Soluble p75 neurotrophic receptor as a reliable biomarker in neurodegenerative diseases: what is the evidence?

Abstract

Neurodegenerative diseases are often misdiagnosed, especially when the diagnosis is based solely on clinical symptoms. The p75 neurotrophic receptor (p75^NTR) has been studied as an index of sensory and motor nerve development and maturation. Its cleavable extracellular domain (ECD) is readily detectable in various biological fluids including plasma, serum and urine. There is evidence for increased p75^NTR ECD levels in neurodegenerative diseases such as Alzheimer’s disease, amyotrophic lateral sclerosis, age-related dementia, schizophrenia, and diabetic neuropathy. Whether p75^NTR ECD could be used as a biomarker for diagnosis and/or prognosis in these disorders, and whether it could potentially lead to the development of targeted therapies, remains an open question. In this review, we present and discuss published studies that have evaluated the relevance of this emerging biomarker in the context of various neurodegenerative diseases. We also highlight areas that require further investigation to better understand the role of p75^NTR ECD in the clinical diagnosis and management of neurodegenerative disorders.

Introduction

Neurodegenerative diseases affect millions of people worldwide and represent a major public health problem. These chronic disorders manifest clinically with motor impairments, cognitive deficits, and psychiatric symptoms. Uncertainties remain about their exact etiology and the underlying molecular pathological mechanisms. They are often misdiagnosed due to reliance on clinical symptom criteria, which often lack specificity.

Neurotrophins belong to a family of proteins that includes nerve growth factor, brain-derived neurotrophic factor (BDNF), and neurotrophins 3 and 4. They are formed by cleavage of precursor proteins called proneurotrophins. They bind tropomyosin receptor kinase (Trk) A, B, and C, respectively, with high affinity and induce neuronal growth. Neurotrophins also have an approximately equal low affinity for the p75 neurotrophic receptor (p75^NTR, also known as nerve growth factor receptor (NGFR)), the first identified member of the tumor necrosis factor receptor superfamily (Bothwell, 1995; Locksley et al., 2001). Named for its molecular mass, p75^NTR is encoded by the NGFR gene located on chromosome 17 (17q21.33). The complete isoform contains six exons and five introns. It encodes a 3.8 kb mRNA with a 5′ untranslated region of approximately 300 nucleotides and a 3′ untranslated region of 2000 nucleotides with a single polyadenylation signal (Johnson et al., 1986). The p75^NTR receptor is a single transmembrane protein of 399 amino acids with an amino-terminal negatively charged extracellular domain (ECD) containing four highly glycosylated tandem cysteine-rich domains (CRD1 to CRD4) (40 amino acids each containing six cysteine residues) involved in ligand binding, and a carboxy-terminal intracellular domain (ICD) of 155 amino acids formed by a juxtamembrane (chopper) domain and a death domain (Johnson et al., 1986; Malik et al., 2021; Bruno et al., 2023). In addition to all neurotrophins, p75^NTR binds pro-neurotrophins, amyloid-β (Aβ), prion protein peptide 106–126, myelin-associated glycoprotein, and rabies virus glycoprotein (Malik et al., 2021).

The p75^NTR receptor is expressed primarily in the developing brain, where it is involved in axonal pruning through the action of neurotrophins (Singh et al., 2008). It is also expressed by Schwann cells in the peripheral nervous system, where it regulates myelination by acting as a coreceptor for myelin-associated glycoprotein binding to the neurite outgrowth protein receptor (Wong et al., 2002). It is weakly expressed in the healthy adult central nervous system, in sympathetic and sensory neurons and in subsets of enteric and parasympathetic neurons (Roux and Barker, 2002). In addition to neuronal tissues, mesenchymal cells express p75^NTR in the limb, kidney, jawbone, tooth, lung muscle, testis, retina, pituitary, and around epithelial structures including hair follicles, salivary glands, perivascular cells, and meninges (Roux and Barker, 2002).

The p75^NTR has no catalytic activity per se. Rather, its activation induces the binding of adaptor proteins to the ICD and involves co-receptors for signal transduction (Schecterson and Bothwell, 2010; Malik et al., 2021). More than thirty partners including small GTPases, ubiquitin ligases and FG nucleoporins interact with p75^NTR and mediate diverse and sometimes opposing cellular effects (Schecterson and Bothwell, 2010; Malik et al., 2021). Ligand binding to p75^NTR induces the cleavage of its ECD by α-secretase A disintegrin and metalloprotease 17 and 10, resulting in the release of p75^NTR ECD into the extracellular environment (DiStefano et al., 1991; Kenchappa et al., 2010; Bao et al., 2018). The soluble form of ECD is produced at high levels during development and after nerve injury (DiStefano et al., 1991). Its precise biological role remains uncertain. It may act as a decoy receptor that scavenges neurotrophins and proneurotrophins in the extracellular space, or it may facilitate p75^NTR-mediated death signaling (Nykjaer et al., 2005). Following proteolytic cleavage of the ECD, the ICD can also be cleaved by the intramembrane ɣ-secretase protein complex and the released ICD translocates to the nucleus where it contributes to the induction of apoptosis (Figure 1; Frade, 2005; Kenchappa et al., 2006; Bruno et al., 2023).

