Author + information
- Received April 4, 2018
- Revision received July 18, 2018
- Accepted August 14, 2018
- Published online December 3, 2018.
- Marianne Brodmann, MDa,∗ (, )
- Martin Werner, MDb,
- Dirk-Roelfs Meyer, MDc,
- Peter Reimer, MDd,
- Karsten Krüger, MDe,f,
- Juan F. Granada, MDg,
- Michael R. Jaff, DOh,
- Henrik Schroeder, MDi,
- for the ILLUMENATE EU RCT Investigators
- aDivision of Angiology, Department of Internal Medicine, Medical University Graz, Graz, Austria
- bDepartment of Angiology, Hanusch Hospital, Vienna, Austria
- cDepartment of Diagnostic and Interventional Radiology, Hubertus-Hospital, Berlin, Germany
- dInstitute for Diagnostic and Interventional Radiology, Academic Teaching Hospital of the University of Freiburg, Karlsruhe, Germany
- eDepartment of Radiology and Interventional Therapy, Vivantes Humboldt Hospital, Berlin, Germany
- fDepartment of Radiology and Interventional Therapy, Vivantes Hospital Spandau, Berlin, Germany
- gCardiovascular Research Foundation, Columbia University Medical Center, New York, New York
- hDepartment of Medicine, Newton-Wellesley Hospital, Newton, Massachusetts
- iCenter for Diagnostic Radiology & Minimally Invasive Therapy, The Jewish Hospital, Berlin, Germany
- ↵∗Address for correspondence:
Prof. Dr. Marianne Brodmann, Division of Angiology, Department of Internal Medicine, Medical University Graz, Auenbruggerplatz 27, A-8036 Graz, Austria.
Objectives The aim of this study was to assess the safety and effectiveness of a next-generation low-dose drug-coated balloon (DCB) designed to optimize the amount of drug transferred into the vessel wall and to maximize the amount of time the drug resides in the vessel wall.
Background Several randomized controlled studies evaluating various DCBs have demonstrated a significantly higher patency rate compared with noncoated percutaneous transluminal angioplasty balloons at 1 year. However, the data are limited and vary by DCB at longer follow-up time points. An earlier generation low-dose DCB failed to demonstrate significant treatment effect at 2 years, raising questions regarding the durability of low-dose DCBs.
Methods In this prospective, multicenter trial, 294 patients were randomized (3:1) to treatment with a DCB or an uncoated percutaneous transluminal angioplasty balloon. Assessments at 2 years included primary patency with duplex ultrasonography, clinically driven target lesion revascularization, and functional outcomes.
Results Primary patency at 2 years was significantly higher in the DCB cohort (75.9% vs. 61.0%; p = 0.025), and the rate of clinically driven target lesion revascularization was significantly lower (12.1% vs. 30.5%; p < 0.001). There were no major limb amputations in either group. The rates of all-cause (6.5% vs. 5.1%; p = 1.00) and cardiovascular-related (1.6% vs. 1.7%; p = 1.00) mortality were similar between groups. Functional improvements over baseline were sustained in both groups, with 60% fewer reinterventions in the DCB group.
Conclusions A sustained treatment effect is achievable with a low-dose DCB with an optimized coating formulation. This trial demonstrated for the first time a statistically significantly higher primary patency rate for a low-dose DCB versus PTA at 2 years. (CVI Drug Coated Balloon European Randomized Clinical Trial; NCT01858363)
- drug-coated balloon
- drug-eluting balloon
- peripheral artery disease
- superficial femoral artery
Devices that deliver antiproliferative drugs to the vessel wall have proved their efficacy for the treatment of peripheral artery disease in the femoropopliteal arteries. Drug-eluting stents have proved to be an effective treatment, as demonstrated by the long-term outcomes of the Zilver PTX randomized clinical studies (1). These results show that drug-eluting stents support sustained safety and effectiveness, including the long-term superiority of drug-eluting stents to uncoated percutaneous transluminal angioplasty (PTA) balloons (2,3). Drug-coated balloons (DCBs) are designed to deliver an antiproliferative drug without the burden of a permanent implant. Stent deployment may limit the options for subsequent treatments such as surgical bypass. Stent fractures pose management challenges, and the burden of treating in-stent restenosis is significant (4).
