Causal association between fibroblast growth factors (FGFs) and gastric cancer: A two-sample Mendelian randomization study.

ObjectiveTo investigate the causal relationships between fibroblast growth factors (FGFs) and gastroesophageal cancers-specifically gastric cancer (GC), cardia cancer (CC), and esophageal cancer (EC)-using a two-sample Mendelian randomization (MR) approach.MethodsBidirectional two-sample MR was conducted using GWAS data of FGFs, FGFRs, and GC, with IVW as the primary method. Univariate MR (UVMR) first identified preliminary associations, followed by multivariable MR (MVMR) to estimate the direct effects of FGFs independent of confounders.ResultsThe IVW results indicated a nominal association between FGF7 and CC (β = -0.5152, P = 0.0402, P-FDR = 0.2008); FGF21 showed a nominal association with both CC (β = -1.0245, P = 0.0175, P-FDR = 0.1750) and GC (β = 1.1361, P = 0.0473, P-FDR = 0.4730). MVMR analysis showed that, after adjusting for cigarettes smoked per day, FGF21 was significantly causally associated with CC (OR = 0.2514); after adjusting for BMI, FGF21 was significantly causally associated with GC (OR = 2.2893). The results of the reverse analysis indicated a nominal association between EC or GC and FGF21 (OR = 0.9739, P = 0.0205, P-FDR = 0.0820. In forward MR analysis, the statistical power for exposure-outcome pairs showing an association exceeded 75%. Sensitivity analyses revealed no evidence of heterogeneity, pleiotropy, or outliers affecting the results.ConclusionsOur findings provide genetic evidence that FGF21 plays a causal role in gastric carcinogenesis, exhibiting a protective effect against CC while increasing the risk of GC independent of key confounders.