JOURNAL OF MAGNETIC RESONANCE IMAGING 38:1472–1479 (2013)

Original Research

Brain Iron MRI: A Biomarker for Amyotrophic

Lateral Sclerosis
Aleksandar Ignjatović, MSc,1 Zorica Stević, MD, PhD,2 Slobodan Lavrnić, MD,3
Marko Daković, PhD,1,3 and Goran Bačić, PhD1,4*

AMYOTROPHIC LATERAL SCLEROSIS, ALS, is the

Purpose: To evaluate the usefulness of MRI detection of
hypointensity areas (iron deposits) in the brain using a
dedicated MRI technique in patients with ALS in estab-
lishing this sign as a potential surrogate biomarker that
correlates with the severity of disease.
Materials and Methods: Forty-six ALS patients and 26
age-matched controls were examined by MRI. The ALS
Functional Rating Scale (ALSFRS) score was determined
before the first MRI examination. The sub-set of 25 ALS
patients was re-examined around 6 months after the first
MRI examination. The MRI examination consisted of rou-
tine T1W, T2W, and FLAIR sequences with the addition of
a thin slice heavily T2* weighted sequence to accentuate
magnetic susceptibility artifacts.
Results: T2*W sequence is superior to any other MRI
sequence in detecting hypointensities in the brain of ALS
patients. Hypointensities were found only in the precentral
gyruses gray matter (PGGM) and were detected in 42
patients. The extent of hypointensities was measured and
scored (0–3) and correlated with ALSFRS (r ¼ 0.545).
Twenty-five patients were re-examined 6 months later, and
the majority of them showed the shift toward higher MRI
scores. No control subjects had hypointensities in PGGM.
Conclusion: The detection of hypointensities in PGGM
appears to be a very promising surrogate MRI biomarker
for ALS due to its simplicity, high sensitivity and specific-
ity, suitability for longitudinal studies, and relationship
with the pathogenesis of the disease.
Key Words: amyotrophic lateral sclerosis; MRI; brain
iron; biomarkers
J. Magn. Reson. Imaging 2013;38:1472–1479.
C 2013 Wiley Periodicals, Inc.
V specificity because they also can be observed, albeit
1
Faculty of Physical Chemistry, University of Belgrade, Serbia
2
Department of Neurology, Faculty of Medicine, University of
Belgrade, Serbia
3
Center for Radiology and Magnetic Resonance, Clinical Center of
Serbia, Belgrade, Serbia
4 these two techniques (9–15). Both decreased frac-
National Cancer Research Center, Belgrade, Serbia
Contract grant sponsor: Ministry of Science of the Republic of Serbia;
Contract grant number: 41005.
*Address reprint requests to: G.B. Faculty of Physical Chemistry,
University of Belgrade, Studentski trg 12-16. 11000 Belgrade, Serbia.
E-mail: ggbacic@ffh.bg.ac.rs
Received July 27, 2011; Accepted February 20, 2013
DOI 10.1002/jmri.24121
View this article online at wileyonlinelibrary.com.
C 2013 Wiley Periodicals, Inc.
V 1472
most frequently encountered primary form of progres-
sive motoneuron disease. It is a devastating neurologi-
cal disorder affecting motoneurons and is
characterized by progressive muscle weakness and at-
rophy. Despite a uniformly fatal outcome, it has a
wide range of survival times, from a few months to a
several decades, with a consistent median of 2–4
years from the onset of symptoms (1). The cause can
be of genetic origin in approximately 20% of cases
(familial ALS, fALS); however, the disease is mostly of
the sporadic type (2). There is no reliable diagnostic
test for ALS, especially in the early stage, and diagno-
sis is mainly based on clinical assessments and relies
on the detection of upper motor neuron (UMN) and
lower motor neuron (LMN) signs in the limb and/or
bulbar territories, together with a history of progres-
sion of symptoms (3). Electromyography can detect
abnormalities of LMN, but there is no reliable tech-
nique for the UMN. Hence, there is a need for the
development of the biomarkers for ALS for both diag-
nostic purposes and especially for the progression of
the disease (an excellent review has been published
recently by Turner et al, (4).
MRI is arguably the best-suited technique for estab-
lishing an in vivo biomarker for the ALS. Traditional
MRI methods, such as T2W or FLAIR images, can
indicate the loss of myelinated fibers by means of
hyperintensity zones in white mater (WM) (5–7). How-
ever, these changes have a low sensitivity and limited
to a lesser extent, in elderly persons free of neurologi-
cal diseases (8). Hence, routine MRI exams are cur-
rently performed to exclude other pathologies rather
than to give a definite diagnosis.
Consequently, researchers turned to more sophisti-
cated MRI techniques, such as diffusion tensor imag-
ing (DTI) and spectroscopy (MRS), often combining
tional anisotropy (FA) on DTI images and the reduced
ratio of N-acetylaspartate to choline (NAA/Cho) and
creatine (NAA/Cr) on MRS have been often found in
various regions of the brain of ALS patients as com-
pared to the control. It has been found that these
changes precede the appearance of T2W-hyperinte-
sities, thus showing their diagnostic potential as an
Brain Iron MRI: A Biomarker for ALS
early indicator of ALS. MR voxel-based morphometry
also has been used to measure regional or global atro-
phy of WM and/or GM (16,17). Recently, a GM perfu-
sion has been studied using arterial spin labeling
(ALS) MRI (18).
