Author + information
- David E. Kandzari, MD∗ ()
- ↵∗Address for correspondence:
Dr. David E. Kandzari, Piedmont Heart Institute, Suite 2065, 95 Collier Road, Atlanta, Georgia 30309.
Over more than a decade there have been continued design changes for drug-eluting stents (DES), including the development of bioresorbable materials, alloy modifications, and novel stent designs. Indeed, the cadence of development has promoted within interventional cardiology an expectation that something new is always around the corner, and with the sometimes-misleading expectation that newer is always better.
In parallel, clinical study has been challenged to keep pace, and comparative DES studies have evolved considerably, shifting focus from angiographic to patient-oriented clinical events, including broader lesion complexities and clinical indications, adopting newer DES standards of comparison and imposing more restrictive statistical criteria. In most instances, however, comparative study has demonstrated statistical noninferiority with the assumption of clinical parity, but not superiority. Contemporary lessons with bioresorbable scaffolding suggest that despite the intuitive appeal of a novel technology, the design of trials for a new therapy to demonstrate only a margin of similarity to a standard can be a shortfall in the pursuit of progress.
In part a consequence of noninferiority trials, a dilemma presented by current DES studies is whether a class effect exists with contemporary stent technologies, or instead the opportunity to differentiate DES is a limitation of trial design. Serial noninferiority comparisons may permit “statistical creep” in which a newer but slightly less efficacious treatment advances to a new standard of comparison (1); yet after multiple comparative trials, newer therapies may be statistically similar to existing ones but ultimately no better than placebo. An issue more relevant is that because adverse outcomes such as stent thrombosis and myocardial infarction (MI) are relatively rare following percutaneous coronary intervention, designing adequately powered studies may be quite challenging. Furthermore, as current DES report the most favorable efficacy and safety outcomes to date, endpoint event rates have migrated even lower over the past decade. Large comparative studies involving newer generation DES demonstrate superior outcomes to conventional bare-metal stents and first-generation DES (2). Against the background of such studies, DES have become perceived by clinicians as the same and viewed by hospitals as an exchangeable commodity, often with pricing rather than data determining inventory.
Despite these limitations, opportunities do exist to demonstrate differences in outcome between DES with the understanding that such differences may not be so readily apparent in a traditional primary endpoint. Expectations for when, where, and how much differences are observed must also be reset. As overall event rates lower, disparities may not occur to the magnitude observed in early studies, and DES may be distinguished by meaningful but less anticipated endpoints such as MI rather than late lumen loss or clinical restenosis (3). The persistence of adverse events with both first-generation and contemporary DES (4,5) presents another example. Specifically, dedicated longitudinal follow-up in clinical trials offers greater insight to the effectiveness of DES, and the accrual of events may amplify the ability to distinguish outcomes between therapies. As an example, it was long-term surveillance of first-generation DES that introduced the risk associated with late stent thrombosis, and detailed patient-level observation for years beyond a trial’s primary endpoint is now a regulatory mandate. Detailed ascertainment of events permits insight to not only annualized estimates of stent thrombosis, but also the persistence of late target lesion revascularization that seems constant with existing DES.
The DUTCH PEERS TWENTE II (DUrable polymer-based sTent CHallenge of Promus ElemEnt versus ReSolute integrity) randomized trial represents best practices in comparative DES study to determine whether outcomes do in fact differ between contemporary everolimus- and zotarolimus-eluting stents (6). Although designed as a noninferiority study, the trial enlisted a broad, unselected patient population representative of routine clinical practice with limited clinical or anatomic exclusion criteria. Despite best intent, trials randomizing “all-comers” without restriction generally enroll a lower-risk population than those patients not randomized but instead followed in a registry (7). In a similar prior study, however, these investigators demonstrated greater clinical risk and lesion complexity among patients included in a parallel registry, but justified the generalizability of their results given that 1-year outcomes were similar to those patients within the randomized population (8). At 1 year in the DUTCH PEERS study, the composite outcome of cardiovascular death, target vessel–related MI, and repeat target vessel revascularization was similar among patients undergoing percutaneous coronary revascularization with zotarolimus-eluting, cobalt alloy–based stents and everolimus-eluting, platinum-chromium stents (6). Now at 5 years, and with remarkably high completeness of follow-up, similar rates of the trial primary endpoint in addition to comparable measures of particular late-term interest, including stent thrombosis and repeat revascularization are reported by Zocca et al. (9) in this issue of JACC: Cardiovascular Interventions.
Altogether, the results affirm observations from other current trials supporting the achievement of best outcomes with percutaneous coronary intervention but also clinical parity among contemporary DES. Second, the report reflects a broad representation of patients with considerable variance in clinical and lesion complexity, many of whom would be systematically excluded from participation in more traditional clinical trials. To this purpose, dedicated follow-up from the study informs expectations in clinical practice, demonstrating for both stent types 5-year rates of repeat target vessel revascularization and definite or probable stent thrombosis of ∼8% and ∼1.5%, respectively.
Consistent with the rapidity of device evolution, the DES studied in the DUTCH PEERS trial have already been succeeded by newer formulations in most geographies, but their relevance to current practice remains. Beyond consideration of stent-related outcomes alone, one advantage of such comprehensive follow-up is insight to the contemporary natural history of patients with coronary artery disease. The marked increase in prescription of oral anticoagulation (16% at 5 years), presumably related to development of atrial fibrillation as the study population ages, is one such example. And the trial also exposes the opportunities for improvement in cardiovascular medicine. Beyond the primary endpoint, adverse cardiovascular events both related and unrelated to the stent territory ensue. Five years following the index revascularization, more than one-half of all events occur between 1 and 5 years, and from a patient perspective, 1 in 5 patients will experience an adverse event. Beyond the first year, repeat revascularization related to the target vessel persists at an annualized rate of approximately 1% per year. What remains unknown from the present study is the rate of recurrent events among patients as well as outcomes in subgroups of especial interest, for example, those with diabetes or presenting with MI.
For patients enrolled in clinical trials, landmark dates include birthdays, anniversaries, and holidays—but typically not primary endpoints. The date may seem quite ordinary and most often unrecognized. But for a clinical trialist, the highly anticipated milestone represents the unveiling of data and discovery. Still, for both patients and investigators, life goes on the day after a primary endpoint, but the responsibility to survey the outcomes of patients is ever more important. It represents not just the beginning of an opportunity to understand the effectiveness of a therapy and its long-term consequences, but from a close distance, to more broadly follow the fate of our patients.
↵∗ Editorials published in JACC: Cardiovascular Interventions reflect the views of the authors and do not necessarily represent the views of JACC: Cardiovascular Interventions or the American College of Cardiology.
Dr. Kandzari has received institutional research grant support from Abbott, St. Jude, Biotronik, Boston Scientific, Medinol, Medtronic, and Orbus Neich; and personal consulting honoraria from Biotronik, Boston Scientific, Medtronic, and Micell Technologies.
- 2018 American College of Cardiology Foundation
- Kandzari D.E.,
- Mauri L.,
- Koolen J.J.,
- et al.,
- BIOFLOW V Investigators
- Kimura T.,
- Morimoto T.,
- Nakagawa Y.,
- et al.
- Claessen B.E.,
- Beijk M.A.,
- Legrand V.,
- et al.
- von Birgelen C.,
- Sen H.,
- Lam M.K.,
- et al.
- Zocca P.,
- Kok M.M.,
- Tandjung K.,
- et al.