Increased cerebrospinal fluid chitinase 3-like 1 and neurofilament light chain in pediatric acquired demyelinating syndromes

https://doi.org/10.1016/j.msard.2018.05.017Get rights and content

Highlights

Abstract

Background

Chitinase 3-like 1 (CHI3L1), neurofilament light chain (NFL) and oligoclonal bands (OCB) in cerebrospinal fluid are associated with central nervous system demyelination in adults. CHI3L1 and OCB are markers of central nervous system inflammation, whereas NFL is a marker of white-matter axonal injury. The aim was to examine whether CHI3L1 and NFL in cerebrospinal fluid are associated with acquired demyelinating syndromes at disease onset in a pediatric population.

Methods

Children (<18 years) referred to hospital for possible neuroinflammatory disease were retrospectively included from 2010 to 2016. Case ascertainment was by review of medical records. NFL and CHI3L1 were measured by enzyme-linked immunosorbent assays. Endpoints were differences in concentrations of cerebrospinal fluid NFL and CHI3L1.

Results

We included 193 children who all underwent cerebrospinal fluid OCB examination as part of their diagnostic work-up and classified these children into 5 groups: acquired demyelinating syndromes (n = 33), normal diagnostic work-up (n = 36), inflammatory neurological disease (n = 50), other neurological disease (n = 55), and systemic inflammatory diseases (n = 19). NFL and CHI3L1 in cerebrospinal fluid differed significantly between the five groups (p = 0.0001). CHI3L1 was significantly higher in acquired demyelinating syndromes than in all other groups, and NFL was significantly higher in acquired demyelinating syndromes than in the other groups except systemic inflammatory disease. Children with acute disseminated encephalomyelitis had significantly higher concentrations of CHI3L1 than did children with multiple sclerosis.

Conclusion

We provide class II evidence that CHI3L1 and NFL are associated with pediatric acquired demyelinating syndromes. CHI3L1 may help distinguishing between acute disseminated encephalomyelitis and multiple sclerosis, but this needs further confirmation.

Introduction

Acquired demyelinating syndromes (ADS) are inflammatory demyelinating diseases of the central nervous system (CNS). They can be either monophasic or relapsing (Krupp et al., 2013, Boesen et al., 2018). Between 2–11% of persons with multiple sclerosis (MS) present before 18 years of age (Renoux et al., 2007). In young children, acute disseminated encephalomyelitis (ADEM) is the commonest manifestation of ADS; during adolescence, other ADS become increasingly common, e.g. MS, transverse myelitis (TM) and optic neuritis (ON), as well as other clinically isolated syndromes (CIS) (Boesen et al., 2018, Krupp et al., 2013, Neuteboom et al., 2008). However, in the initial stages of the disease, it can be difficult to determine the specific type of ADS and whether the ADS is more likely to be monophasic or relapsing. In this regard, recent studies in adults have shown that cerebrospinal fluid (CSF) levels of chitinase 3-like 1 (CHI3L1) and neurofilament light chain (NFL) may be useful biomarkers in adults, but their diagnostic value in childhood ADS is unclear (Comabella et al., 2010, Modvig et al., 2015, Hinsinger et al., 2015, Cantó et al., 2015, Khademi et al., 2013, Arrambide et al., 2016).

To date, the most sensitive and specific investigation for diagnosis and prognostication of ADS is magnetic resonance imagining (MRI) followed by CSF examination (Sadaka et al., 2012). ADEM is characterized by encephalopathy and multifocal neurological deficits including diffuse bilateral T2 white matter lesions on MRI but only presence of CSF oligoclonal bands (OCB) in 0–10% of all cases (Krupp et al., 2013, Mikaeloff et al., 2007, Tenembaum et al., 2002, Dale and Pillai, 2007, Atzori et al., 2009, Alper et al., 2009, Boesen et al., 2018); in contrast, MS is usually non-encephalopathic, typically presents with periventricular MRI lesions, and presence of OCBs in CSF in 35–81% (Krupp et al., 2013, Sadaka et al., 2012, Alper et al., 2009, Polman et al., 2011, Pohl et al., 2004, Chabas et al., 2010, Heussinger et al., 2015). However, studies have reported that 5% to 29% of pediatric ADEM patients progress to MS with a follow-up ranging 0.5–6.6 years (Neuteboom et al., 2008, Mikaeloff et al., 2007, Tenembaum et al., 2002, Dale and Pillai, 2007, Banwell et al., 2011, Suppiej et al., 2008). Accordingly, more sensitive and specific tests are needed.