Regardless of the approach, all of the above MRI
techniques are rather technically demanding and
interpretation of findings is not simple. In addition,
most of the studies have been performed on a rela-
tively small cohort of patients; therefore, it is difficult
to assess which one is a good candidate as a bio-
marker for ALS. Therefore, we investigated another
possibility of using MRI findings as a simple surrogate
marker for ALS.
Elevated amounts of non-heme iron in the brain
have been found in many neurodegenerative disorders
(as confirmed by the postmortem examination) such
as Alzheimer’s disease (AD), Parkinson’s disease (PD),
and Huntington disease (19). This finding opened the
possibility of using the MRI of brain iron as a bio-
marker exploring the iron-based magnetic susceptibil-
ity effect (20). For example, diffusively spread iron
accumulations in AD in certain regions of the brain
(e.g., putamen), can be assessed by measuring the
T2* effect (R2* relaxometry) (21). However, this is not
a very specific method because increased accumula-
tion of iron in these regions also occurs with aging.
Recently a similar approach has been applied to ALS
patients but with little success (22). Namely, it
appears that iron deposits in ALS are localized as
narrow stripes (MRI hypointensities) only in the pre-
central gyruses gray matter (PGGM). These hypointen-
sities were found on T2W images almost 2 decades
ago (23), and postmortem pathological findings
confirmed increased iron deposits in PGGM. Subse-
quently, numerous studies have been performed on
hyperintensities and hypointensities in ALS (24–28),
but most of them were focused on hyperintensities,
regarding hypointensities as a noticeable finding with
doubtful relevance. Some authors speculated that, if
these dark stripes are indeed iron deposits, it might
have a role in propagation of the disease by means of
free radical production, whereas others even ques-
tioned whether dark stripes in PGGM originate from
iron at all and ascribe their origin to the free radicals
themselves (28).
The aim of the study is to investigate the potential
of MRI detection of hypointensities in the brain of ALS
patients as a surrogate biomarker. The study is
directed toward establishing sensitivity, specificity,
and usefulness in longitudinal studies of this sign of
the disease, evaluated by MRI, with the clinical evalu-
ation of the stage of the disease (ALSFRS score).
MATERIALS AND METHODS
This prospective study was approved by the Institu-
tional Review Board of our Center. Informed written
consent was obtained from all patients. Forty-six
patients (15 men and 31 women; ages 38–78 years
[mean, 56 years]) were examined by MRI upon being
diagnosed for ALS using standard clinical tests. The
1473
revised ALS Functional Rating Scale (ALSFRS) score
was determined before the first MRI examination (30).
All patients, except two, had a sporadic ALS. Fourteen
patients had bulbar onset, and 32 limb onset. The
mean duration between the clinical onset of symptoms
and the first MRI examination was 14 6 6 months.
MRI examinations were performed using a 1.5 Tesla
(T) Siemens Avanto MRI scanner. Routine MRI exami-
nation consisted of a standard T1W (510/9.4 ms),
T2W (4800/108 ms), and FLAIR (9900/121/2500 ms)
images with a slice thickness of 5 mm. In addition, a
heavily weighted T2*W sequence (700/25 ms and

700/35 ms, flip angle 20 ), slice thickness 3 and 5
mm, was used on the first five patients for compari-
son of these sequences for visualizing hypointensities

in PGGM. The T2*W (700/35 ms, flip angle 20 , 3 mm
slice thickness) was selected for further examinations
(see Fig. 1). Visual evaluation of the MRI images was
performed independently by two readers, blinded to
the patient’s age and clinical diagnosis, but they were
alerted to note the presence of unusual hypointen-
sities on the images. Then the extent of hypointen-
sities was evaluated by measuring areas: the
hypointensities were manually divided in straight line
segments distributed in a manner which as much as
possible follows their profile. The area of segments
was determined by multiplication of their lengths with
corresponding thicknesses. Total area was obtained
by summation of areas of segments. Correlation
between measured areas and ALSFRS scores was
tested using Spearman’s test. Obtained data were di-
vided to groups using K-means cluster analysis (SPSS
v13) and a semi-quantitative MRI score was estab-
lished. Differences between groups were tested using
one-way analysis of variance (ANOVA) with Bonferroni
correction.
Using the procedure outlined above, we also studied
the following: (i) 25 ALS patients from the original
group were re-examined by MRI approximately 6
months after the first examination (other patients
were either in a poor condition or declined to be re-
examined); (ii) 20 age-matched patients (8 men and
12 women), without observable neurological deficit
(headaches and dizziness) were also examined using
MRI protocols appropriate to their conditions, but
with the addition of the T2*W MRI sequence. Seven
patients having AD or PD were also examined using
additional T2*W.
