Drug-Coated Balloons for Small Coronary Artery Disease: BASKET-SMALL 2 Raban V. Jeger, Ahmed Farah, Marc-Alexander Ohlow, Norman Mangner, Sven Möbius-Winkler, Gregor Leibundgut, Daniel Weilenmann, Jochen Wöhrle, Stefan Richter, Matthias Schreiber, Felix Mahfoud, Axel Linke, Frank-Peter Stephan, Christian Mueller, Peter Rickenbacher, Michael Coslovsky, Nicole Gilgen, Stefan Osswald, Christoph Kaiser, and Bruno Scheller, for the BASKET-SMALL 2 Investigators Dear chairmen, dear ladies and gentlemen. On behalf of my coauthors and the BASKET-SMALL 2 investigators, it is my pleasure to present to you the results of the randomized controlled BASKET-SMALL 2 Trial.

BASKET-SMALL 2 Funding Swiss National Science Foundation Basel Cardiovascular Research Foundation B. Braun Medical AG, Switzerland  No influence of the funding sources on trial design, conduct, or analysis The study was funded by the Swiss National Science Foundation, the Basel Cardiovascular Research Foundation, and the Swiss branch of the manufacturer of the drug coated balloon, B. Braun Medical AG. The funding sources did not have any influence on design, conduct, or analysis of the trial.

Background: Coronary Angioplasty BASKET-SMALL 2 Background: Coronary Angioplasty Acute Vessel Closure (Elastic Recoil, Flow-limiting Dissections) No Dual Antiplatelet Therapy Stents to Prevent Acute Vessel Closure 1977 2000 As most of you may know, the first angioplasty in a coronary vessel was performed 1977 in Zürich, Switzerland. Here you may appreciate the initial angiographic image and the long-term result obtained during a PCI in a remote vessel 23 years afterwards. These images prove that stent-free plain balloon angioplasty indeed may work. However, the results of plain balloon angioplasty were flawed by acute vessel closure due to elastic recoil and flow-limiting dissections. Important is to note, that dual antiplatelet therapy was not used in these times. Consequently, stents were introduced to prevent acute vessel closure. Meier B, N Engl J Med 2001;344:144-5

Background: Drug-Coated Balloons BASKET-SMALL 2 Background: Drug-Coated Balloons Fast and homogenous drug delivery into the vessel wall Highly lipophilic drug & coating matrix Stents are very effective in the prevention of acute vessel closure but have the disadvantages of instent-restenosis and stent thrombosis. For the treatment of instent-restenosis of both bare metal and drug eluting stents, drug coated balloons show good clinical results and currently are the standard treatment. Drug coated balloons are coated with a combination of a highly lipophilic drug and a matrix, for example paclitaxel and iopromide. After contact with the vessel wall, the drug delivery is fast and homogenous. Byrne RA et al, Nat Rev Cardiol 2014;11:13-23

BASKET-SMALL 2 Rationale 2nd-generation drug-eluting stents (DES) are the preferred treatment strategy for de-novo coronary lesions Efficacy of DES is limited in small vessels due to elevated rates of in-stent-restenoses Drug-coated balloons (DCB) are an established treatment strategy for in-stent restenoses in bare metal and drug-eluting stents The efficacy and safety of DCB in de-novo stenoses is unknown The BASKET-SMALL 2 trial was based on the following considerations: Currently, 2nd generation drug eluting stents are the preferred treatment strategy for de novo coronary lesions. However, the efficacy of DES is limited in small vessels due to elevated rates of instent-restenoses. Drug coated balloons are an established treatment strategy for instent-restenoses in both bare metal and drug eluting stents. However, the efficacy and safety of DCB in de-novo coronary stenoses is unknown.

