Meeting NewsPerspective

Mortality not different at 5 years in patients treated with DES vs. controls

Michael Dake

LAS VEGAS — There were no significant differences in mortality at 5 years between patients with peripheral artery disease assigned a drug-eluting stent and those assigned a bare-metal stent or percutaneous transluminal angioplasty, according to new data presented at VIVA 19.

The mortality results did not differ in the intention-to-treat and as-treated analyses, Michael Dake, MD, senior vice president of health sciences and professor of medical imaging, surgery and medicine at the University of Arizona, said during a press conference. The topic has been at the forefront since a meta-analysis showing a long-term mortality signal associated with paclitaxel-coated devices was published in December 2018.

The researchers analyzed the 94% of patients from the 489-patient pivotal trial of the DES (Zilver PTX, Cook Medical) for whom 5-year mortality data were available.

In the intention-to-treat analysis, 21.7% of those assigned the DES died within 5 years compared with 15.2% of controls (P = .08), Dake said.

In the as-treated analysis, in which people who actually received treatment with the DES, which included crossovers and people who had a suboptimal PTA that underwent a secondary randomization to DES, were counted in the DES group, 19.1% of the DES group died within 5 years compared with 17.1% in the control group (P = .6), he said. In all, 40% of patients from the control group were eventually treated with the DES, he noted.

Dake noted that randomization was based only on lesion length, and the DES group had more risk factors than the control group (P < .01), so the numerically higher mortality rates in the DES group could be explained by the risk-factor disparity.

Patients with fewer risk factors had lower mortality rates, and the benefit of DES in freedom from target lesion revascularization was consistent across risk-factor groups, though the greatest benefit was observed in those with fewer risk factors, Dake said at the press conference.

“This raises an interesting question,” Dake said during the press conference. “We know all patients have risk. Who is a high-risk patient? It’s not just someone who has critical limb ischemia. The risk factors need to be further analyzed to determine their relative impact on mortality.”

The randomization process for such trials may need to change, Dake said in an interview.

“There are agreed-upon risk factors for mortality and a different set of risk factors for TLR,” Dake told Cardiology Today. “Some overlap, but most do not. So, in terms of randomization and accounting for balance, it matters what endpoint you are looking at. Remember, none of the trials were powered to look at mortality. In the future, new studies may need to take into account risk factors during randomization.” – by Erik Swain

Reference:

Dake M, et al. Late-Breaking Clinical Trials. Presented at: VIVA 19; Nov. 4-7, 2019; Las Vegas.

Disclosures: The study was funded by Cook Medical. Dake reports he received consultant fees and/or honoraria from Bard Peripheral Vascular, Cook Medical, Novate and W.L. Gore and Associates.

Michael Dake

LAS VEGAS — There were no significant differences in mortality at 5 years between patients with peripheral artery disease assigned a drug-eluting stent and those assigned a bare-metal stent or percutaneous transluminal angioplasty, according to new data presented at VIVA 19.

The mortality results did not differ in the intention-to-treat and as-treated analyses, Michael Dake, MD, senior vice president of health sciences and professor of medical imaging, surgery and medicine at the University of Arizona, said during a press conference. The topic has been at the forefront since a meta-analysis showing a long-term mortality signal associated with paclitaxel-coated devices was published in December 2018.

The researchers analyzed the 94% of patients from the 489-patient pivotal trial of the DES (Zilver PTX, Cook Medical) for whom 5-year mortality data were available.

In the intention-to-treat analysis, 21.7% of those assigned the DES died within 5 years compared with 15.2% of controls (P = .08), Dake said.

In the as-treated analysis, in which people who actually received treatment with the DES, which included crossovers and people who had a suboptimal PTA that underwent a secondary randomization to DES, were counted in the DES group, 19.1% of the DES group died within 5 years compared with 17.1% in the control group (P = .6), he said. In all, 40% of patients from the control group were eventually treated with the DES, he noted.

Dake noted that randomization was based only on lesion length, and the DES group had more risk factors than the control group (P < .01), so the numerically higher mortality rates in the DES group could be explained by the risk-factor disparity.

Patients with fewer risk factors had lower mortality rates, and the benefit of DES in freedom from target lesion revascularization was consistent across risk-factor groups, though the greatest benefit was observed in those with fewer risk factors, Dake said at the press conference.

“This raises an interesting question,” Dake said during the press conference. “We know all patients have risk. Who is a high-risk patient? It’s not just someone who has critical limb ischemia. The risk factors need to be further analyzed to determine their relative impact on mortality.”

The randomization process for such trials may need to change, Dake said in an interview.

“There are agreed-upon risk factors for mortality and a different set of risk factors for TLR,” Dake told Cardiology Today. “Some overlap, but most do not. So, in terms of randomization and accounting for balance, it matters what endpoint you are looking at. Remember, none of the trials were powered to look at mortality. In the future, new studies may need to take into account risk factors during randomization.” – by Erik Swain

Reference:

Dake M, et al. Late-Breaking Clinical Trials. Presented at: VIVA 19; Nov. 4-7, 2019; Las Vegas.

Disclosures: The study was funded by Cook Medical. Dake reports he received consultant fees and/or honoraria from Bard Peripheral Vascular, Cook Medical, Novate and W.L. Gore and Associates.

    Perspective
    Krishna Rocha-Singh

    Krishna Rocha-Singh

    This trial had the most patients of any that comprised the controversial meta-analysis, and the most patients lost to follow-up. The trial design was very problematic in terms of looking for a mortality signal, and approximately 20% of patients were lost to follow-up.

    Adding data from newer trials may help, but those trials often have 3:1 randomization. The control arms are becoming smaller, which becomes statistically problematic.

    Further confusing matters is the FDA language suggesting these devices only be used in patients at high risk for revascularization. The agency has left it up to the doctors to define what that is. We have to consider each patient individually. We are going to have to make a Venn diagram to look at their lesion metrics and their demographic metrics to find out the best technology to use. It may turn out that the highest-risk patients may stand to benefit the most, and their mortality may be driven not by the technology but by their risk factors.

    The question we have to ask is what is going to clear the air. I am concerned that by the time the air gets cleared, I will be retired. And what’s to say there won’t be a similar signal with the devices being developed that use limus drugs?

    There will be more mandates around the trial designs, and the trials will become more expensive. This market has to move on. We can do better. We have to.

    • Krishna Rocha-Singh, MD, FACC, FAHA
    • Chief Scientific Officer
      Prairie Heart Institute of Illinois at St. John’s Hospital
      Springfield, Illinois

    Disclosures: Rocha-Singh reports he has received honoraria from Medtronic, consulted for Alucent Biomedical, Abbott Vascular, Medtronic, ROX Medical and SoundBite Medical, held equity in PQ Bypass and received research grants from Medtronic.

    See more from VIVA