Figure 1.

Regulated proteolysis of the p75 neurotrophic receptor.

1. Mature neurotrophins and their precursors, the pro-neurotrophins, bind and activate the full length p75 neurotrophic receptor (p75^NTR) resulting in the binding of adaptor proteins to its intracellular chopper and signal transduction. 2. Proteases from the A disentegrase and metalloproteinase (ADAM) family, namely ADAM 10 and 17, cleave p75^NTR in the presence or absence of agonists. Cleavage occurs on the stalk region nearby the transmembrane domain, resulting in the release of the extracellular domain fragment in the interstitial fluid. 3. Once the extracellular domain is cleaved, γ-secretase processes the C-terminal fragment to release the p75^NTR intracellular domain fragment, which is further involved in p75^NTR signaling. Created with BioRender.com.

While p75^NTR is weakly expressed in healthy adult neurons, its expression can be greatly increased in neurodegenerative diseases, resulting in increased excretion of its ECD into biological fluids such as blood, cerebrospinal fluid (CSF), and urine. In neurologically healthy individuals, urinary p75^NTR ECD levels are higher in neonates and decline until stabilization at approximately 15 years of age (DiStefano et al., 1991). No evidence of diurnal variation in their levels has been observed (Shepheard et al., 2017), but they may increase again during pregnancy (DiStefano et al., 1991). Levels of p75^NTR ECD in biological fluids have also been assessed in many neurodegenerative diseases, mainly amyotrophic lateral sclerosis (ALS) (Shepheard et al., 2014, 2017, 2022; Jia et al., 2017; Yamada et al., 2021; Shi et al., 2022) and Alzheimer’s disease (AD) (Lindner et al., 1993; Salehi et al., 2000; Hu et al., 2002; Jiao et al., 2015; Yao et al., 2015; Xu et al., 2019), but also in age-related dementia (Messripour et al., 2015), as well as schizophrenia (Zakharyan et al., 2014; Chen et al., 2017; He et al., 2019; Zakowicz et al., 2023) and diabetic neuropathy (Hruska et al., 1993; Humpert et al., 2007). Thus, soluble p75^NTR ECD can be found in readily available fluids at quantifiable levels, which has the potential to be used as a biomarker for these diseases. The purpose of this manuscript is to provide an overview of currently available studies evaluating the potential value of p75^NTR ECD as a biomarker for neurodegenerative diseases.

Retrieval Strategy

We performed a literature review in PubMed from inception to April 2023 using “p75 neurotrophin” OR “NGFR” OR “p75^NTR” AND “blood” OR “serum” OR “plasma” OR “urine” OR “circulation” OR “cerebrospinal fluid” OR “saliva” OR “marker” OR “biomarker” AND “neurodegenerative disease” OR “amyotrophic lateral sclerosis” OR “Alzheimer’s disease” OR “dementia” OR “schizophrenia” OR “diabetic neuropathy” as medical subject headings terms and applied two filters, “humans” and “English”.

Preanalytical and Analytical Aspects of Assessing p75^NTR Extracellular Domain Levels

The p75^NTR ECD has been studied as an index of sensory and motor nerve development, maturation, and injury in multiple biological matrices, including whole blood, blood fractions (i.e., plasma (Humpert et al., 2007; Zakharyan et al., 2014; He et al., 2019; Zakowicz et al., 2023), and serum (Chen et al., 2017; Jia et al., 2017)), urine (Hruska et al., 1993; Lindner et al., 1993; Shepheard et al., 2014, 2017, 2022; Jiao et al., 2015; Messripour et al., 2015), and CSF (Jiao et al., 2015) in the context of various neurodegenerative diseases. The collection of both urine and blood is minimally invasive, thus presenting with a greater biomarker potential transferability to the clinic than the assessment in CSF. p75^NTR ECD levels have mostly been quantified using antibody-based techniques, mainly enzyme-linked immunosorbent assay (ELISA) using either commercialized (Zakharyan et al., 2014; Jiao et al., 2015; Chen et al., 2017; He et al., 2019; Fleury et al., 2022; Zakowicz et al., 2023) or homemade reagents (Hruska et al., 1993; Lindner et al., 1993; Shepheard et al., 2014, 2017, 2022; Jia et al., 2017)). Immunoprecipitation followed by western blotting has also been used (Humpert et al., 2007), but due to its non-quantitative nature it is less likely to be transferred to clinical use. It should be noted that there are currently no standardized procedures or guidelines for the pre-, post- and analytical steps of p75^NTR ECD measurement.

Urine p75^NTR levels appear stable when samples are frozen at –80°C (up to 10 years), but only stable for up to 2 days when stored at room temperature or 4°C (Hruska et al., 1993; Shepheard et al., 2017). They have also been shown to be stable after at least three freeze-thaw cycles (Hruska et al., 1993). No data have been reported on the stability of p75^NTR in other biological fluids (i.e., whole blood, blood fractions, and cerebrospinal fluid), nor have correlations between p75^NTR ECD levels in these different matrices been established.