BASKET-SMALL 2 Design Multicenter, randomized controlled non-inferiority trial (14 centers in Germany, Switzerland, and Austria) Patients undergoing PCI in native coronary arteries <3 mm Initial comparison Sequent Please® DCB (B.Braun Melsungen) vs. Taxus Element® DES (Boston Scientific), then changed to Xience® DES (Abbott Vascular) after 25% of patients Primary Endpoint: Non-inferiority for major adverse cardiac events (MACE; cardiac death, non-fatal myocardial infarction, and target vessel revascularization) @ 12 months Expected MACE rates of 7% for DCB and 10% for DES with non-inferiority margin <4% (upper limit of the two-sided 95% confidence interval of the absolute risk difference) Sample size calculation (based on Xience®): 758 patients BASKET-SMALL 2 is a multicenter, randomized controlled non-inferiority trial that was performed in 14 centers in Germany, Switzerland, and Austria. Inclusion criterion was a PCI in a native coronary artery of less than 3 mm in diameter. The trial was started with the comparison of the Sequent Please paclitaxel-coated balloon and the Taxus Element paclitaxel-eluting stent, to use the same active agent. However, after approximately 25% of patients, the Taxus Element stent was not available on the market anymore due to a litigation. Therefore, the comparator stent was changed to the Xience everolimus-eluting stent and the sample size increased to 758 patients based on expected MACE rates of 7% for Sequent Please and 10% for Xience. The primary endpoint was a composite of cardiac death, non-fatal myocardial infarction, and target vessel revascularization at 12 months. The primary aim of the analysis was non-inferiority of the DCB against a second-generation DES regarding clinical endpoints at 12 months, with non-inferiority being established if the upper limit of the two-sided 95% confidence interval of the absolute risk difference was less than 4%.

BASKET-SMALL 2 Patient Flow 883 patients with percutaneous coronary intervention assessed for eligibility/informed consent obtained Overall, 883 patients were enrolled in the trial and underwent predilation. Successful predilatation was defined as the absence of flow-limiting dissections and/or residual stenoses of more than 30% and was achieved in 758 patients, which were randomized. Overall, 382 patients were randomized into the DCB arm and 376 into the DES arm. After 12 months, more than 95% of patients could be followed up for the primary endpoint.

Clinical Baseline Characteristics BASKET-SMALL 2 Clinical Baseline Characteristics   DCB (n=382) DES (n=376) Age (mean, SD) 67.2 (10.3) 68.4 Male sex (%) 295 (77.2) 262 (69.7) Current/former smoker (%) 226 (60.4) 195 (53.1) Hypercholesterolemia (%) (68.8) 259 (70.0) Arterial hypertension (%) 324 (84.8) 332 (88.8) Family history of CAD (%) 150 (42.6) 128 (38.0) Diabetes mellitus (%) 122   (32.0) 130  (34.9)  Previous MI (%) 160 (41.9) 133 (35.4) Stroke (%)  29 (7.6)  37 (9.8) PAOD (%) 27 (7.1) 26 (6.9) COPD (%) 28 (7.3) 36 (9.6) Renal failure (%) 54 (14.1) 59 (15.7) p=0.0232 In general, clinical characteristics were well balanced in the two randomized arms. Patients were slightly over 65 years old, were mostly men, and had high rates of cardiovascular risk factors including diabetes mellitus in one third of patients. Of note, the percentage of male patients was slightly higher in the DCB group.

Angiographic Baseline Characteristics BASKET-SMALL 2 Angiographic Baseline Characteristics   DCB (n=382) DES (n=376) Target vessel Left anterior descending artery (%) 128 (33.5) 116 (30.9) Left circumflex artery (%) 179 (46.9) 183 (48.7) Right coronary artery (%) 75 (19.6) 77 (20.5) Bifurcation lesion (%) 22 (5.8) 29 (8.0) Procedural success (%; mean, SD) 96 (19) 98 (13) Number of DCB or DES (mean, SD) 1.68 (0.82) 1.26 (0.55) Length of DCB or DES (mm; mean, SD) 23.93 (11.74) 23.18 (12.85) Effective size of DCB or DES (mm; mean, SD) 2.75 (2.14) 2.57 (0.25) Max. inflation pressure (atm; mean, SD) 11.06 (3.54) 13.58 (3.90) Duration of inflation (sec; mean, SD) 48.45 (28.24) 23.36 (18.92) Regarding the baseline angiography, most interventions were done in the left circumflex artery, with a high procedural success in both groups. There were between 1 and 2 devices used per patient, with similar sizes and lenghts in both groups. Of note, inflation pressure was lower and duration of inflation longer in the DCB group.

