Pediatric Neurology 73 (2017) 48e56

Contents lists available at ScienceDirect

Pediatric Neurology
journal homepage: www.elsevier.com/locate/pnu

Original Article

Dexamethasone, Intravenous Immunoglobulin, and

Rituximab Combination Immunotherapy for Pediatric
Opsoclonus-Myoclonus Syndrome
Michael R. Pranzatelli MD *, Elizabeth D. Tate MN
National Pediatric Myoclonus Center, National Pediatric Neuroinflammation Organization, Inc, Orlando, Florida

abstract
BACKGROUND: Although pulse-dose dexamethasone is increasingly favored for treating pediatric opsoclonus-
myoclonus syndrome (OMS), and multimodal immunotherapy is associated with improved clinical response,
there have been no neuroimmunologic studies of dexamethasone-based multimodal disease-modifying therapy.
METHODS: In this observational retrospective study, 19 children with OMS (with or without associated neuro-
blastoma) underwent multibiomarker evaluation for neuroinflammation. Nine children of varying OMS severity,
duration, and treatment status were treated empirically with pulse dexamethasone, intravenous immunoglobulin
(IVIg), and rituximab combination immunotherapy (DEXIR-CI). Another 10 children on dexamethasone alone or
with IVIg at initial evaluation only provided a comparison group. Motor severity (total score) was scored rater-
blinded via videotapes using the validated OMS Evaluation Scale. RESULTS: DEXIR-CI was associated with a 69%
reduction in group total score (P ¼ 0.004) and was clinically well tolerated. Patients given the dexamethasone
combination exhibited significantly lowered B cell frequencies in cerebrospinal fluid (94%) and blood (76%),
normalizing the cerebrospinal fluid B cell percentage. The number of patients with positive inflammatory markers
dropped 87% (P ¼ 0.002) as did the number of markers. Cerebrospinal fluid oligoclonal bands were positive in four
of nine pretreatment patients but zero of six post-treatment patients. In the comparison group, partial response to
dexamethasone alone or with IVIg was associated with multiple positive markers for neuroinflammation despite
an average of seven months of treatment. CONCLUSIONS: Multimechanistic dexamethasone-based combination
immunotherapy increases the therapeutic armamentarium for OMS, providing a viable option for less severely
affected individuals. Partial response to dexamethasone with or without IVIg is indicative of ongoing neuro-
inflammation and should be treated promptly and accordingly.
Keywords: chemokines, CSF immunophenotyping, immunobiomarkers, neuroblastoma, neuroinflammation, OMS, paraneoplastic
syndrome, pediatric neuroinflammatory disorders
Pediatr Neurol 2017; 73: 48-56
Ó 2017 Elsevier Inc. All rights reserved.

Introduction half-life of 190 minutes, and deliverability by a variety of

First synthesized in 1957, dexamethasone is a fluorinated
pharmaceutical steroid that binds to the glucocorticoid
receptor and exerts powerful anti-inflammatory and immu-
nosuppressive properties.1 A number of favorable pharmaco-
kinetic properties, such as 80% to 90% bioavailability, a long
Article History:
Received February 25, 2017; Accepted in final form April 29, 2017
* Communications should be addressed to: Prof. Dr. Pranzatelli; Na-
tional Pediatric Neuroinflammation Organization, Inc; 12001 Research
Parkway; Suite 236; Orlando, FL 32826.
E-mail address: mpranzatelli@omsusa.org
routes, have led to its designation as an essential medication.2-4
Dexamethasone, with its 25-fold greater potency than short-
acting corticosteroids, is used in the management of a bevy
of disorders, including neurological indications.5-11
The therapeutic application of dexamethasone to opso-
clonus-myoclonus syndrome (OMS), a neuroinflammatory
disorder known to be associated with neuroblastoma in at least
half of the patients,12 dates back to 1969,13 7 years after the first
description of OMS, which introduced adrenocorticotropic
hormone (ACTH) as a therapy.14 Dexamethasone is one of
seven steroids tried in OMS, a list that also includes betame-
thasone, hydrocortisone, methylprednisolone, prednisolone,

