Causal Relationship Between Kidney Function and Cancer Risk: A Mendelian Randomization Study

Abstract

Rationale & ObjectivePatients treated with kidney replacement therapy experience a 1.5 to 2-fold increased risk of cancer and cancer mortality compared to the general population. Whether this excess risk extends to people with earlier stages of chronic kidney disease is unclear. This study explored the potentially causal relationship between reduced kidney function and cancer. Study DesignTwo-sample Mendelian randomization (MR). Setting & ParticipantsGenome Wide Association Study (GWAS) summary statistics for estimated glomerular filtration rate (eGFR) (n=567,460) and urinary albuminuria (UACR) (n=127,865) from the CKDGen consortium and cancer outcomes from the UK Biobank (n=407,329). ExposureseGFR and UACR. OutcomesOverall cancer incidence, cancer-related mortality, and site-specific colorectal, lung, and urinary tract cancer incidence. Analytical ApproachUnivariable and multivariable MR were conducted for all outcomes using inverse-variance weighted methods. ResultsThe mean eGFR and median UACR were 91.4 mL/min/1.73m2 and 9.32 mg/g in the CKDGen and 90.4 mL/min/1.73m2 and 9.29 mg/g in the UK Biobank. There were 98,093 cases of cancer, 6,664 colorectal, 3584 lung, and 3,271 urinary tract. There were 15,850 cases of cancer-related death, The genetic instruments for eGFR and UACR comprised 34 and 38 variants, respectively. Genetically predicted kidney function (eGFR and UACR) was not associated with overall cancer risk or cancer death. No associations of genetically predicted eGFR and UACR with overall cancer incidence were observed; odds ratio (95%CI; p-value) of 0.88 (0.40-1.97; p=0.76) and 0.90 (0.78-1.04; p=0.16), respectively. An adjusted generalized additive model for eGFR and cancer demonstrated evidence of non-linearity. There was no evidence of a causal association between eGFR and cancer in a stratified MR. LimitationsTo avoid overlapping samples a smaller GWAS for UACR was used, reducing the strength of the instrument and potentially introducing population stratification. ConclusionsThese findings did not demonstrate a causal association of kidney function with overall cancer incidence or cancer-related death.