RESULTS
Much of the controversy regarding detecting hypoin-
tensities on MRI in ALS patients can be resolved by
the analysis of various MRI sequences used for that
purpose. Figure 1 illustrates the effectiveness of dif-
ferent MRI sequences in detecting hypointensities in
PGGM. The dark stripe is clearly more visible on the
T2*W image (Fig. 1c) than on the standard T2W image
(Fig. 1a). This is due to the superparamagnetic T2*
effect of presumably iron deposits and the fact that
the size of the signal void on T2*W images can greatly
exceed the actual volume that contains the agent
1474
which creates the susceptibility effect. The presence
of dark stripes is virtually imperceptible on FLAIR
images because the bright signal from cerebrospinal
fluid (CSF) is nullified. Yet some authors (28) claimed
that this is the most effective MRI sequence for detect-
ing hypointensities. The recommendation that FLAIR
should be used to detect iron deposits in PGGM is
equivalent to the claim that dark stripes can be spot-
ted more easily on a black background when com-
pared with a white background.
The bottom row in Figure 1 illustrates two more im-
portant points in MRI detection of iron deposits: (i)
reduction of slice thickness from the routine 5 mm to
3 mm is crucial when attempting to visualize hypoin-
tensities in PGGM (compare Fig. 1c and Fig. 1d)
because it reduces the partial volume effect along
curved surfaces, which is especially important when
imaging patients at the early stage of the disease hav-
ing small hypointensities; and (ii) prolongation of TE
is also important because it exaggerates the suscepti-
bility effect (compare Fig. 1d and Fig. 1e).
Henceforth, all evaluations of hypointensities in
PGGM have been performed using a T2*W sequence
with TE ¼ 35. The results are presented in Table 1.
Hypointensities in the PGGM were detected in 42 of
our patients (92%). Four patients having no hypoin-
tensities on MRI were at the very early stage of the
disease (high average ALSFRS of 42). Only one patient
showed pronounced hypointensities in areas other
than PGGM (putamen), based on a routine T2W
sequence covering the entire brain. Virtually none of
the non-ALS patients showed hypointensities in the
PGGM (one patient can probably be classified as the
MRI score 1, see below). None of our AD/PD patients
showed hypointensities in PGGM.
Ignjatović et al.
Figure 1. Different axial MRI
images of the same ALS
patient illustrating the effect
of different MRI protocols and
different slice thickness in
detection of the hypointen-
sities in the precentral
gyruses (iron deposits). The
arrows point to detected hypo-
intensities. a–c: Top row ¼ 5-
mm slices, T2W (a), FLAIR (b),
T2*W, TE ¼ 25 ms (c). d,e:
Bottom row ¼ T2*W, 3-mm sli-
ces, TE ¼ 25 ms (d), TE ¼ 35
ms (e).
Hypointensities were bilateral in the majority of
patients (85%), while in the remaining 15%, they were
exclusively located in the left hemisphere, which is
consistent with reports that the left hemisphere is
more affected than the right one (18,22,29). We found
no significant difference in the appearance of hypoin-
tensities between patients having bulbar or limb
onset, although patients with bulbar onset tend to
have higher MRI score, which is consistent with clini-
cal observations that patients with bulbar onset have
more aggressive disease progression. It is impossible
to speculate about possible differences between fALS
and sALS because we had only 2 fALS patients.
We found correlation between areas of hypointen-
sities in T2*W images and ALSFRS scores (r ¼ 0.545;
P < 0.01 Spearman’s correlation test). Based on the
histogram and cluster analysis (Fig. 2), we divided our
patients into four categories: MRI score 0 ¼ no hypoin-
tensity; MRI score 1 ¼ noticeable (areas bellow 30
mm2); MRI score 2 ¼ clearly visible (areas between 31
and 80 mm2); MRI score 3 ¼ pronounced (areas above
80 mm2). Significant differences were found between
Table 1
MRI Score of Hypointensities in PGGM in ALS Patients at the
Initial MRI Examination*
MRI score No. of patients
0 4
1 11
2 20
3 11
*MRI score 0 ¼ no hypointensity; MRI score 1 ¼ noticeable (areas
bellow 30 mm2); MRI score 2 ¼ clearly visible (areas between 31
and 80 mm2); MRI score 3 - pronounced (areas above 80 mm2).
Brain Iron MRI: A Biomarker for ALS
Figure 2. Histogram of the area of hypointensities in ALS
patients at first MRI examination. Based on cluster analysis,
patients are divided into groups: MRI score 0 ¼ no hypointen-
sity; MRI score 1 ¼ noticeable (area below 30 mm2); MRI
score 2 ¼ clearly visible (area between 31 and 80 mm2); MRI
score 3 ¼ pronounced (area above 80 mm2). Dashed lines
denote group borders.
groups (P < 0.01, one-way ANOVA with Bonferroni cor-
rection). Table 1 shows the number of patients in
each category. Figure 3 shows the relationship
between the ALSFRS score and average area in
groups. There is an obvious pattern of the increased
area of hypointensities with decreased ALSFRS score.
Correlation coefficient obtained using Pearson test is
0.921 (P < 0.01).
