Prolonged DAPT Does Not Increase Mortality, Cancer Risk: More Evidence

A new meta-analysis provides reassurance that long-term antiplatelet therapy does not increase the risk of dying or developing cancer, but reiterates that it is associated with reductions in MI and stroke at the expense of increased bleeding.

In an interview with TCTMD, the study’s lead author, Sammy Elmariah, MD, MPH (Massachusetts General Hospital, Boston), said the meta-analysis shows “there is no association whatsoever with prolonged dual antiplatelet therapy and mortality of any kind or with cancer events, specifically cancer-related mortality or new cancers.”

Concerns were raised after an association between long-term clopidogrel therapy and increased rates of mortality was reported among DES patients in the DAPT Study and among patients with prior lacunar infarct in the Secondary Prevention of Small Subcortical Strokes (SPS3) trial.

In a 2015 safety announcement, however, the US Food and Drug Administration (FDA) said that after evaluation of the DAPT Study and several others it had determined that prolonged clopidogrel “does not increase or decrease overall risk of death in patients with, or at risk for, heart disease.” The cancer concern also arose from an analysis of the DAPT Study that showed that both history of cancer at enrollment and cancer incidence during the trial itself were numerically higher in the continued DAPT group. But the conclusion by the DAPT investigators was that this was possibly a “play of chance,” because the varied types and locations of cancer could not be directly attributed to clopidogrel therapy.

“I think it is likely that it was play of chance in the DAPT trial,” Elmariah noted. “We had a much larger population [in this meta-analysis] than in the DAPT trial, and we see no signal whatsoever in multiple subgroup analyses that would raise concerns that continued dual antiplatelet therapy would be associated with either a mortality risk or a cancer risk.”

In the paper, published online January 8, 2017, in Circulation: Cardiovascular Interventions, Elmariah and colleagues say these findings emphasize the need to select extended antiplatelet therapy “only in those patients where the risks of ischemia are not expected to be fully counterbalanced by the risks of bleeding.”

Less MI, Stroke in Exchange for More Bleeding

The new patient-level meta-analysis included DAPT and SPS3 plus four other trials comparing prolonged versus short-duration clopidogrel plus aspirin or aspirin alone in a combined total of 48,817 randomized patients with cardiovascular and cerebrovascular disease.

When clinical event rates from the trials were pooled, there were no differences between the long-term clopidogrel patients and the aspirin-only patients for the outcomes of all-cause mortality, CV mortality, MI-related mortality, stroke-related mortality, non-CV mortality, or cancer-related mortality. However, the MI rate was 4.05% in the aspirin-only group compared with 3.21% in the long-term clopidogrel group (P < 0.0001). The rate of all stroke was also higher at 3.75% versus 3.04% (P < 0.001), as was ischemic stroke at 3.20% versus 2.47% (P < 0.0001).

Sex, age, prior MI, prior PCI, and prior CABG did not affect the impact of clopidogrel therapy on all-cause, cardiovascular, noncardiovascular, or cancer-related mortality endpoints. Subgroup analyses also determined that the protective effect of long-term clopidogrel was more pronounced in patients younger than age 65 than in those older than age 65, as well as in those who had prior PCI than those who did not.

Both fatal bleeding and major nonfatal bleeding were increased in the prolonged clopidogrel group compared with in patients on aspirin alone, and these differences were consistent across subgroups regardless of age.

Rates of new cancers were similar between the groups, as were cancer events related to both solid and hematologic malignancies.

Consistent With FDA Conclusions

Elmariah and colleagues say the findings are consistent with those of the meta-analysis by the FDA, and with a prior trial-level meta-analysis consisting of 14 randomized trials and data on over 69,000 patients. Adding patient-level data, they say, provides a more detailed picture of cause-specific mortality and the impact of clopidogrel in patient subgroups and clarity on cancer development.

David Kandzari, MD (Piedmont Heart Institute, Atlanta, GA), who was not involved with the latest meta-analysis, told TCTMD that while the findings are reassuring, in that extended clopidogrel therapy was not linked to more all-cause or cardiovascular mortality or to more newly diagnosed cancer or cancer-related death, they are clinically disappointing in the sense that extended DAPT was not associated with improved cardiovascular survival.

He also noted that the reduction in MI and stroke, offset by the increased risk of bleeding, may be essentially a “one-to-one trade-off” at the cost of not impacting survival at all.

To TCTMD, Elmariah said while this is a valid argument, it also should be noted that tremendous strides have been made in reducing cardiovascular mortality, “so it is quite difficult to demonstrate additional reductions. The bulk of deaths that are occurring in these trials are noncardiovascular, so it’s often difficult to tease out small impacts on cardiovascular events.” 

Both the researchers and Kandzari said more work is needed to find a way to differentiate those patients who might derive benefit in terms of avoiding a future MI or potential stroke from those who are at high risk of bleeding.

Part of a more nuanced approach should include use of the DAPT Score, Elmariah suggested, since it can help guide decisions regarding short or prolonged antiplatelet therapy.

“The key problem is not identifying those at risk for bleeding. It’s the need to balance a specific patient’s risk for bleeding with their specific risk for ischemic events,” Elmariah said. “That’s exactly what the DAPT Score aims to do.”

Another issue, Kandzari pointed out, is that the meta-analysis includes patients who were enrolled in trials and may not necessarily reflect real-world patient experience.

“These differences may be even more muted, or they may favor even higher risks of bleeding in a broad, unselected patient population,” he said. “This [analysis] is a good effort forward, but with the caveat that it may not be representative of the entirety of patients observed in clinical practice.”