Plasma GFAP and NfL Predict Long-Term Mortality in Older Adults

The Intersecting Risks of Cognitive Decline and Systemic Mortality

As the aging population grows, clinicians frequently manage patients facing overlapping risks of cognitive decline and cardiovascular mortality. While interventions ranging from intensive blood pressure control targeting a systolic goal of less than 120 millimeters of mercury [1] to targeted pharmacotherapy can influence survival trajectories in older adults, predicting individual long-term outcomes remains challenging. Systemic vascular and neurodegenerative processes often intertwine, complicating risk stratification for both cognitive and cardiovascular endpoints. This remains true even though treatments like cholinesterase inhibitors reduce all-cause mortality in dementia patients (adjusted hazard ratio 0.77, 95 percent confidence interval 0.70 to 0.84) [2] and oral anticoagulants lower mortality (hazard ratio 0.79) in patients with concurrent atrial fibrillation and dementia [3]. Historically, physicians have relied on traditional clinical metrics and broad systemic guidelines for managing dyslipidemia [4] and arterial hypertension [5] to estimate these competing mortality risks. Now, a new study investigates whether emerging blood-based biomarkers designed to track brain health might also serve as predictors of long-term survival, offering a potential tool to refine clinical prognosis and tailor patient care.

Tracking Neurodegeneration in a Biracial Cohort

To determine whether baseline biomarkers of neurodegeneration are associated with long-term mortality risks, researchers analyzed data from the REasons for Geographic and Racial Differences in Stroke cohort study. This national project originally enrolled 30,239 Black and White participants across the continental United States from 2003 to 2007. For this specific analysis, a total of 917 participants had a mean baseline age of 67.4 years (SD 12.1). Within the analyzed sample, 49.4% were female, and 48.6% self-identified as Black, providing a balanced demographic foundation to assess these prognostic tools. The research team measured four specific proteins in baseline plasma from a random sample of participants: neurofilament light chain (NfL), total tau, glial fibrillary acidic protein (GFAP), and ubiquitin carboxy-terminal hydrolase L1 (UCH-L1). These molecules serve as blood-based indicators of structural brain damage. Specifically, NfL reflects axonal injury, GFAP indicates astrocyte activation and neuroinflammation, and UCH-L1 is an enzyme signaling acute neuronal damage. To evaluate the predictive value of these proteins, the investigators utilized cause-specific Cox regression models accounting for competing risks (a statistical method that adjusts for the clinical reality that older patients might die from other conditions, such as heart failure, before developing end-stage dementia). This approach allowed them to calculate hazard ratios (HRs) of outcomes for each biomarker separately. Over a mean follow-up of 11.1 (SD 5.7) years, patient outcomes were tracked rigorously using medical records, the Social Security Death Index, and the National Death Index. During this period, 51.0% (477/935) of participants died, and 9.2% experienced dementia-specific mortality (86/935).

GFAP and NfL Drive All-Cause Mortality Risk

When evaluating the predictive power of these neurodegenerative biomarkers for overall survival, the researchers identified distinct risk profiles. Notably, no associations were observed for total tau, a protein traditionally linked to Alzheimer disease pathology. Instead, the prognostic value for overall survival was driven by markers of astrocyte activation and axonal injury. The analysis revealed that elevated baseline levels of GFAP and NfL strongly predicted reduced survival. In fully adjusted models, the hazard ratio of all-cause mortality per standard deviation increment was 1.93 (95% CI 1.48-2.52) for GFAP. The risk profile for axonal damage was nearly identical. In fully adjusted models, the hazard ratio of all-cause mortality per standard deviation increment was 1.90 (95% CI 1.55-2.32) for NfL. This indicates that for every standard deviation increase in the plasma concentration of these specific proteins, a patient's risk of death nearly doubled over the follow-up period. The researchers also evaluated UCH-L1. While the effect size was smaller compared to GFAP and NfL, it remained a significant predictor of mortality. In fully adjusted models, the hazard ratio of all-cause mortality per standard deviation increment was 1.23 (95% CI 1.09-1.37) for UCH-L1. For clinicians, these data suggest that blood tests originally designed to detect brain injury or neurodegeneration also capture systemic vulnerabilities that drive overall mortality risk in older adults, potentially serving as early warning signals for general physiological decline.

Dementia and Cardiovascular Mortality Link

Beyond all-cause mortality, the researchers evaluated how these biomarkers predicted specific causes of death. The data demonstrated a profound link to fatal cognitive decline. In fully adjusted models, GFAP was associated with dementia-specific mortality, yielding a hazard ratio of 5.66 (95% CI 2.91-11.00). This indicates a more than fivefold increase in the risk of death directly attributed to dementia for patients with elevated baseline levels. Similarly, NfL was associated with dementia-specific mortality in fully adjusted models, with a hazard ratio of 2.72 (95% CI 1.57-4.71). The prognostic utility of these brain-derived proteins extended beyond neurological outcomes to capture systemic vascular risks. In fully adjusted models, GFAP was associated with cardiovascular-specific mortality, demonstrating a hazard ratio of 2.06 (95% CI 1.22-3.49). The relationship was equally evident for axonal damage markers, as NfL was associated with cardiovascular-specific mortality in fully adjusted models, showing a hazard ratio of 2.16 (95% CI 1.66-2.81). For clinicians managing older adults, these findings highlight that structural brain damage markers also reflect severe cardiovascular vulnerability, likely due to shared microvascular pathologies affecting both organ systems. Ultimately, the study establishes that plasma biomarkers of neurodegeneration, particularly GFAP and NfL, were associated with increased risk of all-cause, dementia-specific, and cardiovascular-specific mortality in a biracial cohort. Because these blood tests capture overlapping risks for both the brain and the heart, the researchers note these associations should be considered when assessing links between these biomarkers and other outcomes, as well as when used in clinical practice. As these assays become more widely available to practicing physicians, they carry dual prognostic weight. A positive biomarker result may soon prompt clinicians to evaluate patients not only for impending cognitive decline but also for severe cardiovascular disease, allowing for more aggressive, multidisciplinary preventive care.

References

1. Group TSR. A Randomized Trial of Intensive versus Standard Blood-Pressure Control. New England Journal of Medicine. 2015. doi:10.1056/nejmoa1511939

2. Truong C, Recto C, Lafont C, Canoui-Poitrine F, Belmin JB, Lafuente-Lafuente C. Effect of Cholinesterase Inhibitors on Mortality in Patients With Dementia: A Systematic Review of Randomized and Nonrandomized Trials.. Neurology. 2022. doi:10.1212/WNL.0000000000201161

3. Wang D, Xu X, Han X, et al. Clinical benefits of oral anticoagulants in atrial fibrillation patients with dementia: a systematic review and meta-analysis. Frontiers in Cardiovascular Medicine. 2023. doi:10.3389/fcvm.2023.1265331

4. Mach F, Baigent C, Catapano AL, et al. 2019 ESC/EAS Guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk. European Heart Journal. 2019. doi:10.1093/eurheartj/ehz455

5. Williams B, Mancia G, Spiering W, et al. 2018 ESC/ESH Guidelines for the management of arterial hypertension. European Heart Journal. 2018. doi:10.1093/eurheartj/ehy339