0887-8994/$ e see front matter Ó 2017 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.pediatrneurol.2017.04.027
 M.R. Pranzatelli, E.D. Tate / Pediatric Neurology 73 (2017) 48e56
prednisone, and triamcinolone.15 Other early reports of its use
in OMS made similar observations.16,17 More recently, interest
in dexamethasone has been renewed as a pulse-dose therapy
for OMS in anecdotal reports.18-21 Partial response and relapse
remain clinical challenges in OMS.22
In recent years, multimodality has been the vanguard of
immunotherapy for many neuroinflammatory disorders.
Cerebrospinal fluid (CSF) B cell expansion, the first biomarker
of disease activity in OMS,23 lead to the novel application
of rituximab after a phase I clinical drug trial (ClinicalTrials
.gov NCT00244361).24 Neuroinflammation in OMS is multi-
component, occurring much earlier than previously
appreciated. Consequently, the National Pediatric Myo-
clonus Center has advocated a biomarker-driven therapeutic
approach using rapid deployment of multimodal, multi-
mechanistic, combination immunotherapy. We introduced
this technique in OMS as part of front-loaded ACTH-in-
travenous immunoglobulin (IVIg)-rituximab combination
therapy (FLAIR-CI).25 Increased immunosuppression with
these agents has a favorable effect on developmental
outcome in OMS.26 However, when the cost of ACTH spiked
and some insurance companies began denying it, we
substituted dexamethasone in individuals with mild or
moderate severity, still also giving IVIg and rituximab. One
aim of the present exploratory study was a retrospective
analysis of observations on the feasibility and clinical utility
of dexamethasone-IVIg-rituximab combination immuno-
therapy (coined DEXIR-CI) in OMS.
Markers for neuroinflammation have expanded the
diagnostic yield of lumbar puncture beyond reliance on
leukocytosis,27 which is minimal or absent in OMS.28 Those
best able to discriminate OMS from noninflammatory
pediatric neurological disorders are CSF B cell percentage,23
positive CSF oligoclonal bands (OCB),28 and the concentra-
tions of chemokines CXCL1329 and CXCL10 in CSF30 and
CCL22 in serum.31 The rationale for inclusion of inflamma-
tory markers, such as chemokines, in this study is their
capacity to orchestrate inflammation through chemo-
attraction of B cells (CXCL13, CXCL10), T helper type 1
(CXCL10), or T helper type 2 cells (CCL22) into the CNS.
Although such a panel is not exhaustive, each component has
been shown to have a high probability of revealing neuro-
inflammation in this disorder.23,28-31 Another aim of the
present study was to evaluate the yield of the commercial
and research immunobiomarkers used in identifying neu-
roinflammation in OMS, their relation to partial clinical
response, and whether DEXIR-CI reversed the abnormalities.
Methods
Study design
The purpose of the study was to retrospectively analyze clinical and
neuroimmunologic observations on two groups of children with OMS,
one of which went on to be treated with DEXIR-CI. On the basis of our
previous research, we wished to address several questions:
(1) Is partial response to treatment indicative of ongoing neuro-
inflammation in OMS? Accordingly, the type and number of CSF/
serum inflammatory markers in children with persistent OMS
symptoms and signs despite previous immunotherapy were analyzed.
(2) Is dexamethasone with or without IVIg sufficient to reliably protect
children with OMS from failure to remit? Observations from another
49
group of children on dexamethasone alone or with IVIg at initial
evaluation provided a comparison.
(3) Do more rigorous neuroinflammation diagnostic tools (beyond
routine CSF studies) that are available to the clinician make a dif-
ference in understanding treatment failure and identifying patients
