Genetically Determined Telomere Length and Risk for Haematologic Diseases: Results from Large Prospective Cohorts and Mendelian Randomization Analysis

Abstract The causal direction and extent of the link between telomere length, epigenetic age acceleration (EAA), and the occurrence of haematological malignancies and benign haematological disorders remain uncertain because of the inherent susceptibility of observational studies to confounding and reverse causation. We conducted two-sample single-variable Mendelian randomisation (SVMR) and multivariable Mendelian randomisation (MVMR) analyses using summary statistics from genome-wide association studies (GWAS) to explore potential associations among telomere length, EAA, and multiple haematologic diseases. We employed an independent validation dataset and utilised various Mendelian randomisation (MR) methods with distinct model assumptions to verify the validity of our findings. Additionally, we performed MVMR analysis based on Bayesian model averaging (MVMR-BMA) to determine whether telomere length, in isolation from EAA, was the true causal factor in the development of haematologic diseases. We obtained 59 GWASs on haematologic diseases from FinnGen, with 182–27,371 cases and 88,536– 376,651 controls. Increased telomere length due to germline genetic variation was generally associated with an increased risk for 10 of 21 haematological malignancies. Genetically predicted telomere length and EAA were not directly associated with the risk of nearly all benign haematological disorders. MVMR-BMA analysis showed that telomere length had the strongest association with the risk of haematologic diseases compared with the five EAA.Our study outcomes suggest the plausibility of potential clinical applications of telomere length, such as serving as a valuable risk prediction tool or as a targeted intervention for the prevention of haematologic diseases.