As shown in Table 1, assessment of normal ranges of p75^NTR ECD in healthy subjects included in the various published studies has yielded variable results. Levels of p75^NTR ECD are reported in units that reflect the different methods and assay conditions used, as well as the different types of biological fluids examined, making it essential to establish local reference ranges that cannot currently be derived from the published literature. This may also explain the wide range of levels reported in neurodegenerative diseases (Table 2), which may hinder the development of robust cut-offs for use as diagnostic or prognostic biomarkers.

Table 1.

Summary of p75^NTR assays and results in human healthy control samples

Sample types	p75NTR fragments	Assay reagents	Healthy participant characteristics	Reported values	References
Urine	p75NTR ECD	Homemade ELISA	n = 19 Age: 63.3–83.9 yr	0–2 ng/µL (visually deduced)	Lindner et al., 1993
	p75NTR ECD	Homemade ELISA	n = 18 men Age: 39 ± 9 yr	85.5 ± 20.3 ng/mg creatinine	Hruska et al., 1993
			n = 27 women Age: 35 ± 9 yr	84.1 ± 24.0 ng/mg creatinine
	p75NTR ECD	Homemade ELISA and commercial ELISA (R&D Systems, Minneapolis, MN, USA)	n = 12 6 women, 6 men Age: 59.6 (40–71) yr	2.6 ± 0.2 ng/mg creatinine	Shepheard et al., 2014
	p75NTR ECD	Homemade agarose gel diffusion assay	n = 12 Group 1 (52–63 yr) Group 2 (64–77 yr) Group 3 (78–87 yr)	Undetectable	Messripour et al., 2015
	p75NTR ECD	Homemade indirect quenching fluoro-immunoassay	n = 12 Group 1 (52–63 yr)	Group 1: 106 ± 18 ng/mg creatinine
			Group 2 (64–77 yr)	Group 2: 173 ± 12 ng/mg creatinine
			Group 3 (78–87 yr)	Group 3: 202 ± 37 ng/mg creatinine
	p75NTR ECD	Homemade ELISA	n = 45 23 women, 22 men Age: 50.0 ± 13.0 yr	3.6 ± 1.4 ng/mg creatinine	Shepheard et al., 2017
	p75NTR ECD	Homemade ELISA	n = 97 41 women, 56 men Age: 55.95 ± 13.19 yr	2.49 ± 2.07 ng/mg creatinine	Jia et al., 2017
	p75NTR ECD	Homemade ELISA	n = 21 11 women, 10 men Age: 60.6 ± 6.9 yr	3.2 ± 1.0 ng/mg creatinine	Shepheard et al., 2022
Plasma	p75NTR FL, ICD and ECD	Western Blot	n = 25	ICD: 0.139 ± 0.024 AU	Humpert et al., 2007
		Primary antibodies	Age: 59 ± 7 years	ECD: 0.069 ± 0.018 AU
		anti-NGFR ECD (1:400, Biosource International, Camarillo, CA, USA)	4 women, 21 men	FL: 0.130 ± 0.040 AU (visually deduced)
		anti-NGFR ICD (1:7500, Upstate, Chicago, IL, USA)
		anti-NGFR FL (NS)
		Secondary antibodies (1:4000, Santa Cruz Biotechnologies, Heidelberg, Germany)
	Not specified	Human NGFR ELISA kit, RayBiotech, Inc., Norcross, GA, USA	n = 120 (among a total of n = 250 77 women, 173 men Age: 44 ± 9 yr)	7.01 (6.89–7.14) pg/mL	Zakharyan et al., 2014
	Not specified	Human p75NTR ELISA Kit (mlbio, Shanghai, China)	n = 29 12 women, 17 men Age: 20 ± 1 yr	3.89 ± 0.46 ng/mL	He et al., 2019
	p75NTR ECD	ELISA (R&D Systems, Cat# DY367)	n = 607 358 women, 249 men Age: 64.3 ± 7.8 yr	169 (68) pg/mL, range 44–8040 pg/mL	Fleury et al., 2022
	p75NTR ECD	ELISA (R&D Systems, Cat# DY367)	n = 34 19 women, 15 men Age: 15 ± 1.81 yr	938.21 ± 1635.76 pg/mL	Zakowicz et al., 2023
Serum	p75NTR ECD	p75NTR-ECD ELISA assay (R&D Systems)	n = 129 53 women, 76 men Age: 71.7 ± 9.6 yr	Serum: 29.36 ± 20.21 pg/mL	Jiao et al., 2015
	Not specified	DuoSet human ELISA Development system (R&D Systems, Cat# DY367)	n = 30 14 women, 16 men Age: 36 ± 13 yr	1250 ± 250 pg/mL (visually deducted)	Chen et al., 2017
CSF	p75NTR ECD	p75NTR-ECD ELISA assay (R&D Systems)	n = 129 53 women, 76 men Age: 71.7 ± 9.6 yr	CSF: 661.7 ± 248.2 pg/mL	Jiao et al., 2015

Data are expressed as mean ± standard deviation, median (interquartile range) or min-max. AU: Arbitrary unit; CSF: cerebrospinal fluid; ECD: extracellular domain; ELISA: enzyme-linked immunosorbent assay; FL: full length; ICD: intracellular domain; NGFR: nerve growth factor receptor; p75^NTR: p75 neurotrophic receptor.