for more intensive immunotherapy? Flow cytometric immunophe-
notyping and quantification of OCB were analyzed.
(4) Did DEXIR-CI redress neuroinflammation and have clinical benefit?
Before and after clinical and neuroimmunological observations
made in patients treated empirically with DEXIR-CI were compared.
(5) Was the combination therapy well tolerated? Observations available
on safety and outcome were gathered, which may be useful to the
design of a randomized clinical trial.
Patients
Patient characteristics were documented (Table 1). The 19 children with
OMS reported in the present study were referred or recruited to the Na-
tional Pediatric Myoclonus Center, an international center specialized for
pediatric-onset OMS, and examined by two OMS experts. Parents signed
consent for enrollment of their child in a case-control study approved by
the local Institutional Review Board (SIU School of Medicine, Springfield,
IL). In this translational research, which was part of a multiplex chemokine
profiling project, additional CSF and serum for research purposes was
collected at the time of the diagnostic lumbar puncture and venipuncture.
Published pediatric noninflammatory neurological control data from Dr.
Pranzatelli’s laboratory were used for comparison in neuroimmunologic
studies,23,28-31 because CSF sampling is not performed in healthy children.
Briefly, the types of disorders included ataxia, developmental delay, head-
ache, movement disorders, seizures, and miscellaneous disorders.
Nine children were treated with DEXIR-CI. The choice of treatment
modality was empirical, not part of a clinical drug trial. Because of the
very high morbidity rate in OMS, we introduced into our clinical practice
retesting for neuroinflammation after immunotherapy,25 subsequently
known as testing for freedom from disease activity32 or no evidence of
disease activity.33 Parents were reconsented.
The other ten children, arriving on dexamethasone monotherapy or
dexamethasone and IVIg, provided a comparison group for analysis only
at initial evaluation, not as part of any subsequent treatment they may or
may not have received. Treatment and longitudinal data were analyzed
retrospectively via exempt review status granted by Western Institu-
tional Review Board (Wallup, WA).
DEXIR-CI treatment
Dexamethasone pulses of 7 mg/kg twice daily were given orally for
three consecutive days (21 mg/kg/day total), repeated monthly. IVIg was
1 g/kg on each of two consecutive days (2 g/kg total) for induction, then 1
to 2 g/kg monthly. Rituximab was infused IV at 300 to 375 mg/m2 on
each of four consecutive weeks uninterrupted by IVIg. The order of the
agents varied according to local logistics. Rituximab was the third agent
delivered in 57% of the patients. As prophylaxis against gastritis and acid
reflex, parents were instructed to give the children oral ranitidine during
the days of the pulses. Sulfamethoxazole/trimethoprim was adminis-
tered as intermittent antimicrobial prophylaxis.
Biomarker assessments
Flow cytometry was used to measure the frequency of CSF and blood
lymphocyte subsets per our standardized protocol23 in the Pathology
Department at St. John’s Hospital (Springfield, IL). Chemokine enzyme-
linked immunoabsorbent assays for CXCL10, CXCL13, and CCL22 were
performed in the investigator’s laboratory as per the manufacturers’
instructions and our previous publications.29-31 OCB were measured in
CSF and parallel serum samples using isoelectric focus with immuno-
fixation at ARUP Lab (Salt Lake City, UT).28 Immunoglobulins in CSF and
serum were quantified in the clinical laboratory. In patients on IVIg,
blood for immunoglobulin testing was drawn at the 4-week trough, just
before the next monthly dose, in an effort to minimize false positive
50 M.R. Pranzatelli, E.D. Tate / Pediatric Neurology 73 (2017) 48e56

