analysis did not confirm noninferiority, underscoring sensitivity to treatment adherence and procedural crossover.
Generalizability also deserves attention. The trial enrolled de novo coronary lesions in vessels 2 to 5 mm in diameter, but results may not extend equally to complex anatomies, heavy calcification, diffuse disease, bifurcations, chronic total occlusions, left main disease, or high-risk clinical subsets. As with any device trial, outcomes will likely depend on lesion preparation, operator experience, and careful patient selection.
From a mechanistic perspective, the rationale for an SEB is compelling. Sirolimus has strong antiproliferative effects, and biodegradable polymer microreservoirs may provide more durable and controlled local delivery than earlier balloon technologies. If this platform proves capable of preserving acute efficacy while reducing long-term implant-related complications, it could meaningfully shift PCI practice in selected lesions. But that promise requires longer follow-up and corroboration across lesion subsets and procedural contexts.
Clinical implications and future directions
For now, this trial suggests that an SEB with provisional stenting is a credible alternative strategy in selected de novo coronary lesions, especially when operators want to reduce the amount of implanted hardware. It should not yet be viewed as a universal replacement for DES. Rather, it may occupy a niche in anatomies where acceptable lesion preparation and vessel expansion can be achieved without a permanent scaffold.
The most clinically relevant next questions are whether longer follow-up will reveal fewer late adverse events, whether certain lesions or patient groups derive particular benefit, and whether the higher revascularization rate can be reduced through better lesion preparation, imaging guidance, or refinement of the balloon platform. Ongoing 5-year follow-up will be critical to determine whether early noninferiority is sustained and whether longer-term superiority can be demonstrated.
Conclusion
In this large randomized trial, an SEB-based PCI strategy with provisional DES was noninferior to systematic DES implantation for 1-year target vessel failure in the intention-to-treat analysis. Safety outcomes were low and similar, but clinically driven target vessel revascularization was modestly higher with the SEB strategy, and the per-protocol analysis did not confirm noninferiority. The results support the SEB approach as a promising stent-sparing option for selected de novo coronary lesions, while underscoring the need for longer follow-up and careful patient selection before broad adoption.
Funding, registration, and references
The study was registered at ClinicalTrials.gov under NCT04859985. The abstract and trial report identify the SELUTION DeNovo Investigators and were published in Circulation on June 15, 2026.
Reference: Spaulding C, Krackhardt F, Bogaerts K, Abdelaal E, Alfonso F, Briguori C, Bruch L, Cruden N, Den Hartog AW, Garot P, Godin M, Hildick-Smith D, Johnson T, Ladwiniec A, Linke A, Maart CA, Mashayekhi K, Meier P, Meunier L, Morgan K, O’Kane P, Puymirat E, Rissanen TT, Sabaté M, Schmitz T, Toth GG, Trevelyan J, Wanczura P, Marcus W, Wykrzykowska JJ, Urban P, Eccleshall S, SELUTION DeNovo Investigators. Sirolimus-Eluting Balloon With Provisional Stenting Versus Systematic Drug-Eluting Stent Implantation to Treat De Novo Coronary Lesions: A Randomized, Open-Label, Noninferiority Trial. Circulation. 2026-06-15. PMID: 42290366. URL: https://pubmed.ncbi.nlm.nih.gov/42290366/