This TSMR analysis examines effect of 5 aging indicators on 24 gastrointestinal diseases. The findings suggest that longer TL may reduce the risk of Alcoholic liver disease and Cirrhosis, but may increase the risk of Cholelithiasis, Cholecystitis, and Pancreatic Cancer. Additionally, genetically predicted HorvathAge acceleration appears to decrease the risk of Chronic gastritis and Gastric cancer. The reverse MR analyses we conducted were designed to eliminate the possibility of reverse causation in the above results. Accidentally, we found that the occurrence of celiac disease and ulcerative colitis may reduce TL, the occurrence of celiac disease may accelerate GrimAge and PhenoAge, and the occurrence of cholangitis may accelerate GrimAge and HannumAge.
A published MR study using UKB data found a significant association between longer telomere length (TL) and a reduced risk of alcoholic liver disease and cirrhosis (Beta: 0.710; CI: 0.570–0.890)13. Our results, using the FinnGen database, are consistent with this finding but avoid bias due to sample overlap. The possible explanation for inducing hepatocyte senescence and cirrhosis is the absence of telomerase activity and telomere shortening due to chronic liver injury, ongoing cell division, and liver regeneration19. Longer telomeres provide a replicative advantage to cells and reduce the risk of pathogenesis compared to shorter telomeres. However, our results differ from those of Daniele et al. (2019), who conducted a study on PANDoRA and found an association between short telomere length and pancreatic ductal adenocarcinoma20. The association between TL and pancreatic cancer is controversial, with one study finding that telomere lengthening was associated with an increased risk of pancreatic cancer, two studies finding that both telomere lengthening and telomere shortening were associated with an increased risk of pancreatic cancer, and two studies using genetic risk scores reporting no evidence of an association with pancreatic cancer risk8,21–25. Observational studies have not yet explored the association between longer TL and Cholelithiasis and Cholecystitis.
This study covered the four most common epigenetic clocks, and the results confirmed that the genetically predicted HorvathAge clock is associated with a reduced risk of chronic gastritis and gastric cancer. However, no study has yet explored the association between epigenetic clock acceleration and these two diseases. The HorvathAge clock is the most widely used epigenetic clock. It is built from 353 age-related CpGs that are automatically selected by a mathematical algorithm. The correlation between HorvathAge and chronological age can be as high as 0.96. This clock is supported by data from various areas of research, including cancer, Alzheimer's disease, aging, and lifestyle7.
In the reverse MR, we relaxed the instrumental variable criteria but did not reduce the instrumental variable strength because there were insufficient instrumental variables for 24 gastrointestinal diseases. We did not find reverse causality for the above associations, but we also found that the development of UC may shorten TL, an RCT by Sho demonstrated the mechanism of shortening TL in the pathogenesis of UC, and our study suggests that there may be a positive feedback mechanism that contributes to shortening TL during the development of UC26. It was found that Celiac disease causes TL shortening and acceleration of GrimAge and PhenoAge, and the development of cholangitis may cause acceleration of GrimAge and HannumAge, and a study showing a strong independent correlation between epigenetic senescence and inflammation may explain part of the reason, but more studies are needed to reveal the cause27.
Epigenetic clocks may have greater validity in assessing ageing. However, they may have less specificity compared to telomere length. This study shows that there is a discrepancy between the results of the EC and the TL as indicators of ageing. Two possible explanations for this inconsistency exist. Firstly, there is a significant difference in the correlation of EC and TL with age. The Berlin Aging Study demonstrated a strong positive correlation between DNA methylation age and chronological age, as well as a weak negative correlation between DNA methylation age (EC) and leukocyte telomere length28. In addition, the EC has emerged as a new biomarker of ageing, which is often adjusted for the different factors that influence the life span. For instance, HorvathAge is a multi-tissue and cell type EC constructed by assessing DNA methylation levels in 51 different tissues and cell types29. GrimAge, on the other hand, consists of 7 plasma protein markers based on DNA methylation and smoking pack-years30.
his study has strengths in that MR studies are characterized by a reduced susceptibility to confounders and reverse causality, thereby enhancing the reliability of the results. Additionally, our study used exposures and outcomes from different populations of the same ethnicity, effectively avoiding sample overlap. However, a notable limitation of this study lies in the limited sample size of EC, resulting in a limited number of suitable instrumental variables. In terms of reverse causality, we adjusted the p-value threshold for the instrumental variable related to the epigenetic clock to 5 × 10 − 6, aiming to increase the pool of instrumental variables, as the initial number was insufficient for the respective diseases. Nonetheless, we employed the F statistic to filter out weak instrumental variables, ensuring the fulfillment of the correlation assumption. Hence, it is believed that this adjustment is unlikely to substantially affect the pertinence of the instrumental variables and the dependability of the results in this study. On the other hand, in our MR analyses of epigenetic clocks in relation to gastrointestinal disorders, power was insufficient, especially in the EC analyses. Therefore, our findings may not reflect a true effect.