Contrast of results from ASPREE participants previously taking or not taking regular aspirin prior to trial commencement.

Conference:

SAPC ASM 2019 - Exeter

Talk Code:

1B.3

Presenter:

Mark Nelson

Co-authors:

MR Nelson (1, 2), RL Woods (2), G Polekhina (2), CM Reid (2, 3), AM Murray (4), B Kirpach (5), RC Shah, (6), M Ernst (7), E Richmond (8), JE Lockery (2), R Wolfe (2), AM Tonkin (2), N Stocks (9), JJ McNeil (2) on behalf of the ASPREE Investigator Group.

Author institutions:

1. Menzies Institute for Medical Research, University of Tasmania, Hobart TAS, Australia. 2. School of Public Health and Preventive Medicine, Monash University, Melbourne VIC, Australia. 3. School of Public Health, Curtin University, Perth WA, Australia. 4. Division of Geriatrics, Department of Medicine, Hennepin County Medical Center and University of Minnesota, Minneapolis MN, U.S. 5. Berman Cen

Problem

The ASPREE study, an N of 19,114 randomised controlled trial of low dose aspirin vs. placebo for disability free survival in the elderly, showed no benefit and possible harm. This has been interpreted as only relevant to the initiation of the drug. Clinicians and patients alike need guidance on the safety of stopping aspirin in those who are already taking the drug without overt cardiovascular disease.

Approach

An intention to treat analysis of regular aspirin users prior to trial entry randomised to placebo arm (cessation) or aspirin arm (continuation) for a median of 4.7 years. Cox proportional-hazards models were used to compare results between the aspirin group and the placebo group. Hazard ratios were calculated for death from any cause and for death related to specific causes. Cumulative incidences were used to show the risk of death related to each specific cause, with stratification according to trial group and with allowance for the competing risk of death from the other causes.

Findings

2094 were taking aspirin prior to study entry of whom 1053 were randomised to aspirin and 1041 to placebo. There were 70 deaths in the aspirin arm and 79 in the placebo arm (Hazard Ratio = 0.86, 95% Confidence Interval from 0.62 to 1.19). There were differing trends in cancer deaths (38 vs. 32, HR=1.15 and 95%CI: 0.72, 1.85) compared to both CVD deaths (11 vs 27, HR=0.24 and 95%CI: 0.05, 1.14) and coronary heart disease deaths (5 vs 14, HR=0.35 and 95%CI: 0.12, 0.96). Excess cancer deaths were driven by lung cancer (11 vs. 6). Cumulative incidence plots for cancer mortality suggests that adverse risk is reduced quickly on cessation compared with continued aspirin use. Cumulative incidence for CVD death and events showed a delayed adverse effect for ceasing aspirin. Cumulative incidence for major bleeding showed no difference. Limitation of possible survival bias as 4-week placebo run in phase may have to CVD events not captured in-trial.

Consequences

Ceasing low dose aspirin after the age of 70 in those without a clinical indication for use (i.e. CVD secondary prevention) showed both adverse (CVD) and beneficial (cancer) effects. The findings need to be interpreted in light of the main findings and suggest aspirin cessation may be a reasonable strategy in an age group with a high drug burden.

Submitted by:

Mark Nelson

Funding acknowledgement:

National Institutes for Aging (US), National Cancer Institute (US), National Health and Medical Research Council of Australia.