Table 2.

Summary of p75^NTR ECD levels variation in neurodegenerative diseases

Neurodegenerative diseases	Sample types	Assay reagents	Patients characteristics	Healthy participants (controls) characteristics	p75NTR ECD levels variation in patients vs. controls	References
ALS	Urine	Homemade ELISA and commercial ELISA (R&D Systems, Minneapolis, MN, USA)	n = 28 17 men, 11 women Age: 44–82 yr	n = 12 6 men, 6 women Age: 40–71 yr	Increased	Shepheard et al., 2014
		Homemade ELISA	n = 54 28 men, 26 women Age: 64.1 ± 13.2 yr	n = 45 22 men, 23 women Age: 50.0 ± 13.0 yr	Increased	Shepheard et al., 2017
		Homemade ELISA	n = 101 57 men, 44 women Age: 54.78 ± 10.61 yr	n = 97 56 men, 41 women Age: 56.81 ± 11.98 yr	Increased	Jia et al., 2017
		ELISA (Biosensis)	n = 70 50 men, 20 women Age: 63.9 ± 7.9 yr	n = 43 31 men, 12 women Age: 62.4 ± 7.2 yr	Increased	Yamada et al., 2021
		Homemade ELISA	n = 46 33 men, 13 women Age: 66.0 ± 11.7 years	n = 21 10 men, 11 women Age: 60.6 ± 6.9 yr	Increased	Shepheard et al., 2022
Alzheimer’s disease	Blood	qPCR	n = 26 13 men, 13 women Age: 71.8 ± 8.3 yr	n = 28 12 men, 16 women Age: 69.7 ± 9 yr	No significant difference	Xu et al., 2019
	Serum	ELISA (R&D Systems)	n = 156 73 men, 83 women	n = 129 76 men, 53 women	Increased	Jiao et al., 2015
	CSF		Age: 70.8 ± 15.2 yr	Age: 71.7 ± 9.6 yr	Decreased
		ELISA	n = 29	n = 27	Decreased	Yao et al., 2015
	Urine	Homemade ELISA	n = 62	n = 19	Increased for mild AD	Lindner et al., 1993
			20 men, 42 women (n = 31 mild AD & n = 31 moderate to severe AD) Age: 74 ± 1.2 yr	5 men, 14 women Age: 68.0 ± 2.8 yr	Decreased for moderate to severe AD group
Age-related dementia	Urine	Homemade agarose gel diffusion assay	n = 12 Group 1 (age: 52–63 yr) Group 2 (age: 64–77 yr)	n = 12 Group 1 (age: 52–63 yr) Group 2 (age: 64–77 yr)	Increased	Messripour et al., 2015
		Homemade indirect quenching fluoro-immunoassay	Group 3 (age: 78–87 yr)	Group 3 (age: 78–87 yr)	Increased
Schizophrenia	Plasma	Human NGFR ELISA kit, RayBiotech, Inc., Norcross, GA, USA	n = 120 patients with chronic schizophrenia receiving haloperidol	n = 120 age- and sex-matched	Decreased	Zakharyan et al., 2014
			n = 25 antipsychotic-free with a first psychotic episode		Decreased
		human p75NTR ELISA Kit (mlbio, Shanghai, China)	n = 30 confirmed schizophrenic patients (i.e., had a first psychotic episode) 21 men, 9 women Age: 20 ± 4 yr	n = 29 17 men, 12 women Age: 20 ± 1 yr	Increased	He et al., 2019
			n = 30 clinical high-risk individuals 21 men, 9 women Age: 20 ± 4 yr		Decreased
		ELISA (R&D Systems, Cat# DY367)	n = 45 20 men, 25 women Age: 15 ± 2 yr	n = 34 15 men, 19 women Age: 15 ± 2 yr	No significant difference	Zakowicz et al., 2023
	Serum	DuoSet human ELISA Development system (R&D Systems, Cat# DY367)	n = 34 15 men, 19 women Age: 34 ± 11 yr	n = 30 16 men, 14 women Age: 36 ± 13 yr	decreased	Chen et al., 2017
Diabetic neuropathy	Urine	Homemade ELISA	n = 98 65 men, 33 women Age: 58 ± 12 yr (men) Age: 63 ± 12 yr (women)	n = 45 18 men, 27 women Age: 39 ± 9 yr (men) Age: 35 ± 9 yr (women)	Increased	Hruska et al., 1993
	Plasma	Western blot	n = 80 59 men, 21 women Age: 58 ± 6 yr	n = 25 21 men, 4 women Age: 59 ± 7 yr	ICD: Increased ECD: Decreased FL: No significant difference	Humpert et al., 2007

ECD: Extracellular domain; ELISA: enzyme-linked immunosorbent assay; FL: full length; ICD: intracellular domain; NGFR: nerve growth factor receptor; p75^NTR : p75 neurotrophic receptor.

