Depression is a mental disorder characterized by biological and psychosocial diversity, hindering the identification of effective diagnostic biomarkers. Brain-derived neurotrophic factor (BDNF), a neurotrophic factor, has garnered attention recently, with its blood levels, genetic polymorphisms, and cerebrospinal fluid concentrations correlating with depression severity and treatment response. In particular, changes in the mBDNF/proBDNF ratio, p75NTR, and tropomyosin receptor kinase B concentrations have been suggested to be associated with neuroplasticity. The Val66Met polymorphism and epigenetic changes in the BDNF gene may also influence treatment responsiveness. Standardization of measurement methods, artificial intelligence analysis, and multifactorial integration are expected to facilitate the application of personalized medicine for depression treatment in the future.
Major depression ranks among the most common mental disorders worldwide, affecting millions of people, and there is an urgent need to elucidate its etiology and treatment methods. Given its clinical heterogeneity and diverse etiological hypotheses, depression still lacks an effective and reliable biomarker. Various potential biomarkers have been explored over the past few decades, including neurotrophic factors, cytokines, endocrine markers, neuroimaging findings, genetic polymorphisms, epigenomics, and gut microbiota. In particular, brain-derived neurotrophic factor (BDNF), interleukin 6, C-reactive protein, cortisol, hypothalamic-pituitary-adrenal (HPA) axis dysfunction, or neuroimaging findings, such as reduced hippocampal volume or decreased prefrontal cortex activity, have been reported with reproducibility in many studies. However, none of these candidates have demonstrated sufficient robustness or versatility for clinical application, and thus diagnostic, predictive, or treatment response markers for major depression remain to be established [1, 2].
Traditional operational diagnostic criteria based on the Diagnostic and Statistical Manual of Mental Disorders or International Classification of Diseases are symptom-based classifications limited by biological heterogeneity [3, 4]. As a result, patients with different pathophysiological backgrounds may be grouped under the same diagnosis, diluting the statistical power to detect between-group differences. In addition, various confounding factors such as medication history, comorbidities, lifestyle factors, variability in sample collection timing, and analytical techniques hinder reproducibility.
Furthermore, many candidate biomarkers are not disease-specific but are associated with other factors such as stress response, inflammation, and aging. Therefore, their clinical utility in differential diagnosis is limited. Meanwhile, multi-modal analysis integrating multiple indicators is gaining attention through the application of artificial intelligence and machine learning, and the possibility of disease subtype classification and treatment response prediction is being explored [5–7]. However, even with these approaches, careful consideration is still needed for clinical implementation, owing to issues such as the black-box nature of algorithms and the lack of external validity. The current diagnosis of depression, which is still based on patients’ subjective complaints and psychiatrists’ behavioral observations, is clearly inaccurate and becoming outdated. We expect that the immediate adoption of comprehensive evaluations combining rigorous measurements using artificial intelligence analysis, omics analysis including BDNF, neural network analysis, and the use of new devices will advance the individualization of depression diagnosis and treatment.
This review is organized around the potential for blood BDNF as a diagnostic biomarker for depression, as a stress biomarker, and as a therapeutic biomarker for depression (including prehension with genetic polymorphisms), as well as recent topics (including relevance to industrial psychiatry). BDNF may contribute to psychiatry and occupational psychiatry as one key to personalized medicine in the future.
Significance of BDNF in Depression as a Biological MarkerThe identification of biomarkers that elucidate the biological basis of depression and enable objective prediction of diagnosis and prognosis remains a critical challenge in psychiatry. Unfortunately, no biologic markers for depression with high specificity and sensitivity have been identified, and their clinical utility is currently limited [8]. One potential biomarker is BDNF, a neurotrophic factor belonging to the nerve growth factor family, which plays a central role in the development, differentiation, survival, and maintenance of plasticity of neurons in the central and peripheral nervous systems [9]. BDNF is synthesized as proBDNF (precursor BDNF) and is cleaved within or outside cells to form mature BDNF. Mature BDNF binds to the high-affinity tropomyosin receptor kinase B (TrkB), activating the MAPK/ERK, PI3K/Akt, and PLCγ pathways to promote intracellular signal transduction and regulate neuronal survival, differentiation, and synaptic plasticity. Of note, proBDNF induces cell death and synaptic pruning through the p75NTR receptor, and the functions of BDNF are precisely regulated by maintaining a balance between proBDNF and BDNF [10].
