Dual Perspectives on DAPT

Consensus on optimal duration of dual antiplatelet therapy after stenting remains elusive.

The optimal duration of dual antiplatelet therapy after PCI with stenting is one of the hottest topics in interventional cardiology in recent years. However, despite a high volume of new research emerging at major medical conferences and in the top scientific journals, there is still no consensus in the interventional cardiology community.

The question remains: shorter or longer? Shorter duration of dual antiplatelet therapy (DAPT) has been associated with lower rates of bleeding. Meta-analyses suggest there also may be a link between shorter DAPT and lower all-cause mortality. On the other hand, longer duration of DAPT has been associated with lower rates of MI and stent thrombosis.

No algorithms have been put forth to identify which patients are most likely to benefit from DAPT duration of less than 1 year, 1 year, or more than 1 year. Guidelines committees in the United States and Europe are in the process of incorporating new research from studies including the DAPT trial, but it is too early to forecast how guidelines might be revised. For now, cardiologists must use their clinical judgment to decide on the appropriate DAPT duration for each patient who has received a coronary stent.

“Prolonged DAPT is something you might consider in selected patients, and those would be patients in whom the potential ischemic benefits of prolonged DAPT outweigh the potential bleeding risks. The problem is, we do not yet know how to identify those patients,” Gregg W. Stone, MD, professor of medicine at Columbia University, director of cardiovascular research and education at the Center for Interventional Vascular Therapy at New York Presbyterian Hospital/Columbia University Medical Center, and co-director of medical research and education at the Cardiovascular Research Foundation, told Cardiology Today’s Intervention. “We all have our educated intuition as to what makes a high ischemic-risk patient as opposed to a high bleeding-risk patient, but it would be nice to get more data out of these large databases to come up with a formulaic method of prediction.”

The Current Standard

The standard duration of DAPT after PCI with stenting has been 1 year, but that standard has evolved over time and was never definitively proved as optimal, experts told Cardiology Today’s Intervention.

Early trials with bare-metal stents, such as CREDO and PCI-CURE, indicated that 9 months to 1 year of DAPT provided more clinical benefit than 30 days of DAPT, Dean J. Kereiakes, MD, FACC, FSCAI, medical director of The Christ Hospital Heart and Vascular Center and the Carl and Edyth Lindner Center for Research and Education at The Christ Hospital, Cincinnati, said in an interview.

In the first studies of DAPT with drug-eluting stents, DAPT duration was 3 months in the SIRIUS trial of a sirolimus-eluting stent (Cypher, Cordis) and 6 months in the TAXUS IV trial of a paclitaxel-eluting stent (Taxus, Boston Scientific). Those became the early standards, said Stone, a member of the Cardiology Today’s Intervention Editorial Board.

Then, in 2006, concerns emerged about late stent thrombosis associated with these first-generation DES, and the FDA convened a special meeting on the topic.

“We all agreed it was a concern,” Stone said. “Without any data to suggest that it would help to prolong DAPT or be worth the increased risk for bleeding, the FDA felt that it was reasonable to increase DAPT duration to 1 year, or even longer, if the patient was tolerating the drugs without side effects or bleeding. Most societal guidelines concurred.”

Gilles Montalescot, MD, PhD, professor of cardiology and head of the department of cardiology at Pitié-Salpêtrière Hospital in Paris, noted that 1-year DAPT was already the standard of care for patients with ACS. “Patients with DES were at continuously higher risk for MI and death after the conversion from DAPT to single antiplatelet therapy, presumably due to late stent thrombosis,” Montalescot said. “Then it was recommended [to utilize] DAPT for 1 year after implantation based on expert consensus, assimilating DES in stable patients to ACS treatment.”

Kereiakes said a compelling presentation at that 2006 FDA meeting was made by Eric L. Eisenstein, DBA, from Duke University Medical Center, Duke Clinical Research Institute and Duke Translational Medicine Institute. Eisenstein and colleagues found that in an observational study of patients receiving stents at Duke Heart Center, extended use of DAPT with aspirin and clopidogrel in patients with DES reduced risk for death or a composite of death/MI. In patients with BMS, they found that DAPT duration did not affect those outcomes. The data were subsequently published in JAMA.