Figure 4 illustrates changes between the first and
second MRI examination. Table 2 shows results
obtained for 25 patients at the second MRI examina-
tion performed approximately 6 months after the first
one. Upon re-examination, all patients showed hypo-
intensities, and 17 patients showed an increased MRI
Figure 3. Correlation between the MRI score and ALSFRS in
ALS patients at the initial MRI examination. One-way ANOVA
with Bonferroni correction was used to compare groups. Sig-
nificant differences (P < 0.01) were found between groups.
Pearson correlation value is 0.921 (P < 0.01).
1475
Figure 4. Progression of the extent of hypointensity on MRI
images (T2*W; TE ¼ 35, slice ¼ 3mm) in one ALS patient. a:
Initial examination; ALSFSR ¼ 39, MRI score ¼ 2; area ¼ 40
mm2. b:) At 6 months later; ALSFRS ¼ 30, MRI score ¼ 3;
area ¼ 94 mm2.
score. Three patients from category 1 and 5 patients
from category 2 remained in the initial broad catego-
ries, but the increase of the area of hypointensities
was noticed in all cases (16% average area increase).
Correlation with ALSFRS based on the same criteria
used to form Figure 3 for the subset of re-examined
patients is shown in Figure 5. Again, a pattern of the
increased area of hypointensities with decreased
ALSFRS score is observed, indicating the ability of
MRI T2*W to monitor the progression of the disease.
Hyperintensities in the subcortical white matter
were found in 27 patients (58%). The number and
extent of hyperintensities showed a weak correlation
to the ALSFRS as well as to the MRI T2*W score,
except that all patients having the MRI T2*W score
equal to 3 also had pronounced hyperintensities.
DISCUSSION
Biomarkers can be defined as “an objective measure-
ment that acts as an indicator of normal biological
processes, pathogenic processes, or pharmacological
responses to therapeutic intervention” (UK Medical
Research Council). The definition is broad enough to
encompass various potential biomarkers, but any
successful biomarker should fulfill additional criteria:
to have a high specificity and sensitivity, both in diag-
nosis and longitudinal studies. Ideally, such a
Table 2
Change of MRI Score in 25 ALS Patients Between the First and
Second Examination
MRI score No. of patients
0 ! 0 0
0 ! 1 2
0 ! 2 2
1 ! 1 3
1 ! 2 3
1 ! 3 0
2 ! 2 5
2 ! 3 7
3 ! 3 3
1476
Figure 5. Correlation between the MRI score and ALSFRS in
ALS patients at re-examination. Pearson correlation value is
0.992 (P < 0.04).
biomarker also should contribute to the better under-
standing of the pathogenesis of the disease. We
believe that MRI detection of hypointensities in PGGM
fulfils most of the above criteria to be qualified as a
surrogate in vivo biomarker for ALS.
Sensitivity
Hypointensities in PGGM were found in 42/46 of our
patients (92%), which is considerably higher than in
other studies where the detected incidence was
around 40% (7,25–28). Table 2 and Figure 4 show
that the incidence and intensity of dark stripes
increases with the prolongation of the disease, con-
sistent with findings in other studies (23,27). Conse-
quently, the duration of the disease has to be taken
into account when analyzing these results. In our
study, the time since onset was 14 6 6 months, while
in quoted studies it was 20 6 5 months. Therefore, the
differences in sensitivity are not due to the different
time period but due to the different MRI techniques
used. Here, we used a technique dedicated to enhance
T2* artifacts, whereas in other studies, the authors
were more focused on detecting hyperintensities in
WM and used standard techniques (T2W and/or
FLAIR) to visualize these changes, while detecting
hypointensities was of the secondary importance. The
only study performed using routine T2W with a 5-mm
slice thickness and having similar sensitivity (13/15),
has been performed on the group of patients having
an extremely long period of around 32 months
between the onset and MRI examination (23), i.e.,
when these structures are pronounced even on a rou-
tine T2W MRI. There is little doubt that sensitivity of
detecting iron can be substantially enhanced by using
higher magnetic fields due to the well known fact that
the prominence of T2* artifacts increases with the
strength of magnetic field (20,31). The MRI with
higher field strength also enables the efficient use of
specialized techniques such as susceptibility induced
Ignjatović et al.
phase contrast, which should produce images with
high spatial resolution (32) and more precise anatomi-
cal localization of iron deposits.
Detection of hyperintensities in our ALS patients of
58% is close to the average value of around 53%
found in other studies (6,7,23,25–27). This is a much
lower sensitivity than sensitivity of hypointensity
signs and is also lower than in DTI or MRS studies,
especially at the early stage of disease. Therefore, this
approach should be ruled out as an MRI surrogate
biomarker, and its use as a reliable sign of ALS for
diagnostic purposes is questionable.