TABLE 1.
Clinical/Demographic Characteristics of the Groups

Parameter DEXIR-CI DEX/IVIg Comparison Group

No. % Mean  S.D. No. % Mean  S.D.
N 9 10
Gender
Male 6 67 6 60
Female 3 33 4 40
Age (years) 3.0  1 2.5  1.7
Age category
Infant 0 0 0 0
Toddler 4 44 8 80
Youngster 5 56 2 20
OMS onset (years) 1.9  0.8 1.7  1
OMS duration (years) 0.9  1 0.7  0.9
Duration category
Acute 4 44 4 40
Subacute 3 33 3 30
Chronic 2 22 2 20
Total score 14  3 13  9
Severity category
Mild 6 67 5 56
Moderate 1 11 3 33
Severe 2 22 1 11
Tumor
Yes 5 56 7 70
No 4 44 3 30
Prior treatment
None 3 33
Steroids, IVIg 2 22
IVIg 2 22
A, S, I 1 11
S, I, C 1 11
Treatment time (months) 0.8  1 0.6  0.8
0 3 33
<3 2 22
3-12 3 33
>12 1 11
Abbreviations:
A ¼ ACTH
C ¼ Cyclophosphamide
DEX ¼ Dexamethasone
DEXIR-CI ¼ Dexamethasone-IVIg-rituximab combination immunotherapy
I or IVIg ¼ Intravenous immunoglobulins
S ¼ Corticosteroids
There was one missing total score value in the DEX/IVIg comparison group, hence severity category calculations are based on n ¼ 9.
There were no statistically significant differences in means or frequencies between the two groups. Age categories could not be compared with chi-square because of zero values.

findings in patients on chronic IVIg infusions. Given that the half-life of two-tailed t tests, using Bonferroni corrections for multiple statistical
IVIg is about 21 days, this approach seems reasonable. comparisons where indicated. Pretreatment versus post-treatment data
were analyzed as the means  standard error of the mean by paired t
tests. For statistical significance determination, a ¼ 0.05. Biomarker
Clinical assessments
levels were determined to be high or low by case-control data. The
designation of “increased” biomarker level denotes a concentration or
“Total score” (TS), a summation of subscores on a videotaped 12-item
percentage greater than the 75th interquartile range and “low” means
validated scale, was used as a measure of OMS severity.34 Each scale item
less than the 25th interquartile range. Frequency analyses of clinical/
is scored 0 to 3 according to written guidelines by a trained blinded rater
demographic data in the DEXIR-CI group compared with the
as part of our standard clinical practice. Total Score designates mild (TS
dexamethasone and IVIg group were made by chi-square or Fisher’s
0 to 12), moderate (TS 13 to 24), or severe (TS 25 to 36) categories.
exact test.
Detailed reporting of this OMS Evaluation Scale has been made in pre-
vious publications,34,35 including a videotape.23,25 It is reprinted with
permission of the publisher as an online Supplemental Table for the
convenience of the reader who may not have access to those sources. Results
OMS duration was categorized as acute (0 to 3 months), subacute (3 to
12 months), and chronic (more than 12 months). Clinical observations on DEXIR-CI

