Multiple Sclerosis Journal - October 2017 - 1445

N Dubuisson, F Puentes et al.
treatments. In SPMS, reduction in NfH levels have
been observed following treatment with the sodiumchannel blocker lamotrigine (a putative neuroprotectant) based on adherence, defined using detectable
serum levels, between 12 and 24 months, implying a
treatment lag of 12 months.16
Neurofilament antibodies
Biological plausibility. A bystander outcome in the
inflammatory response in MS is the generation of
autoantibodies to neuronal antigens. Neurofilament
antibodies form as a humoural immune response to
continuous Nf release following axonal injury. This
phenomenon has been observed in several disease
subtypes of MS.62,63 Both immunoglobulin G (IgG)
and immunoglobulin M (IgM) responses against
different neurofilament proteins occur, anti-NfL,
anti-NfM and anti-NfH.
Strength and specificity of association. Higher
levels of intrathecal IgG antibodies to NfL protein
are found in MS compared with other neurodegenerative disorders64 and are elevated even during the
progressive phase of the disease.62,65-67 Anti-NfM
intrathecal production of IgM and IgG antibodies are
also elevated in MS patients (relapsing-remitting,
primary and secondary progressive). Increased levels
of specific antibodies to NfH have also been demonstrated in CIS developing clinically definite MS.68
Consistency of findings. Although CSF anti-NfL,
-NfM and -NfH antibodies are elevated in MS, there
is no clear association with disease severity or the
various MS subtypes.64,66,69 Moreover, low levels of
antibodies have been detected in healthy controls,67
particularly anti-NfM antibodies which have been
detected in non-immune-mediated neurological disorders, such as headache and chronic fatigue syndrome.64 This will limit their usefulness as a
disease-specific biomarker.

Biological gradient. Reports of a relationship
between anti-NfL levels and disease onset and
progression in MS have been contradictory.66,67,69
However, there does appear to be treatment response
to natalizumab therapy, with a reduction in levels.69
Tau protein
Biological plausibility. Tau is another cytoskeleton
protein found primarily in neurons, its role being
to stabilize microtubules, thus facilitating fast axonal transport.70 Like neurofilament proteins, Tau is
released during axonal transection and is a potential
biomarker in MS, other neurodegenerative disorders, and central nervous system (CNS) injury as a
whole.71
Strength and specificity of association. CSF tau is
higher in MS compared to healthy controls,72-74
although the finding is not specific for MS.75 Levels
are substantially higher in those with prior relapses.76
CSF and serum measurements have been evaluated
in stroke77 and head injury,78 as well as MS,79 where
they are also elevated. In general, serum levels are
much lower than in the CSF.80
Consistency of findings. The report of elevated tau
in MS has not been a consistent finding, with three
studies reporting no significant differences between
MS and healthy controls.43,81,82 Conflicting findings
have been reported with relapses versus stable
subjects as well, with one study showing no difference observed between the two groups,83 while
another group found a significant positive correlation with the number of relapses before the lumbar
puncture.76

The utility of anti-Nf antibodies in sera remains to be
demonstrated. Ehling et al.67 noted that anti-NfL IgG
levels were elevated in both the serum and CSF in
MS. However, Bartos et al.64 found higher levels of
anti-NfM in sera, unrelated to CSF anti-NfM levels,
implying that disease status cannot be fully assessed
by isolated serum measurements alone.

Temporal sequence of association. Two groups have
demonstrated a lower CSF tau level in SPMS compared to RRMS, with a decrease in levels over the
course of disease. The low levels indicate completed
axonal degeneration in SPMS.84,85 In contrast, Brettschneider et al.86 and Kapaki et al.72 detected higher
levels in progressive MS (PPMS and SPMS) than
RRMS, suggesting continued axonal damage in progressive subtypes. Neither age nor disease duration
was grossly different between the sub-populations.
Therefore more work is required to look at this in
greater detail.

Temporal sequence of association. Anti-NfL and
anti-NfH levels have been correlated with disease
duration and EDSS levels;66 although in a study by
Bartos et al.,64 a similar correlation was not noted for
anti-NfM levels.

Biological gradient. CSF tau levels correlate with the
development brain atrophy84,87 and EDSS.88 Serum
tau has been evaluated pre- and post-mitoxantrone
treatment in MS (RRMS and SPMS), but there was no
significant change.79

journals.sagepub.com/home/msj

1445

https://journals.sagepub.com/home/msj

Table of Contents for the Digital Edition of Multiple Sclerosis Journal - October 2017