Micro-simulation model to predict effects of low-dose aspirin on colorectal cancer, cardiovascular disease and safety outcomes among low-dose aspirin users

Abstract

Abstract Background Regular use of low-dose aspirin (LDA) has been shown to reduce the risk of cardiovascular disease (CVD). Recent evidence indicates that regular LDA use also reduces the risk of colorectal cancer (CRC). Use of LDA has also been associated with an increased risk of gastrointestinal (GI) bleeding, peptic ulcers and intracranial hemorrhage (ICH). Purpose We aim to evaluate the population-level benefits and risks of daily and/or regular LDA among individuals taking LDA for primary or secondary CVD prevention in the United Kingdom (UK) accounting for the evidence on the reduced CRC risk. Methods Individual-level state transition modelling was used to predict the impact of LDA on CRC, CVD, safety events (GI bleeding, ICH and symptomatic peptic ulcers) and related deaths in a UK population. Individual event histories were simulated for hypothetical cohorts of 1 million adults aged 50–59 years and aged 60–69 years indicated to take LDA for primary or secondary CVD prevention. The QRISK CVD prediction score was used to simulate adults indicated to use LDA for CVD prevention. Model parameters were informed based on published scientific literature on CVD, CRC and safety outcomes mimicking the UK epidemiology and prevention guidelines. Monte Carlo simulation was used to account for parameter uncertainty. Results In the cohort of subjects for which low-dose aspirin use was initiated between 50–59 years, the decrease in incidence rates (IRs) (per 100,000person years) of non-fatal CVD was smaller when low-dose aspirin use was initiated for primary prevention compared to initiation for secondary prevention (−203 [95% UI, −277 to −115] versus −794 [95% UI, −997 to −536]). Similar results were obtained for fatal CVD (−97 [95% UI, −136 to −60] versus −381 [95% UI, −502 to −257]). Whether low-dose aspirin treatment was initiated for primary or secondary CVD prevention did not greatly influence the changes in the IR of the non-CVD outcomes. The changes in IR for fatal CRC were −46 [−69, −31] and −44 [−67, −26] for primary and secondary CVD prevention while the changes in IR for fatal bleeding were 5 [1, 9] and 5 [1, 10]. Similar results were obtained when low-dose aspirin use was initiated between 60–69 years. Conclusions In all simulation settings considered, low-dose aspirin related reductions in non-fatal and fatal CVD outcomes were larger in case of secondary CVD prevention compared to primary CVD prevention. These reductions were larger than the increases in fatal safety events. This favorable benefit-risk profile is more pronounced in the case of secondary CVD prevention. The results from the simulation model can be used to inform discussions with patients about the potential benefits and risk of LDA initiation. Funding Acknowledgement Type of funding source: Private company. Main funding source(s): Bayer AG