Statistical analysis Dexamethasone-based combination treatment was

associated with a 69% reduction in group Total Score
Statistical analyses were computed using GraphPad Prism software (P ¼ 0.004) as assessed 8.1  2 months later (Figure A).
(La Jolla, CA). Independent groups were analyzed as the means  S.D. by Although none of the Total Scores were in the trace range
 M.R. Pranzatelli, E.D. Tate / Pediatric Neurology 73 (2017) 48e56
pretreatment and the group was heterogeneous as to
severity, all but one was post-treatment, a reflection of
greater homogeneity (Figure B). DEXIR-CI was clinically
well tolerated with the exception of reversible behavior
complaints, such as irritability or sleepiness, during the
pulse period and up to a few days afterward.
Immunologic observations on DEXIR-CI
The dexamethasone-based combination immuno-
therapy reduced biomarker-measured evidence of neuro-
inflammation. Before treatment, there was a range of CSF B
cell frequencies from normal/intermediate to increased
(Figure C). Patients treated with dexamethasone-based
combination immunotherapy exhibited significantly low-
ered B cell frequencies in CSF (95%) compared with the
control range (Figure D). In blood, the frequency of B cells
also dropped (75%) inclusive of patients who were not yet
repopulated and those who had started to repopulate from
rituximab (Figure F).
The number of patients with positive inflammatory
markers dropped 87% (P ¼ 0.002) (Table 2). CSF OCB were
positive in four of nine pretreatment but zero of six post-
treatment patients. Two of the four patients with
treatment-lowered serum IgM concentrations (who had the
lowest IgM) also had low serum IgA levels with normal IgG
(on IVIg). The IgA levels, which were not low enough for
true IgA deficiency, are seen more often in children with
autoimmune diseases, or might have been a treatment
effect.
Neuroinflammation in the dexamethasone and IVIg comparison
group
To determine the relative contribution of dexamethasone
or dexamethasone and IVIg to the neuroimmunologic effects
of DEXIR-CI, the 10 other patients on dexamethasone
monotherapy or the combination of dexamethasone and IVIg
at the time of initial evaluation were analyzed (Table 3).
Clinical characteristics of the dexamethasone monotherapy
subgroup and the dexamethasone with IVIg subgroup were
similar and were therefore combined to counter the small
subgroup sample sizes. Mean OMS duration was subacute
(8.4 months), and the patients had been on treatment for an
average of 7.2 months. OMS severity was rated at the low-
moderate level (TS ¼ 13).
The combined comparison group displayed multiple
markers of neuroinflammation, including expansion of CSF
B cells, increased concentrations of CSF CXCL13 and CXCL10,
and serum CCL22. In contrast, CSF neuroinflammation was
not predicted by blood B cell frequency or the concentration
of serum immunoglobulins.
The immunomarker frequencies were also informative
(Table 4). Eighty percent of patients displayed an elevated B
cell percentage, 78% had increased CXCL13, and 86% had
increased CXCL10. Serum CCL22 was elevated in 60%. Data
were also analyzed for the number of positive immuno-
markers per person of five possible markers. The mean
number of positive markers for the group was 2.6  0.4. The
group frequency of positive immunomarkers of five
possible markers was 66  21%. The type of positive
markers and the number of positive markers were similar in
51
the DEXIR-CI group pretreatment to the dexamethasone
comparison group.
Relapse and course
Five patients relapsed whereas four did not. The number
of relapses was zero in 56%, one in 33%, and two in 22%. In
75%, the relapse was mild (one OMS motor sign) and in 25%,
moderate (two motor signs). There were no severe relapses
(three motor signs). Relapse etiology was illness-related in
75% and treatment tapering-triggered in 25%. The relapse
remitted spontaneously in two patients, was treated with
an increase in ACTH dose in one patient, and ACTH and IVIg
were started in another patient. The patient with an initial
TS of 30 still had a TS of 16 after DEXIR-CI. He was the only
one in the group to relapse twice and had to be put on ACTH
because of prolonged drooling, poor expressive language,
paucity of words, and the motor signs of OMS. At last
contact, three (33%) of the patients were off immuno-
therapy, one (11%) was on dexamethasone, four (44%) were
on both dexamethasone and IVIg, and one (11%) was on
ACTH and had received cyclophosphamide and another
course of rituximab.
Discussion
In this proof-of-concept exploratory study, there were
several novel findings. Specific results are addressed indi-
vidually in the following data synthesis for each study
question in the order that it was posed in Study Design.
First, partial response to treatment was indicative of
ongoing neuroinflammation in all 19 patients. However,
treating physicians would be caught unaware of the “smol-
dering” inflammatory process25 if only insensitive routine
CSF studies are done or if CSF immunodiagnostics are not
ordered at all. In sum, the data support the argument that
OMS is too serious to be managed only clinically without
availing the patient of biomarker diagnostics, although a
prospective study would be required for validation. The data
also would support the testable approach of taking a second
look after treatment, especially when there are clinical
problems, to be sure that neuroinflammation is gone. The
retest approach has gained ground in the field of multiple
sclerosis in adults.32,33
Second, the study shows that dexamethasone mono-
therapy or the combination of dexamethasone and IVIg does
not afford sufficient protection against neuroinflammation in
some patients. The clinical parallel is that the long-term
outcome of children with OMS treated with steroid
monotherapy of any type is notoriously poor22; the same
limitations apply to ACTH monotherapy.36 Ideal therapies
afford the best protection for the most patients. Still popular
in some practices, the monotherapies do not meet that
standard.
Third, more rigorous neuroinflammation diagnostic tools
that are available to clinicians evaluating children with OMS
do make a difference in understanding inadequate treat-
ment response and identifying patients for more intensive
immunotherapy. We have advocated the use of biomarkers
of disease activity to interrogate the capacity of any
immunotherapeutic agent for OMS to reverse or eliminate
neuroinflammation.25 The diagnostic tools to measure CSF B
52 M.R. Pranzatelli, E.D. Tate / Pediatric Neurology 73 (2017) 48e56

