Author + information
- Received May 4, 2016
- Revision received June 6, 2016
- Accepted June 20, 2016
- Published online September 26, 2016.
- Xin Jia, MDa,
- Jiwei Zhang, MDb,
- Baixi Zhuang, MDc,
- Weiguo Fu, MDd,
- Danming Wu, MDe,
- Feng Wang, MDf,
- Yu Zhao, MDg,
- Pingfan Guo, MDh,
- Wei Bi, MDi,
- Shenming Wang, MDj and
- Wei Guo, MDa,∗ ()
- aChinese PLA General Hospital, Beijing, China
- bRenji Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China
- cXiyuan Hospital, China Academy of Chinese Medical Sciences, Beijing, China
- dZhongshan Hospital Fudan University, Shanghai, China
- eThe People’s Hospital of Liaoning Province, Shenyang, China
- fThe First Affiliated Hospital of Dalian Medical University, Dalian, China
- gThe First Affiliated Hospital of Chongqing Medical University, Chongqing, China
- hThe First Affiliated Hospital of Fujian Medical University, Fuzhou, China
- iThe Second Hospital of Hebei Medical University, Shijiazhuang, Hebei Province, China
- jThe First Affiliated Hospital, Sun Yat-sen University, Guangzhou, China
- ↵∗Reprint requests and correspondence:
Dr. Wei Guo, Vascular Surgery Department, Fuxing Road 28, Beijing 100853, China.
Objectives The aim of this study was to investigate the efficacy and safety of a new paclitaxel-coated balloon catheter in the treatment of stenotic or occluded femoropopliteal arteries.
Background The incidence of restenosis can be reduced by the use of drug-coated balloons. However, dose, coating composition, and technology are decisive for efficacy.
Methods Two hundred Chinese patients with peripheral arterial occlusive disease were prospectively randomized to treatment with new paclitaxel-coated or standard uncoated balloon catheters. The primary endpoint was angiographic late lumen loss at 6 months, measured by a blinded core laboratory. Secondary angiographic endpoints (6 months) and specific clinical endpoints (1 year) were binary restenosis, ankle-brachial index, Rutherford stage, clinically driven target lesion revascularization, and amputation.
Results Patients’ mean age was 66 years, 74% were men, 31% were smokers, and 55% had diabetes. Patients were in Rutherford stages 2 through 5, with a mean lesion length of 150 mm; 25% had in-stent restenosis, 55% had occlusion or partial occlusion, and 20% underwent provisional stenting. Late lumen loss at 6 months was available for 89%, and clinical follow-up was available for >95% per group. Mean late lumen loss was 0.05 ± 0.73 mm with coated balloons and 1.15 ± 0.89 mm with uncoated balloons (p < 0.001). Correspondingly, the rates of restenosis were 22.5% and 70.8% (p < 0.001). After 1 year, the rates of target lesion revascularization were 7.2% and 39.6% (p < 0.001), and Rutherford class and ankle-brachial index improved more markedly in the coated group (p < 0.046 and p = 0.023, respectively). One major amputation was recorded in the control group. No coating-related adverse events were observed for doses of up to 43 mg paclitaxel per patient.
Conclusions In this medium-sized trial with long superficial femoral artery lesions, the use of paclitaxel-coated balloon catheters markedly improved angiographic and clinical outcomes of interventions despite advanced disease in the majority of patients.
The treatment of stenotic and occluded arteries by minimally invasive endovascular methods provides fast relief from symptoms with low risk to many patients and contributes to cost reduction. Several methods are available (1), and the choice depends on the patient, the vessel territory, the kind of disease, and physicians’ experience and preferences. All interventional endovascular methods cause vessel injury, which may result in neointimal proliferation and restenosis soon after treatment. The use of drug-coated balloons (DCBs) offers a convenient approach for the prophylactic delivery of a high local concentrations of antiproliferative drugs (2), while the overall dose is less than levels that are associated with risk for systemic adverse effects (3). For some products, convincing clinical results have been published (4–8), but further information is desirable. The questions that remain to be answered pertain to the efficacy of some products on the market outside the United States and application modes, as well as the use of DCBs in specific patient populations, vessel territories, and lesion types, including severe disease such as chronic occlusions and long lesions (9–11).
