Background: Epidemiological and laboratory-based studies have provided conflicting evidence for a role of ghrelin in colorectal cancer (CRC) development. We conducted two-sample Mendelian randomization (MR) analyses to evaluate evidence for an association of circulating ghrelin and CRC risk overall and by sex, cancer subsite and age at diagnosis. Methods: Genetic instruments proxying plasma total ghrelin levels were obtained from a recent genome-wide association study of 54,219 participants. Summary data for CRC risk were obtained from a recent meta-analysis of several genetic consortia (up to 73,673 cases and 86,854 controls). A two-sample MR approach and several sensitivity analyses were applied. Results: We found no evidence for an association of genetically-predicted plasma total ghrelin levels and CRC risk (0.95, 95% confidence interval: 0.81-1.12; R2 of ghrelin genetic instruments: 4.6%), with similarly null results observed when stratified by sex, anatomical subsite, and for early-onset CRC. Conclusions: Our study suggests that plasma ghrelin levels are unlikely to have a causal relationship with overall, early-onset, and sex- and cancer subsite-stratified CRC risk. Impact: This large-scale analysis adds to the growing body of evidence that plasma total ghrelin levels are not associated with CRC risk.