In this aspect, two CSF biomarkers are of particular interest: CHI3L1 (also known as YKL40) and NFL. CHI3L1 is produced by macrophages, neutrophils, and astrocytes in the inflamed CNS and plays a role in inflammation and tissue response to injury (Lee et al., 2011, Bonneh-Barkay et al., 2010, Di Rosa et al., 2013). In adults, CHI3L1 has been shown to: a) be elevated in ADS, b) serve as a prognostic biomarker for conversion from monophasic demyelination to multiple sclerosis, and c) predict development of cognitive and physical disability in MS (Comabella et al., 2010, Modvig et al., 2015, Hinsinger et al., 2015, Cantó et al., 2015). NFL forms the basis of the neurofilament fiber abundant in myelinated axons. If elevated, NFL is a sensitive marker of white matter axonal injury (Khademi et al., 2013). In adults, increased levels of NFL in CSF following CIS is a risk factor for MS and provides prognostic information (Modvig et al., 2015, Khademi et al., 2013, Arrambide et al., 2016). In children, increased NFL has been associated with opsoclonus-myoclonus syndrome, CNS Langerhans cell histiocytosis, progressive encephalopathy, and infectious and inflammatory CNS disorders (Pranzatelli et al., 2014, Shahim et al., 2013, Osterlundh et al., 2008). However, it has not been studied in childhood ADS.

Our aim was to determine the diagnostic value of CHI3L1 and NFL in the CSF at onset in distinguishing pediatric ADS from non-ADS and, subsequently, ADEM from MS.

Section snippets

Study population

Children (<18 years) were retrospectively included from June 1, 2010 to July 1, 2016. The children were admitted for diagnostic work-up of possible neuroinflammatory disease including CSF analysis for OCB in the capital region of Copenhagen at either Rigshospitalet/Copenhagen University Hospital or Hvidovre Hospital. All CSF samples were obtained during the initial diagnostic work-up (acute phase). We accessed the CSF samples through a biobank at the Neuroimmunology Laboratory,

Baseline characteristics

Among all children (n = 193), 56% were girls and the median age at onset was 13.8 years (range: 1.3–17.9 years) (Table 1). This was similar for children with ADS (median: 15.0; range 2.5–17.9 years) and for the children of the control group (median: 15.5, range: 10.8–17.7 years). For children with monophasic ADS the median follow-up was 3.3 years (range = 0.9‒7.7 years).

The CSF analyses revealed that 23% had IgG OCB in CSF, 21% had an increased IgG index, and 26% had pleocytosis, which was

Discussion

In children (<18 years) referred to hospital for diagnostic work-up for neurological disease including CSF examination of IgG OCB, we show that CHI3L1 and NFL are significantly higher in those with ADS compared with those without ADS. Furthermore, CHI3L1 ≥ 283 µg/L distinguished ADEM and MS with 75% sensitivity and 93% specificity. To our knowledge, this is the first study investigating CHI3L1 and NFL in the CSF as diagnostic biomarkers for pediatric ADS.

In corroboration with our findings,

Conclusions

This study provides class II evidence that increased CSF concentrations of CHI3L1 and NFL are associated with ADS in children referred to hospital for diagnostic evaluation. Further, CHI3L1 may help in distinguishing between ADEM and MS at onset, although this needs to be further validated. Future studies should investigate the relationship between increased CSF concentrations of CHI3L1 and NFL and establish the optimal combination of biomarkers including cut-off levels to best predict the

Financial disclosure statements

Dr. Boesen has served on scientific advisory board for Teva; has received speaker honoraria for lecturing from Novartis and support for congress participation from Teva, Novartis and Roche.

Dr. Jensen reports no disclosures.

Dr. Magyari has served on the scientific advisory board for Biogen Idec, Novartis, and Merck Serono; has received honoraria for lecturing from Biogen Idec, Merck Serono, Novartis, and Genzyme; has received support for congress participation from Biogen Idec, Novartis,

Acknowledgements

We thank medical technologist Ulla Abildtrup for her invaluable help in performing the ELISAs, and the Danish MS Society and Karen A. Tolstrups Foundation for financial support.