It is difficult to compare sensitivity or correlation
with clinical findings obtained in this study with
results obtained using other studies of MRI surrogate
markers (DWI/DTI or MRS), because authors used
different clinical signs, in addition to ALSFRS, to diag-
nose and score ALS (Appel ALS scale, disease dura-
tion, presence or severity of UMN dysfunction, Alswort
spasticity scale, Norris scale score, etc.). Some DTI
studies reported correlation of decreased FA with the
selected clinical score in the range r ¼ 0.3–0.63 (9,13–
15,29) using different methods of correlation. How-
ever, no correlation with ALSFRS has been found in
any of these studies and in few others (22,33). On the
other hand, it appears the information from DTI anal-
ysis might be limited only to the early stage of disease
and is not useful in longitudinal studies because no
changes in FA or MD have been observed with the
progression of the disease (34,35). A similar conclu-
sion can be drawn from MRS studies of ALS patients.
The correlations of the reduction of NAA/Cho and
NAA/Cr ratios in various parts of the brain with the
clinical sign of the choice are in the range r ¼ 0.2–0.65
(11,15,36–39), but some studies failed to correlate
MRS findings with any score for the ALS (13,40).
Again, only in one study the correlation of MRS find-
ings with ALSFRS (r ¼ 0.5) has been reported (11). In
longitudinal studies, a weak correlation has been
found between MRS parameters and duration/sever-
ity of disease (39,41). An interesting MRS study where
concentration of myo-inositol (Ins) in precentral
gyruses has been measured (42) showed that the
decreased concentration of Ins correlates with the se-
verity of the disease, but also correlates with the
appearance of hypointensities in PGGM. A relatively
high correlation between the perfusion (ALS MRI) in
certain parts of the brain and ALSFRS (up to r ¼ 0.85)
has been found (25), but the study has been per-
formed on a limited number of patients with no con-
trols, so it would be premature to evaluate this MRI
technique as a potential surrogate marker for ALS.
There is no doubt that DTI/DWI and MRS have a
potential of being a surrogate marker for ALS, but
questionable sensitivity, particularly in the longitudi-
nal studies is a serious drawback. From the entire
above discussion, it appears that MRI measurement
of hypointensities in PGGM is a very promising MR
marker in terms of sensitivity, correlation with clinical
findings, and successful in follow-up studies. It is
also technically not demanding and requires little
scanning time so it can be easily incorporated into
any clinical protocol. Of importance is also that we
Brain Iron MRI: A Biomarker for ALS
demonstrated that measurements of hypontensities
as biomarkers for ALS can be successfully performed
on 1.5T scanners, which are currently much more
widely available than 3T scanners. Larger introduc-
tion of high field MRI scanners into clinical practice
should only increase the value of determining hypoin-
tensities as biomarkers in ALS patients.
Specificity
It appears that hypointensities in PGGM are highly
specific for ALS. None of our control patients had that
sign and other authors reported its incidence of
approximately 4% in control groups (23,26,27). In an
extensive study (8) on hyper- and hypointensities in
various neurological disorders (including ALS but pre-
dominantly AD), a higher percentage of hypointen-
sities in PGGM has been reported (approximately
15%), but the majority of patients contributing to this
high overall percentage were older than 70 years old.
Increased iron content seems to be regularly
detected in almost all neurological disorders, but also
in elderly persons with no neurological diseases. In a
majority of diseases and in older patients, increased
iron content is generally diffusively spread in certain
regions of the brain (e.g., putamen and substantia
nigra) and cannot be easily visualized as a single sus-
ceptibility artifact due to the insufficient resolution of
clinical scanners. However, these changes can be
detected and iron content can be potentially quanti-
fied by measuring R2* maps in these regions, because
it has been confirmed that the values of R2* nicely
correlate to the amount of iron determined chemically
on postmortem specimens (43). The use of R2* as a
surrogate marker has been used in studies of the AD
and PD patients with some success (21,44,45), but
has been unsuccessful in ALS patients (22) because
no increased values of R2* have been found in the
comparison to the control group. These results seem
to confirm the results of this study that the elevated
iron content in ALS patients is almost completely
packed as narrow stripes in PGGM, so that in terms
of specificity this is the most specific MR surrogate
marker for ALS.
What Is so Special About Precentral Gyruses in
ALS and Why Would Iron Accumulate in PGGM?
It is difficult to speculate why hypointensities (iron
deposits) are located solely in PGGM of ALS patients
and connect these findings to the pathogenesis of
ALS. However, apart from measurements of hypoin-
tensities, most all other MRI techniques such as mor-
phometry, DTI, and MRS showed that first or most
pronounced changes occur in precentral gyruses
(13,15,16,42,46). The same is also true for some non-
MRI studies, such as SPECT (single photon emission
computed tomography) (47). According to Fukunaga
(48), iron in cerebral cortex is colocalized with myelin-
ated axons in cerebral layers and one can speculate
that accumulation of iron in the sixth layer could
explain the appearance of dark stripes at the interface
1477
of gray and white matters that are parallel to the
gyrus profile (see also Note Added in Proof).
Are Iron Deposits the Sole Origin of
Hypointensities in PGGM in ALS Patients?