A limited amount of data are available on the mid- and long-term treatment effect, or treatment durability, of DCBs. The IN.PACT SFA randomized trial of the IN.PACT Admiral DCB, coated with 3.5 μg/mm2 of paclitaxel, reported a sustained treatment effect with a statistically higher patency rate in the DCB arm compared with the PTA arm through 3 years (5).
DCBs containing a lower drug dose density have been designed in an attempt to reduce the likelihood of potential local and distal drug toxicity. However, clinical outcomes in some studies involving low-dose DCBs thus far have been modest. The LEVANT 2 study, comparing a 2 μg/mm2 DCB (Lutonix) with standard PTA, produced a 24-month primary patency outcome of 58.6% (vs. 53.0% in the PTA arm, p = 0.05) (6). These results call into question the sustainability of the biological effect of DCBs characterized by lower doses than the clinically proven 3.5 μg/mm2. To date, no low-dose DCB has shown a statistically significant treatment effect over an uncoated PTA balloon beyond 1 year.
Promising outcomes with a different low-dose DCB, Stellarex (Philips Spectranetics, Colorado Springs, Colorado), were observed in the single-arm ILLUMENATE first-in-human study. The results showed core laboratory–adjudicated primary patency rates of 89.5% and 80.3% at 1 and 2 years, respectively (7). Although this first-in-human study suggests that durability of this particular low-dose DCB is possible, the data require validation in a randomized controlled trial. The ILLUMENATE EU RCT (European Randomized Clinical Trial) is designed to assess the safety and effectiveness of the Stellarex DCB. The 1-year outcomes, as previously published (7), demonstrate superior 12-month patency rates of Stellarex over PTA (89.0% vs. 65.0%; log-rank p < 0.001). Presented here are the 2-year results.
Trial methodology in detail and results through 1 year have been previously reported (8). Briefly, the ILLUMENATE EU RCT is a prospective, randomized, multicenter, single-blind study. The study is composed of patients with claudication and rest pain due to ≥70% stenosis of the superficial femoral and/or popliteal artery (P1 segment). The study was approved by the appropriate medical ethics committee at each site and was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained before study procedures were performed.
Independent core laboratories analyzed all images, including duplex ultrasonography (VasCore, Massachusetts General Hospital, Boston, Massachusetts) and angiography (SynvaCor, Springfield, Illinois). Core laboratory staff members remained blinded to treatment assignment. A blinded independent clinical events committee adjudicated all adverse events. An independent data safety and monitoring board monitored the study for safety. Data were monitored for accuracy, with 100% source document verification.
Patients were screened and evaluated for the inclusion and exclusion criteria, which included a clinical assessment, medical and medication history, angiography, and assessment for study eligibility. Eligible patients received dual antiplatelet therapy before the procedure per hospital standard of care (antiplatelet medication 3 days pre-procedure or a loading dose on the day of the intervention).
Following the procedure, patients were maintained on a 75-mg clopidogrel bisulfate daily regimen (or ticlopidine hydrochloride if unable to tolerate clopidogrel) for a minimum of 1-month post-procedure. Patients received 100 mg of acetylsalicylic acid daily throughout the 5-year length of the study.
Clinical evaluations at 2 years post-procedure included duplex ultrasonography of the target lesion(s), ankle-brachial index, Rutherford-Becker classification, walking impairment questionnaire, walking distance (either by treadmill or 6-min walk test), and adverse events. Patient follow-up is ongoing for 5 years.
Patency was defined as the absence of target lesion restenosis determined by duplex ultrasound peak systolic velocity ratio of ≤2.5 and freedom from clinically driven target lesion revascularization (CD-TLR) at 12 months. Any retreatment by an invasive procedure, including atherectomy, angioplasty, stenting, endarterectomy, bypass, or thrombolysis, performed to open or increase the luminal diameter of the target lesion was considered a target lesion revascularization.
