460P - Mendelian randomization study showed no causality between metformin use and lung cancer risk

Background

Reduced lung cancer incidence was observed after the use of metformin. However, previous reports from observational study can be biased by confounders, inverse causality and so on, keeping the causality between observed decreased lung cancer risk and metformin use uncertain. Mendelian randomization (MR) analysis is a novel approach for the estimation of causality, eliminating the effect of confounders by using single-nucleotide polymorphisms (SNPs) as instrument variables. Growth Differentiation Factor 15 (GDF 15) is a robust biomarker of metformin use. To verify the potential preventive effect of metformin on lung cancer, we aimed to assess the causality, using SNPs of GDP 15.

Methods

We utilized the genome-wide association study (GWAS) summary data of GDF 15 and lung cancer. Two-sample MR was used to investigate the causality. In detail, Inverse-Variance Weighted, MR-Egger and Weighted Median methods were conducted. Single SNP analysis and leave-one-out analysis were performed to assess whether the result was driven by a single SNP. We also conducted the MR-Egger regression to evaluate the pleiotropy.

Results

The result of this study did not support a causality between metformin use and lung cancer incidence (OR = 0.988, 95% CI = 0.914-1.069, P-value = 0.768), which was not driven by single SNP. (Table) Directional horizontal pleiotropy did not seem to exist in the MR-Egger regression (intercept β = −0.02, P = 0.69).

Table: 460P Mendelian randomization estimates of the causality between metformin utilization and lung CaExpo

Odds ratio.

Confidence interval.

Conclusions

This study indicated no causality between metformin and lung cancer. The potential use of metformin in lung cancer prevention need further verification in randomized controlled trials. Further studies are warranted to investigate the shared and modifiable risk factors of metformin use and lung cancer contributing to the observed reverse association, which will be beneficial to the control of lung cancer incidence.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

National Key R&D Program of China (2016YFC0905500, 2016YFC0905503); Science and Technology Program of Guangdong (2017B020227001); Science and Technology Program of Guangzhou (201704020072).

Disclosure

All authors have declared no conflicts of interest.