Amyotrophic Lateral Sclerosis

ALS is one of the neurodegenerative diseases in which p75^NTR has been the most studied as a potential biomarker. It is characterized by the degeneration of motor neurons leading to progressive loss of function and ultimately death. The diagnosis of ALS is based on clinical symptoms and the exclusion of other similar diseases. The current method used to monitor ALS is the Revised ALS Functional Rating Scale, a validated questionnaire based on motor, respiratory and bulbar functions that correlates with quality of life (Cedarbaum et al., 1999). However, questionnaires can be subjective to both the patient and the clinician and are subject to high levels of inter- and intra-individual variability. Reliable biomarkers that provide objective information about ALS progression are still lacking. Drug development for the treatment of ALS has been largely unsuccessful, resulting in very few therapeutic options. Repeated failures in ALS drug development have been attributed, in part, to delays in diagnosis and the lack of reliable biomarkers (Katyal and Govindarajan, 2017). Here, we review the potential of p75^NTR as a diagnostic, prognostic, and pharmacological biomarker for ALS.

p75^NTR as a diagnostic biomarker in ALS

In 2014, a study showed that urinary p75^NTR ECD levels were higher in ALS patients than in both healthy controls and patients with Parkinson’s disease or multiple sclerosis (Shepheard et al., 2014). This finding was confirmed in other studies using age- and sex-matched controls and in Chinese and Japanese cohorts of ALS patients, and was recently validated in a meta-analysis (Jia et al., 2017; Shepheard et al., 2017; Yamada et al., 2021; Shi et al., 2022). Using receiver operating characteristic (ROC) curves, Shepheard et al. (2014) showed that p75^NTR levels could discriminate ALS patients from healthy volunteers with 100% specificity and 93% sensitivity (area under the curve (AUC) of 1.0) and from Parkinson’s disease and multiple sclerosis with 79% specificity and 93% sensitivity (AUC of 0.96). Another study in a Japanese cohort also reported that urinary p75^NTR levels can discriminate ALS patients from healthy controls, although with a weaker ROC curve AUC value of 0.66 (Yamada et al., 2021). Taken together, these results suggest that urinary p75^NTR ECD may aid in the diagnosis of ALS when used in combination with other emerging ALS biomarkers such as urinary N-terminal titin fragment (a marker of muscle damage), neopterin (a marker of inflammation and immune activation), or blood neurofilament light chain (a marker of axonal damage) (Yamada et al., 2021; Shepheard et al., 2022). While none of these biomarkers is specific for ALS, the combination of multiple biomarkers improves diagnostic performance over their use in isolation (Yamada et al., 2021). Prospective studies evaluating the efficacy of these biomarkers compared to current symptom-based diagnosis are still lacking.

We recently reported high plasma p75^NTR ECD levels in approximately 15% of 1280 adults without cognitive impairment from the BEL-AGE cohort, which consisted of 673 patients with coronary artery disease, a risk factor for neurodegenerative disease, and 607 controls without coronary artery disease (Fleury et al., 2022). As p75^NTR ECD has been shown to occur before the onset of symptoms in mouse models (Shepheard et al., 2014), future follow-up and monitoring of this cohort will shed some light on whether the observed increase in p75^NTR ECD levels precedes the onset of ALS (and other neurological diseases with elevated p75^NTR levels) in humans.

p75^NTR as a prognostic biomarker in ALS

In addition to its potential diagnostic value, p75^NTR levels may also have prognostic value. Indeed, urinary p75^NTR ECD levels have been reported to be higher in ALS patients suffering from the aggressive (i.e., rapidly progressive) form of the disease compared to those with the slower progressive form (Shepheard et al., 2014; Jia et al., 2017). In support of these findings, Jia et al. (2017) further demonstrated that urinary p75^NTR ECD levels were higher in patients with advanced stages of ALS. However, even newly diagnosed patients (6 months or less since diagnosis) had significantly higher urinary p75^NTR ECD levels than healthy controls or patients with either Parkinson’s disease or multiple sclerosis (Jia et al., 2017). Taken together, these findings support that urinary p75^NTR ECD could help differentiate ALS from similar neurological disorders, but also contribute to the stratification of ALS stages.

