SWEDEPAD: No Mortality Signal With Paclitaxel in PAD

An unplanned interim analysis from a study that was halted in the aftermath of concerns about paclitaxel shows no difference in rates of long-term all-cause mortality between patients with PAD treated with coated or uncoated devices, regardless of the extent of limb disease.

The SWEDEPAD trialists, along with those from the BASIL-3 study, paused recruitment following publication of a meta-analysis in late 2018 by Konstantinos Katsanos, MD, PhD (Patras University Hospital, Rion, Greece), and colleagues. The controversial meta-analysis reported a 68% relative increase in risk of all-cause death with paclitaxel-coated devices versus uncoated devices at 2 years and a 93% relative risk increase by 5 years in those with femoral and/or popliteal artery disease.

SWEDEPAD consists of two parallel studies: SWEDEPAD 1 and SWEDEPAD 2, which are focused on patients with critical limb ischemia (CLI) and intermittent claudication, respectively. The new data show no significant difference in all-cause death in paclitaxel-treated versus uncoated device-treated patients at a mean follow-up of 2.5 years in either SWEDEPAD 1 (HR 1.04; 95% CI 0.90-1.21) or SWEDEPAD 2 (HR 1.18; 95% CI 0.72-1.93).

“Because this interim analysis does not show a significantly higher incidence of death resulting from the use of paclitaxel-coated devices than from the use of uncoated devices, we believe that equipoise remains, and recruitment has recently resumed, with enrollment of patients in both the chronic limb-threatening ischemia cohort and the intermittent claudication cohort,” the SWEDEPAD investigators, led by Joakim Nordanstig, MD, PhD (Gothenburg University, Sweden), write in the paper published online last week ahead of print in the New England Journal of Medicine.

It probably is the closest thing to an answer to Katsanos’ meta-analysis in terms of a prospective trial. Sahil Parikh

Commenting for TCTMD, Sahil Parikh, MD (NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY), characterized SWEDEPAD as an important data set because it has many characteristics that lend weight and reassurance regarding the lack of mortality signal. Those characteristics include a large and inclusive patient population and follow-up of up to 4 years. Plus, the socialized nature of the Swedish health system means there is little chance of the kinds of missing data problems that plagued many of the paclitaxel pivotal trials.

“It probably is the closest thing to an answer to Katsanos’ meta-analysis in terms of a prospective trial,” Parikh said. “This is reassuring in that the data suggests that there is no signal, albeit at a shorter time point than 5 years.” He added that the resumption of SWEDEPAD based on these interim findings is an important step forward for the researchers and for clinicians.

“It hopefully will boost their own enrollment, and it will allow us to kind of exhale a little bit, because there's been a lot of confirmatory data that's been released this year,” Parikh noted, citing the lack of a mortality signal in the ongoing Medicare analyses requested by the US Food and Drug Administration, individual patient-level data, and multiple reconciled databases from the pivotal trials.

For his part, Katsanos told TCTMD that he agreed that the SWEDEPAD results are “quite reassuring.”

“However, the results relate more to the CLI population, because as acknowledged by the authors, there were few deaths among patients with intermittent claudication, and as a result, the confidence interval for that particular group of patients was wide, implying ongoing uncertainty about the direction and magnitude of the effect, if any,” he said in an email.

Continuous Mortality Rate With No Differences

According to Nordanstig and colleagues, they had two primary reasons for wanting to publish their mortality data ahead of completing the trial, despite not having prespecified that they would do so.

“First, we sought to reduce patients’ and physicians’ concerns regarding the safety of paclitaxel-coated devices, and second, we considered the data to be important to support completion of ongoing trials investigating the efficacy of such devices in peripheral artery disease,” they write.

At the time of the unplanned interim analysis, SWEDEPAD had enrolled 2,289 patients (1,149 in the drug-coated-devices group and 1,140 in the uncoated-devices group). A total of 11 peripheral devices from nine manufacturers were used in the trial.

In the overall population, 25.1% of patients died during the follow-up period, including 25.5% in the drug-coated device group and 24.6% in the non-coated group (HR 1.06; 96% CI 0.92-1.22). As with the Katsanos meta-analysis, there were no differences at 1 year between the two treatment groups overall (HR 1.03; 95% CI 0.77-1.37) or by PAD severity.

In Kaplan-Meier analysis, the mortality rate was continuous throughout the entire follow-up period in the overall population and in the CLI group. However, the event rate was higher in the latter 2 years of the trial than in the first 2 years in the intermittent claudication cohort.

“In the overall population, the mortality rate per 100 patient-years was 10.4% among patients who received treatment with drug-coated devices and 9.8% among those who received treatment with uncoated devices,” the investigators write.

Where Are We Now?

Given the lack of a mortality signal in this trial and others, as well as the fact that the majority of paclitaxel device trials were neither designed nor powered to address mortality, the SWEDEPAD researchers suggest that “the reported mortality signal may have been caused by attrition bias that was potentiated in the meta-analysis or may even have been the result of chance.”
 

The SWEDEPAD results are a very welcome and reassuring addition to the inconsistent body of evidence around paclitaxel use in the peripheral arteries. Konstantinos Katsanos

The researchers acknowledge the potential for biases in SWEDEPAD due to their open-label study design, as well as the few deaths and wide confidence intervals in the claudicant group, as pointed out by Katsanos. There also was much greater use of low-dose paclitaxel devices in SWEDEPAD than in the trials included in the Katsanos meta-analysis, as well as variation in treatment effect across centers participating in SWEDEPAD.

“Still, contrary to real-world analyses that are contaminated by a myriad of confounders and unknown selection bias, the SWEDEPAD results are a very welcome and reassuring addition to the inconsistent body of evidence around paclitaxel use in the peripheral arteries,” Katsanos told TCTMD.

As recently as late November, an FDA scientist said the agency planned to assess new data as they were published and update their advice to clinicians when appropriate.

To TCTMD, Parikh said he wasn’t sure if the new SWEDEPAD data would carry enough weight with the FDA for them to modify their position on paclitaxel-based devices at this time, but they do likely strengthen the position of the vascular research community that trials of paclitaxel-based devices should continue to completion and that physicians should be able to use the devices in consultation with their patients.

“From my standpoint as a clinician, I've gone back to using drug-eluting technologies pretty much as I had been prior to the [Katsanos] meta-analysis,” he commented. “When I have a conversation with patients about this, I say that the preponderance of the evidence suggests that there's no danger and that I think it will prevent repeat procedures. In the current environment of COVID-19, and people's unwillingness to come back for repeat procedures, that seems to have made more of an impact on patients than any theoretical mortality risk, especially now that we're not even sure that risk is really true.”