DAPT Analysis Eases Concerns Over Slightly Higher Death Rate Seen With Extended Thienopyridine Use
LONDON, England— Analysis of data from the DAPT Study shows that the 0.5% higher rate of all-cause death among patients continuing thienopyridines beyond 12 months is largely attributable to cancer, much of it preexisting, rather than to bleeding. The findings were presented August 31, 2015, at the European Society of Cardiology Congress.
Published late last year, the DAPT Study was designed in response to a 2006 request by the FDA. It enrolled 25,682 patients within 72 hours of receiving an FDA-approved stent (89% DES) and randomized 9,961 event-free DES patients at 12 months to continue receiving aspirin plus a thienopyridine or switch to a placebo. By 30-month follow-up, patients on extended DAPT were more likely to experience moderate or severe bleeding but had fewer MACCE due to a reduction in MI.
“However, a focus of concern,” said Dr. Mauri, “was that there was a 0.5% mortality difference, higher in the thienopyridine arm.” Specifically, the all-cause death rate was 2.0% in patients who continued on DAPT and 1.5% in those receiving aspirin alone (P = .05).
This disparity, she noted, is what prompted the current effort to better understand the causes of death in DAPT, especially given that continued thienopyridine use did not increase the risk of fatal bleeding in the trial. “We wanted to look more closely to see if any prior bleeding could be related to subsequent mortality and [have contributed to] these results,” Dr. Mauri said.
For the current analysis, Dr. Mauri and colleagues adjudicated and analyzed deaths after randomization for all subjects, both BMS- and DES-treated (n = 11,648). A blinded, independent Clinical Events Committee (CEC) reviewed all possible primary endpoint events, classifying deaths as cardiac, vascular, or noncardiovascular, or due to fatal bleeding, according to Bleeding Academic Research Consortium criteria. A second CEC then reviewed all postrandomization deaths to tease out any relationships with prior clinically evident bleeding and/or trauma, or with malignancy or complications from related treatments.
Full follow-up data were available on 94% of the original cohort.
Bleeding Not the Explanation for Difference
Analysis of deaths between 12 and 30 months showed a nonsignificant increase in mortality of 0.4% with continued DAPT vs aspirin alone (P = .07). Extending that window to 33 months (all patients switched to aspirin alone at 30 months) did not change the absolute difference in mortality but, with the addition of more events, a trend emerged (P = .05).
During the randomized-treatment period (12-30 months), noncardiovascular deaths were more common with continued DAPT than with aspirin (0.92% vs 0.49%; P = .01) but cardiovascular death rates were similar. During the aspirin-only period (30-33 months), that increase in noncardiovascular deaths disappeared (0.11% vs 0.16%; P = .58). Combining both time frames resulted in all-cause death rates of 2.22% for continued DAPT and 1.76% for aspirin alone (P = .05).
Adjudicated by cause, there were no differences between study arms in deaths (12-33 months) attributed to bleeding with or without cancer and/or trauma. Nor was there a disparity with trauma-related death without bleeding.
Imbalance in Cancer at the Time of Enrollment
History of cancer at enrollment and cancer incidence were both numerically higher in the continued DAPT group (9.1% vs 9.4% and 2.0% vs 1.6%, respectively). That group also had a higher rate of all cancer-related death (0.6% vs 0.3%; P = .02). Most of these deaths occurred in the absence of bleeding.
When the CEC looked at the time of diagnosis, it “found that there was an imbalance in the numbers of patients who were enrolled with cancer already known at the time of enrollment who subsequently died,” Dr. Mauri said, noting that excluding these patients in a sensitivity analysis resulted in a lack of difference in cancer-related death between trial arms.
Importantly, she added, cancers “occurred in disparate locations” between the continued DAPT vs aspirin-alone groups: lung (10 vs 9), prostate (5 vs 0), pancreas (4 vs 0), hematologic (2 vs 1), and other (9 vs 6; P = NS for all). “The top [locations] are the ones where you would most expect to see cancer in patients of this age category—solid tumors—with no difference in hematologic malignancies,” Dr. Mauri explained. “It’s also worth noting that among those patients who died, in many cases the cancer was already metastatic at the time of diagnosis.”
The Play of Chance?
Among the study’s limitations is that low event rates may translate into lack of statistical power to detect mortality differences, she noted. In addition, patients with serious bleeding events or at high risk were not eligible for randomization.
Supporting the current findings, Dr. Mauri said, is a recent meta-analysis showing that “mortality and cancer-related death have not been increased across prior large randomized trials of thienopyridine therapy with extended follow-up.”
“Whether the cancer-related deaths that were observed are truly related to the treatment—particularly because they weren’t mediated by bleeding—is uncertain, and it’s possible that this could be a chance finding,” Dr. Mauri said. Nonetheless, she emphasized, “in patients with limited life expectancy due to cancer, we should be cautious and consider the risks of antiplatelet therapy very carefully.”
The overall message, she concluded, is that continued DAPT should still be considered for MI prevention in patients who have taken 12 months of therapy after stenting.
Asked about potential mechanisms for promoting cancer, Dr. Mauri replied that it is difficult to come up with a good answer when looking at rare events. Data in the literature are inconsistent, showing both that antiplatelet agents could be protective and that they could be “detrimental to cancer. I don’t think there’s convincing evidence on either side,” she commented.
One clue that the play of chance could be responsible, Dr. Mauri suggested, is that “cancers were not all of one type and don’t follow one cell line. Many of them were already quite advanced at the time of diagnosis.”
Mauri L. DAPT Study: causes of late mortality with dual antiplatelet therapy after coronary stents. Presented at: European Society of Cardiology Congress; August 31, 2015; London, England.
- DAPT Study Shows Benefit of Long-Term Treatment After DES Implantation
- DAPT Substudy: No Apparent Benefit of Long-term DAPT in BMS Patients
- Despite New Findings, Optimal DAPT Duration Still Up in the Air
- The DAPT Study was supported by Abbott, Boston Scientific, Bristol Myers Squibb-Sanofi, Cordis, Daiichi Sankyo, the Department of Health and Human Services, Eli Lilly, and Medtronic.
- Dr. Mauri reports relationships with multiple pharmaceutical and device companies.