“The early trials and Eric Eisenstein’s data from Duke suggested 1-year DAPT, but that was not proven by any single, large-scale, randomized, adequately powered trial,” said Kereiakes, who also is professor of medicine at The Ohio State University and a member of the Cardiology Today’s Intervention Editorial Board. “But that is where [the] 1-year [recommendation] came from, and then people tried to whittle it down to shorter durations. There were some data that longer is better, but it was expert consensus that 1 year made sense.”

Spotlight on DAPT Trial

To try to provide an answer as to the optimal duration of DAPT, the FDA, the Harvard Clinical Research Institute, and eight stent and pharmaceutical manufacturers collaborated on the DAPT trial, the first study adequately powered to detect an association between DAPT duration and ischemic MACCE and stent thrombosis.

The study included a total of 11,648 randomized patients: 9,961 patients with DES and 1,687 with BMS. Each participant received aspirin and a thienopyridine — usually clopidogrel, but sometimes prasugrel (Effient, Daiichi Sankyo/Eli Lilly) — for 1 year after stenting. At 1 year, patients who did not have an adverse event and were adherent to thienopyridine therapy were randomly assigned to continue thienopyridine therapy an additional 18 months or to placebo.

Among patients with DES, compared with DAPT for 1 year, those who received DAPT for 30 months had lower rates of stent thrombosis (0.4% vs. 1.4%; HR = 0.29; 95% CI, 0.17-0.48), MACCE (4.3% vs. 5.9%; HR = 0.71; 95% CI, 0.59-0.81) and MI (2.1% vs. 4.1%; HR = 0.47; P < .001), but a higher rate of moderate/severe bleeding (2.5% vs. 1.6%; P = .001). The long-term DAPT group had a higher rate of all-cause mortality than the short-term group, but it was not statistically significant (2% vs. 1.5%; HR = 1.36; 95% CI, 1-1.85; P = .052). Among patients with BMS, results were similar, but not statistically significant.

“Across the board, we saw consistent benefit,” DAPT investigator Laura Mauri, MD, MSc, interventional cardiologist at Brigham and Women’s Hospital, associate professor of medicine at Harvard Medical School and chief scientific officer at Harvard Clinical Research Institute, told Cardiology Today’s Intervention. “It didn’t matter whether you presented with MI for your stent procedure. It didn’t matter whether you had other risk factors for MI. When we think about who might benefit, it doesn’t necessarily follow the traditional idea that you would just give [prolonged DAPT to] high-risk patients.”

A majority of the MIs prevented were not related to the stent, according to Kereiakes, co-principal investigator on the trial with Mauri.

“The magnitude of the increment of MI is more than twice the magnitude of the increment in stent thrombosis,” he said. “The reason is that the majority of infarcts were not related to the stent. What we saw in the DAPT trial was that 55% of the MIs prevented by DAPT were not related to stent thrombosis. Stopping DAPT is a hazard that I don’t think has ever been demonstrated as graphically as it was in the DAPT trial.”

An unexpected finding in both treatment groups was an elevated risk for MI and stent thrombosis within the first 3 months after thienopyridine discontinuation.

“Not only was there a dramatic increase in MI and stent thrombosis between 12 and 15 months in the group randomly assigned to placebo, but between 30 and 33 months, when the longer treatment group comes off their thienopyridine, a timeframe where we would have thought that patients would be more stable and not so prone to thrombosis, there is again a highly significant increase in stent thrombosis and a threefold greater increment in MI,” Kereiakes said. “The hazard of thienopyridine discontinuation was still there. This type of information has public health care ramifications.”

Trials of Short-Term DAPT

Although the DAPT trial demonstrated advantages of DAPT duration for more than 1 year, other trials have examined advantages of DAPT duration for less than 1 year.