Brain-derived neurotrophic factor (BDNF) is highly expressed in the hippocampus, prefrontal cortex and amygdala, and is deeply involved in memory formation, learning, and emotional regulation [11]. In particular, BDNF contributes to the molecular basis of neuroplasticity in the hippocampus by regulating the movement of AMPA receptors in the postsynaptic region and modulating synaptic strength during long-term potentiation [12]. Furthermore, the Val66Met polymorphism of the BDNF gene has been associated with prefrontal cortex function and stress vulnerability, suggesting its possible usefulness in understanding individual differences in depression onset [13].
Brain-derived neurotrophic factor (BDNF) expression increases in response to exercise, calorie restriction, and environmental stimuli, making it a promising therapeutic target for mental disorders and neurodegenerative diseases [14]. Reduced BDNF levels in the blood of patients with depression and increased BDNF levels following antidepressant administration have been reported [15]. In animal models, stress and electroconvulsive stimulation have been shown to alter mRNA and protein expression levels in the hippocampus. In humans, a correlation between reduced blood BDNF levels in patients with depression and recovery following treatment has been reported [16, 17]. In the future, elucidating the molecular pathophysiology of BDNF and its clinical application will deepen the understanding of depression and its treatment.
Several issues should be considered when interpreting blood BDNF levels. First, BDNF measurement is technically unstable, as serum BDNF levels are influenced by numerous factors, including post-collection processing time, storage conditions, centrifugation conditions, time of blood collection, and the subject’s smoking history, exercise habits, sleep patterns, and inflammatory status. As a result, inter-study reproducibility is low, and consistent conclusions are difficult to achieve even through meta-analyses [18–20].
Second, whether blood BDNF levels truly reflect the state of BDNF in the brain remains uncertain. While BDNF is primarily expressed in the brain, it is also produced in peripheral tissues such as platelets, lymphocytes, vascular endothelial cells, and muscles. The decrease in blood BDNF concentrations in depression and depressive states can be considered to reflect abnormalities in endothelial cells or reduced endothelial cell activity [21–25]. In a 2016 review article, Miller and Raison stated that depression is not merely an abnormality of neurotransmitters, but a multisystemic disease involving abnormalities in the immune system, inflammation, and vascular endothelial cell function [26]. In fact, depression and depressive states often coexist with autoimmune and inflammatory diseases [27, 28]. A previous study reported that blood p75NTR concentrations are elevated and blood TrkB concentrations are reduced in patients with depression [29]. In rats, an imbalance between the full-length type of tropomyosin receptor kinase B (TrkB-FL) and the dominant negative type (TrkB-T1) has been suggested to inhibit BDNF signaling. TrkB-FL signaling was also recovered after antidepressant treatment [30, 31]. These findings suggest differences in BDNF signaling between the central nervous system and peripheral tissues, making interpretation challenging.
Third, some limited amounts of BDNF cross the blood-brain barrier into the bloodstream. Although studies on the relationship between BDNF concentration in the brain and blood have yielded positive results in a few rat samples, these results lack reproducibility. In rats subjected to chronic stress, blood BDNF concentrations decreased, but no correlation with brain BDNF levels (mRNA, protein) was observed. The study furthermore reported no correlation between BDNF concentration in the cerebrospinal fluid and that in blood. These results suggest that blood BDNF does not meet the essential criteria for a biomarker that “visualizes brain state” in understanding the pathophysiology of depression.
Fourth, the fact that depression is a heterogeneous syndrome rather than a homogeneous single disease limits the usefulness of a single biomarker [32]. Of note, BDNF reduction may be limited to specific subtypes (e.g., melancholic type) or may fluctuate through interactions with the inflammatory system, hindering its application as a uniform indicator [33–35]. In addition, genetic polymorphisms of BDNF (particularly the Val66Met polymorphism) have been reported to influence blood BDNF levels and treatment responsiveness. Failure to control for genetic variability may lead to confusion in causal interpretations. Recent Mendelian randomization studies on BDNF have argued that low blood BDNF levels provide only weak evidence as a state marker rather than a risk factor for depression onset [36].
