Predictive vs Descriptive Biomarkers

Descriptive Biomarkers

Descriptive biomarkers reflect current biological state. They help characterize ageing processes but do not necessarily forecast outcomes for an individual, and are often closer to prognostic markers than predictive tools in the clinical sense. [1] [2]

Predictive Biomarkers

Predictive biomarkers estimate future risk, such as likelihood of disease, disability, or mortality. Their value depends on accuracy, calibration, and actionable interpretation, particularly when used to estimate individual-level outcomes. [1]

Why the Distinction Matters

A biomarker can be strongly descriptive without being predictive. For example, a marker may track age without predicting who will develop disease, which is why longitudinal validation against outcomes is essential. [1] [4]

Use in Research and Practice

Descriptive markers are often used to understand mechanisms, while predictive markers guide clinical decisions. Many ageing biomarkers are still primarily descriptive, with stronger evidence for population-level associations than for individual prediction. [3] [5]

Summary

Descriptive biomarkers characterize ageing, while predictive biomarkers estimate future outcomes. Both are valuable, but they serve different purposes and require different evidence. [3] [4]

References

1. Sechidis, K., Papangelou, K., Metcalfe, P. D., Svensson, D., Weatherall, J., & Brown, G. (2018). Distinguishing prognostic and predictive biomarkers: an information theoretic approach. Bioinformatics, 34(19), 3365-3376. https://academic.oup.com/bioinformatics/article/34/19/3365/4991984
2. Oldenhuis, C. N. A. M., Oosting, S. F., Gietema, J. A., & de Vries, E. G. E. (2008). Prognostic versus predictive value of biomarkers in oncology. European Journal of Cancer, 44(7), 946-953. https://www.sciencedirect.com/science/article/abs/pii/S0959804908002335
3. Zhavoronkov, A., Barzilai, N., Kaeberlein, M., Snyder, M. P., Sebastiano, V., Gladyshev, V. N., et al. (2023). Biomarkers of aging for the identification and evaluation of longevity interventions. Cell, 186(18), 3758-3775. https://pmc.ncbi.nlm.nih.gov/articles/PMC11088934/
4. Belsky, D. W., Kritchevsky, S. B., Cuervo, A. M., Niedernhofer, L. J., Gladyshev, V. N., et al. (2024). An expert consensus statement on biomarkers of aging for use in human intervention studies. Journal of Gerontology: Biological Sciences, 80(5), glae297. https://academic.oup.com/biomedgerontology/article/80/5/glae297/7930267
5. Jylhava, J., Pedersen, N. L., & Hagg, S. (2021). Ranking biomarkers of aging by citation profiling and associative statistics. Frontiers in Genetics, 12, 686320. https://pmc.ncbi.nlm.nih.gov/articles/PMC8176216/