The only report connecting hypointensities in PGGM
in ALS patients with post-mortem histopathology sec-
tions of PGGM, albeit on the limited number of
patients, confirmed intense iron deposits in this area
(23). From the MRI point of view, there is no known
structure other than iron deposits that can produce
susceptibility artifact in that section of the brain. So it
has been taken for granted by the majority of authors
(24–27) that hypointensities in the PGGM are the fin-
gerprint of the accumulated iron. Nevertheless, some
authors claim that the hypointensities in PGGM are
not due to iron deposits and vaguely offered an alter-
native source such as free radicals (28). The enlarge-
ment of the hypointensity area with prolongation of
the TE as demonstrated here is an unequivocal proof
that hypointensities in PGGM are due to the suscepti-
bility artifacts. There are no known free radicals
(unpaired electron species) that can produce (super)-
paramagnetic susceptibility artifacts on MRI, unless
certain organic compounds having delocalized elec-
trons are clumped within solid particles and inten-
tionally introduced into the subject to measure e.g.
tissue oxygenation using in vivo EPR spectroscopy
(49,50). Another suggestion (25) that hypointensities
might be a consequence of senescent changes such as
increased lipofuscin granules (wear-and-tear pig-
ments) is probably worth considering when hypoin-
tensities are found in aged patients. However, it is
difficult to imagine that these can correlate with
ALSFRS or significantly change during the 6-month
follow-up period as found here.
What Is the Relevance of Iron Deposits in ALS?
The elevated amounts of iron deposits have been
found in numerous neurological diseases, but the
mechanism for this iron mismanagement is still
unclear and can be related to numerous metabolic
processes (51) including superoxide-mediated release
of iron from ferritin (52). Free iron (unlike iron within
soluble stable ferritin) when converted into hemosid-
erin and other oxyhydroxide derivatives becomes
more likely to exchange electrons with surrounding
molecules, and produce free radicals (in particular
hydroxyl radical •OH) by means of the Fenton reac-
tion. Such mechanisms have been considered as a
major culprit in the pathophysiology of different neu-
rological diseases including ALS (19); however, there
is no direct in vivo evidence.
In conclusion, the importance of establishing that
there is a free or loosely bound iron in the brain of
ALS patients is invaluable in designing new therapeu-
tic drugs for ALS, because some studies demonstrated
the effectiveness of chelation therapy that removes
iron in PD and AD patients (53,54). Iron chelation
therapy has been successfully applied in mutant
SOD1 G93A mice model of ALS (55).
1478
NOTE ADDED IN PROOF
Kwan et al recently published a similar study (using
3T and 7T scanners) on a limited number of patients
and demonstrated that the motor cortex hypointen-
sities are due to increased iron accumulation (56).
REFERENCES
1. Beghi E, Logroscino G, Chio A, et al. The epidemiology of ALS and
the role of population-based registries. Biochim Biophys Acta
2006;1762:1150–1157.
2. Bruijn LI, Miller TM, Cleveland DW. Unraveling the mechanisms
involved in motor neuron degeneration in ALS. Annu Rev Neuro-
sci 2004;27:723–749.
3. Brooks BR, Miller RG, Swash M, Munsat TL. El Escorial revisite-
d:revised criteria for the diagnosis of amyotrophic lateral sclero-
sis. Amyotroph Lateral Scler Other Motor Neuron Disord
2000;1:293–299.
4. Turner MR, Kiernan MC, Leigh PN, Talbot K. Biomarkers in
amyotrophic lateral sclerosis. Lancet Neurol 2009;8:94–109.
5. Ellis CM, Simmons A, Dawson JM, Williams SCR, Leigh PN, Dis-
tinct hyperintense MRI signal changes in the corticospinal tracts
of a patient with motor neuron disease. ALS and other motor neu-
ron disorders 1999;1:41–44.
6. Hecht MJ, Fellner F, Fellner C, Hilz MJ, Heuss D, Neundorfer B.
MRI-FLAIR images of the head show corticospinal tract altera-
tions in ALS patients more frequently than T2-, T1- and proton-
density-weighted images. J Neurol Sci 2001;186:37–44.
7. Zhang L, Ulug AM, Zimmerman RD, Lin MT, Rubin M, Beal MF.
The diagnostic utility of FLAIR imaging in clinically verified amyo-
trophic lateral sclerosis. J Magn Reson Imaging 2003;17:521–
527.
8. Ngai S, Tang YM, Du L, Stuckey S, Hyperintensity of the precen-
tral gyral subcortical white matter and hypointensity of the pre-
central gyrus on fluid-attenuated inversion recovery: variation
with age and implication for the diagnosis of amyotrophic lateral
sclerosis, AJNR Am J Neuroradiol 2007;28:250–254.
9. Elis CM, Simmons A, Jones DK, et al. Diffusion tensor MRI
assesses corticospinal tract damage in ALS. Neurology 1999;53:
1051–1058.
10. Hong YH, Lee KW, Sung JJ, Chang KH, Song IC. Diffusion tensor
MRI as a diagnostic tool of upper motor neuron involvement in
amyotrophic lateral sclerosis. J Neurol Sci 2004;227:73–78.
11. Wang S, Poptani H, Woo JH, et al. Amyotrophic lateral sclerosis:
diffusion-tensor imaging and chemical shift MR imaging at 3.0 T.