Statistical outcomes were assessed using the intent-to-treat principle. Continuous data are summarized using descriptive statistics: number, mean, and SD. Categorical variables are summarized using frequency counts and percentages. For ordinal-scaled variables, a combination of these presentations was used as appropriate: frequency and percentage of observations within a category and means and SDs of the scores of the categories. For categorical and ordinal variables, percentages were calculated on the basis of nonmissing data.
Binary endpoints are presented as counts and percentages, with exact 95% confidence intervals calculated using the Clopper-Pearson method. Superiority was tested using the Fisher exact or chi-square contingency table method. Kaplan-Meier survival analyses and associated log-rank tests are also presented for the outcomes of patency, safety, and CD-TLR. Subgroup analyses present the frequencies and percentages for patency by selected subgroups with a treatment group interaction test from a logistic regression model.
The ILLUMENATE EU RCT included 294 patients randomized to treatment with the Stellarex DCB (222 patients, 254 lesions) or PTA (72 patients, 79 lesions) at 18 centers in Germany and Austria between December 2012 and April 2015. Baseline characteristics and results through 12 months were previously reported (8). Baseline patient and procedural characteristics were similar between groups, as shown in Tables 1 and 2⇓⇓. There were no significant differences between groups except in reference vessel diameters, which were slightly larger on average in the DCB cohort (5.02 ± 0.79 mm vs. 4.77 ± 0.69 mm; p = 0.012). A logistic regression analysis indicated reference vessel diameter to be significantly associated with outcomes of both primary effectiveness and safety endpoints, but no statistical evidence of interaction effects was found, and the impact on the treatment group comparisons was minimal (9). Mean lesion lengths were 7.2 and 7.1 cm in the DCB and PTA groups, respectively. Total occlusions represented 19% of lesions in each group.
Follow-up compliance was 98%, with visits completed at 2 years for 181 of 185 eligible DCB patients and 93% of eligible PTA patients (54 of 58). Prior to the 2-year follow-up visit, 13 patients (5.9%) died, 15 (6.8%) withdrew from the study, and 9 (4.1%) were lost to follow-up in the DCB arm, and 3 (4.2%) died, 10 (13.9%) withdrew, and 1 (1.4%) was lost to follow-up in the PTA arm. At 24 months, 91.2% of patients in the DCB cohort (165 of 181) and 88.9% in the PTA group (48 of 54) were on antiplatelet medications.
Effectiveness outcomes through 2 years
Primary patency through the 2-year follow-up window (790 days) was significantly higher for the DCB group, at 75.9% (145 of 191 lesions) in comparison with 61.0% (36 of 59 lesions) for PTA (difference 14.9%; 95% confidence interval: 1.1% to 28.7%; p = 0.025). Similarly, the primary patency Kaplan-Meier estimates at 730 days showed sustained treatment effectiveness at 75.2% for the DCB group versus 61.2% for the PTA group (log-rank p = 0.004 through 2 years) (Figure 1). The median follow-up time for primary patency was 1,035 days (interquartile range: 711.5 to 1,129.5 days) for patients randomized to the DCB arm and 1,040 days (interquartile range: 694.5 to 1,137.0 days) for those in the PTA arm. For lesions in which restenosis developed, the average time to loss of patency was 440.5 ± 216.7 days in the DCB arm (n = 46) and 311.4 ± 138.8 days in the PTA arm (n = 23).
Stents were placed in 15% of lesions in the DCB cohort and 11% of lesions in the PTA cohort. When these lesions were excluded, the 2-year patency rate for the nonstented lesions was 77.1% (121 of 157) in the DCB cohort and 60.0% (30 of 50) in the PTA cohort (p = 0.0180). The CD-TLR rates for the nonstented lesions were 9% (14 of 156) and 31.4% (16 of 51), respectively.
Subgroup analyses of primary patency through 2 years (Table 3) indicated similarly high patency for patients with diabetes (75.4%) as for those without diabetes (76.2%) following treatment with the Stellarex DCB. Additionally, no evidence of a difference in treatment effect (DCB vs. PTA) was seen between men and women (treatment-by-sex interaction p = 0.585) or diabetic status (treatment-by-diabetes interaction p = 0.300).