The present study may help advance our understanding of the use of DCBs in femoropopliteal lesions because of its design, the enrollment of a sufficiently large number of patients for subgroup analysis, the high proportion of long lesions and total occlusions, and the specific population.
Study design, data quality, and ethical considerations
The aim of this prospective, multicenter randomized controlled clinical study was to compare the efficacy, safety, and tolerance of paclitaxel-coated balloon catheters with commonly used uncoated balloon catheters in a typical patient population affected by significant peripheral arterial occlusive disease. After passage of the lesion with a guidewire, each patients was randomly assigned (1:1) to treatment with either an Orchid paclitaxel-coated peripheral balloon catheter (Acotec Scientific, Beijing, China) or an Admiral Xtreme peripheral balloon catheter (Medtronic, Minneapolis, Minnesota) by a central randomization computer system. The Orchid DCB is coated with paclitaxel (3 μg/mm2), and magnesium stearate is used as a carrier. Magnesium stearate displays lubricant activity at very low content of the composition. This carrier can reduce friction and minimize drug loss during insertion and transit to the target area.
Patients were treated according to the standardized study protocol and followed up with interim clinical examinations for 12 months. The primary endpoint was late lumen loss (LLL) in the treated vessel segment after 6 months, or earlier in case of clinically required target vessel revascularization.
Because of the difference in appearance and packaging between the test and control products, a complete double-blind design could not be achieved. However, angiographic parameters, including the primary efficacy endpoint, were evaluated by the quantitative angiography center at the Chinese PLA General Hospital Vascular Center, which was blinded to the treatment performed in the individual patients (DCBs vs. uncoated balloons).
The software used for measurement was the QVA system (Pie Medical Imaging, Maastricht, the Netherlands). Videos from each study site were identified by random numbers. Trial monitoring was performed by Beijing Bionovo Pharmaceutical Technology Development, and data management and statistical analysis of the trial were performed by the Medical Research & Biometrics Center, National Center for Cardiovascular Diseases, China.
The trial was conducted in conformity with the Declaration of Helsinki and the provisions for the conduct of clinical trials of medical devices issued by the China Food and Drug Administration. The study was approved by the Chinese PLA General Hospital Ethics Committee and the local ethics committees of participating hospitals. Patients were informed about the purpose, treatment, and conduct of the study and their right to withdraw their consent at any time. Patients were enrolled after providing written informed consent.
Patients, treatment, and follow-up
Adult patients could participate in the study if they provided written informed consent; if they had significant claudication, rest pain, or ischemic lesions typical of Rutherford stages 2 to 5; if they had severe (≥70%) stenosis or total occlusion in the superficial femoral artery or popliteal artery; if the total lesion length did not surpass 40 cm; and if inflow lesions, if any, were successfully treated. Exclusion criteria were intervention of lesions beyond single superficial femoral artery at the same time; significant renal insufficiency (creatine level >150 μmol/l); acute thrombosis; complete lack of runoff vessels; and relevant safety issues, such as pregnancy or lactation, life expectancy <2 years, and allergy to aspirin, heparin, clopidogrel, paclitaxel, or contrast agent.