Fifth, changes in the patterns of BDNF vary depending on the type, dose, and duration of antidepressant treatment, which undermines its reliability as a clinical indicator. Considering these points, BDNF may not be a diagnostic marker for depression itself but may indicate the involvement of the immune and inflammatory systems in the pathophysiology of depression. Moreover, it can be considered an indicator of plasticity changes induced by therapeutic interventions. In fact, antidepressants and anti-inflammatory drugs have been reported to alter blood BDNF levels [14, 37–39]. Therefore, future studies should construct multivariate models combining blood BDNF with other biological markers, such as inflammatory factors, brain imaging findings, and cognitive function scores. However, even such an approach may only modestly represent a subset of depression phenotypes. Furthermore, through comprehensive studies that consider the reorganization of disease classifications, such as those using RDoC (Research Domain Criteria), the meaning of BDNF should be defined in a more context-dependent manner.
In humans, the BDNF gene is located on chromosome 11 (11p14.1) and has a complex structure with multiple promoters and splicing sites, enabling precise transcriptional regulation in various brain regions and developmental stages. BDNF not only promotes the survival, differentiation, and synapse formation of neurons, but also plays a role in long-term potentiation in the hippocampus, which is involved in memory and learning, and is considered a core component of the molecular basis of plasticity [40–42]. BDNF, which is secreted in a neuroactivity-dependent manner, primarily binds to the TrkB receptor, activating intracellular signaling pathways such as MAPK/ERK, PI3K/Akt, and PLC-γ, thereby enhancing synaptic function and cell survival signals. Of particular interest is the BDNF gene polymorphism Val66Met (rs6265), which is a single-nucleotide polymorphism where the valine (Val) at codon 66 is replaced by methionine (Met). This polymorphism is known to reduce the extracellular secretion of BDNF and is associated with memory function, emotional regulation, stress responses, and various mental disorders such as depression, schizophrenia, and anxiety disorders; it is considered one of the genetic vulnerability markers [43–46]. The Val66Met polymorphism has also shown potential association with responsiveness to antidepressant and cognitive behavioral therapies and is attracting attention as a molecular basis for future personalized medicine; however, consistent results have not yet been obtained [47–50].
The BDNF gene is also important from an epigenetic perspective. Animal experiments and human studies have reported that chronic stress or traumatic experiences increase DNA methylation in the BDNF promoter region, resulting in reduced transcriptional activity [51, 52]. The BDNF gene polymorphisms, particularly the Val66Met polymorphism (rs6265), have been extensively reported to be associated with the risk of developing depression [53]. This substitution reduces Ca++-dependent secretion of BDNF, thereby inhibiting synaptic plasticity and hippocampal neurogenesis [54, 55]. Consequently, this impaired secretion is thought to increase susceptibility to depression by altering the function of the hippocampus and prefrontal cortex, which are involved in memory and emotional regulation. Epidemiological studies have reported that individuals carrying the Met allele have a higher probability of having a history of depressive episodes, particularly in interaction with environmental factors, such as childhood adversity or chronic stress (gene-environment interaction); however, these findings remain within the realm of hypothesis. In other words, the Val66Met polymorphism may not directly confer a strong risk for depression but could function as a vulnerability factor during stress exposure. Gene-environment interaction (G × E) may partially contribute to the onset of depression through abnormal stress responses in neural circuits or hyperactivity of the hypothalamic-pituitary-adrenal (HPA) axis [46, 56, 57].
Functional neuroimaging studies have identified amygdala hyperactivity and reduced functional connectivity with the prefrontal cortex in individuals with the Met allele, suggesting that these may serve as neural substrates for emotional regulation disorders [58, 59]. Met allele carriers are also reported to show blunted responses to antidepressants and potentially require longer-term interventions or alternative therapies [60, 61]. However, reproducibility is limited, and ethnic differences and variability in study designs make it difficult to draw definitive conclusions about the association between the Val66Met polymorphism and depression onset. In particular, the frequency of the Met allele is higher in East Asian populations, suggesting that the risk structure may differ from that in Western populations [62, 63]. From an epigenetic perspective, methylation of the BDNF promoter region is associated with depressive symptoms, and the interaction between the Val66Met polymorphism and epigenetic modifications [64–66] is an important area for future research.
