Second-Generation Stents & New Frontiers
Is newer better?One aspect of the recent registries, such as the E-Five Registry, and randomized trials, such as SPIRIT IV and ENDEAVOR IV, is the revelation of longer-term data with larger numbers of patient on the various DES platforms, explains Mark Turco, MD, director of the Center for Cardiac and Vascular Research at the Washington Adventist Hospital in Takoma Park, Md.
The plethora of recent data, including the “encouraging results” with low-frequency, late-occurring events such as stent thrombosis, allows interventionalists to make inferences about potential differences between the various DES, says David E.
Kandzari, MD, director of interventional cardiology research at Scripps Clinic in La Jolla, Calif.
The results show that the crisis from ESC.06, which raised concerns about the correlation between DES and stent thrombosis, has “nearly subsided,” notes David Cox, MD, associate director of the cath lab and director of interventional cardiology research at Lehigh Valley Hospital in Allentown, Pa. “As a result, drug-eluting stents are approaching utilization levels similar to their height of popularity in 2006,” he says.
“Conversely, in the bare-metal stent era, we routinely switched platforms, believing that they all produced similar outcomes,” Kandzari explains. “Now, we are focused on stent safety in addition to efficacy. Ultimately, when choosing a DES, we should focus on the impact a chosen stent will have on the patient over the next 15 years, as opposed to the next 15 minutes of the procedure, regarding deliverability.”
“The exciting results of SPIRIT IV showed that when compared with [Boston Scientific’s] Taxus paclitaxel-eluting stent, [Abbott’s] Xience V everolimus-eluting stent is a very talented platform for safety and efficacy,” says Turco, adding that caveats remain for diabetics as no significant differences in the diabetic population were seen.
The trial results have clearly influenced the U.S. market, where Xience is becoming the preferred stent platform. In fact, the current estimates from Millennium Research Group and Guidepoint Global are that Xience/Promus (the latter of which is sold by
Boston Scientific) holds between 55 and 60 percent of the U.S. market, while Medtronic’s Endeavor maintains about 10 to 15 percent. For first-generation stents, Taxus holds a similar U.S. market share to Endeavor, but it is slipping, and Cordis’ Cypher consumes the remainder of the market.
At TCT.09, the Endeavor zotarolimus-eluting stent platform also showed some strong long-term data, according to Turco. “We are learning that as you get out further along with the Endeavor platform, there is a durability for both safety and efficacy that was not reflected in the early angiographic results,” he says.
Cox speculates that labs which continue to use the first-generation stents may have a contractual relationship with those companies. However, the choice to use first-generation stents is not endangering patients, as these stents also have positive long-term data, Cox says. “Most cath labs will switch over to second-generation products due to the perception of their safety benefits, as well as their unquestionable superior deliverability,” he says.
While Boston Scientific’s market share of conventional DES may slide, it sells the only FDA-approved stents for longer lesions and smaller vessels, so it will continue to dominate those markets.
What’s left? “We need more clinical trial data to assess stenting higher risk patients for off-label usages, such as left main, bifurcation disease, smaller vessels and longer lesions,” Turco says.
Game-changing outcomes?Mainly based on consistent data across several trials, the adoption shift continues to move toward the second-generation stents. The COMPARE trial found that Xience has a superior safety and efficacy profile compared to the second-generation Taxus
Liberte paclitaxel-eluting stent in unselected patients undergoing PCI. The difference was attributable to a lower rate of stent thrombosis (less than 1 percent vs. 3 percent), MI (3 vs. 5 percent) and target vessel revascularization (2 vs. 6 percent). Based on the results, Smits et al recommended that paclitaxel-eluting stents “should no longer be used in everyday clinical practice” (Lancet 2010;375(9710):201–209).
While Kandzari commends the design of the all-comers trial, he defines the author’s concluding statement as “stark,” despite it being supported by their findings.
Turco says that COMPARE is only one trial, and no single trial should alter clinical practice, especially if it is based on a single-center’s experience. The COMPARE results also do not mirror the TAXUS ATLAS trial results, says Turco, of which he was the principal investigator.
Turco acknowledges the data have pushed interventionalists away from paclitaxel, but he remains unconvinced that there are large safety differences between the two stents given the lack of power in these trials to determine this. “Yet, the efficacy data is clearly stronger for everolimus, based on results of SPIRIT and COMPARE,” he says.
SPIRIT IV randomized 3,690 patients to Xience or Taxus Express (the first-generation paclitaxel DES). The primary endpoint of target lesion failure through one year was 4.2 percent in the Xience arm and 6.8 percent for Taxus—a statistically significant difference. Though both stents had very low rates of stent thrombosis, Xience fared better than Taxus up to one year: 0.29 percent versus 1.06 percent—also statistically significant.