Urinary p75^NTR ECD levels have been shown to correlate with the Revised ALS Functional Rating Scale (Jia et al., 2017; Shepheard et al., 2017, 2022). More precisely, Yamada et al. (2021) showed that urinary p75^NTR ECD specifically correlated with trunk, upper limb, and lower limb motor function as assessed by the revised ALS Functional Rating Scale. While the respiratory domain had a P value of 0.05, the bulbar domain did not correlate with urinary p75^NTR levels (Yamada et al., 2021). However, no difference in urinary p75^NTR ECD levels was reported whether the onset of ALS was spinal or bulbar (Shepheard et al., 2014; Jia et al., 2017). The variation in time of the revised ALS functional rating scale (ΔFRS), which decreases with ALS progression, has a prognostic value in terms of predicting the life expectancy from diagnosis. Interestingly, the patients with higher p75^NTR ECD levels are reported to have a shorter life expectancy than those with lower levels (Shepheard et al., 2017; Shi et al., 2022). Shepheard et al. (2022) recently published a longitudinal study in which they followed ALS patients and recorded the Revised ALS Functional Rating Scale and urinary p75^NTR ECD levels. They showed that p75^NTR levels were not relevant in predicting life expectancy in ALS patients in a multivariate model that included ΔFRS, meaning that urinary p75^NTR ECD levels did not improve the prognostic value of a model that already included ΔFRS (Shepheard et al., 2022).

p75^NTR as a pharmacological biomarker in ALS

Urinary p75^NTR levels have been shown to increase with ALS progression, suggesting that p75^NTR could be used as a tool to monitor ALS progression (Shepheard et al., 2022). Accordingly, p75^NTR may be an interesting pharmacodynamic biomarker that reflects treatment efficacy. In a recent phase 1 clinical trial, Henderson et al. (2021) evaluated the efficacy and safety of the CD14-targeting monoclonal antibody atibuclimab in 10 ALS patients. The antibody was administered in two dosing regimens (high and low dose) multiple times during the first 4 days of the 33-day follow-up period. Interestingly, urinary p75^NTR ECD decreased by 40% from baseline levels on day 4 of the high-dose treatment (Henderson et al., 2021). This change was not statistically significant (P = 0.051), likely due to the limited sample size (n = 7 ALS patients on the high-dose regimen). It is also worth noting that urinary p75^NTR ECD levels were not significantly lower 29 days after the last antibody injection compared to baseline. Further studies are needed to assess the impact of ALS treatment on p75^NTR ECD levels and thus its relevance as a pharmacological biomarker in ALS.

Alzheimer’s Disease

Another neurodegenerative disease in which p75^NTR has been studied as a potential biomarker is AD. It is the most prevalent form of dementia characterized by the buildup of Aβ plaques, Tau protein phosphorylation and secretion and progressive decline of cognitive function (Selkoe, 2001). AD is a heterogenous and multifactorial disease resulting in a broad range of genotypic and phenotypic presentations, a complex profile of cognitive impairment and different rates of progression. Demographic factors such as age, sex, ethnic differences and many others further contribute to the heterogeneity of the disease (Duara and Barker, 2022). This pathophysiological heterogeneity introduces uncertainty regarding the specificity and sensitivity of current biomarkers and diagnostic tools.

The most commonly used diagnostic tool for AD is the Mini-Mental State Exam, a screening tool for cognitive impairment (Devenney and Hodges, 2017). Several factors influence the reliability of the Mini-Mental State Exam score, including language/communication barrier, culture, and education levels, which together can lead to a delayed diagnosis, i.e. months to years following the onset of symptoms (Devenney and Hodges, 2017). Finding new biomarkers that are less dependent on the environment and that could improve the stratification of cognitive impairment in AD is of utmost importance.

In AD, Aβ stimulates astrocyte synthesis of p75^NTR resulting in hippocampal neuronal death (Sáez et al., 2006). In parallel, p75^NTR has also been reported to regulate Aβ deposition and Tau phosphorylation (Wang et al., 2011; Shen et al., 2019). However, there are some discrepancies in the literature regarding its levels in AD patients. Indeed, while Hu et al. (2002) have reported increased expression of p75^NTR in cortical neurons of AD patients, Salehi et al. (2000) saw a decrease in p75^NTR in the brain of AD patients. Notwithstanding, several studies have explored whether circulating levels of p75^NTR ECD could be used as biomarkers of cognitive health in the context of AD. Indeed, Aβ could reduce A disintegrin and metalloprotease 17 expression, and thus p75^NTR ECD shedding, therefore resulting in the decrease of its soluble level in AD (Yao et al., 2015).