Radiology 2006;239:831–838.
12. Wong JCT, Concha L, Beaulie C, Johnston W, Allen PS, Kaira S.
Spatial profiling of the coricospinal tract in amyotrophic lateral
sclerosis using diffusion tensor imaging. J Neuroimaging
2007;17:234–240.
13. Charil A, Corbo M, Filippi M, et al. Structural and metabolic
changes in the brain of patients with upper motor neuron disor-
ders: a multiparametric study. Amyotroph Lateral Scler 2009;10:
269–279.
14. Agosta F, Pagani E, Petrolini M, et al. MRI predictors of long-term
evolution in amyotrophic lateral sclerosis. Eur J Neurosci
2010;32:1490–1496.
15. Pyra T, Hui B, Hanstock C, et al. Combined structural and neuro-
chemical evaluation of the corticospinal tract in amyotrophic lat-
eral sclerosis. Amyotrophic Lateral Scler 2010;11:157–165.
16. Grosskreutz J, Kaufmann J, Fradrich J, Dengler R, Heinze HJ,
Peschel T. Widespread sensorimotor and frontal cortical atrophy
in amyotrophic lateral sclerosis. BMC Neurol 2006;6:17.
17. Mezzapesa DM, Ceccarelli A, Dicuonzo F, et al. Whole-brain and
regional atrophy in amyotrophic lateral sclerosis. AJNR Am J
Neuroradiol 2007;28:255–258.
18. Rule R, Schuff N, Miller R, Weiner M, Gray matter perfusion cor-
relates with disease severity in ALS, Neurology 2010;74:821–827.
19. Stankiewicz J, Panter SS, Neema M, Arora A, Batt CE, Bakshi R.
Iron in chronic brain disorders: imaging and neurotherapeutic
implications. Neurotherapeutics 2007;4:371–386.
20. Schenck JF, Zimmerman EA. High-field magnetic resonance
imaging of brain iron: birth of biomarker? NMR Biomed
2004;17:433–445.
Ignjatović et al.
21. Bartzokis G, Sultzer D, Cummings J, et al. In vivo evaluation of
brain iron in Alzheimer disease using magnetic resonance imag-
ing. Arch Gen Psychiatry 2000;57:47–53.
22. Langkammer C, Enziger C, Quasthoff S, et al. Mapping of iron
deposition in conjunction with assessment of nerve fiber tract in-
tegrity in amyotrophic lateral sclerosis. J Magn Reson Imaging
2010;31:1339–1345.
23. Oba H, Araki T, Ohtomo H, et al. Amyotrophic lateral sclerosis:
T2 shortening in motor cortex at MR imaging. Radiology
1993;189:843–846.
24. Ishikawa K, Nagura H, Yokota T, Yamonauchi H. Signal loss in
the motor cortex on magnetic resonance imaging in amyotrophic
lateral sclerosis. Ann Neurol 1993;33:218–222.
25. Cheung G, Gawal MJ, Cooper PW, Farb RI, Ang LC, Amyotrophic
Lateral Sclerosis: correlation of clinical and MR imaging findings,
Radiology 1995;194:263–270.
26. Basak M, Erturk M, Oflazoglu B, et al. Magnetic resonance imag-
ing in amyotrophic lateral sclerosis. Acta Neurol Scand 2002;105:
395–399.
27. Hecht MJ, Fellner F, Fellner C, et al. Hyperintense and hypoin-
tense MRI signals of the precentral gyruses and corticospinal
tract in ALS: a follow-up examination including FLAIR images. J
Neurol Sci 2002;199:59–65.
28. Hecht MJ, Fellner C, Schmid A, Neundorfer B, Fellner FA. Corti-
cal T2 signal shortening in amyotrophic lateral sclerosis is not
due to iron deposits. Neuroradiology 2005;47:805–808.
29. Wong JCT, Concha L, Beaulleu C, Johnston W, Allen PS, Kalra S.
Spatial profiling of the corticospinal tract in amyotrophic lateral
sclerosis using diffusion tensor imaging. J Neuroimaging
2007;17:234–240.
30. Brooks BR, Miller RG, Swash M, Munsat TL. El Escorial revisited:
revised criteria for the diagnosis of amyotrophic lateral sclerosis.
Amyotroph Lateral Scler Other Motor Neuron Disord 2000;1:293–
299.
31. Duyn JH. High-field MRI of brain iron. Methods Mol Biol 2011;
711:239–249.
32. Duyn JH, van Gelderen P, Li TQ, de Zwart JA, Koretsky AP,
Fukunaga M. High-field MRI of brain cortical substructure based
on signal phase. Proc Natl Acad Sci U S A 2007;104:11796–
11801.
33. Toosy AT, Werring DJ, Orrell, RW, et al. Diffusion tensor imaging
detects corticospinal tract involmenent at multiple levels in amyo-
trophic lateral sclerosis. J Neurol Neurosurg Psychiatry 2003;74:
1250–1257.
