Telomere length and chronological age across the human lifespan: A systematic review and meta-analysis of 414 study samples including 743,019 individuals

Abstract

Telomere attrition is a proposed hallmark of aging. To evaluate the association of telomere length (TL) with chronological age across the human lifespan, we conducted a systematic review and meta-analysis of 414 study samples comprising 743,019 individuals aged 0-112 years. We examined both cross-sectional and longitudinal data, and evaluated the impact of various biological and methodological factors including sex, health status, tissue types, DNA extraction procedures, and TL measurement methods. The pooled corrected correlation between TL and age from cross-sectional samples was -0.19 (95%CI: -0.22 to -0.15), which weakened with increased chronological age (β = 0.003, p < 0.001). Z-score change rates of TL across the lifespan showed a gradual decrease in shortening rate until around age 50 and remained at a relatively stable rate towards the elderly period. A greater attrition rate was observed in longitudinal than cross-sectional evaluations. For TL measured in base pairs, the median change rate of TL was -23 bp/year in cross-sectional samples and -38 bp/year in longitudinal samples. Methodological factors including TL measurement methods and tissue types impacted the TL-age correlation, while sex or disease status did not. This meta-analysis revealed the non-linear shortening trend of TL across the human lifespan and provides a reference value for future studies. Results also highlight the importance of methodological considerations when using TL as an aging biomarker.

Keywords: Human; Meta-analysis; Telomere length.

Figure 1.. Flowchart of the paper screening process.

The paper using the UK biobank cohort (Codd et al., 2022) was included in a later process.

Figure 2.. Caterpillar plot of corrected correlations from cross-sectional samples.

The black dots and the gray horizontal lines are effect sizes and 95% confidence intervals from individual samples. The black diamond is the pooled correlation. The two influential cases were excluded from this plot. Correlations were corrected for the age range restriction to reduce the noise caused by the heterogeneity of age range across different samples.

Figure 3.. Change trend of corrected correlations across age fitted by a time-varying effect model in cross-sectional samples.

The solid black line is the point estimate. The dashed lines are the boundaries of the 95% confidence interval. The two influential cases were excluded from this plot. Correlations were corrected for the age range restriction to reduce the noise caused by the heterogeneity of age range across different samples.

Figure 4.. Forest plot of pooled corrected correlations stratified by demographic, disease, and methodological factors.

CVD, cardiovascular disease; PTSD, post-traumatic stress disorder; qPCR, quantitative polymerase chain reaction; qFISH, quantitative fluorescence in situ hybridization; PBMC, peripheral blood mononuclear cells. The two influential cases were excluded from this plot. Correlations were corrected for the age range restriction to reduce the noise caused by the heterogeneity of age range across different samples.

Figure 5.. Absolute telomere length change trend across the lifespan in longitudinal studies.

Age and TL center values (mean, median, or midrange) at baseline and follow-up from longitudinal studies were connected for each study in the line plot to show the change trend of TL as a function of age.