A 36
B 36
P = 0.004
30 30

Total Score
Total Score 24 24

18 18

12 12
-69%
6 6

0 0
Pre Post Pre Post

C D
10 P = 0.007 10
% of CSF Lymphocytes

% of CSF Lymphocytes

8 8

6 6

4 4

2 2
-95%
0 0
Pre Post Pre Post

E F
50 P = 0.0004 50
% of Blood Lymphocytes

% of Blood Lymphocytes

40 40

30 30

20 -75% 20

10 10

0 0
Pre Post Pre Post
FIGURE.
Box and whisker graph of DEXIR-CI treatment effect on (A) total score, (C) CSF B cell percentage, and (E) circulating B cells. The median is the line through
the box, and 25th and 75th interquartile ranges are the upper and lower sides of the box, respectively. The plus symbol signifies the mean. The error bars are
Tukey error bars. Individual patient responses are shown for (B) total score, (D) CSF B cell percentage, and (F) circulating B cells. CSF, cerebrospinal fluid;
DEXIR-CI, dexamethasone, IVIg, and rituximab combination immunotherapy.
 M.R. Pranzatelli, E.D. Tate / Pediatric Neurology 73 (2017) 48e56 53

TABLE 2. TABLE 3.
Before and After Immunologic Observations Cross-Sectional Data From 10 Other Children With Symptomatic OMS on Dexa-
methasone Pulses or Dexamethasone and IVIg at Initial Evaluation
Biomarkers Mean  SEM Range Decrease P
Immunomarkers Parameter Dexamethasone Dexamethasone Subgroups
B % (CSF) Monotherapy with IVIg Combined
Before 3.8  0.9[ 0.9-8.4 Combined
After 0.2  0.1 0-0.7 95% 0.007*,y Clinical
OCB (CSF) N 4 6 10
Before 2.9  1[ 0-8 Gender (M/F) 4 M/0 F 2 M/4 F d
After 0 100% 0.048 Onset (years) 1.9  1.4 1.6  0.5 d
CXCL13 (CSF) Duration (years) 1.3  1.2 0.36  0.23 d
Before 6.4  2[ 0-10 Tumor (yes/no) 3 Y/1 N 4 Y/2 N d
After 2.6  1[ 0-7.3 59% NS Treatment time 1.1  1 0.23  0.2 d
CXCL10 (CSF) Total score* 12  8 14  10 d
Before 157  52 39-381 Immunomarkers
After 208  21[ 161-303 d NS B cell % (CSF) 3.0  2.7[ 6.1  3[ 4.9  3[
CCL22 (serum) OCB (CSF) 0.3  0.6 1.4  2 1.0  2
Before 1359  235[ 617-2430 CXCL13 (CSF) 5.1  3.7[ 11  11[ 8.4  8[
After 1007  42 653-1148 26% NS CXCL10 (CSF) y
450  171[ 383  192[
No. markers CCL22 (serum) y 1570  600[ 1447  588[
Before 2.6  0.4 1-4 Blood immune cells
After 0.33  0.2 0-2 87% 0.002*,y B cell % 19  9 22  7 21  8
Blood immune cells Serum Ig
B% IgM 100  52 90  37 94  42
Before 27 3 16-44 IgG 598  228 1215  568 941  535
After 6.6 2 0-13 75% 0.0004*,y IgA 67  60 28  8Y 46  43
Serum Ig
Abbreviations:
IgM
CSF ¼ Cerebrospinal fluid
Before 112  20 46-223 IVIg ¼ Intravenous immunoglobulin
After 36  8Y 6-50 75% 0.02* OCB ¼ Oligoclonal band
IgG OMS ¼ Opsoclonus-myoclonus syndrome
Before 1053  117 628-1650 [ Greater than 75th percentile interquartile range for control subjects.
After 1099  115 786-1720 NS Y Less than reference range.
IgA Clinical/demographic information on the combined subgroups is shown in Table 1.
* Total score categories: 0 to 6 absent/trace; 7 to 12 mild; 13 to 24 moderate; 25 to
Before 70  21 8-197
36 severe.
After 52  16 8-141 26% NS y
Too few data in cell for subgroup statistics, but included in combined data.
Abbreviations:
CSF ¼ Cerebrospinal fluid
NS ¼ Not statistically different
OCB ¼ Oligoclonal band
Pos. ¼ Number of positive inflammatory markers of five possible markers.
result is the first evidence, besides our report on CXCL13,29 of
SEM ¼ Standard error of the mean. the effects of dexamethasone-based combination immuno-
[ Greater than 75th percentile interquartile range for control subjects. therapy on CSF and blood B lymphocytes and CSF OCBs.25
Y Less than reference range.
Although the rationale and data on the effectiveness of
Biomarker concentrations are expressed as pg/ml for CXCL13, CXCL10, and CCL22,
and mg/dl for IgM, IgG, and IgA. multimodal immunotherapy have become increasingly
* P < 0.05 on paired t test. Only paired data were used in the statistical analysis. compelling, the contrarian view is that the agents should be
y
Significant after Bonferroni corrections: 0.05/4 ¼ 0.01 for CSF and 0.01 for blood/ administered stepwise and added only if escalation is clini-
serum.
cally necessary over a six- to 12-month period. An in-depth
analysis of the contrasting approaches is available
elsewhere.25
cell frequency by flow cytometry and CSF OCBs exist at Fifth, DEXIR-CI was well tolerated, but this was not a
major medical centers, awaiting routine implementation. clinical trial, so safety data are limited. Observations on
More data on possible biomarker panels will facilitate this outcome also were limited, but relapses were observed in
purpose. Some of the high-information chemokine/cyto- some treated patients. The relapse rate of 56% is consistent
kine research tests may eventually become clinically with that reported previously in OMS.22 The question of
available. whether the relapse rate would be lower in children treated
Fourth, DEXIR-CI redressed neuroinflammation and initially with the combination therapy in a timely manner is
exhibited clinical benefit even in the 67% of patients who had crucial, but the answer awaits a clinical trial. Some patients
only partially responded to corticosteroids, IVIg, or other still had a difficult time with relapse, requiring use of ACTH
agents previously. It was associated with reduction in the or other agents.
clinical severity of OMS and both the number of positive Although the specific mechanisms by which dexametha-
immunobiomarkers of disease activity and the number of sone effects on CSF neuroinflammation in OMS are
patients with positive biomarkers. The study extends our unknown, the drug exerts many potentially relevant phar-
previous research by measuring a panel of biomarkers across macodynamic effects on the CNS. In experimental animals, it
patients, demonstrating multiple positive biomarkers of suppresses the messenger RNA and protein expression of C-C
disease activity in children symptomatic with OMS, whether motif chemokine ligand 2 (CCL2) in activated microglia,
they had been previously treated with immunotherapy. This which inhibits microglia migration.37 It also suppresses
54 M.R. Pranzatelli, E.D. Tate / Pediatric Neurology 73 (2017) 48e56