Jiao et al. (2015) investigated AD patients and age-matched controls with no neurological disorders, with respect to their levels of p75^NTR ECD in serum and CSF as measured by ELISA. The study also investigated Aβ proteins, namely Aβ₄₀ and Aβ₄₂, total Tau protein and phosphorylated Tau (pTau181). Interestingly, serum p75^NTR ECD levels were significantly higher in the AD group than in the control group. Conversely, CSF p75^NTR ECD levels were significantly lower in the AD group compared to the control group (Jiao et al., 2015). Moreover, p75^NTR ECD levels were weakly but positively correlated with Aβ₄₂ and pTau181 levels in serum, while CSF p75^NTR ECD levels were negatively correlated with CSF pTau181 and total Tau. A second study from the same group confirmed the observed decrease of p75^NTR ECD in CSF of AD patients compared to age-matched controls (Yao et al., 2015). In a follow-up study, Xu et al. (2019) investigated whether p75^NTR mRNA in whole blood could be a potential AD biomarker. Its levels were compared between AD patients and controls and correlated with age, sex, the Mini-Mental State Exam score and ApoE genotype. No significant difference was observed between the groups and there was no correlation with the previously mentioned parameters. These findings suggest that, unlike protein levels, whole blood p75^NTR mRNA is not an adequate biomarker for AD diagnosis (Xu et al., 2019). In an attempt to unravel the molecular mechanism underpinning the interaction between p75^NTR and Aβ proteins, Spuch and Carro (2011) showed that in the choroid plexus, where CSF is produced, p75^NTR bound to Aβ crosses the choroid plexus barrier and reaches blood circulation. Whether this brain to blood transport of p75^NTR can explain the above-mentioned correlations between Aβ₄₂, pTau181 and p75^NTR levels in blood and CSF remains unknown.

Finally, Lindner et al. (1993) reported that p75^NTR ECD levels in urine vary in a U-shaped manner that could be relevant for AD patient stratification. In their study, they compared non-demented, mild AD and moderate to severe AD patients. While p75^NTR ECD levels were significantly higher in mildly demented patients, they were significantly lower in moderate to severe AD patients when compared to non-demented controls. Moreover, urine p75^NTR ECD levels seemed to decrease with age because when standardized to age and compared to the non-demented controls, only mild AD patients still had higher levels of p75^NTR ECD, but severe AD patients (older than control) showed no significant difference in p75^NTR ECD levels (Lindner et al., 1993).

Taken together, p75^NTR ECD could be an interesting biomarker for AD diagnosis and severity assessment. A better distinction between different stages and severities of AD would help draw better conclusions, calling for further research in this field.

Age-Related Dementia

p75^NTR utility as an emerging biomarker has been evaluated, although less extensively, in other neurodegenerative diseases such as age-related dementia. The diagnosis of dementia relies essentially on personal and family questionnaires as well as on behavioral criteria (Gale et al., 2018). A pilot study including 12 patients with dementia (aged between 52 and 87 years) and 12 age-, sex- and weight-matched healthy volunteers aimed to assess whether urine p75^NTR ECD levels could be used as a reliable biomarker (Messripour et al., 2015). p75^NTR ECD was measured in the first urine of the morning by indirect quenching fluoro-immunoassay performed on the same day and expressed as a function of the urine creatinine level, and by agarose gel diffusion. p75^NTR ECD was undetectable by agarose gel diffusion in control samples while it varied between (+) and (+++) in patient samples. Its concentration was approximately 2-fold greater in patients compared to control individuals with a positive correlation with age in both groups. Increased levels of p75^NTR ECD might be the result of the upregulation of p75^NTR in the brains of demented patients (DiStefano et al., 1991). Results of this pilot study should be confirmed in a large well-designed clinical trial to demonstrate the clinical relevance of increased urine p75^NTR ECD level in patients as a biomarker for age-related dementia diagnosis and evaluation of its progression.

Schizophrenia

Schizophrenia is another severe neurodegenerative disease with uncertain etiology and physiopathology to this day (Archer, 2010; Owen et al., 2011). Its accurate diagnosis is important for treatment, but no reliable biomarker has yet been approved making clinical criteria (i.e., symptomatic description without objective validation) the only diagnostic method currently available. The BDNF pathway proteins are implicated in synaptic plasticity, neuronal development and synaptogenesis, and play key roles in response to stress and anxious stimuli (McAllister et al., 1999). Consequently, they can have major roles in schizophrenia pathogenesis thus contributing to uncover novel objective diagnostic biomarkers and new target-oriented therapies.

In a single center prospective study, Zakharyan et al. (2014) measured plasma p75^NTR ECD by ELISA in Armenian individuals among which half had chronic schizophrenia and were receiving haloperidol, 25 had a first psychotic episode and were antipsychotic-free and 120 were age- and sex-matched healthy volunteers. Both treated and non-treated patients had significantly lower median p75^NTR ECD levels than controls. No difference was observed between schizophrenic patients receiving haloperidol or not. Moreover, this study also assessed the relationship between a p75^NTR single nucleotide polymorphism and p75^NTR ECD plasma levels. The rs2072446 single nucleotide polymorphism substituting serine for leucine at position 205 has been positively associated with schizophrenia. Indeed, a significant increase in p75^NTR ECD levels was seen in CC homozygotes (i.e. wild-type) compared to carriers of the minor T allele (CT+TT), both in patients and in controls (Zakharyan et al., 2014).

Another single-center prospective study assessed p75^NTR ECD levels in serum samples from 34 schizophrenic patients and age- and sex-matched healthy controls using ELISA. Levels of p75^NTR ECD were significantly decreased in patients, as were proBDNF and TrkB. ROC curve analysis revealed that combining serum levels of p75^NTR ECD with tissue plasminogen activator, plasminogen activator inhibitor 1, BDNF, proBDNF and TrkB was more accurate and powerful than its use in isolation for the diagnosis of schizophrenia (Chen et al., 2017).