Safety outcomes through 2 years
Superiority of the Stellarex DCB over PTA was maintained through 2 years. The safety endpoint, defined as the composite of freedom from device- and procedure-related death through 30 days and freedom from target limb amputation and CD-TLR through 2 years (790 days), was 87.9% (167 of 190) for DCB and 69.5% (41 of 59) for PTA, for a difference of 18.4% (95% confidence interval: 4.9% to 32.2%; p = 0.001). As shown in Figure 2, the Kaplan-Meier estimates for freedom from a primary safety event through 2 years were superior in the DCB cohort (88.9% for DCB vs. 71.7% for PTA, log-rank p < 0.001).
Major adverse events through 2 years were significantly lower in the DCB group, because of differences in CD-TLR (Table 4). The rate of CD-TLR through the end of the 2-year visit window (790 days) was 12.1% for DCB versus 30.5% for PTA (p < 0.001). The treatment effect was not only maintained at 2 years but increased in magnitude compared with 1 year (difference in CD-TLR between DCB and PTA of 18.4% at 2 years vs. 10.8% at 1 year). The treatment differences in CD-TLR at 2 years were similar between patients with diabetes (18.5% difference [11.9% DCB vs. 30.4% PTA]) and those without diabetes (18.4% [12.2% DCB vs. 30.6%]), as well as between men (18.0% difference [11.3% DCB vs. 29.3% PTA]) and women (19.3% difference [14.0% DCB vs. 33.3% PTA]). As shown in Figure 3, the freedom from CD-TLR was significantly higher in the DCB group (88.9% vs. 71.8%; log-rank p < 0.001).
All-cause mortality rates were no different between groups, at 6.5% (13 of 199) for DCB and 5.1% (3 of 59) for PTA (p = 1.00). Per clinical events committee adjudication, no deaths were procedure or device related.
Clinical and functional outcomes through 2 years
Clinical and functional improvements over baseline were sustained through 2 years and remained similar between treatment groups, but significantly more CD-TLRs were required in the PTA group (30.5% in the PTA arm vs. 12.1% in the DCB arm, p < 0.001) to maintain outcomes. Rutherford-Becker classification was improved from baseline for 87.8% of DCB patients and 84.9% of PTA patients at 2 years. Distance walked, as measured by the 6-min walk test (n = 77) or treadmill test (n = 123), improved for 74.2% of DCB patients and 66.7% of PTA patients at 2 years. Improvements in the walking impairment questionnaire composite score at 2 years were seen in 82.1% of DCB patients and 94.7% of PTA patients, and improvements in ankle-brachial index at 2 years were achieved for 78.7% of DCB patients and 83.3% of PTA patients.
The ILLUMENATE EU RCT demonstrated a durable treatment effect of the Stellarex DCB, with no indication of deterioration in superior patency and clinical benefit through 2 years. These outcomes validate the first-in-human results (7) and for the first time demonstrate a low-dose DCB with sustained outcomes on the basis of objective, blinded, core laboratory assessment of patency.
DCB performance is driven by the total dose of drug effectively transferred and retained in the tissue. Coating configuration is critical for DCB performance, and it varies on a spectrum from pure crystalline to pure amorphous. These components continue to be modified as DCBs evolve.
Highly crystalline coatings deposit particles with low solubility on the vessel wall, resulting in higher and longer tissue residency time. With highly crystalline coatings, the potential for tissue toxicity and particle embolization rates are both increased. First-generation DCBs display a high crystalline coating structure, resulting in high paclitaxel tissue levels over time (10). Particle solubility plays a key role in maintaining paclitaxel tissue levels and the antirestenotic effect over time (10). Recent clinical studies suggest that with the original nominal dose (3.5 μg/mm2), vessel patency was improved and maintained compared with plain PTA up to 3 years (5).
Lowering paclitaxel dose is a beneficial technological advancement. It not only decreases the early overshoot in the transfer dose but also reduces any theoretical concerns of particulate embolization, especially in critical territories such as distal vessels with poor runoff. The holy grail in DCB technology development has been in obtaining a comparable antirestenotic effect as seen with first-generation DCB technologies but at a lower dose. This approach would still maintain a high treatment effect with limited drug embolization, which is meant to maintain a high safety profile in the rare but real higher risk clinical and anatomic circumstances, such as single vessel runoff in the presence of ischemic wounds.