TABLE 4.
Frequency of Patients With Positive Immunomarkers by Type or Number of Markers Compared by Group

Marker DEXIR-CI Dexamethasone With/Without IVIg

Period # Pos. # Neg. % Pos. # Pos. # Neg. % Pos.
Type of marker
CSF B cells %
Before 5 3 62 8 2 80
After 0 8 0
OCB
Before 4 4 50 2 6 25
After 0 8 0
CXCL13
Before 5 2 71 7 2 78
After 2 5 29
CXCL10
Before 2 4 33 6 1 86
After 3 3 50
CCL22
Before 3 4 43 3 2 60
After 0 7 0
Number of markers
Marker positive
Before 9 0 100 8 1 89
After 2 7 22
1 Marker
Before 7 2 78 8 2 80
After 1 9 11
3 Markers
Before 5 4 56 5 4 56
After 0 9 0
Abbreviations:
CSF ¼ Cerebrospinal fluid
DEXIR-CI ¼ Dexamethasone-IVIg-rituximab combination immunotherapy
IVIg ¼ Intravenous immunoglobulin
OCB ¼ Oligoclonal bands
Lower n values for particular markers indicate missing data.

production of reactive oxygen species, nitric oxide, and in-
flammatory responses of activated microglia.38 In vitro,
dexamethasone diminishes the late proinflammatory and
cytotoxic effects of amyloid beta-protein in cerebrovascular
smooth muscle cells.39 Pretreatment with dexamethasone
prevents tumor necrosis factor aeinduced blood-brain
barrier disruption and neuroinflammation in astroglial-
endothelial cell cocultures.40 It inhibits the uptake-2
high-capacity monoamine transport system in cultured rat
cerebellar granule neurons in vitro.41 In experimental ani-
mals, it reduced inflammatory mediators in the CSF caused
by acute Lyme neuroborreliosis and prevented neurode-
generative changes.42
Although dexamethasone use is common, the optimal
dose of dexamethasone for OMS and other neuro-
inflammatory disorders has not been established, as it is not
based on biomarker metrics, but rather on dose data
derived from hematologic disorders in adults. The adult
literature suggests that more frequent dexamethasone
cycles/month and total cycles can be evaluated in OMS.
Expressed as milligrams per kilogram, the adult dose
computes to 21 mg/m2 for men and 25 mg/m2 for women,
with each pulse repeated once, twice, or three times a
month for one to six monthly cycles.43-46 A case seriese
based dosing for pediatric-onset OMS is 20 mg/m2 for three
to four days once monthly.19-21 Although dexamethasone
has minimal mineralocorticoid activity, adverse effects with
long-term use, such as bone loss, cataracts, easy bruising,
muscle weakness, and thrush, would need to be weighed.
Nonmedical factors also have entered the treatment
discussion. Unlike ACTH (sold as Acthar gel), which has
become astonishingly expensive since 2008,47 dexametha-
sone has remained inexpensive. It is available in most areas
of the world, whereas ACTH currently is not. The high cost
of ACTH has dissuaded many child neurologists from pre-
scribing it for OMS, but the track record for prednisone,
prednisolone, and pulse-dose methylprednisolone is
poor.22 Hence the need for viable alternatives, which we
hope our study addresses. In severe OMS, however, data on
the effectiveness of dexamethasone are limited. In contrast,
FLAIR-CI has been demonstrated to be successful for in-
duction in severe OMS, so we use it in those patients.25 In
our practice, any patient with severe OMS who is failing the
dexamethasone-based combination in one month would be
switched to the ACTH-based combination.
Strengths of the present study included the careful search
for multiple inflammatory markers in CSF (or serum) at initial
evaluation of symptomatic children with OMS, using several
methodologies to create a “neuroinflammation panel.” The
CSF data for both DEXIR-CI and dexamethasone and IVIg
treatment are novel. Patients were evaluated by two OMS
experts, a case-control design was used for cross-sectional
data, and both clinical and neuroimmunologic response ob-
servations were made. Steps were taken to minimize
 M.R. Pranzatelli, E.D. Tate / Pediatric Neurology 73 (2017) 48e56
false-positive immunoglobulin quantification in patients with
chronic IVIg transfusions.
Limitations of the study included its small sample size,
the retrospective analysis of treatment observations (not a
clinical drug trial), the focus on diagnostic rather than on
nondiagnostic aspects of OMS, and the lack of long-term
follow-up. Responses in acute, immunotherapy-naive OMS
may differ from those in partial responders. The order of
immunotherapeutic agent administration, which varied
according to local logistics, could have affected the
outcome, although there are no data either way.
Conclusions
Our exploratory study suggests that DEXIR-CI is useful
for some children with OMS. Pending further evaluation, it
is probably best reserved for less severely affected patients,
unless ACTH is not available, as the effectiveness of ACTH-
based combination immunotherapy in severe OMS is well
established. Recognizing the limits of existing evidence-
based guidelines, it is likely that the dose and dosing of
dexamethasone will undergo significant permutations to
improve upon therapeutic performance. Dexamethasone
dose-ranging studies should be conducted before a large
trial of dexamethasone-based therapy is undertaken.
Although there is no cure for OMS, substantial gains can be
made with facile use of modern, simultaneously deployed
immunotherapy coupled with immunobiomarker mea-
