Clarifying blood-based brain-derived neurotrophic factor measurement in Alzheimer's disease research: A call for pre-analytic standardization

Brain-derived neurotrophic factor (BDNF) is a neurotrophin from a family of proteins that support neuronal survival, development, and function. ¹ Its gene is expressed in the cortex, hippocampus, and basal forebrain, which are essential for higher cognition, memory, and learning. BDNF enhances neurogenesis, synaptic growth, and plasticity, and facilitates neurotransmission and hippocampal long-term potentiation. ² Importantly, BDNF is first produced as pro-BDNF, which is then processed into mature BDNF, with the two isoforms exerting opposing effects on neuronal health. ² It crosses the blood-brain barrier bidirectionally, ³ making peripheral detection feasible. Lower serum BDNF levels have been associated with increased risks for Alzheimer's disease (AD) ⁴ and other neurodegenerative and psychiatric conditions, suggesting that reduced trophic support may contribute to neuronal degeneration and cognitive decline.

In their recent study, Kueck et al. ⁵ address persistent inconsistencies in pre-analytics in BDNF findings in AD and cognitive impairment (CI) research. Prior meta-analyses have consistently reported lower serum BDNF levels in individuals with AD compared to healthy controls,^6–9 and longitudinal studies suggest that higher serum BDNF may confer a protective effect against the risk of developing AD. ⁴ However, despite its neurotrophic significance and relevance, BDNF has received comparatively less attention due to inconsistent findings and interpretive challenges, particularly in the preclinical mild cognitive impairment (MCI) phase, largely driven by non-standardized laboratory pre-analytical procedures and uncontrolled clinical variables.

This study uniquely compares BDNF levels in platelet-rich plasma (PRP) versus platelet-poor plasma (PPP) and examines their relationship to the ATN (amyloid, tau, neurodegeneration) biomarkers. ¹⁰ The authors found that BDNF was consistently higher in PRP than PPP across all diagnostic groups, consistent with a prior report. ¹¹ Notably, the two BDNF measures were uncorrelated (R² = 0.013). PPP BDNF, but not PRP BDNF, was significantly lower in the cognitively impaired group compared to healthy controls. Furthermore, in the CI group, PPP BDNF positively correlated with pTau217, pTau181, and GFAP, but not with Aβ_42/40 or NfL, suggesting BDNF may serve a compensatory role in response to tau pathology and astroglial activation. These findings reinforce the notion that free-circulating BDNF, rather than platelet-derived BDNF, may be more biologically relevant in neurodegenerative contexts. ⁹

Particularly, as the author indicated, BDNF may reflect early compensatory changes in preclinical AD. The positive associations between PPP BDNF and pTau markers in CI individuals, but not in healthy controls, support this directional hypothesis. By performing such an association, this study advanced previous findings that support compensatory mechanisms, including reported positive association between BDNF and high-sensitivity C-reactive protein. ¹² This finding is also consistent with the hypothesis that during the initial stages of cognitive impairment, BDNF may increase as part of an adaptive neuroprotective mechanism against accumulating insults.^12,13 Here, the authors posited that hyperphosphorylated tau disrupts synaptic function and microtubule stability, leading to neuronal stress. This may trigger a compensatory upregulation of neurotrophic factors like BDNF, which supports synaptic plasticity and neuronal repair.

Blood-based BDNF measurement is highly sensitive to pre-analytical conditions, and methodological inconsistencies across studies have contributed substantially to data variability and the contradictory findings that have hindered the field over the past decades. Most importantly, BDNF levels differ by sample type: serum reflects total BDNF, including platelet-stored contributions released during clotting as well as upon other stimulations, whereas PPP better captures free circulating BDNF, which is more reflective of steady-state physiology and considered more central nervous system (CNS)-relevant, though not exclusively CNS-derived. The authors implemented key steps to reduce such laboratory confounding: blood samples were collected in 10 mL EDTA tubes, immediately centrifuged at 1500 × g for 10 min at 4°C to generate PRP, followed by a second high-speed centrifugation to obtain platelet-PPP. EDTA tube employed in this study effectively prevents clotting and platelet degranulation, which are common sources of artifact in serum samples. By reporting detailed pre-analytic variables, often overlooked in prior studies, the authors enhance the transparency and replicability of their findings. Additional factors such as selection bias, assay variability, batch effects, and detection sensitivity were also reported, enhancing transparency. However, whether laboratory staff was blinded to sample diagnostic status was not specified—a potential source of bias.

Another major strength of the study is the use of the Simoa^® BDNF Discovery Kit, which accordingly to the assay manual, seems to selectively measure mature BDNF and avoids confounding by pro-BDNF, an isoform with opposing, neurotoxic effects. ⁹ This is critical, as many commercial ELISAs fail to differentiate BDNF isoforms, a limitation noted in prior reviews.^9,14

Despite these strengths, the study has certain limitations that could have impeded its interpretations. While laboratory-related pre-analytical variables were rigorously controlled, there remains room for improvement in addressing clinical variables; The cognitively impaired (CI) group was relatively small (n = 25) compared to the healthy controls (n = 74) and diagnostically heterogeneous, including individuals with probable AD, amnestic and non-amnestic MCI, and frontotemporal dementia, all with Clinical Dementia Rating scores of 0.5 or 1, indicating early-stage impairment. This heterogeneity could obscure group differences and complicate the interpretation of BDNF-pathology associations.