In summary, the BDNF Val66Met polymorphism is a strong candidate for higher vulnerability to depression. However, its effects are mediated through interactions with environmental factors, and risk assessment using a single factor may have low efficacy. In the future, a multi-layered approach integrating brain imaging, biomarkers, epigenetics, and environmental factors is expected to enable more accurate risk assessment and individualized prevention and treatment strategies.
Numerous studies in humans have reported that psychosocial stress affects blood BDNF concentrations. For example, blood BDNF concentrations were significantly reduced in workers with long working hours or interpersonal stress, such as workers with burnout tendencies or high subjective stress, mothers raising children, and patients with post-traumatic stress disorder. In a cohort study involving healthcare workers and employees in Germany, Hellweg et al reported lower blood levels of BDNF in workers exposed to chronic stress. Similarly [67], Bath et al reported an association between workaholic tendencies and low BDNF levels in business people in Taiwan [68]. Alves et al investigated the relationship between childcare stress and BDNF levels among Japanese mothers with young children [69]. The results confirmed that BDNF levels were significantly lower in the high-stress group. Mothers exhibiting depressive tendencies showed a more pronounced decrease in BDNF levels. Another study reported that long working hours and interpersonal stress in the workplace may influence BDNF levels [70]. Furthermore, temporary decreases in blood BDNF levels have been observed following acute stress exposure (e.g., the Trier Social Stress Test). However, the responsiveness of BDNF to acute stress shows high individual variability and is believed to be influenced by psychological characteristics and stress tolerance. Similar to the relationship between depression and serum BDNF levels, any clinical application of serum BDNF for stress diagnosis has not yet been achieved and requires several issues to be addressed, such as standardizing measurement accuracy and controlling various confounding factors. Therefore, future research should longitudinally examine the relationship between workplace stress and BDNF, and intervention evaluations using BDNF as a biomarker are required. Psychiatrists should be involved in workplace mental health not only through psychosocial support but also from a neurobiological perspective.
BDNF Gene Polymorphisms and Antidepressant ResponsivenessNumerous clinical studies have been conducted on the relationship between the Val66Met polymorphism and antidepressant response. Individuals carrying the Met allele (Val/Met or Met/Met) tend to have poorer antidepressant response compared with Val/Val homozygotes. This difference is particularly evident in standard monoamine antidepressant treatments, such as SSRIs and SNRIs. Meta-analysis results also support this trend, with Kishi et al reporting that patients with the Val/Val genotype demonstrated significantly higher treatment response and remission rates than those with the Met allele [71]. In addition, increase in BDNF concentration is suppressed in Met allele carriers following treatment initiation, which may be associated with delayed onset of antidepressant effects or inadequate symptom improvement [72–74]. Further, the Val66Met polymorphism has been reported to be associated with not only treatment response but also treatment side effects and cognitive function. For example, some studies suggest that Met allele carriers exhibit more pronounced sexual dysfunction and cognitive impairment associated with antidepressants, potentially necessitating careful follow-up and treatment selection. However, reports on the relationship between the Val66Met polymorphism and treatment response are inconsistent [75–77].
The frequency of the Met allele is higher in East Asian populations in particular, and risk structures and pharmacokinetics may differ compared with those of Western populations, making it essential to consider ethnic differences in analyses [62, 78]. In addition, factors such as the type of antidepressant, dosage, treatment duration, and the presence of concomitant therapies may influence the results [71, 79]. Furthermore, the Val66Met polymorphism does not independently determine antidepressant responsiveness; interactions with other genes (e.g., SLC6A4, HTR2A, COMT) and environmental factors (e.g., childhood adversity, chronic stress, lifestyle habits) are important. Recently, based on the G × E (gene × environment) model, the Val66Met polymorphism has been hypothesized to mediate stress vulnerability and antidepressant response. Also, epigenetic changes, such as DNA methylation in the BDNF promoter region, may modify the expression or function of genotypes, necessitating comprehensive molecular-level evaluation [58, 80].
In summary, biomarkers for depression remain at the “candidate” stage at present, and cannot directly support diagnostic, prognostic, or treatment-selection decisions. Going forward, the identification of clinically useful biomarkers is anticipated through the use of biological phenotyping-based stratification, longitudinal data, standardized measurement methods, and cross-disease comparative studies.