34. Blain CRV, Williams VC, Johnston C, et al. A longitudinal study
of diffusion tensor MRI in ALS. Amyotrop Lateral Scler 2007;8:
348–355.
35. Agosta F, Rocca MA, Valsasina P, et al. A longitudinal diffusion
tensor MRI study of the cervical cord and brain in ALS patients.
J Neurol Neurosurg Psychiatry 2009;80:53–55.
36. Sarchielly P, Pelliccolli GP, Tarducci R, et al. Magnetic resonance
imaging and (1)H-magnetic resonance spectroscopy in amyotro-
phic lateral sclerosis. Neuroradiology 2001;43:189–197.
37. Elis CM, Simmons A, Andrews C, Dawson JM, Williams SCR,
Leigh PN. A proton magnetic resonance spectroscopic study in
ALS. Correlation with clinical findings. Neurology 1998;51:1104–
1109.
38. Rooney WD, Miller RG, Gelinas D, Schuff N, Maudsley AA, Weiner
MW. Decrease in N-acetylaspartate in motor cortex and cerebro-
spinal tract in ALS. Neurology 1998;50:1800–1805.
39. Pohl C, Block W, Karitzky J, et al. Proton magnetic resonance
spectroscopy of the motor cortex in 70 patients with amyotrophic
lateral sclerosis. Arch Neurol 2001;58:729–735.
40. Yin H, Lim CCT, Ma L, et al. Combined MR spectroscopic imaging
and diffusion tensor MRI visualizes corticospinal tract degenera-
tion in amyotrophic lateral sclerosis. J Neurol 2004;251:1249–
1254.
41. Suhy J, Miller RG, Rule R, et al. Early detection and longitudinal
changes in amyotrophic lateral sclerosis by (1)H MRSI. Neurology
2002;58:773–779.
42. Bowen BC, Pattany PM, Bradley WG, et al. MR imaging and local-
ized proton spectroscopy of the precentral gyrus in amyotrophic
lateral sclerosis. AJNR Am J Neuroradiol 2000;21:647–658.
43. Langkammer C, Krebs N, Goessler W, et al. Quantitative imaging
of brain iron: a postmortem validation study. Radiology
2010;257:455–462.
Brain Iron MRI: A Biomarker for ALS
44. Graham JM, Paley MN, Grunewald RA, Hoggard N, Griffiths PD.
Brain iron deposition in Parkinson’s disease imaged using the
PRIME magnetic resonance sequence. Brain 2000;123:2423–
2431.
45. Michaeli S, Oz G, Sorce DJ, et al. Assessment of brain iron and
neuronal integrity in patients with Parkinson’s disease using
novel MRI contrasts. Mov Disord 2007;22:334–340.
46. Sage CA, Peeters RR, Gorner A, Robberecht W, Sunaert S. Quan-
titative diffusion tensor imaging in amyotrophic lateral sclerosis.
Neuroimaging 2007;34:486–499.
47. Kumar M, Abdel-Dayem HM, Three generations of amyotrophic
lateral sclerosis in a family: SPECT brain perfusion findings. Clin
Nucl Med 1999;24:539–540.
48. Fukunaga M, Li TQ, van Gelderen P, et al. Layer-specific varia-
tion of iron content in cerebral cortex as a source of MRI contrast.
Proc Natl Acad Sci U S A 2010;107:3834–3839.
49. Bačić G, Liu KJ, O’Hara JA, et al. Oxygen tension in murine tu-
mor: a combined EPR and MRI study. Magn Reson Med 1993;30:
568–572.
50. Liu KJ, Bačić G, Hoopes PJ, et al. Assessment of cerebral pO2 by
EPR oximetry in rodents: effects of anesthesia, ischemia, and
breathing gas. Brain Res 1995;685:91–98.
1479
51. Li X, Jankovic J, Le W. Iron chelation and neuroprotection in
neurodegenerative diseases. J Neural Transm 2011;118:473–
477.
52. Toshihiko Y, Makoto T, Akemi S, Shunsaku H. Activated micro-
glia cause superoxide-mediated release of iron from ferritin. Neu-
rosci Lett 1995;190:21–24.
53. Perez CA, Tong Y, Guo M. Iron chelators as potential therapeutic
agents for Parkinson’s disease. Curr Bioact Compd 2008;4:150–
158.
54. Kupershmidt L, Weinreb O, Amit T, et al. Neuroprotective and
neuritogenic activities of novel multimodal iron-chelating drugs
in motor-neuron-like NSC-34 cells and transgenic mouse model
of amyotrophic lateral sclerosis. FASEB J 2009;23:3766–3779.
55. Wang Q, Zhang X, Chen S, Zhang S, Youdium M, et al. Prevention
of motor neuron degeneration by novel iron chelators in
SOD1(G93A) transgenic mice of amyotrophic lateral sclerosis.
Neurodegener Dis 2011;8:310–321.
56. Kwan JY, Jeong SY, van Gelderen P, et al. Iron Accumulation in
deep cortical layers accounts for MRI signal abnormalities in
ALS: correlating 7 Tesla MRI and pathology. PLoS ONE 2012;7:
e35241.