Death Card

The aim of most biomarker studies is to identify susceptibility to a particular disease or responsiveness to a given therapy. But what if biomarkers for all-cause mortality could be found? How would that change the game? Researchers from Estonia and Finland decided to put that theory to the test by analyzing over 17,000 samples for 106 biomarkers using high-throughput molecular profiling by nuclear magnetic resonance (NMR) spectroscopy (1).

Candidate biomarkers included 85 lipoprotein lipid measures, four proteins, and 17 metabolites (amino acids, glycolysis precursors, and other small molecules) all of which were measured in plasma from the Estonian Biobank cohort and then replicated and validated with serum from the FINRISK study.

After some pretty intense statistical analysis (see Figure 1), which compared biomarker data with the number of individuals who had died five years after the sample was taken, the group arrived at a surprising conclusion. They seemed to have discovered four biomarkers predictive of all-cause mortality: albumin, alpha-1-acid glycoprotein, citrate, and the size of very-low-density lipoprotein (VLDL) particles.

“We were very surprised. Although we were looking for such biomarkers we did not expect to find such strong predictors; by combining the four biomarkers into a single risk marker we produce a stronger predictor of short-term risk of death than if someone has had cancer,” says co-author Johannes Kettunen of the Institute for Molecular Medicine in Finland (IMMF). Kettunen elaborated further in an IMMF news clipping: “What is especially interesting is that these biomarkers reflect the risk for dying from very different types of diseases, such as heart disease or cancer. They seem to be signs of a general frailty in the body. Next, we aim to study whether some kind of connecting factor between these biomarkers can be identified.”

Low circulating albumin levels have already been associated with increased mortality from a number of causes. A high level of alpha-1-acid glycoprotein has been linked with mortality in the elderly and in cardiovascular mortality and cancer prognosis. VLDL particle size was inversely associated with risk of death in this study – and that association became stronger with alpha-1-acid glycoprotein’s inclusion in the multivariate model. High citrate levels constitute the fourth factor. “The mechanisms underlying how citrate is associated with short-term risk of death among ambulatory people nonetheless remain elusive,” according to the authors of the paper.

The authors admit that molecular coverage with NMR spectroscopy is somewhat limited compared with mass spectrometry, which could potentially extend the ability of such studies to offer insight into risk assessment and disease pathway elucidation. But Kettunen stands by their chosen methodology: “The key element of this kind of research is the rigorous demand for statistical significance and replication in an independent study setting. Mass spectrometry would be a good complementary method and has great potential to reveal new information; however, it does not yet meet the other two key criteria for large-scale epidemiological studies: cost and high-throughput.”

All samples were taken from non-fasting individuals – comparison with fasting biomarker concentrations would make for an interesting future study.

“We must now understand the mechanisms underlying these results to clarify the utility of these circulating biomarkers in the guidance of disease screening and targeted prevention based on molecular profiles,” Kettunen concludes.