Author Contributions

Dr. Boesen, Dr. Sellebjerg, and Dr. Jensen have contributed to the conception and design of the study, acquisition and analysis of data, and drafting a significant portion of the manuscript or figures. Dr. Magyari, Dr. Uldall, Dr. Born, and Dr. Blinkenberg have contributed to the conception and design of the study as well as commented the final version of the manuscript. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

Funding

The study was supported financially by the Danish MS Society (grant numbers: A29625 and A31526), Teva, Novartis, and Genzyme, but none of the sponsors took part in designing or analyzing data derived from the study.

References (36)

  • D.J.A. Callen et al.

    Role of MRI in the differentiation of ADEM from MS in children

    Neurology Mar 17.

    (2009)
  • E. Cantó et al.

    Chitinase 3-like 1: prognostic biomarker in clinically isolated syndromes

    Brain J. Neurol. Apr.

    (2015)
  • D. Chabas et al.

    Younger children with MS have a distinct CSF inflammatory profile at disease onset

    Neurology Feb 2.

    (2010)
  • M. Comabella et al.

    Cerebrospinal fluid chitinase 3-like 1 levels are associated with conversion to multiple sclerosis

    Brain J. Neurol. Apr.

    (2010)
  • R.C. Dale et al.

    Early relapse risk after a first CNS inflammatory demyelination episode: examining international consensus definitions

    Dev. Med. Child Neurol. Dec.

    (2007)
  • F. Deisenhammer et al.

    Guidelines on routine cerebrospinal fluid analysis. Report from an EFNS task force

    Eur. J. Neurol. Sep.

    (2006)
  • M. Di Rosa et al.

    Evaluation of CHI3L-1 and CHIT-1 expression in differentiated and polarized macrophages

    Inflammation Apr.

    (2013)
  • M.P. Gorman et al.

    Increased relapse rate in pediatric-onset compared with adult-onset multiple sclerosis

    Arch. Neurol. Jan.

    (2009)
  • Cited by (16)

    • Diagnostic Value of Oligoclonal Bands in Children: A Nationwide Population-Based Cohort Study

      2019, Pediatric Neurology
      Citation Excerpt :

      However, we reviewed the medical record in children with MS initially coded with a non-MS diagnostic code (e.g., optic neuritis) during 2008 to 2015. Taken together, children included in the chart review and children coded as MS (G35) presumably cover all registered cases of ADS during 2008 to 2015 in Denmark.10,15,33 The medical-record-validated diagnoses overruled the ICD-10 codes from the National Patient Register.

    • Pediatric autoimmune encephalitis in Denmark during 2011–17: A nationwide multicenter population-based cohort study

      2019, European Journal of Paediatric Neurology
      Citation Excerpt :

      We reviewed their medical records to estimate the incidence of “antibody negative but probable AIE”.2 For the chart review we collected data on date of onset, symptoms, treatment, clinical evolution and whether the children fulfilled the diagnostic criteria for AIE proposed by Graus et al.2 Further, we used a medical record-validated reference cohort of children suspected of neurological disease (n = 596) during 2008–2016 from our previous publications to determine the positive predictive value of ICD-10 codes used for anti-NMDAR encephalitis.25–27 We grouped children into diagnostic groups based on hospital diagnoses in the National Patient Register from two months before and one year after the date of AIE antibody testing.

    • How minimal variations in neuronal cytoskeletal integrity modulate cognitive control

      2019, NeuroImage
      Citation Excerpt :

      Only the latter seems to be modulated by the integrity of the neuronal cytoskeleton, but this functional nexus explains a staggering amount of behavioral performance variation, especially given that we only assessed healthy variation at the molecule level. This finding is of high relevance for research in clinical contexts using NF-L, in which the neurofilament parameter may be used for biomarker purposes since there are strong links between NF-L and the integrity of white matter structure as shown in studies on demyelinating diseases (Boesen et al., 2018; Khalil et al., 2018; Kuhle et al., 2016b, 2017; Melah et al., 2016). The current results show that even within a healthy control sample strong variations in behavioral or neuropsychological performance can be explained by tiny variations in sNF-L concentrations.

    • Neurofilament light chain: serum reference intervals in Danish children aged 0-17 years

      2023, Scandinavian Journal of Clinical and Laboratory Investigation
    • Prognostic Markers in Multiple Sclerosis

      2022, Neuroscience and Behavioral Physiology
    View all citing articles on Scopus
    View full text