Epidemiological research has revealed a galaxy of biomarkers, such as genes, molecules or traits, which are associated with increased risk of atherosclerotic cardiovascular diseases (ASCVD). However, the etiological basis remains poorly characterized. Mendelian randomization (MR) involves the use of observational genetic data to ascertain the roles of disease-associated risk factors and, in particular, differentiate those reflecting the presence or severity of a disease from those contributing causally to a disease. Over the past decade, MR has evolved into a fruitful approach to clarifying the causal relation of a biomarker with ASCVD and to verifying potential therapeutic targets for ASCVD. In this review, we selected high-quality MR studies on ASCVD, examined the causal relationship of a series of biomarkers with ASCVD, and elucidated the role of MR in validating biomarkers as a therapeutic target by comparing the results from MR studies and randomized clinical trials (RCTs) for the treatment of ASCVD. The good agreement between the results derived by MR and RCTs suggests that MR could be performed as a screening process before novel drug development. However, when designing and interpreting a MR study, the assumptions and limitations inherent in this approach should be taken into account. Novel methodological developments, such as sensitivity analysis, will help to strengthen the validity of MR studies.