Paclitaxel-Based Therapies for Patients With Peripheral Artery Disease

Peripheral artery disease affects >12 million people in the United States.¹ Multisocietal guidelines recommend endovascular treatment for patients with intermittent claudication with anatomic disease involving the superficial femoral artery and popliteal artery locations^1,2 after the failure of exercise therapy. Over the past decade, dozens of trials have been performed investigating the role of paclitaxel (PTX)-coated balloons and stent technologies in patients with superficial femoral artery/popliteal disease for symptoms of intermittent claudication and critical limb ischemia.³

A recent article by Katsanos et al, “Risk of Death Following Application of Paclitaxel-Coated Balloons and Stents in the Femoropopliteal Artery of the Leg: A Systematic Review and Meta-Analysis of Randomized Controlled Trials,”³ concluded that “there is increased risk of death following application of paclitaxel-coated balloons and stents in the femoropopliteal artery of the lower limbs.” The authors performed a retrospective meta-analysis that combined published data on PTX-coated balloons and stents from recent randomized clinical trials. The investigators reviewed 28 studies (drug-eluting stents [DES]=4, drug-coated balloons=24) that had a 1-year follow-up, 12 that had a 2-year follow-up, and 2 (only 2 in published format) with a 5-year follow-up. These findings are unexpected and call into question the safety of PTX-based therapies that have already been used in tens of thousands of patients since gaining Food and Drug Administration approval.

PTX was discovered in 1963 during a screening program developed by the National Cancer Institute. It was isolated from the Pacific yew Taxus brevifolia, an evergreen found in the Pacific northwest.⁴ Several years later, PTX was shown to have anticancer properties; a mechanism of action was identified in 1979.⁴ For the past several decades, oncologists have tested and used PTX for treatment of many different cancers that are resistant to standard chemotherapy.⁴ The dose used for patients with cancer (100–175 mg/m² intravenously over 3–24 hours) is much higher than what was used in peripheral artery disease (PAD) trials (2–3.5 μg/mm²), differing in mode of delivery.⁴

PTX-coated stents have been used to treat coronary artery disease since the mid-2000s. The majority of the trials reported 1-year follow-up in major adverse cardiovascular events and stent restenosis. The TAXUS IV randomized, clinical trial demonstrated the superiority of PTX DES (n=662) to bare metal stents (n=652) with 9-month angiographic follow-up.⁵ Furthermore, there have been several additional coronary trials using PTX-coated stents in comparison with sirolimus-eluting stents, and none of them observed a significant difference in mortality between the 2 treatments with restenosis significantly lower in the DES cohorts than in the bare metal stent cohorts.

There have been 2 PTX-based superficial femoral artery randomized, clinical trials that have published 5-year data. Both trials showed superiority of the PTX-based therapy over the control arm at 5 years; however, both are limited by a significant number of patients being lost to follow-up or withdrawing from the study. For example, the Zilver PTX stent trial demonstrated superiority in patients treated with “>40% relative risk reduction for restenosis and target lesion revascularization through 5 years for the overall drug eluting stent (DES) in comparison with standard care and for provisional DES in comparison with provisional bare metal stent.” In this article, the authors reported that, of the 477 patients enrolled, ≈28% (133) of the patients were lost to follow-up or withdrew. Additional real-world safety and efficacy of the stent was reported in a large postmarket approval study.

With respect to the published results for PTX-coated balloons (PCBs), similar results were observed. In a randomized, clinical trial comparing PCBs (n=48) with controls (n=54), patients treated with PCBs had lower target-lesion revascularization (21%) than in the control group (56%, P=0.0005). In addition, a small group of patients had both duplex and angiographic follow-up, and, in this group, there was a lower rate of binary restenosis than in the PCB arm in comparison with controls (17% versus 54%, respectively; P=0.04). In this study, fewer than one-third of the patients (n=31) were available for 5-year follow-up. One major weakness of the present meta-analysis is that it does not define mortality rates for individual PCB and DES participants, which leaves unanswered the question of whether the outcomes are the same for all PTX devices or are there meaningful differences?

The authors’ observation from the meta-analysis may be biologically plausible given that the dosing, mode of delivery, and the pharmacokinetics of PTX-based PAD devices differ from systemic use for cancer or that used in coronary artery disease. The differences in the PTX therapy in patients with PAD may explain why thousands of patients have been treated with PTX for cancer and coronary artery disease with safe outcomes. It would be very easy to dismiss this meta-analysis as one full of “lies, damn lies, and statistics.” Nevertheless, the question becomes what do we do now for our patients, the physician community, regulatory agencies, and payers?

Here are our recommendations for moving forward expeditiously to establish data validity and confidence in these devices or establish the evidence necessary to have them removed from the market. First, a detailed analysis of the specific causes of death in all patients enrolled in all PTX-based trials for PAD treatment needs to be performed. Second, an in-depth analysis of patient-level data needs to be undertaken to evaluate any difference in mortality among total procedural PTX doses. Such an analysis would include the total dose of PTX per patient, which can be determined by reviewing the number, length of devices, the frequency of ipsilateral reinterventions, and contralateral treatments performed. The question of total dose arises from the heterogeneity of devices on the market, each with a proprietary drug delivery platform with different drug dose and device pharmacokinetics. Third, all ongoing PTX-based clinical trials need to determine and monitor the cause of mortality and possible link to PTX therapy. Fourth, all sponsors of clinical trials should share their data publicly so that the results can be validated independently with robust statistical analyses. Fifth, there remains the possibility that not all patients enrolled in these trials (PTX-based therapy and control) were followed up in the same way, resulting in a surveillance bias. For example, did patients with better outcomes require less closer follow-up and thus worse optimal medical management and higher mortality rates than what has been observed in other therapeutic interventional PAD trials? Finally, we would recommend mandatory participation in a registry to collect outcomes of these devices going forward.

The alarm raised by the publication of a harm signal tempered by contrary data of safety in another vascular bed and much greater doses used in a different disease process creates an opportunity for us to do best by our patients. We hope that these unexpected adverse results invoke the best influences in medicine: our prime rule of doing no harm and our scientific method for determining truth.