In the JournalsPerspective

Another study finds no evidence of increased long-term mortality with paclitaxel-coated devices

Eva Freisinger

Patients who underwent endovascular revascularization with a paclitaxel-based drug-eluting device did not have an increased mortality risk over an 11-year follow-up period, according to a safety analysis published in the European Heart Journal.

“Our findings now show on a large-scaled, real-world analysis on almost 65,000 patients that paclitaxel-based devices do not increase long-term mortality after its peripheral application,” Eva Freisinger, MD, senior physician in the department of cardiology at University Hospital Muenster in Germany, told Healio.

The debate on paclitaxel-coated devices was spurred in late 2018 after a meta-analysis found that the devices were associated with a long-term death risk in patients with peripheral artery disease. The FDA confirmed the findings in its own analysis.

The FDA’s Circulatory System Devices Panel met in June 2019 to discuss the signal of a late mortality risk associated with the devices, which resulted in the agency allowing the devices to remain on the market but with updated product labeling. Clinical trials were also recommended to be designed to follow patients long term and to improve adjudication of deaths. The latest action by the FDA was in August 2019, when the agency advised clinicians to discuss the risks of paclitaxel-coated devices with patients.

The present study is consistent with U.S. retrospective analyses from CMS data that have not shown any elevated mortality risk associated with paclitaxel-coated devices.

German health insurance data

Researchers analyzed data from 64,771 patients (mean age, 72 years; 45% women) who underwent 107,112 endovascular revascularization procedures with 23,137 drug-eluting devices between 2007 and 2015. Patients were categorized by their first endovascular revascularization procedure: drug-eluting stent (n = 676), drug-coated balloon (n = 2,648), bare-metal stent (n = 28,290) and percutaneous transluminal angioplasty (n = 33,157).

Follow-up was conducted until death or until December 2017 for a median of 92 months.

Although the use of paclitaxel-based DES were associated with increased hazards after 4 years, the link with increased long-term mortality was not statistically confirmed for up to 11 years after application, with an HR between 0.64 and 1.1 (P for all > .057).

During the first year, the use of DCBs was linked with decreased mortality (HR = 0.92; P < .001). There was an indifferent correlation in the years after its application (P for all > .202).

Potential implications

Freisinger told Healio that she thinks that this evidence may aid in the agency’s decision on the use of these devices. “We expect that official authorities like the FDA and The Federal Institute for Drugs and Medical Devices will soon react on these new results and adapt their statements on paclitaxel-based drug-eluting devices,” she said. “Based on our results, drug-eluting devices should be able to be applied where indicated without concern of late-term mortality associated with its use.”

Even those these results may have an impact on decision-making, more research is needed in this area, Freisinger said. “The whole debate on this safety issue points at the need of critical surveillance of medical devices even after its introduction on the market and its regular use in everyday practice,” Freisinger added. “Administrative databases are a suitable approach to answer these demands for real-world evidence and will be of use for similar questions in the future. Particularly high-risk cohorts that are usually excluded from the pivotal trials should be object of further investigation.” – by Darlene Dobkowski

For more information:

Eva Freisinger, MD, can be reached at eva.freisinger@ukmuenster.de.

Disclosures: Freisinger reports she received grants from Bayer and Pfizer. Please see the study for all other authors’ relevant financial disclosures.

Eva Freisinger

Patients who underwent endovascular revascularization with a paclitaxel-based drug-eluting device did not have an increased mortality risk over an 11-year follow-up period, according to a safety analysis published in the European Heart Journal.

“Our findings now show on a large-scaled, real-world analysis on almost 65,000 patients that paclitaxel-based devices do not increase long-term mortality after its peripheral application,” Eva Freisinger, MD, senior physician in the department of cardiology at University Hospital Muenster in Germany, told Healio.

The debate on paclitaxel-coated devices was spurred in late 2018 after a meta-analysis found that the devices were associated with a long-term death risk in patients with peripheral artery disease. The FDA confirmed the findings in its own analysis.

The FDA’s Circulatory System Devices Panel met in June 2019 to discuss the signal of a late mortality risk associated with the devices, which resulted in the agency allowing the devices to remain on the market but with updated product labeling. Clinical trials were also recommended to be designed to follow patients long term and to improve adjudication of deaths. The latest action by the FDA was in August 2019, when the agency advised clinicians to discuss the risks of paclitaxel-coated devices with patients.