Antiplatelet therapy with 75 mg/day clopidogrel and 100 mg/day aspirin was started 3 days before the intervention. The heparin dose during the intervention was 0.5 to 0.6 mg/kg body weight and, if required, an additional half dose after 2 h. After passage of the lesion with the guidewire, patients were randomized to 1 of the treatment groups. Except for the use of uncoated versus paclitaxel-coated catheters, treatment in both groups was basically identical. Pre-dilation of the lesion with uncoated balloon catheters was not mandatory but recommended. The length of the balloon had to be sufficient to cover at least 1 cm distal and proximal to the lesion. If the lesion length (plus 2 cm) surpassed the balloon length, an overlap of 1 cm between balloon-treated segments was recommended. As well, 30 to 60 s of inflation time was recommended. Post-dilation as well as the implantation of self-expanding stents was allowed in case of an unsatisfactory primary result or flow-limiting dissection. In addition to angiograms of the treated vessel segment before and after treatment, radiographs showing the inflated balloons in at least 2 orthogonal views were documented. Post-procedural therapy included aspirin and clopidogrel for 6 months.
Clinical examinations were performed before the intervention, before discharge from the hospital, 12 weeks after the procedure, and at 6 and 12 months. A control angiographic study was scheduled at 6 months (20 to 36 weeks after the index procedure). Used balloon catheters were collected and analyzed for residual paclitaxel using high-performance liquid chromatography with ultraviolet detection (2).
The primary endpoint was LLL, defined as the difference between the minimal luminal diameter after the procedure and at 6-month follow-up angiography or at the time of clinically driven target lesion revascularization (TLR).
Secondary efficacy endpoints were angiographic indicators of restenosis of the treated lesion, such as minimum luminal diameter and restenosis rate of the target lesion (stenosis ≥50% of reference diameter) and clinical indicators such as clinically driven TLR, changes in Rutherford classification and ankle-brachial index (ABI), and major amputation of the treated leg. At 12-month follow-up, primary patency, defined as freedom from TLR and restenosis as determined by Doppler ultrasound peak systolic velocity ratio ≤2.4, was recorded.
Safety evaluation included adverse events, such as thrombosis, allergy, and bleeding, occurring during and after the intervention and before discharge, clinically significant changes in laboratory test results, and major adverse events, such as all-cause death and major amputation of the target limb.
Analysis of the data for all endpoints was performed for the per protocol population. Continuous variables are expressed as mean ± SD. Categorical variables are expressed as frequencies or ratios. Normally distributed continuous variables were compared by using 2-sided Student t tests, and continuous variables not normally distributed were compared by using the Wilcoxon rank test. Categorical variables were compared using the 2-sided likelihood ratio chi-square test or Fisher exact test. For the intergroup comparison of the primary endpoint, the Cochran-Mantel-Haenszel chi-square test with the central effect adjusted was used. The McNemar test was used for the intragroup comparison of categorical data, the paired Student t test was used for intragroup comparison of normally distributed continuous data, and the Wilcoxon signed rank test was used for intragroup comparison of continuous data that were not normally distributed.
All statistical analysis were performed at a 2-sided significance level of 0.05. The statistical analysis software used was SAS version 9.4 (SAS Institute, Cary, North Carolina).
Overall, 200 patients were enrolled between April 2013 and June 2014 in random order at 10 clinical sites: 100 patients each in the control arm of the study treated with conventional uncoated balloon catheters and in the investigational arm treated with a similar paclitaxel-coated balloon catheter (Figure 1). Patient populations in the 2 groups were very similar, with no statistically significant differences (Table 1). In both groups, more than 50% of patients had diabetes, 40% were classified in Rutherford stage 4 or 5, 25% had involvement of the popliteal artery, and 25% underwent the intervention for treatment of in-stent restenosis. The mean lesion length was 152 ± 109 mm in the control group and 147 ± 110 mm in the DCB group (p = 0.78). Within the scope of the study, only 1 lesion per patient was treated. Balloon diameters ranged from 3 to 6 mm, the length of the uncoated Admiral balloons ranged from 20 to 150 mm, and the length of the drug-coated Orchid balloons ranged from 40 to 300 mm. If the treated vessel segment surpassed the length of a single DCB, a second catheter had to be used. Therefore, in many patients, more than 1 coated balloon was used. All but 2 patients in the control arm (uncoated balloon from a different manufacturer) were treated according to the protocol. In more than 80% of patients, crossover access was chosen (Table 2), and in >90% of patients, stenotic or occluded segments were pre-dilated. In 52% of patients in the control group and 54% in the DCB group, total occlusions were treated. No device failures were observed. Stents were implanted in 21% and 19% of patients, respectively. Still, residual stenosis of 35% in the control group and 33% in the DCB-treated patients was observed.
