Aging is a complex process characterized by various phenotypic changes, including epigenetic modifications, inflammation, and metabolic dysregulation. Recent advancements in the field of epigenetics have led to the development of DNA methylation (DNAm) clocks, which estimate biological age and predict age-associated outcomes based on DNAm profiles.
In the study titled ‘Ensemble DNA methylation clock demonstrates Immune-metabolic aging signatures associated with mortality,’ researchers aimed to address the limitations of existing DNAm clocks that primarily rely on linear models. These linear models may not adequately capture the heterogeneity and non-linearity of survival-related DNAm signatures.
The authors constructed a heterogeneous stacked ensemble survival model using DNAm data from the Framingham Heart Study. Initially, they identified 190 CpG loci through elastic net Cox regression, which were then used to build a survival prediction model. This model was created by fusing five complementary survival models with the aid of a neural network meta-learner.
To evaluate the prediction power of their survival model, the researchers conducted external validation. They found that their model exhibited strong performance in predicting all-cause mortality, significantly outperforming the PhenoAge model and showing statistical comparability to GrimAge.
It is important to note that the performance estimates derived from this study were based on cohorts of European ancestry and were validated in postmenopausal women aged 50-79 years. Therefore, the applicability of these findings may be limited to demographically similar populations.
What this study does not show
This study does not demonstrate clinical utility in predicting health outcomes beyond mortality, nor does it validate the model in diverse populations.
Evidence scorecard
- Analytic validity: Established
- Clinical validity: Partial
- Clinical utility: Not established
The evidence class for this study is primary_research, and the human status is unclassified.