Many studies have reported that serum and plasma BDNF concentrations are reduced in patients with depression. BDNF has a large molecular size and cannot cross the blood-brain barrier. Therefore, serum or plasma BDNF is unlikely to reflect BDNF dynamics within the brain [21–25]. In this regard, the cause of the decrease in blood BDNF levels in patients is worth exploring. Platelets contain a large amount of BDNF and play a role in transporting this factor to the brain and other tissues. Platelet dysfunction has been reported in patients with depression, and thus BDNF secretion may be insufficient. Specifically, platelet activation or aggregation abnormalities may influence BDNF levels in the blood. This deficiency in platelet activation may be caused by abnormal secretion of stress hormones or enhanced inflammatory responses [81, 82]. Regarding endothelial cell dysfunction, these cells maintain vascular health and promote appropriate blood flow. In patients with depression, chronic stress and inflammation impair vascular endothelial function, which leads to systemic circulatory failure and reduced BDNF supply to the brain. Furthermore, reduced production of nitric oxide from the vascular endothelium impairs vascular dilation, thereby obstructing BDNF transport.
Through these mechanisms, blood BDNF levels in patients with depression decrease, impairing the plasticity and repair functions of brain neural circuits, thereby exacerbating depressive symptoms [83–85]. Many studies have reported that exercise increases BDNF levels. In particular, aerobic exercises, such as walking, jogging, and cycling, increase blood BDNF levels. Acute exercise results in a temporary increase in BDNF levels after a single session, while chronic exercise lasting several weeks or more leads to a sustained increase in baseline blood BDNF levels. Studies in patients with depression have also reported that regular exercise intervention (≥30 min, ≥3 times a week) increases blood BDNF levels and improves depressive symptoms [86–88]. Diet is also an important factor influencing BDNF. A high-fat, high-sugar “Western-style diet” and a diet centered on processed foods have been suggested to reduce BDNF levels. In contrast, the Mediterranean diet may contribute to increased BDNF levels. Among these, omega-3 fatty acids (particularly docosahexaenoic acid and eicosapentaenoic acid), polyphenols, vitamin D, zinc, magnesium, calorie restriction, and intermittent fasting are of particular interest. Conversely, refined carbohydrates, saturated fatty acids, and excessive alcohol consumption are associated with reduced BDNF levels [89, 90].
Antidepressant treatment is reported to restore blood BDNF levels, which occurs in parallel with clinical improvement. Therefore, blood BDNF is not only considered a “state marker” for depression but also holds promise as a “treatment response marker.” When examining the relationship between pharmacotherapy and BDNF in detail, major antidepressants, including selective serotonin reuptake inhibitors (SSRIs), serotonin-norepinephrine reuptake inhibitors (SNRIs), and noradrenergic and specific serotonergic antidepressants, have been shown to significantly increase blood BDNF levels within a few weeks of treatment, and this change correlates significantly with symptom improvement. In particular, an increase in BDNF levels within 2–4 weeks of treatment initiation may serve as a predictive factor for subsequent improvement in depressive symptoms. Therefore, monitoring BDNF dynamics in the early stages of treatment may help predict treatment response and determine early intervention strategies [91–93].
In contrast, baseline BDNF levels are reported to be low in non-responsive or treatment-resistant patients with depression, and the increase after treatment is limited. In particular, cases with minimal changes in serum BDNF levels may serve as a basis for reconsidering treatment strategies from an early stage or exploring complementary approaches, such as ketamine. According to recent studies, ketamine and its derivatives rapidly activate the BDNF pathway, restoring plasticity and exerting acute antidepressant effects, suggesting that BDNF plays a central mediatory role in this mechanism [94–96].
Blood proBDNF
Brain-derived neutrotrophic factor (BDNF) is synthesized as proBDNF, a precursor, and converted to mBDNF by proteases. Reports indicate that serum proBDNF levels are elevated in patients with depression, suggesting that a decrease in the mBDNF/proBDNF ratio may be associated with the pathophysiology of depression. The neuroplasticity hypothesis of depression suggests that chronic stress or inflammation may suppress the production or function of mBDNF, leading to a predominance of proBDNF, which in turn may contribute to neural atrophy in the hippocampus and prefrontal cortex. SSRIs and electroconvulsive therapy are also reported to reduce proBDNF levels and increase mBDNF levels, suggesting a potential association between treatment response and changes in the mBDNF/proBDNF ratio [97–100].