Patients with lesion lengths up to 40 cm were treated with up to 4 coated balloons per lesion. The mean paclitaxel dose per patient was 13 mg, and the highest dose was 43 mg. Mean residual paclitaxel on balloons after deployment was 11.4 ± 5.8% of the original dose.
A total of 36 adverse events from intervention to discharge from the hospital after treatment were reported, none serious, and the majority of events were judged to be unrelated to the specific treatment. Adverse events were randomly distributed between patients treated with the uncoated (n = 22) and coated (n = 14) balloon catheters and ranged from reactions to contrast medium administration to hemorrhage at the puncture site. The most frequent events were unrelated common infections. Although in some patients various hematologic abnormalities were observed, there were no changes suggesting myelosuppression, which is known to be associated with high-dose paclitaxel treatment in patients with tumors.
During the period between treatment and 6-month follow-up, 2 patients in each treatment group died of causes unrelated to peripheral arterial occlusive disease (Figure 1, Table 3); 1 patient in the DCB group died shortly after the 6-month angiographic control, but the death was reported at 6 months. Two patients in the control group were excluded because of treatment with nonassigned balloon, and 1 patient in the DCB group was lost to follow-up after 6 months. In each group (control and DCB), 89% of patients underwent angiographic follow-up, allowing LLL measurement at 6 months, and for 99% and 98% of patients, respectively, clinical information was available.
Overall, patients treated with the paclitaxel-coated balloons had clearly better angiographic and clinical outcomes up to 6-month follow-up (Figure 2). The primary endpoint of the study, LLL, averaged 0.05 ± 0.73 mm in the DCB group compared with 1.15 ± 0.89 mm in the control group. Binary restenosis rates were 22.5% and 70.8%, respectively. The TLR rate mirrored LLL: 6.1% of patients in the DCB group required repeat treatment of the target lesion compared with 38.8% in the control group. Corresponding statistically significant differences were found for minimal luminal diameter after 6 months or at the time of reintervention. Of concern, vascular positive remodeling was seen in both arms on the basis of a distribution analysis of LLL (Figure 3), in which the value of LLL was negative, and this finding was more frequent in the DCB group.
Clinical outcomes at 6 months were more favorable in patients treated with paclitaxel-coated balloons. Rutherford class was significantly lower in these patients (p < 0.001). The proportion of patients whose Rutherford class improved was 79% in the coated-balloon group and 57% in the uncoated-balloon group (p < 0.01). ABI was higher in patients treated with coated balloons (p < 0.001), and the improvement over baseline ABI was greater (p < 0.001) than in the patients treated with uncoated balloons. In addition, the difference between the treatment groups with regard to the incidence of clinically driven TLR was statistically highly significant in favor of patients treated with coated balloon catheters (6 of 99 patients vs. 38 of 98 patients, p < 0.001). It is interesting that 6 patients in the control group had TLR before 6-month control angiography compared with 1 patient in the group treated with the DCB.
Subgroup analysis was performed for sex, diabetes, lesion length, total occlusions, and treatment of in-stent restenosis (Table 4). The number of patients per subgroup ranged from 20 to 51. In each of the subgroups, LLL following treatment with DCBs was significantly less than LLL following treatment with conventional balloons (p < 0.001).
No difference was found between the two treatments in terms of safety, defined as the occurrence of major adverse events, both for the sum of death from any cause and major amputation of the treated leg taken together and for each parameter separately (Table 3). Neither clinical chemistry nor hematology revealed a difference between treatment groups. Adverse events were either unrelated to the treatment or known and pre-defined risks of percutaneous transluminal angioplasty, such as bleeding at the puncture site or deposition of thrombotic material in patients not responding to anticoagulant agents.
