433P - Association between aspirin and cancer risk: A Mendelian randomization analysis

Background

Previous researches have reported on the protective effect of aspirin on different types of cancers. However, evidence from observational studies can be biased by multiple confounding factors. Also, it remains unclear whether aspirin use is casually related to a decreased cancer risk. Consequently, we conducted Mendelian randomization analyses, using single nucleotide polymorphisms (SNPs) as instrumental variables to investigate the casual effect of aspirin on different types of cancers.

Methods

Mendelian randomization analyses were conducted using pooled statistics from corresponding researches and several large-scale consortia. Inverse-variance weighted, MR-Egger and weighted median methods were utilized to evaluate the possible causal relationship between aspirin and the observed reduction in cancer risk. Results were shown by odds ratios (OR) and their corresponding 95% CIs. Analyses were conducted using the package “TwoSampleMR” in R.

Results

A total of 19 cancers corresponding to 344,392 cases and 5,424,758 controls were included in the final analysis. In contrast to the lower risk for various cancer types reported in conventional observational studies, our MR results only showed a decreased risk in lung cancer [odds ratio (OR) 0.0418; 95% confidence interval (CI) 0.0031-0.5638; P=0.0168] and squamous cell lung cancer [OR=0.002; 95% CI 1.2145×10^-5-0.3009; P=0.0153]. Insignificant results were observed in other types of cancers including lung adenocarcinoma, breast cancer, ovarian cancer, colon cancer, kidney cancer, liver cancer, esophageal cancer, oral cancer, pancreas cancer, rectal cancer, malignant melanoma, non-melanoma skin cancer, Hodgkin’s lymphoma as well as non-Hodgkin’s lymphoma.

Conclusions

The potential effect of aspirin in cancer prevention still needs to be interpreted with cautions and aspirin is still too early to be recommended as a universal chemoprotective agent for primary cancer prevention. Further randomized controlled trials with a more rigorous design are warranted in the future.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.