Association between Alcohol Misuse and Leukemia: A Mendelian Randomization Study

Background: Leukemia comprises a spectrum of hematologic malignancies characterized by abnormal proliferation of white blood cells. Given the challenging nature of leukemia treatment, preventative measures targeting modifiable risk factors, such as alcohol misuse, are crucial. Although alcohol misuse is a known risk factor for several cancers, its relationship with leukemia is complex. Previous studies have suggested a potential protective effect of alcohol against leukemia, although findings are inconsistent and limited by biases and small sample sizes. This study uses genome-wide association studies (GWAS) to elucidate the causal relationship between alcohol consumption and leukemia.

Methods: This study employed a two-sample Mendelian Randomization (MR) analysis to investigate the causal relationship between alcohol misuse and leukemia risk, using GWAS datasets. SNPs were selected as instrumental variables (IVs) based on relevance, independence, and exclusivity. Alcohol misuse data from the UK Biobank, categorized participants into four drinking levels. Leukemia risk GWAS data from the Figgen database focused on four subtypes of leukemia. Statistical methods included MR-Egger, weighted median, Inverse Variance Weighted (IVW), simple mode, and weighted mode. Heterogeneity and pleiotropy were assessed using Cochran's Q test, funnel plots, MR-Egger intercept test, and MR-PRESSO global test, with a leave-one-out sensitivity analysis for robustness. Analyses were conducted using R (version 4.3.2).

Results: The MR analysis revealed that genetically predicted light drinking was significantly inversely associated with Acute Myeloid Leukemia (AML), Chronic Myelogenous Leukemia (CML), Acute Lymphoblastic Leukemia (ALL), and Chronic Lymphocytic Leukemia (CLL), suggesting a protective effect. Specifically, light drinking showed strong inverse associations with AML (OR = 0.06, 95% CI [0.01, 0.26], P < 0.001), CML (OR = 0.028, 95% CI [0.01, 0.22], P < 0.001), ALL (OR = 0.04, 95% CI [0.01, 0.22], P < 0.001), and CLL (OR = 0.23, 95% CI [0.11, 0.50], P < 0.001). However, no significant causal relationships were found between heavy or extensive/high-risk drinking and these leukemia subtypes. Sensitivity analyses, including horizontal pleiotropy tests, heterogeneity assessments, and leave-one-out analysis, confirmed the robustness of these findings. No evidence of pleiotropy was detected, and the instrumental variables demonstrated minimal heterogeneity, ensuring the validity of the causal inferences drawn from this study.

Conclusion: This inaugural MR study reveals that genetically predicted light alcohol misuse is associated with a decreased risk of the four most common types of leukemia, while no significant associations were found with heavy or high-risk alcohol intake. This suggests a J-shaped curve, where low alcohol intake may offer protective effects. Despite significant findings, limitations include the low prevalence of leukemia and challenges in achieving high statistical power. Future research with larger GWAS datasets is needed to validate these results. The study's strengths include large sample sizes and an MR approach, minimizing observational biases. These findings suggest moderate alcohol consumption could aid in leukemia prevention, but balanced messaging on risks is crucial.

No relevant conflicts of interest to declare.