Trials published since 2011 that have explored DAPT duration of 3 or 6 months compared with 1 year or longer include EXCELLENT, PRODIGY, RESET, OPTIMIZE, SECURITY, ITALIC and ISAR-SAFE (see Table). The latter two were presented alongside the DAPT trial results at the 2014 American Heart Association Scientific Sessions.

In ITALIC, results revealed no difference in the primary outcome, a composite of death, MI, urgent target vessel revascularization, stroke and major bleeding at 1 year, in patients assigned 6-month DAPT vs. those assigned 2-year DAPT after everolimus-eluting stent (EES) implantation (6-month DAPT, 1.6%; 2-year DAPT, 1.5%; HR = 1.072; 95% CI, 0.517-2.221).

In ISAR-SAFE, findings showed no difference in the composite primary outcome of death, MI, stent thrombosis, stroke and TIMI major bleeding at 9 months after randomization in patients assigned 6-month DAPT vs. those assigned 1-year DAPT after DES implantation (6-month group, 1.5%; 12-month group, 1.6%; HR = 0.91; 95% CI, 0.55-1.5). The results for both were consistent with most of the other trials of short-term DAPT duration.

A problem, Kereiakes said, is that those trials “are underpowered, have few events and have variable adherence to the assigned treatments. The entirety of the combined ISAR-SAFE and ITALIC studies have only 12 total stent thromboses in the randomized treatment arms. There is nothing that can be said with confidence regarding the impact of DAPT duration on stent thrombosis from those two trials.”

Meta-Analyses Highlight Risks

Teams of investigators also undertook meta-analyses with hopes of drawing further conclusions from the trial data. Three such analyses were recently published.

One meta-analysis of 10 studies — ARCTIC-INTERRUPTION, DAPT, DES-LATE, EXCELLENT, ISAR-SAFE, ITALIC, OPTIMIZE, PRODIGY, RESET and SECURITY — demonstrated that shorter DAPT was associated with lower all-cause mortality vs. longer DAPT (HR = 0.82; 95% CI, 0.69-0.98; number needed to treat = 325), with no significant heterogeneity across trials. Compared with longer duration, shorter DAPT also lowered risk for major bleeding (HR = 0.58; 95% CI, 0.47-0.72), but increased risk for MI (HR = 1.51; 95% CI, 1.28-1.77) and definite/probable stent thrombosis (HR = 2.04; 95% CI, 1.48-2.8). There was no difference in stroke outcomes (HR = 1.03; 95% CI, 0.78-1.36).

Another meta-analysis covered four trials — EXCELLENT, OPTIMIZE, PRODIGY and RESET — and had a primary outcome of MACE, defined as cardiac death, MI or definite/probable stent thrombosis. At 1 year, compared with prolonged DAPT, short-term DAPT was associated with similar rates of MACE (HR = 1.11; 95% CI, 0.86-1.43), but lower rates of bleeding (HR = 0.66; 95% CI, 0.46-0.94).

A third meta-analysis of the same 10 trials included in the mortality analysis indicated that, compared with longer DAPT, shorter DAPT was associated with increased stent thrombosis (OR = 1.71; 95% CI, 1.26-2.32), but that this was attenuated in patients who received a second-generation DES (OR = 1.54; 95% CI, 0.96-2.47) vs. those who received a first-generation DES (OR = 3.94; 95% CI, 2.2-7.05; P for interaction = .008). In addition, shorter DAPT was linked to a lower rate of clinically significant bleeding (OR = 0.63; 95% CI, 0.52-0.75) and a numerically lower rate of all-cause mortality (OR = 0.87; 95% CI, 0.74-1.01) compared with longer DAPT.

“We found that first-generation DES are associated with more late events like death, MI and stent thrombosis,” Roxana Mehran, MD, FACC, FACP, FCCP, FESC, FAHA, FSCAI, an author of that meta-analysis and professor of medicine in cardiology at Icahn School of Medicine at Mount Sinai and director of interventional cardiovascular research and clinical trials at Mount Sinai Heart at The Mount Sinai Hospital, said in an interview. “Within the DAPT study itself, results did not show a stent effect; in the meta-analysis, with larger numbers of patients, we did see an important P interaction when we looked at first- vs. second-generation DES. That is something we have to think about.”