Recently, He et al. (2019) measured p75^NTR ECD levels in plasma samples of confirmed schizophrenic patients (i.e. had a first psychotic episode), clinical high-risk individuals and age- and sex-matched healthy controls using ELISA. Unlike in the two previously mentioned studies (Zakharyan et al., 2014; Chen et al., 2017), p75^NTR ECD was significantly higher in the first group compared to the controls, but it was lower in the high-risk group of patients (He et al., 2019). Authors postulated that neurodegenerative processes might be more active in the prodromal stage resulting in a significant decrease of p75^NTR ECD, which subsequently induces a compensatory mechanism leading to the increase of plasma level of the cleaved receptor in the fully developed stage of schizophrenia. However, the cross-sectional design of the study was not amenable to test this hypothesis.

Finally, a recent study compared peripheral blood levels of many proteins, including p75^NTR ECD, between early-onset schizophrenia-spectrum adolescents and healthy controls (Zakowicz et al., 2023). No difference was observed in p75^NTR ECD levels between both groups, either at admission, or at 6 to 8 weeks of follow-up. However, p75^NTR ECD level was moderately correlated with the age at the onset of first suicidal thoughts, as well as with the number of suicide attempts, indicating a potential value as a monitoring biomarker of the disease course and treatment responsiveness (Zakowicz et al., 2023). Further replication cohort studies with larger datasets including drug-naïve and treated patients are needed to confirm these preliminary results. The distinction between the different types of schizophrenia should also be considered.

Diabetic Neuropathy

Diabetic patients are at increased risk of neurodegenerative diseases (Xu et al., 2011; Cheng et al., 2012; Yue et al., 2016; Zhang et al., 2017). Yet, no reliable biomarker exists to identify diabetic patients most likely to develop neuropathic complications.

Hruska et al. (1993) assessed p75^NTR ECD levels using a home-developed ELISA technique on urine samples, as a function of creatinine levels. Urine specimens were collected from healthy volunteers, diabetic and diabetic with neuropathy patients. p75^NTR ECD levels were significantly increased in diabetic patients with neuropathy when compared to either the respective sex-matched diabetic patients without neuropathy or the controls. This may result from the increase of p75^NTR concentration on Schwann cells in diabetic patients with neuropathy, leading to an enhanced turnover or more available receptors for cleavage (Sobue et al., 1988). Results of this study suggest that p75^NTR ECD levels could be used as a biomarker for diabetic neuropathy. It is worth mentioning that a thorough neurological characterization of included patients was lacking.

A second single-center clinical study assessed p75^NTR immunoreactivity by western blot analysis in albumin-depleted plasma samples from type 2 diabetes patients with normal renal function, screening for peripheral diabetic neuropathy and from sex- and age-matched healthy controls (Humpert et al., 2007). Three distinct p75^NTR fractions were analyzed: the full-length receptor, ICD and ECD. While immunoreactivity of the full-length receptor was similar, that of the ICD was significantly higher and of the ECD was lower in diabetic patients compared to controls. No correlation of either of these subunits was observed with the neuropathy disability score, arguing against the use of p75^NTR as a biomarker of peripheral neuropathy in diabetic patients (Humpert et al., 2007).

Considering the limited number of studies, further trials using more accurate measurement techniques of blood and/or urine p75^NTR levels are still needed to clearly establish the influence of hyperglycemia on p75^NTR expression and shedding as well as its reliability as a new biomarker that allows early diagnosis and prediction of associated neuropathies in diabetic patients.

Conclusion

Many studies on the potential use of p75^NTR ECD as an emerging easy-to-measure biomarker for neurodegenerative diseases have been conducted. They concur that p75^NTR could be interesting and relevant. However, these studies raise many questions, with large variations within a single disease population, as well as between different neurological diseases. Moreover, the specificity of p75^NTR ECD as a single biomarker for a particular neurodegenerative disease remains arguable and no diagnostic tool based on this biomarker is available yet. A combination of p75^NTR ECD with other biomarkers could be a better approach, a paradigm that remains to be assessed and compared against traditional diagnostic methods currently used.

The published studies are often limited by the small number of participants and therefore require validation with larger datasets. Furthermore, the diversity of the techniques used to measure p75^NTR ECD levels as well as the sample types selected add to the discrepancies seen in the literature, highlighting the need for a standardized protocol for p75^NTR quantification. A standardization of quantitative ELISA kits would be valuable, as homemade kits may significantly differ from one group to another, hindering the reproducibility of results. Equally important, sample type (whole blood, blood fraction, urine or CSF) introduces important variability making the comparison difficult between different studies. Further investigations are thus crucial to determine if p75^NTR ECD in different biological fluids can truly be used as a diagnostic biomarker of one or more neurodegenerative diseases and whether it is correlated with the disease progression and the response to the treatment.

Additional file: Open peer review report 1 ^{(82.7KB, pdf)} .