Mendelian randomization study of the relationship between blood and urine biomarkers and lung cancer

Introduction Identifying suitable biomarkers is crucial for exploring the pathogenesis, early screening, and therapeutic monitoring of lung cancer. This study aims to analyze comprehensively the associations between lung cancer and biomarkers in blood and urine. Methods Bidirectional two-sample Mendelian randomization (MR) was used to evaluate the potential causal relationships between blood and urine biomarkers and lung cancer. We obtained Single nucleotide polymorphisms (SNPs) related to lung cancer from the 2021 Finnish database of genome-wide association studies, including small cell lung cancer (SCLC), total non-small cell lung cancer (NSCLC), lung adenocarcinoma (LAC), and lung squamous cell carcinoma (LSCC).Data on blood and urine biomarkers were derived from the UK Biobank cohort, comprising 376,807 participants. Results We found a potential inverse causal relationship between total bilirubin and SCLC (β=-0.285, P=0.015, FDR=0.12). Urate was inversely associated with NSCLC (β=-0.158, P=0.004, FDR=0.036*). Serum calcium showed a possible inverse relationship with lung squamous cell carcinoma (β=-0.256, P=0.046, FDR=0.138), while urinary creatinine was positively associated (β=1.233, P=0.024, FDR=0.216). Non-albumin proteins (β=-0.272, P=0.020, FDR=0.180) and total protein (β=-0.402, P=0.009, FDR=0.072) were inversely related to lung squamous cell carcinoma. The AST/ALT ratio was positively associated with lung adenocarcinoma (β=0.293, P=0.009, FDR=0.072). Our reverse Mendelian randomization study found a positive causal association between small cell lung cancer and serum creatinine (β=0.022, P=0.002, FDR=0.018*), while it was inversely associated with the estimated glomerular filtration rate(eGFR)(β=-0.022, P=0.003, FDR=0.027*). A positive causal relationship was also observed with cystatin C (β=0.026, P=0.005, FDR=0.045*) and glycated hemoglobin HbA1c (β=0.013, P=0.014, FDR=0.028*). A negative causal relationship was observed with Gamma_glutamyltransferase (β=-0.013, P=0.019, FDR=0.152). For non-small cell lung cancer, a negative causal relationship was found with albumin (β=-0.024, P=0.002, FDR=0.016*), while a potentially positive causal relationship was observed with cystatin C (β=0.022, P=0.006, FDR=0.054). Possible negative causal relationships were also observed with phosphate (β=-0.013, P=0.008, FDR=0.072) and urinary potassium (β=-0.011, P=0.012, FDR=0.108), while a potential positive causal relationship was observed with C-reactive protein (β=0.013, P=0.040, FDR=0.280).Regarding lung squamous cell carcinoma, an inverse causal relationship was found with eGFR (β=-0.022, P=9.58e-06, FDR=8.62×10-5*), while a positive causal relationship was observed with serum creatinine (β=0.021, P=1.16e−4, FDR=1.05×10-3*). Potential positive causal relationships were observed with Urate (β=0.012, P=0.020, FDR=0.180), urea (β=0.010, P=0.046, FDR=0.141), and glycated hemoglobin HbA1c (β=0.020, P=0.049, FDR P=0.098), whereas a potential negative causal relationship was observed with sex hormone-binding globulin(SHBG) (β=-0.020, P=0.036, FDR=0.108).Lastly, adenocarcinoma was found to have a positive causal association with alkaline phosphatase (β=0.015, P=0.006, FDR=0.033*). Conclusion Our study provides a robust theoretical basis for the early screening and therapeutic monitoring of lung cancer and contributes to understanding the pathogenesis of the disease.