A subsequently developed DCB aimed to improve the safety profile by decreasing the total paclitaxel and modifying coating solubility toward a more amorphous (soluble) coating formulation (11). Experimental studies suggested that although particulate loss was decreased (11,12), long-term paclitaxel retention and its overall biological effect against restenosis were reduced (13). Clinical data in the superficial femoral artery territory appear to correlate with the experimental findings (6,14).
The Stellarex DCB was designed to optimize drug uptake into the arterial tissue and maximize retention over time to facilitate a durable treatment effect. The paclitaxel drug itself is a combination of both crystalline and amorphous states. This hybrid formulation was created to balance the need for coating stability during device transit and prompt bioavailability upon balloon delivery, attributes of the amorphous formulation, while providing sustained drug tissue release over time, by the crystalline paclitaxel. The excipient is polyethylene glycol. It is a polymer with a relatively high molecular weight of 8,000 g/mol, which makes the coating highly stable and adaptable to balloon deformation and has a slower dissolution rate; both characteristics minimize coating loss during tracking.
Although cross-trial comparisons should be avoided, in the absence of head-to-head trials, decision making on clinical strategy and technology adoption is still frequently based on clinical evidence. To avoid misleading conclusions, careful attention should therefore be taken in checking consistency not only in patient population across trials (clinical and anatomic characteristics), but also in study methodology and endpoint definitions and reporting methods. Patency is an objective binary endpoint often assessed by a blinded independent core laboratory. These factors make it a good candidate for comparison purposes, but differences in follow-up visit windows and reporting practices continue to make even this objective endpoint somewhat challenging to compare from one trial to the next (15). With regard to longer term 2-year follow-up data, we can refer to 2 published and peer-reviewed DCB trials, the IN.PACT SFA (16) and LEVANT 2 (14,17) trials. A summary of patency data (Kaplan-Meier estimates and observed proportional rates) reported in these trials is in Table 5.
A Kaplan-Meier analysis offers the advantage of showing event distribution over time and adjusting the denominator for missing data. However, in trials with a follow-up-driven endpoint, such as primary patency, and for an interim analysis (e.g., assessing 2-year rates when most patients have yet reached 3-year follow-up), this method can generate important random variances in the final estimates due to the concentration of patient censoring and depending on when the last event occurs relative to that. For this reason, a common compromise is made in accepting Kaplan-Meier estimates truncated at the nominal time point (e.g., at day 730 for the 2-year follow-up window ending at day 790), when a significant number of patients and events are normally lost from neglecting the second half of the follow-up window. There are also differences in Kaplan-Meier methods used across trials. Vardi et al. (15) demonstrated a range of more than 10% when using different methods to calculate Kaplan-Meier estimates, using the same primary patency rate. Another way to present patency data is by a proportion of lesions analyzed: the number of patent lesions divided by the number of evaluated lesions. This has the advantage of including data generated through the full follow-up window (typically), and these methods remain less subject to random variances compared with Kaplan-Meier estimates.
Although patency is an important and objective metric and a more direct indicator of DCB efficacy, assessments of functional status and quality of life are of paramount importance to the patients being treated. Within this trial, functional assessments such as distance walked and walking impairment questionnaire score improvements were sustained through 2 years and maintained at similar levels in the DCB cohort, with significantly fewer reinterventions compared with the control arm (12.1% vs. 30.5%, p < 0.001). This latter variation has a potential impact on the cost-effectiveness of Stellarex compared with PTA.
As previously reported, physicians were not blinded to treatment due to visible coating on the DCB catheter. The CEC, DSMB, and core lab personnel were blinded to treatment. Functional outcome assessments conducted at 12 and 24 months, regardless of whether or not there was a reintervention. The revascularization rates necessary to maintain functional outcomes need to be considered when interpreting these data; 60% fewer reinterventions were required in the DCB group to maintain the observed improvements through 2 years.