“SPIRIT IV results were more compelling than COMPARE results, but COMPARE is important as it assessed the second-generation paclitaxel stent,” Cox states. Also, since the trials were presented within moments of each other at TCT.09, the combination was “pretty appealing for Xience,” he adds.
The three-year results for ENDEAVOR IV, also presented at TCT.09, showed a strong but statistically non-significant trend favoring Endeavor compared with Taxus for target vessel failure rates (12.4 vs. 16.1 percent). Kandzari et al also found that rates of death and MI were significantly lower with Endeavor, about half that of the Taxus-treated patients—1.6 versus 2.3 percent for cardiac death and 2.2 versus 4.9 for MI. Although not a pre-specified endpoint, the trial also highlighted a significant difference in very late stent thrombosis.
Despite these outcomes, there are still physicians who witness durable, positive long-term outcomes with Taxus, and those who still believe paclitaxel may be the drug of chose in diabetic patients. In SPIRIT IV, diabetics (1,185) were the only subgroup for which Xience did not show a marked reduction in target lesion revascularization over Taxus: 6.5 versus 6.9 percent.
As a result of its potential in diabetics, Taxus should not be completely removed from inventory shelves, say Cox and Turco.
Left main: A new PCI frontier?While interventionalists are intrigued by the recent safety and efficacy data, questions remain if those data can be extrapolated from selected trial populations to more complex disease states that are not indicated for stent usage. Clinicians don’t seem to be waiting for a definitive answer. The two-year results from the real-life TAXUS ARRIVE registry showed that about 64 percent of all PCI procedures are off-label.
Richard J. Shemin, MD, a cardiothoracic surgeon at UCLA Ronald Reagan Medical Center in Los Angeles, attributes the proliferation of stenting for left main disease to the “enthusiastic nature” of interventionalists, as the techniques they employ are “sometimes well ahead of the clinical evidence,” noting that this has been at the root of much of the tension of the modern era in the treatment of ischemic heart and coronary heart disease between medical therapy, catheter-based interventions and surgical approaches.
The proliferation of using PCI to treat left main began at this last round of conferences, Turco explains, where the SYNTAX data revealed that unprotected left main percutaneous intervention can be performed “fairly safely” compared with CABG.
Kandzari, who co-authored a 2009 JACC white paper encouraging a change to the guidelines, adds that the studies suggest that the hard endpoints of death, MI and stroke are similar in the treatment of left main disease, comparing PCI to CABG (2009;54(17):1576-1588).
The LE MANS registry, which enrolled 252 patients after unprotected left main PCI, found that MACCE occurred in 4.8 percent of patients during the 30-day follow up period, including a 1.5 percent mortality incidence. During a mean follow-up of 3.8 years, Buszman et al found there were 25.4 percent MACCE and 13.9 percent deaths. The five- and 10-year survival rates were 78.1 percent and 68.9 percent, respectively (J Am Coll Cardiol 2009;54:1500-1511).
“Our 12-year experience with left main stenting reflects its continuous progress,” Buszman et al wrote.
“In my opinion, the standard of therapy for left main is still CABG,” says Shemin. In a Circulation editorial, Shemin wrote that surgeries for left main have a “distinct advantage in that they can ignore the complexity and location of the left main coronary lesion, because bypass grafts are placed distally to the left anterior descending and circumflex coronary arteries…[and] complete revascularization is easily accomplished (2008;118:2326-2329).
“With the change in guidelines [released late in the year at AHA.09] from a Class III to a Class IIb for left main disease, we are moving toward a world of greater acceptance of left main intervention,” Turco says. “However, we need to be very careful about saying that every patient with left main disease should undergo an intervention, and every operator should be performing left main PCI. A careful look at the differences in left main PCI in high SYNTAX score patients versus lower score patients clearly supports a more modified approach that is carefully thought out.”
SYNTAX provides context“With the backdrop of SYNTAX, which found favorable outcomes for PCI with the left main subgroup, especially in moderate- to low-risk patients, compared with surgery, we have come to a state of clinical equipoise for the disease,” Kandzari says.
SYNTAX provides a “slightly better understanding of when we can get away with treating patients with PCI versus CABG in the multi-vessel subgroup,” Turco says.
The outcomes in interventional studies tend to be driven by repeat revascularization. Yet, some have argued that if only the hard endpoints of death and MI are assessed, there is very little difference between PCI and surgery. As a result, Kandzari suggests the trial has been misrepresented to only favor surgery.
Shemin acknowledges the biases of both surgeons and interventionalists, and as a result, they break down the MACE events to favor their specialty.
