SAFE-DCB Demonstrates Consistent Efficacy at 1 Year in Real-world Femoropopliteal Disease Patients

WASHINGTON, DC—New registry data show promise for the use of the Lutonix drug-coated balloon (Bard) in a population of femoropopliteal disease patients with a high comorbidity burden and complex lesions.

The findings come amidst a months-long controversy over paclitaxel-coated balloons and stents, stemming from the now famous “Katsanos meta-analysis” pointing to higher late mortality with these devices as compared with plain balloon angioplasty. In fact, as Nicolas Shammas, MD, MS (Genesis Health Institute, Davenport, IA), presented the 1-year SAFE-DCB findings of real-world use of the Lutonix device in a late-breaking clinical trial session last week at CRT 2019, a day-long town hall on the safety concerns was simultaneously taking place elsewhere at the meeting.

Because of the high comorbidity burden observed in the population of the SAFE-DCB registry, Shammas told TCTMD that the 7.1% mortality they observed at 1 year was “in many ways reassuring.”

However, “this is not a trial powered to look at mortality. That was not the intention of that registry. This was a postmarket registry started years before even the controversy of mortality was even there. We didn't even have any controversy whatsoever when the first patient was enrolled,” he said, noting that there was not a high level of adjudication done for every death in SAFE-DCB, though they did have a medical monitor looking to see if these events were device-related.

Ultimately, the SAFE-DCB data are “very consistent with already published Lutonix studies demonstrating the excellent safety endpoint freedom from TLR,” Shammas said in his presentation.

One-Year Results

For the study, Shammas and colleagues enrolled 1,005 patients with superficial femoral and popliteal artery lesions from 74 US centers to be treated with the DCB between April 2015 and September 2016. More than half were diabetic (54.1%) or had CAD (52.4%), and more than one-third (35.1%) had critical limb ischemia. Chronic total occlusion (CTO) lesions were present in 25.3% of patients, 60% of patients had moderate or severe calcification, and the mean target lesion length and preintervention stenosis were 75 mm and 86.7%, respectively.

More than three-quarters of lesions (77.2%) underwent predilatation, while 49.6% were treated with atherectomy and 12.2% with cutting or scoring balloons. Acute procedural success was achieved in 93% of patients with a mean residual stenosis of 11.54%; 13.9% of lesions required a stent.

At 1 year, freedom from TLR—the primary efficacy endpoint defined as the first revascularization procedure after the initial study procedure—was 89.5%. Subgroup analyses showed no difference in this endpoint regardless of sex, CTO status, lesion length, lesion prep, proper technique used, and diabetes.

The primary safety endpoint of freedom from composite of device and/or procedure-related perioperative (≤ 30 days) death, target limb major amputation (above the ankle), and target vessel revascularization was 98.2%. By 1 year, 71 patients had died (7.1%), most commonly from cardiac arrest, MI and respiratory failure, sepsis or septic shock, and unknown causes. Freedom from target vessel revascularization at 1 year was 86.9%.

Future Due Diligence

Looking forward, Shammas said he would be “strongly in favor” of additional scrutiny to adjudicate deaths in the 2- and 3-year outcomes of SAFE-DCB. “It has to be brought up to the sponsors and their willingness to spend more money and all that stuff to do all that adjudication,” he commented. “I will definitely convey that to them and try to kind of raise the bar in investigating these mortalities a little bit more at year 2 and 3, but I can't tell you if they will respond in a positive way or not.”

It’s also possible that the US Food and Drug Administration might mandate additional adjudication as well, Shammas said.

“What the meta-analysis did is raise awareness of a potential problem. I don't want people to think that the meta-analysis is coming out with a definitive link or causality or anything like that because absolutely not,” he added. “We don't know at this point in time that the mortalities were attributed to the paclitaxel. There is no causality that we can establish. We don't even understand the mechanism of it. We also really don't have any type of patient-level analysis.”

Because of this, it will be “tough” to go back and analyze mortalities thought to be associated with paclitaxel in a retrospective manner, Shammas said. However, “the point is from now on, can we become a little bit more diligent about looking at the cause of mortality, adjudicate that a little bit more in detail and start diving into the patient-level analysis to see really if there is a link? I really think it's our obligation to do that. It's really important.”

Primary Patency Needed?

Following the presentation of the results, discussant D. Christopher Metzger, MD (Wellmont CVA Heart Institute, Kingsport, TN), questioned if primary patency was measured at all in SAFE-DCB.

“Yes. The primary patency was an addendum to the protocol, [but] there was only a smaller percentage of patients who ended up actually adhering to receiving the patency analysis at 1 year,” Shammas replied. “There were about 500 patients who didn't have a duplex ultrasound at 1 year, and this is currently being looked at and hopefully we'll be able to provide that data.”

Additionally, Metzger said, “you reported that 200 were lost to follow-up at 1 year—that was more than 20% of the patient population lost to follow-up at 1 year—and then even worse, 50% got their duplex analysis and then the report is freedom from TLR, which to me is a nebulous thing. It means either the patient didn't want another procedure or the doctor didn't want to do another procedure, and we lost at least 20% of the patients. So how comfortable are we with that? You can put a chart up that says 89% freedom from TLR, but we lost a lot and freedom from TLR does not mean freedom from restenosis as you know.”

“Absolutely,” Shammas replied. “Freedom from TLR, I completely agree, is not the strongest hard endpoint we can use. Patency would have been a lot [better], but that’s how the study was designed initially. When we realized that patency should be there, it was incorporated in the protocol.”

Panelist Mehmet Cilingiroglu, MD (Arkansas Heart Hospital, Little Rock), said that he is currently serving as PI of the DIAMOND-PAD study, which is using optical coherence tomography—not ultrasound—to image PAD patients with moderate to severe calcification before and after treatment with the Diamondback Orbital Atherectomy System (CSI). “So the things that you are describing—how do they evolve, how the plaque really happens—we already enrolled 20 patients, and it will be very interesting [to see the results],” he said. “We also want to know what happens to DCB that have dissections at 1 year.”

Shammas agreed that “it’s very hard to see what happens inside the vessel” in their IVUS-based analyses compared with angiographic readings. “In fact, when you do the IVUS, we had about four to six times more dissections than we've seen in an angiogram,” he said.

Panelist Gunnar Tepe, MD (Hospital of Rosenheim, Germany), commented on the “really striking” high rate of atherectomy in SAFE-DCB as well as the fact that vessel preparation, which is talked about “all the time,” didn’t have any effect on the primary efficacy endpoint. “Do you have an explanation for that?” he asked.

It’s possible that vessel preparation gives the most bang for its buck in complex lesions and that SAFE-DCB didn’t include enough of these complex lesions to demonstrate the value of atherectomy, Shammas answered. “When you translate vessel prep into simpler lesions, shorter lesions, or a lot less CTO, or a lot less calcification, you may not get the exact benefit that you really want to see from the vessel prep,” he said. “This is how we designed the JET-RANGER study; we designed it purely for very complex disease, long lesions, CTOs, and severe calcification because we believe the value of vessel prepping is really predominantly in these complex lesions. I don't think we have that complex lesion subset here that may reflect the value of atherectomy at this point.”