Another meta-analysis of 69,644 patients in 14 trials found that compared with aspirin alone or DAPT of 6 months or less, continued treatment with DAPT was not associated with a difference in all-cause mortality (HR = 1.05, 95% credible interval, 0.96-1.19), CV mortality (HR = 1.01, 95% credible interval, 0.93-1.12) and non-CV mortality (HR = 1.04, 95% credible interval, 0.9-1.26).

What can be learned from all of this research is that “safe interruption of DAPT 6 months after implantation is possible, if we accept that the data were consistent in the studies performed with duration shorter than 1 year,” said Montalescot, a member of the Cardiology Today’s Intervention Editorial Board. “Prolongation of DAPT beyond 1 year after DES implantation is also possible. This has been tested in patients selected after 1 year of follow-up. The strategy of using DAPT to reinforce secondary prevention is more sensible in patients at high ischemic risk, although the benefit was observed across almost all subgroups. Long-term DAPT reduces stent-related and non–stent-related thrombotic events, at the cost of more bleeding complications. Selection of patients for the duration of DAPT is going to be important.”

Bleeding vs. Thrombosis

The question that an interventional cardiologist must ask, then, is what should be feared more: bleeding events or thrombotic events?

Unfortunately, there is no consensus on this, either. Some experts said they believe that patients should receive long-term DAPT after stenting, except for those at high bleeding risk. Some said patients should receive short-term DAPT after stenting, except for those at high thrombotic risk. And some said therapy should be individualized based on patient risk factors.

“If patients do not bleed, particularly on clopidogrel, then they should probably be maintained on DAPT indefinitely,” Kereiakes said. “We have at times likened it to statin therapy because the benefit appears to continue to accrue over time. But it is not without hazards, and we don’t know who is at greatest risk. There is a lot of discussion about individualized therapy, but there is not one algorithm that [has been] proven effective.”

Stone, who was a co-author of two of the meta-analyses, said “our conclusion from the entire data was that the default should be shorter rather than longer DAPT as giving the best net overall outcomes for patients, because of the side effects of prolonged DAPT. However, it needs to be done on a case-by-case individualized basis where you carefully look at the potential benefits of ischemic complications and stent thrombosis compared with the potential risks for bleeding. Unfortunately, the characteristics that predict a high rate of atherosclerosis and ischemic MACE also predict bleeding. It’s not all the same factors, but there is often an overlap.”

Also of great concern is the increased rate of all-cause mortality observed in the long-term DAPT groups in the DAPT trial and one of the meta-analyses, Stone said. “That is particularly concerning because you are preventing MI and stent thromboses with prolonged DAPT, but it did not translate to any reduction in cardiac death. On the other hand, there was a substantial increase in bleeding and a concomitant increase in non-CV mortality, such that all-cause mortality was increased.”

Several factors could be in play. Clinicians may better treat MI and stent thrombosis than bleeding complications, late stent thromboses may be less clinically relevant than early stent thromboses, and the excess deaths in the DAPT trial are mostly attributed to bleeding, cancer and trauma, “which do mechanistically make sense that they may be related to being on a second antiplatelet agent,” Stone said.

A mechanistic link, however, is uncertain. Kereiakes noted that fatal bleeding was rare in the DAPT study (0.02% in both arms) and attributed the mortality imbalance in the DAPT trial to chance, since cancer-related mortality accounted for the differences seen and was not bleeding-related in almost all cases; by chance, more patients had cancer in the longer-term DAPT group than in the shorter-term group. “No one legitimately thinks that either the drugs or the stent were causing cancer,” he said.

Physicians should consider several factors. “Stronger and longer antiplatelet therapy works in coronary patients, in particular, if they have presented with a thrombotic event such as ACS or stent thrombosis,” Montalescot said. “In low-risk patients at high bleeding risk, short treatment should prevail, as safety must be considered first.”