The present study is consistent with U.S. retrospective analyses from CMS data that have not shown any elevated mortality risk associated with paclitaxel-coated devices.

German health insurance data

Researchers analyzed data from 64,771 patients (mean age, 72 years; 45% women) who underwent 107,112 endovascular revascularization procedures with 23,137 drug-eluting devices between 2007 and 2015. Patients were categorized by their first endovascular revascularization procedure: drug-eluting stent (n = 676), drug-coated balloon (n = 2,648), bare-metal stent (n = 28,290) and percutaneous transluminal angioplasty (n = 33,157).

Follow-up was conducted until death or until December 2017 for a median of 92 months.

Although the use of paclitaxel-based DES were associated with increased hazards after 4 years, the link with increased long-term mortality was not statistically confirmed for up to 11 years after application, with an HR between 0.64 and 1.1 (P for all > .057).

During the first year, the use of DCBs was linked with decreased mortality (HR = 0.92; P < .001). There was an indifferent correlation in the years after its application (P for all > .202).

PAGE BREAK

Potential implications

Freisinger told Healio that she thinks that this evidence may aid in the agency’s decision on the use of these devices. “We expect that official authorities like the FDA and The Federal Institute for Drugs and Medical Devices will soon react on these new results and adapt their statements on paclitaxel-based drug-eluting devices,” she said. “Based on our results, drug-eluting devices should be able to be applied where indicated without concern of late-term mortality associated with its use.”

Even those these results may have an impact on decision-making, more research is needed in this area, Freisinger said. “The whole debate on this safety issue points at the need of critical surveillance of medical devices even after its introduction on the market and its regular use in everyday practice,” Freisinger added. “Administrative databases are a suitable approach to answer these demands for real-world evidence and will be of use for similar questions in the future. Particularly high-risk cohorts that are usually excluded from the pivotal trials should be object of further investigation.” – by Darlene Dobkowski

For more information:

Eva Freisinger, MD, can be reached at eva.freisinger@ukmuenster.de.

Disclosures: Freisinger reports she received grants from Bayer and Pfizer. Please see the study for all other authors’ relevant financial disclosures.

    Perspective
    Mitchell W. Krucoff

    Mitchell W. Krucoff

    The good side of the Freisinger report is that this large dataset with 11-year follow-up is the longest follow-up reported in this large a cohort. So this does add importantly to our knowledge base.

    And of course, all additions to knowledge in this space have implications for clinical practice because the question of the late paclitaxel mortality signal compared to significant early benefit, which drug-eluting devices clearly have compared to any other therapeutic options in these patients is a critical public health concern. PAD is an important area of human suffering.

    The Freisinger report does not address what was raised at panel and what continues to be the center of ongoing concern, which is dealing with a signal that comes from multiple independent randomized trials. The Freisinger report is a very large dataset with very long-term follow-up, but it is not randomized.

    The key thing to appreciate there is that we still have very little insight into what, if any, plausible biological mechanism would mediate this kind of a finding. Nonrandomized patients can be statistically matched to some degree in lieu of randomized treatment assignment, but if you don’t know what the mechanism of the mortality signal is that you’re trying to account for, then the statistical methods for matching patient cohorts could also leave out critical variables and not match patients appropriately to the question we are trying to answer.

    So despite the very large and thorough approach Freisinger and colleagues report, the bottom line is that with no clue at this point as to what the biologically plausible mechanism of a late mortality impact might be, the ability to be sure you’re covering the right ground with statistical analyses of nonrandomized cohorts remains a big question mark in terms of the relevance of these findings to the safety signal emerging from randomized trials.

    Most recent discussions on this issue, including from the FDA, have focused on all sides of how we go forward on behalf of these patients. Careful gathering of high-quality data and follow-up in prospective randomized trials with paclitaxel devices is probably the most important way that we’re going to build additional knowledge and confidence as to whether this particular safety signal is either a very bizarre statistical anomaly — in other words, noise — or whether it is a signal of a biological mechanism that we clearly don’t understand. Of course, expedited development of drug-eluting devices that do not use paclitaxel potentially could also provide more effective alternatives altogether.

    • Mitchell W. Krucoff, MD, FACC, FAHA, FSCAI
    • Professor of Medicine/Cardiology
      Duke University Medical Center
      Director, Cardiovascular Devices Unit and eECG Core Laboratories
      Duke Clinical Research Institute

    Disclosures: Krucoff reports he works with all manufacturers of paclitaxel-coated devices through the Duke Clinical Research Institute to conduct research.