FDA Panel Deluged With Conflicting Paclitaxel-Device Data in PAD

Patrice Wendling

June 19, 2019

The US Food and Drug Administration's (FDA's) Circulatory System Devices Panel advisory committee unanimously agreed that there is a late mortality signal in patients with peripheral arterial disease (PAD) treated with paclitaxel-coated balloons and stents but couldn't say why or whether there is a class effect, at least based on the existing data.

"All of our industry sponsors have indicated that in carrying these studies out longer there's missing data and what we as a panel say 'is what makes it very difficult for us to interpret the data,' " panel chair Richard Lange, MD, MBA, president, Texas Tech University Health Sciences Center, El Paso, said. "So what we're left with is incomplete data, very good analysis, a signal, a smoking gun but no bullet or dead bodies at this particular point."

The 2-day meeting was called in response to observations of excess mortality starting at 2 years through 5 years in patients with PAD treated with paclitaxel-coated devices first detected in a meta-analysis of 28 randomized controlled trials (RCTs) published last December by Katsanos et al.

In March, the FDA warned physicians that "alternative treatment options should generally be used for most patients" after its preliminary review showed about a 50% increase in mortality with paclitaxel-coated devices among three pivotal trials (IN.PACT SFA 1 and 2, Zilver PTX, and LEVANT 2) with 5-year follow-up data.

Today's updated meta-analysis from the agency confirms that signal, finding a 57% increased mortality risk at 5 years with paclitaxel-coated devices vs uncoated devices for the as-treated patients in both fixed effect and random effects analyses (risk ratio [RR], 1.57; 95% confidence interval [CI], 1.16 - 2.13). Mortality risk estimates were 46% and 41%, respectively, when follow-up data available only through 3 years was added from a fourth pivotal trial, ILLUMENATE.

A similar trend was observed for the intention-to-treat population. A sensitivity analysis conducted to evaluate the impact of missing 5-year mortality data — which ranged from 2.7% to 26% — also suggested an increased mortality risk with the paclitaxel-coated devices.

Subgroup analyses showed no difference in mortality risk based on patient sex, although there was some evidence of a differential treatment effect between patients inside the United States versus those outside the United States in the IN.PACT SFA study only.

Industry Pushback

Industry representatives countered with five of their own analyses, none of which showed excess mortality. For example, Medtronic's pooled IN.PACT IDE and Japan analysis of as-treated patients showed a 5-year cumulative incidence of mortality of 14.7% for the IN.PACT Admiral drug-coated balloon (DCB) vs 12.0% for the percutaneous transluminal angioplasty (PTA; HR, 1.40; 95% CI, 0.76 - 2.57; P = .29).

Additionally, Philips' patient-specific, pooled meta-analysis of RCTs with less than 5% loss to follow-up showed no increased mortality signal at 3 years between the Stellarex DCB and PTA (9.3% vs 9.9%; P = .928). The addition of non-RCTs to the analysis showed a mortality rate of 7.9% for the DCB.

"We're subjected to a forest of dueling numbers," remarked panelist John W. Hirshfeld Jr, MD, professor of medicine, University of Pennsylvania Medical Center. "The problem is that the numbers presented by industry and the numbers presented by FDA are not the same. As a consequence, we have a conundrum in terms of what weight to put on each analysis we see."

Bran Zuckerman, MD, director of the FDA's Division of Cardiovascular Devices, said industry has "stepped up to the bar" in terms of providing additional data and analysis but "I would be careful in using the most up-to-date data analyses. They have not been checked by the FDA," and "the FDA has had a lot of back and forth with each company to verify data; so I would take for what it is. But the FDA data we believe is verifiable."

Other Views

Speaking during the afternoon's public hearing, Konstantinos Katsanos, MD, PhD, Patras University Hospital, Rion, Greece, lead author of the original meta-analysis, noted that only Cook Medical has provided patient-level data for further review. Based on those data from the ZILVER PTX trial, he showed a counterintuitive negative interaction between paclitaxel and competing risk factors, where the hazard of death was highest for patients with the fewest risk factors and lowest for those with the highest.

Eric A. Secemsky, MD, Beth Israel Deaconess Medical Center, Boston, Massachusetts, provided data from a recent Medicare analysis, which he said is important, as these are the patients seen in everyday practice. Among the roughly 152,000 patients, more than half of whom had chronic limb ischemia, the cumulative incidence of death at a median follow-up of 799 days was 43.1% in patients treated with paclitaxel-coated devices and 47.6% in those treated with uncoated devices.

Throughout the day, panel members and industry representatives stressed the need to look not only at statistical significance but the practical significance of the findings.

Daniel Clair, MD, Chair, Department of Surgery, University of South Carolina, Columbia, noted that PAD afflicts more than eight million Americans and that, following the FDA's March 2019 warning letter, healthcare systems have unilaterally decided to pull the devices off the shelves.

"For many physicians, the fear of potential lawsuits has led to these products being essentially unusable in the current environment," he said. "For our patients, this means a dramatically increased risk for intervention failure and risks related to reintervention, potentially limb loss in more severe cases."

Col. Todd Rasmussen, MD, Walter Reed National Military Medical Center, Bethesda, Maryland, said the analyses by Katsanos et al and the FDA team have brought to the fore several critical areas of study, such as the dose relationship, length of lesion, and inclusion of long-term patient data.

"However, with regard to magnitude, I find the findings to be statistically significant but not particularly practically significant as I sit in my office and talk to my patients," he said. "I would not underestimate our patients' ability to access and assess data and to make their own decisions."

The panel will reconvene tomorrow for further discussions including the dose–response relationship between paclitaxel and mortality and potential strategies to address the crisis.

US FDA Circulatory System Devices Panel advisory committee meeting, Gaithersburg, Maryland, June 19-20, 2019. Agenda

Follow Patrice Wendling on Twitter: @pwendl. For more from theheart.org | Medscape Cardiology, join us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.