Mauri noted that patients enrolled in the DAPT trial “had a much higher likelihood to benefit from ischemic risk reduction” because patients at high bleeding risk, such as those on warfarin or other oral anticoagulants, were not included. “They clearly are a high-risk population for bleeding, and we have to think about strategies to shorten or simplify therapy,” she said. However, “because the DAPT study showed large reductions in stent thrombosis and MI, many of which were not even related to the stent, this will change practice toward keeping our patients who can tolerate DAPT on it for a longer period of time.”

Until a formula is devised, a physician’s knowledge of a specific patient will be paramount, said Mehran, associate medical editor of Cardiology Today’s Intervention. “For example, one could think about a patient who has multiple stents, diffuse disease and peripheral vascular disease as one who might receive an important benefit from prolonged DAPT, but some of those same patients are also at high risk for bleeding complications. It is an intricate balance.”

Next Wave of Research

The next wave of research will attempt to better identify which patients benefit most from short-term DAPT and which benefit most from long-term DAPT. Researchers for one analysis presented at the 2015 American College of Cardiology Scientific Sessions in March and published in the Journal of the American College of Cardiology reported that results of the DAPT trial were consistent regardless of whether the patient was stented because of MI.

There will be more research into whether stent type matters, experts said.

According to Stone, newer-generation DES are associated with lower rates of stent thrombosis than first-generation DES, and fluoropolymer-coated EES appear at least just as safe as BMS in that regard. “When you look at the ischemic benefit of prolonged DAPT, it was clearly less with EES than the other types of stents,” he said. “There was no significant reduction in MACE with EES compared with the other stent types, and while there was a reduction in stent thrombosis with prolonged DAPT with EES, the absolute reduction was only 0.4% over 1.5 years of extra therapy. With contemporary DES, I do not believe prolonged DAPT should be the standard of care.”

Kereiakes said a subgroup analysis of participants in the DAPT trial who had an EES is being prepared. “The absolute numbers may be less for the EES as far as stent thrombosis, but the appropriate thing to compare in our trial was not the absolute figures, because the stent was not randomly assigned,” he said. “In addition, the individual stent treatment groups are underpowered and the demographics of the individual stent treatment groups vary. To compare the magnitude of the treatment effect is more appropriate.” The DAPT study EES subgroup included nearly 5,000 randomized patients, more than the combined total EES-treated group in other studies of DAPT duration, Mauri said.

Another analysis being prepared concerns the discontinuation issue, Kereiakes said.

“We are trying to determine not only who should stay on DAPT for 30 months or longer, but who can be safely discontinued,” he said. “Is there a simple profile that could tell the clinician that a certain patient is at higher risk for stent thrombosis or MI if they discontinue their thienopyridine vs. at relatively low risk? We are aggressively pursuing a discontinuation analysis to try to provide guidance to clinicians based on real data.”

Mauri said she would like to see more research on how to reduce MI risk in patients who have been stented, but also have other plaques. “We can revisit how we reduce risk in patients who received coronary stents to prevent them from having progression outside the treated area, which is what our patients think about when they come back for follow-up after a year,” she said. “It’s not so much about the stent, it’s about how they can avoid having another heart attack in the future.”

More to Be Learned

The knowledge base on DAPT duration has increased drastically since 2011, but so much more is still to be learned. In the meantime, clinicians will have to rely on their own judgement, as the guidelines have not yet been updated, and there is little consensus on what any updated guidelines might say.

“The guidelines should and will incorporate the findings from the new research, but we may see differences between U.S. and European guidelines,” Montalescot said.

At the moment, “the big answer is that one size doesn’t fit all,” Mehran said. “We cannot make recommendations across the board for all patients to be prescribed a uniform optimal duration of DAPT. In the future, first and foremost, we have to understand the mandatory duration of DAPT to protect patients against stent thrombosis. Beyond that, we need to understand the patients who will benefit from prolonged DAPT for protection against ischemic complications with the least hazard of bleeding.” – by Erik Swain