However, Shemin is hesitant to dismiss repeat revascularization as an important endpoint. “Are we harming our patients by first attempting an interventional approach, and then mopping up with surgery later? In this era of comparative effectiveness, which requires an assessment in the durability of procedures, we should re-evaluate the cost and risk associated with repeat revascularization,” he says.
Kandzari concurs that where PCI falls short in left main is in the very complex patients, and in the need for revascularization, despite the use of DES. He also points out a current shift in interventional trial endpoints away from MACE to target lesion failure, which is a “more stent-specific endpoint that evaluates a cardiovascular death, repeat target lesion revascularization related to the stent territory and MI that is attributable to the treated vessels. The FDA is attempting to refine the endpoints to be more stent specific.”
In favoring surgery for this disease state, surgeons also can point to the longer-term CABG outcomes, which can be “measured by the decades,” Shemin says. “Most of the interventional studies—for both practical and economic reasons—usually examine outcomes within a two-year window.”
Many cardiologists are awaiting the EXCEL trial, which is examining how the Xience Prime platform performs in left main PCI compared with CABG in 2,500 patients over the course of three years. The primary composite endpoint is death, MI and stroke. The trial, sponsored by Abbott, has four co-principal investigators: two surgeons and two interventionalists.
EXCEL is the first trial powered to evaluate left main treatment specifically, and a surgeon and interventionalist team will decide the best treatment option. “PCI for left main will never become a Class 1 indication until it has been more definitively evaluated in an appropriately designed and adequately sized trial,” says Kandzari, who is EXCEL’s U.S. PCI leader.
Which left main patients?In certain patient populations, left main PCI might be an acceptable treatment, yet Turco acknowledges a few anatomic caveats. “Distal left main coronary stenting is a trickier procedure than proximal and mid-shaft left main because we still do not have a perfect bifurcation technology that is going to treat a distal left main, extending into the circumflex and LAD distribution,” he says.
“Most interventionalists feel comfortable employing PCI for ostial, proximal and mid-shaft left main with informed patient consent,” Cox says. “The 2009 guideline change took out much of the legal concern, but it failed to differentiate anatomic specifications, which remain a consideration for the operator.”
“While stent thrombosis could be catastrophic in the left main, it is exceptionally infrequent,” Kandzari says. “Restenosis, if it occurs, often occurs elsewhere from the left main. As a result, treating the isolated left main percutaneously is not only technically feasible, but is associated with favorable long-term outcomes.”
Shemin acknowledges that peri-procedural risks associated with surgery are “slightly higher” compared with PCI, but the overall results have “long-term durability.” He adds: “Surgeons feel so strongly about left main disease because it is clear-cut that we save lives and we’re not just palliating the disease.”
“Performing left main PCI ad-hoc is not yet ready for prime time,” concurs Cox, who also was cautious about operator volume for such complex procedures. “The procedures require more IVUS, patient-informed consent, a consultation with a surgeon and one’s interventional colleagues, as well as a SYNTAX-level intervention.”
“Technique definitely plays a role,” Kandzari says. “Intuitively, we believe achieving complete lesion coverage, as well as adequate stent expansion and apposition, is important for the treatment of any coronary lesion with PCI, but it is especially important in the left main.”
Next generation of game changersThe next breakthrough in the U.S. interventional realm may not require the emergence of a new technology. “The next major step is when data indicate when we can safely reduce our patients’ dependence on dual-antiplatelet therapy,” Turco says. “The next product that shows we can safely stop the recommendation of clopidogrel or prasugrel within three months of implantation will have a huge advantage, but it is important to remember we need long-term trial results, not just 12-month data for safety.”
Due to the U.S. regulatory thresholds, there most likely will not be any new, revolutionary technologies for the next four to five years. “However, we may see iterations of technology with more deliverable platforms,” Turco notes.
The nearest-term step toward a new DES could be a metallic-based platform with a bioresorbable polymer, which will elute the drug in a controlled fashion, and the polymer also will resorb in parallel to the drug, such as the Nevo program from Cordis,
Kandzari suggests. NEVO II, a global, randomized, non-inferiority trial of approximately 2,000 patients with coronary artery disease, comparing the Nevo sirolimus-eluting stent to Xience, was launched in March 2009.
Also, international research may provide answers before the U.S. market is able to react. “In the next three to five years, we will learn what role biodegradable stents will play in our practice versus biodegradable polymers,” Cox says. Interestingly, Abbott has the only mature biodegradable stent in the pipeline, while the other stent companies are focused on biodegradable polymers.
In the interim, Turco encourages physicians to re-examine the SPIRIT data to see “how good it really is.” Based on the recent second-generation DES, many interventionalists are questioning how much more improved these current therapies can get.