Twelve-month results did not indicate much change (Table 2). The difference in TLR between patients treated with drug- coated balloons and those treated with conventional uncoated balloons remained almost unchanged. Primary patency, Rutherford class, and ABI all indicated statistically better outcomes for patients treated with the coated balloons.
Drug coating of medical devices has significantly improved long-term outcomes of angioplasty in several vessel territories (12–14), while improvement in others is under debate (15,16). However, devices and their coatings differ from one another. Whereas large controlled trials have become the standard when investigating coronary interventions, sufficiently large, prospective multicenter randomized studies indicating angiographic as well as clinical benefits in a relevant patient population are still scarce for other vessel territories and have only recently been published (17,18). In these larger trials, primary endpoints were measured by Doppler ultrasound, which is less precise than the angiographic control used in smaller trials.
The present study, in which we investigated a new paclitaxel-coated balloon catheter in femoropopliteal lesions, is unique in that it meets the aforementioned criteria with regard to study design and enrolled a patient population that comes closer to clinical practice than that of any comparable study conducted before: approximately 40% of the patients had advanced disease corresponding to Rutherford class 4 or 5, a quarter of the patients had popliteal involvement, more than 50% of lesions were chronic total occlusions, and the mean lesion length was 15 cm, longer than in any comparable trial.
The paclitaxel dose on the balloons was 3 μg/mm2 balloon surface, close or identical to other coatings for which satisfactory efficacy has been shown in randomized trials. Furthermore, a wide range of investigational balloon lengths of up to 30 cm was available for the trial, reducing the need for overlapping and the risk for untreated gaps between balloons. Treatment was the same as in clinical practice, with the possible exception of the initial inflation time, which was approximately 3 min in both treatment groups. Pre-dilation was not mandatory but was performed in more than 90% of procedures. The study protocol also allowed post-dilation with prolonged inflation time as well as provisional stenting when flow-limiting dissection occurred or residual stenosis was unacceptably high.
Provisional stenting was performed in 20% of cases, which differs from the LEVANT (Lutonix Paclitaxel-Coated Balloon for the Prevention of Femoropopliteal Restenosis) 2 study, in which patients with flow-limiting dissection or clinically significant residual stenosis were excluded from randomization (19), and from the IN.PACT SFA (Randomized Trial of IN.PACT Admiral Drug-Eluting Balloon vs. Standard Percutaneous Transluminal Angioplasty for the Treatment of Atherosclerotic Lesions in the Superficial Femoral Artery and/or Proximal Popliteal Artery) (17) and THUNDER (Local Taxane With Short Exposure for Reduction of Restenosis in Distal Arteries) (19) studies, with about 10% stenting in each. A reason for the higher stenting rate may be the longer target lesion length and the high percentage of total occlusions in the present study.
Vascular positive remodeling was found in both groups, especially in the DCB group, which is similar in the PACIFIER (Paclitaxel-Coated Balloons in Femoral Indication to Defeat Restenosis) study (2). Positive remodeling is a typical phenomenon for vessel compensatory response to injury caused by the balloon dilation, potentially resulting in mild luminal enlargement. It is caused by the disruption of the diseased intima-media complex during balloon dilation, followed by a vascular healing process. In patients treated with percutaneous transluminal angioplasty, the luminal gain achieved by positive remodeling is frequently minimized by neointimal proliferation. Therefore, LLL values are typically positive, indicating net lumen loss. Negative LLL values corresponding to a net luminal gain are rarely observed in percutaneous transluminal angioplasty arms. In patients treated with DCB, the paclitaxel effect of inhibiting the proliferation of smooth muscle cells effectively reduces this lumen loss because of neointimal proliferation. Consequently, we found more patients with a net luminal gain in the DCB arm, which underlines the treatment effect of the coating.