For the first time, the ILLUMENATE EU RCT demonstrated the durability of a low-dose DCB in humans. Previously, the sustainability of the biological effect was not supported beyond 12 months on the basis of randomized data. The implications of these findings are important as they may support the use of this technology in vascular territories never explored before and could help expand the therapeutic options for current generation DCB technologies.
WHAT IS KNOWN? DCBs are an effective endovascular option for treatment of lesions in the femoropopliteal artery. Differences in coating characteristics affect the amount of drug that is absorbed in the arterial wall and the amount of time it remains there.
WHAT IS NEW? The ILLUMENATE EU RCT is the first study to demonstrate a significantly higher primary patency rate through 2 years with a low-dose DCB compared with an uncoated balloon.
WHAT IS NEXT? Additional studies of the use of DCBs for more complex lesions, such as in-stent-restenosis and lesions that cannot be adequately pre-dilated, are needed. Additional research is necessary on the use of DCBs in conjunction with atherectomy devices.
The authors thank Teresa Yurik, MS, for statistical support and Meghan Schadow, MS, for assistance with medical writing. The authors thank Marie Allouis, PhD, for trial management and all of the investigators and site coordinators for their dedication to this project.
This study was sponsored by Philips Spectranetics. Additional funding for data analysis and medical writing assistance with manuscript preparation was provided by Philips Spectranetics. Dr. Brodmann is a consultant for Spectranetics, BARD, Medtronic, Intact Vascular, Surmodics, Bayer Healthcare, and Shockwave. Dr. Granada has received research grants from Spectranetics Philips, Medtronic, and Boston Scientific. Dr. Jaff is an uncompensated advisor to Abbott Vascular/Boston Scientific/Cordis, a Cardinal Health Company/Medtronic; is a consultant to Philips/Volcano, Micell, Vactronix, and Venarum; and is an equity investor with Primacea, PQ Bypass, Embolitech, Vascular Therapies, Gemini, Sano V, and EFemoral. Dr. Schroeder has received speaking honoraria and consulting fees from Spectranetics and Philips. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- clinically driven target lesion revascularization
- drug-coated balloon
- percutaneous transluminal angioplasty
- Received April 4, 2018.
- Revision received July 18, 2018.
- Accepted August 14, 2018.
- 2018 American College of Cardiology Foundation
- Dake M.D.,
- Ansel G.M.,
- Jaff M.R.,
- et al.
- Dake M.D.,
- Ansel G.M.,
- Jaff M.R.,
- et al.
- ↵Dake M. The Zilver PTX randomized trial of paclitaxel-eluting stents for femoropopliteal artery disease: 5-year results. Presented at: VIVA 2014; Las Vegas, Nevada.
- Tosaka A.,
- Soga Y.,
- Iida O.,
- et al.
- Schneider P.A.,
- Laird J.R.,
- Tepe G.,
- et al.
- ↵Laurich C. Oral presentation. Presented at: SVS; June 18, 2015; Chicago, Illinois.
- Schroeder H.,
- Werner M.,
- Meyer D.R.,
- et al.
- Krishnan P.,
- Faries P.,
- Niazi K.,
- et al.
- Granada J.F.,
- Stenoien M.,
- Buszman P.P.,
- et al.
- Schorn I.,
- Malinoff H.,
- Anderson S.,
- et al.
- Kolodgie F.D.,
- Pacheco E.,
- Yahagi K.,
- Mori H.,
- Ladich E.,
- Virmani R.
- Gongora C.A.,
- Shibuya M.,
- Wessler J.D.,
- et al.
- Vardi M.,
- Lei L.,
- Shangguan S.,
- Doros G.
- Laird J.R.,
- Schneider P.A.,
- Tepe G.,
- et al.
- ↵Lutonix 035 Drug Coated Balloon PTA Catheter Instructions for Use. Available at: http://www.bardpv.com/eifu/uploads/DW5190-01r7-Lutonix-DCB-IFU-with-GeoAlign.pdf. Accessed November 18, 2018.
- IN.PACT Admiral Instructions for Use. Available at: http://manuals.medtronic.com/content/dam/emanuals/cardio/M052624T001_Rev1H_view_color.pdf. Accessed November 18, 2018.