P75NTR and TrkBR
Recent studies have reported elevated p75NTR concentrations in the blood of patients with depression. This finding suggests that pathways involved in neurotoxicity are activated by chronic stress or increased inflammatory cytokines, and excessive activation of the proBDNF-p75NTR pathway may be associated with neurotoxicity and atrophy in the hippocampus and prefrontal cortex. In addition, p75NTR is closely associated with glial cell activation, supporting the glial-derived inflammation hypothesis [97, 101]. In contrast, a previous study indicated that TrkB concentrations are significantly reduced in patients with depression, suggesting that impaired BDNF-TrkB signaling may be associated with reduced neuroplasticity, decreased motivation, and cognitive dysfunction in patients with depression [102]. In particular, an imbalance between the TrkB-FL and TrkB-T1 inhibits BDNF signaling. Reports have also confirmed the recovery of TrkB-FL following antidepressant treatment, suggesting its potential as a biomarker for treatment response. Clinically, the blood levels of these receptors are expected to be useful as auxiliary indicators for classifying the pathology of depression, assessing severity, and predicting treatment response. However, receptor expression may differ between the central nervous system and peripheral tissues, and further verification is needed to determine how well blood levels reflect brain states [103]. Of note, confounding factors such as inflammation, metabolic status, and age must be considered. Based on the above, elevated p75NTR and reduced TrkB are important biological marker candidates associated with neuroplasticity disorders in depression. Analyzing the mBDNF/proBDNF/p75NTR/TrkB network combining these markers may serve as an important strategy for achieving personalized medicine [104].
BDNF Concentration in Cerebrospinal Fluid
Brain-derived neurotrophic factor (BDNF) concentration in cerebrospinal fluid has also garnered attention as a biomarker for depression. Cerebrospinal fluid BDNF may more directly reflect BDNF activity within the brain. A previous study reported that cerebrospinal fluid BDNF concentration is reduced in patients with depression, suggesting that impairments in BDNF signaling within the central nervous system may contribute to the pathophysiology of depression [105].
Occupational Psychiatry and BDNF
Recent occupational health research has begun to incorporate biomarkers such as BDNF to objectively assess the impact of job stress. Cross-sectional studies of healthcare workers, first responders, and office employees have found associations between low serum BDNF levels and higher scores on burnout, depression, and anxiety scales [106–108]. Moreover, longitudinal studies suggest that individuals with persistently low BDNF levels are more vulnerable to developing stress-related disorders in the face of adverse occupational conditions [109, 110]. Conversely, certain work-related interventions appear to influence BDNF levels positively. Physical activity programs, which are increasingly adopted in occupational wellness strategies, have been shown to increase peripheral BDNF concentrations [111, 112]. Mindfulness-based stress reduction (MBSR), cognitive behavioral therapy (CBT), and workplace social support interventions may also indirectly elevate BDNF expression through reduced allostatic load and improved emotion regulation [113–115].
Biomarkers are important for enhancing the diagnosis and treatment of depression. BDNF has garnered particular attention as a potential biomarker for depression, owing to its functions and regulatory mechanisms. However, other biomarkers are also crucial for understanding the high heterogeneity and complex pathophysiology of depression and for developing effective treatments. This includes the identification of biological subtypes, development of personalized treatment, prediction of treatment response, and identification of indicators for disease progression. Recent studies have shown that blood levels of BDNF, genetic polymorphisms, and cerebrospinal fluid levels of BDNF are associated with symptom severity and treatment response in patients with depression. These findings suggest that depression treatment can be customized for individual patients, representing the first step toward the realization of personalized treatment for depression. However, the clinical application of biomarkers, such as BDNF, requires improvements in their specificity and sensitivity, as well as a deeper understanding of the biological heterogeneity of depression. In addition, identifying other new biomarkers for more precise diagnosis and treatment is an urgent priority, and integration of the biomarkers is necessary. Future research should advance large-scale cohort studies and international collaborative research to deepen our understanding of the effectiveness of biomarkers, including BDNF, in various ethnic groups and regions, as well as the interaction between environmental and genetic factors. The development of biomarker detection methods using the latest artificial intelligence technology will also contribute to research advancement.
Not applicable
Not applicable
The author has no conflict of interest related to this paper.
Not applicable
The author performed all tasks related to the writing of this paper.
Not applicable
Not applicable