The 6-month safety and efficacy results indicate a substantial beneficial effect of the coating: as in other trials of paclitaxel-coated balloon catheters in the superficial femoral artery, there was no difference in safety compared with uncoated balloon catheters. Local administration of paclitaxel inhibits neointimal proliferation, which was convincingly shown in the 6-month angiographic control by reduced LLL as the primary endpoint and increased minimal luminal diameter, reduced binary restenosis rate, and increased primary patency. The biological effect displayed on radiography translates into equally significant improvements in Rutherford class and ABI and a largely reduced need for repeat TLR.
Like every clinical trial, this study had limitations. The most significant limitation is the problem of blinding. In this study, the mode of randomization precluded the selection of patients according to treatment arm. However, as in other trials of DCBs, the coating is visible. Although follow-up was not necessarily done by the physician who treated the patient, no specific measures were taken to blind the person who performed repeat angiography, made the decision about TLR, or reported other efficacy parameters and adverse events, whereas quantitative evaluation of the angiograms was performed by a core laboratory blinded to the type of treatment the patients had received.
In the recent discussion of potential bias regarding the decision to perform TLR, the ratio of TLR to binary restenosis rate has been brought up. In this trial, the ratio was 6 TLRs to 20 binary restenoses in the DCB group and 38 TLRs to 63 binary restenoses in the control arm. Taking the distribution of LLL into account, this is an expected ratio, because more of the restenoses in the control group were of higher grades, whereas a higher proportion of the binary restenoses in the coated-balloon group were closer to the 50% stenosis limit (Table 2) and did not necessarily require revascularization.
The present study demonstrates the effective inhibition of restenosis following angioplasty with paclitaxel-coated balloons in a broad range of patients and lesions, with no recognizable adverse effects.
WHAT IS KNOWN? The use of DCBs for femoropopliteal occlusive disease is associated with significant reductions in late restenosis, and the coating technology is decisive.
WHAT IS NEW? This study has verified that a new paclitaxel-coated balloon can markedly improve angiographic and clinical outcomes despite advanced diseases in the majority of investigated subjects.
WHAT IS NEXT? Real-world clinical trials of the new DCB are needed to confirm the outcomes.
This study was supported by an unrestricted grant from Acotec. All authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Jia, Zhang, and Zhuang contributed equally to this work. Drs. Guo and Fu have overall responsibility.
- Abbreviations and Acronyms
- ankle-brachial index
- drug-coated balloon
- late lumen loss
- target lesion revascularization
- Received May 4, 2016.
- Revision received June 6, 2016.
- Accepted June 20, 2016.
- American College of Cardiology Foundation
- Rogers J.H.,
- John R.,
- Laird J.R.
- Scheller B.,
- Speck U.,
- Abramjuk C.,
- Bernhardt U.,
- Böhm M.,
- Nickenig G.
- Freyhardt P.,
- Zeller T.,
- Kröncke T.J.,
- et al.
- Werk M.,
- Albrecht T.,
- Meyer R.-D.,
- et al.
- Scheinert D.,
- Duda S.,
- Zeller T.,
- et al.
- Scheinert D.,
- Schulte K.-L.,
- Zeller T.,
- Lammer J.,
- Tepe G.
- Liistro F.
- Laird J.R.,
- Schneider P.D.
- Scheinert D.
- Jaff M.R.
- Stabile E.,
- Virga V.,
- Salemme L.,
- et al.
- Fanelli F.,
- Cannavale A.,
- Boatta E.,
- et al.
- Katsanos K.,
- Karnabatidis D.,
- Kitrou P.,
- et al.
- Liistro F.,
- Porto I.,
- Angioli P.,
- et al.
- Zeller T.,
- Baumgartner I.,
- Scheinert D.,
- et al.
- Tepe G.,
- Laird J.,
- Schneider P.,
- et al.