Primary Endpoint (Non-Inferiority MACE 12 Months) BASKET-SMALL 2 Primary Endpoint (Non-Inferiority MACE 12 Months) And this is the result for the primary endpoint: With 7.33 and 7.45%, rates of the combined clinical endpoint in the DCB and the DES groups were similar after 12 months. Since the upper limit of the 95% confidence interval of the difference did not cross the predefined boundary of 4%, non-inferiority was established. Results were similar for the per protocol and the full analysis sets. PPS, per protocol set; FAS, full analysis set.

BASKET-SMALL 2 MACE (12 Months) HR 0.97, 95% CI 0.58 to 1.64; p=0.9180 This is the result for the primary endpoint, MACE. Rates for the combined clinical endpoint were low and similar over 12 months for the two randomized groups, DCB in red and DES in blue.

Myocardial Infarction BASKET-SMALL 2 Single Components of MACE (12 Months) Cardiac Death Non-fatal Myocardial Infarction Target-vessel Revascularization 3.1 vs. 1.3%; HR 2.33, 95% CI 0.82 to 6.61; p=0.1131 1.6 vs. 3.5%; HR 0.46, 95% CI 0.17 to 1.20; p=0.1123 3.4 vs. 4.5%; HR 0.75, 95% CI 0.36 to 1.55; p=0.4375 This is the result for the single components of the primary endpoint. Rates of all cardiac death, non-fatal myocardial infarction, and target vessel revascularization were low and similar over 12 months for the two randomized groups.

Major Bleeding (BARC ≥3, 12 Months) BASKET-SMALL 2 Major Bleeding (BARC ≥3, 12 Months) 1.1 vs. 2.4%; HR 0.45, 95% CI 0.14 to 1.46; p=0.1834 This is the result for major bleeding that was defined as all events classified as Bleeding Academic Research Consortium types 3, 4 and 5. Event rates were low and similar after 12 months between the two groups.

MACE (12 Months) for Device Subgroups BASKET-SMALL 2 MACE (12 Months) for Device Subgroups 15.0 vs. 6.8%; HR 2.08, 95% CI 0.61 to 7.07; p=0.2404 12.8 vs. 5.7%, HR 2.04, 95% CI 0.88 to 4.76; p=0.0987 This subgroup analysis of the primary endpoint shows the results for the different devices or combinations of devices used. In red and dark blue you can see the results for patients that were treated with DCB only and with Xience stents, while patients treated with a combination of DCB and DES are shown in yellow and patients treated with Taxus stents are shown in light blue. Rates for the combination of DCB and DES were numerically higher than for DCB only, without statistical significance. The same applies for the comparison of Taxus and Xience stents.

BASKET-SMALL 2 Conclusions First large randomized controlled trial testing the efficacy and safety of a paclitaxel-iopromide-coated DCB vs. second-generation DES in a large all-comer population regarding clinical endpoints DCB are non-inferior to DES in lesions of small native coronary arteries regarding MACE up to 12 months, with similar event rates for both treatment groups  Small native coronary artery disease may safely be treated with DCB after successful predilatation No permanent material is implanted in the coronary artery potentially translating in a reduced number of very late adverse events  Long-term follow-up is needed Ladies and gentlemen, let me conclude that BASKET-SMALL 2 is the first large randomized controlled trial testing the efficacy and safety of a paclitaxel-iopromide-coated DCB against second-generation DES in a large all-comer population regarding clinical endpoints. The study shows that DCB are non-inferior to DES in lesions in small native coronary arteries regarding MACE up to 12 months, with similar and low event rates for both treatment groups. Therefore, small native coronary artery disease may safely be treated with DCB after successful predilatation. Based on the fact that no permanent material is implanted in the coronary artery, one might expect a lower number of late adverse events in the DCB group. However, long term follow-up is necessary to confirm this hypothesis.

              BASKET-SMALL 2 Thank you! I want to thank all centers and investigators that contributed to this trial.

The study has been published today in the Lancet and can be found online. Thank you for your attention.