Furthermore, although the authors adjusted for age, sex, and education, they did not fully account for psychiatric comorbidities or psychotropic medication use, both of which are known to influence peripheral BDNF. Specifically, depression has been associated with lower serum BDNF levels, while antidepressant use has been shown to increase them, ¹⁵ introducing potential confounding effects. While these issues were acknowledged as limitations, their impact on interpretation remains significant and deserves closer attention. This issue is relevant, as prior meta-analyses have also reported conflicting findings, including no significant changes in BDNF levels in individuals with MCI.^6–9

Aligning with the authors’ suggestion, this study supports the potential of BDNF as a treatment-responsive biomarker. Prior research shows that exercise and psychosocial interventions can elevate plasma and serum BDNF and enhance cognition,^5,11, ¹⁶ highlighting its promise for tracking therapeutic effects, particularly in early-stage interventions. However, the interpretation of serum BDNF increases should be approached with caution, as they may largely reflect changes in platelet BDNF content rather than CNS-derived levels. In contrast, plasma (particularly PPP) BDNF is thought to better capture free circulating levels more relevant to CNS physiology. We thus encourage future research to employ PPP to reflect free circulating BDNF and to avoid serum and PRP.

The potential of integrating BDNF into the ATX(N) framework is intriguing, as a synaptic plasticity marker, it may fit within the “N” (neurodegeneration) or “X” (emerging biomarkers) categories. Plasma BDNF may serve as a stage-specific marker, reflecting compensatory upregulation during prodromal phases and depletion with advanced pathology. Support for this hypothesis comes from both observational studies and meta-analyses linking higher BDNF to reduced dementia risk,^4,6–9 alongside recent reports of BDNF's association with tau and astroglial biomarkers in MCI.^17,18 To advance this integration, large community-based cohorts with harmonized protocols are essential, enabling examination of BDNF in relation to established ATN markers and critical moderators such as APOE4, BDNF Val66Met, and sex. Leveraging longitudinal datasets, including ADNI, UK Biobank, and other large-scale consortia, would provide the statistical power necessary to test these hypotheses and determine whether BDNF offers unique predictive or mechanistic value beyond the existing ATN markers. Given the heterogeneity in pre-analytical variables and the resulting conflicting findings, the clinical utility of blood-based BDNF currently relies on comparisons with control groups, and no standardized reference values have yet been established. Only with evidence from large longitudinal datasets and methodological standardization may BDNF transition from a promising molecule to a clinically meaningful biomarker in AD.

In summary, to minimize confounding from platelet-derived BDNF, studies should adopt BDNF-specific pre-analytic protocols that account for clinical variables (e.g., psychiatric comorbidities, antidepressant use) and laboratory factors, including the use of EDTA-treated blood, platelet-poor plasma, mature BDNF-specific assays, and rapid, chilled processing (Table 1). Standardization across these domains is essential to reduce measurement variability and improve comparability and replicability across studies.

OPEN IN VIEWER

Table 1.

Recommended pre-analytic standards for blood-based BDNF measurement.

A. Laboratory Variables
Component	Standard / Recommendation
Anticoagulant	Use EDTA tubes to prevent clotting and platelet degranulation.
Sample Type	Use platelet-poor plasma (PPP) to reflect free circulating BDNF; avoid serum and PRP. 5
Processing Time	Process blood samples immediately whenever possible, or within 30 min of collection to minimize platelet activation. 5
Centrifugation – Step 1	1500 × g for 10 min at 4°C to obtain platelet-rich plasma (PRP). 5
Centrifugation – Step 2	10,000 × g high-speed centrifugation to obtain platelet-poor plasma (PPP).
Temperature Control	Maintain processing at 4°C to limit platelet activation.
Assay Specificity	Use assays that detect mature BDNF only (e.g., Simoa®, Aviscera, or R&D Quantikine).
Isoform Distinction	Ensure assay distinguishes mature BDNF from pro-BDNF, which has opposing effects. 9
Analytical Rigor	Report batch/assay lot difference (if any), processing time, intra- and inter-assay variability, dynamic range, and detection sensitivity. 5

B. Clinical Variables
Component	Standard / Recommendation
Study Design	Minimize/address sample selection bias, perform power analysis to ensure adequate sample size, appropriate study design fitting to the purposes, e.g., matched case-control or longitudinal epidemiological cohort study.
Sample Blinding	Label samples using de-identified coding to blind laboratory staff to participants’ clinical data, especially diagnostic status and cognitive scores.
Psychiatric Comorbidities	Account for depression, anxiety, and other mental health conditions associated with reduced serum and plasma BDNF levels. 15
Medication Use	Document psychotropic medication use, particularly antidepressants, which are known to increase blood-based BDNF levels, most consistently in serum, 15 with more variable effects in plasma. 19
Diagnostic Clarity	Reduce heterogeneity by clearly defining diagnostic categories and disease stages (e.g., AD, MCI, non-AD dementias).

Table 1 illustrates that both laboratory and clinical factors jointly shape the reliability of blood-based BDNF measurement. On the laboratory side, variables such as anticoagulant choice, centrifugation speed, processing time, and assay specificity can substantially alter measured levels, especially given the sensitivity of platelets. On the clinical side, comorbidities, psychotropic medication use, and diagnostic heterogeneity introduce additional variability that may obscure disease-related effects. Thus, the table emphasizes that without rigorous standardization across both domains, peripheral BDNF findings are at risk of inconsistency and misinterpretation.