surements and heightened clinical awareness of neuro-
inflammation as a tractable target.
Dr. Pranzatelli is Founder and President of the National Pediatric Neuroinflammation
Organization, Inc, a Florida charitable organization and a registered 501(c)(3)
nonprofit, and Adjoint Professor of Neurology at the University of Colorado School of
Medicine. The authors thank all participating children and their families, referring
and treating physicians, and Nathan R. McGee, who performed the cytokine assays.
Study Funding: Biomarker studies were supported in part by donations from fam-
ilies to Dr. Pranzatelli’s research on OMS. There was no targeted funding.
Supplementary data
Supplementary data related to this article can be found at
http://dx.doi.org/10.1016/j.pediatrneurol.2017.04.027.
References
1. Czock D, Keller F, Rasche FM, Häussler U. Pharmacokinetics and
pharmacodynamics of systemically administered glucocorticoids.
Clin Pharmacokinet. 2005;44:61-98.
2. Dexamethasone. Available at: https://www.drugs.com/monograph/
dexamethasone/html. Accessed February 24, 2017.
3. Jackson RK, Irving JA, Veal GJ. Personalization of dexamethasone
therapy in childhood acute lymphoblastic leukaemia. Br J Haematol.
2016;173:13-24.
4. 19th WHO Model List of Essential Medicines (April 2015). WHO;
2015.
5. Chu CC, Hsing CH, Shieh JP, Chien CC, Ho CM, Wang JJ. The cellular
mechanisms of the antiemetic action of dexamethasone and
related glucocorticoids against vomiting. Eur J Pharmacol. 2014;
722:48-54.
6. Hockey B, Leslie K, Williams D. Dexamethasone for intracranial
neurosurgery and anaesthesia. J Clin Neurosci. 2009;16:1389-1393.
7. Kostaras X, Cusano F, Kline GA, Roa W, Easaw J. Use of dexameth-
asone in patients with high-grade glioma: a clinical practice
guideline. Curr Oncol. 2014;21:e493-e503.
55
8. Nagakumar P, Doull I. Current therapy for bronchiolitis. Arch Dis
Child. 2012;97:827-830.
9. Patnaik SK, Upreti V, Dhull P. Steroid responsive encephalopathy
associated with autoimmune thyroiditis (SREAT) in childhood.
J Pediatr Endocrinol Metab. 2014;27:737-744.
10. Warris LT, van den Heuvel-Eibrink MM, den Hoed MA, Arsen FK,
Pieters R, van der Akker EL. Does dexamethasone induce more
neuropsychological side effects than prednisone in pediatric acute
lymphoblastic leukemia? A systematic review. Pediatr Blood Cancer.
2014;61:313-318.
11. Zafren K. Prevention of high altitude sickness. Travel Med Infect Dis.
2014;12:29-39.
12. Boire A, Khakoo Y. Paraneoplastic syndromes. In: Swaiman A,
Ferriero S, Finkel G, Pearl S, eds. Swaiman’s Pediatric Neurology:
Principles and Practice 6e, Chapter 118. Oxford: Elsevier; 2017.
13. Christoff N. Myoclonic encephalopathy of infants. A report of two
cases and observations on related disorders. Arch Neurol. 1969;21:
229-234.
14. Kinsbourne M. Myoclonic encephalopathy of infants. J Neurol Neu-
rosurg Psychiatry. 1962;25:271-276.
15. Pranzatelli MR. The neurobiology of opsoclonus-myoclonus. Clin
Neuropharmacol. 1992;15:186-228.
16. Berg BO, Albin AR, Wang W, Skoglund R. Encephalopathy associated
with occult neuroblastoma. J Neurosurg. 1974;41:567-572.
17. Rivner MH, Jay WM, Green JB, Dyken PR. Opsoclonus in hemophilus
influenza meningitis. Neurology. 1982;32:661-663.
18. Ertle F, Behnisch W, Al Mulla NA, et al. Treatment of
neuroblastoma-related opsoclonus-myoclonus-ataxia syndrome
with high-dose dexamethasone pulses. Pediatr Blood Cancer. 2008;
50:683-687.
19. Oguma M, Morimoto A, Takada A, et al. Another promising
treatment option for neuroblastoma-associated opsoclonus-
myoclonus syndrome by oral high-dose dexamethasone pulse:
lymphocyte markers as disease activity. Brain Dev. 2012;34:
251-254.
20. Rostásy K, Wilken B, Baumann M, et al. High pulsatile dexametha-
sone therapy in children with opsoclonus-myoclonus syndrome.
Neuropediatrics. 2006;37:291-295.
21. Wilken B, Baumann M, Bien CG, et al. Chronic relapsing opsoclonus-
myoclonus syndrome: combination of cyclophosphamide and
dexamethasone pulses. Eur J Paediatr Neurol. 2008;12:51-55.
22. Pranzatelli MR, Tate ED. Trends and tenets in relapsing and progressive
pediatric opsoclonus-myoclonus. Brain Dev. 2016;38:439-448.
23. Pranzatelli MR, Travelstead AL, Tate ED, et al. B- and T-cell markers
in opsoclonus-myoclonus syndrome: immunophenotyping of CSF
lymphocytes. Neurology. 2004;62:1526-1532.
24. Pranzatelli MR, Tate ED, Verhulst SJ, et al. Pediatric dosing of rit-
uximab: serum concentrations in opsoclonus-myoclonus syndrome.
J Pediatr Hematol Oncol. 2010;32:e167-e172.
25. Pranzatelli MR, Tate ED. Opsoclonus myoclonus syndrome. In:
Swaiman K, Ashwal S, Ferriero D, et al., eds. Swaiman’s Pediatric
Neurology: Principles and Practice 6e, Chapter 120. Oxford: Elsevier;
2017.
26. Mitchell WG, Wooten AA, O’Neil SH, Rodriquez JG, Cruz RE,
Wittern R. Effect of increased immunosuppression on develop-
mental outcome of opsoclonus myoclonus syndrome (OMS). J Child
Neurol. 2015;30:976-982.
27. De Graaf MT, de Jonste AH, Kraan J, Boonstra JG, Sillevis Smitt PA,
Gratama JW. Flow cytometric characterization of cerebrospinal fluid
cells. Cytometry B Clin Cytom. 2011;80B:271-281.
28. Pranzatelli MR, Slev PR, Tate ED, Travelstead AL, Colliver JA,
Joseph SA. Cerebrospinal fluid oligoclonal bands in childhood
opsoclonus-myoclonus. Pediatr Neurol. 2011;45:27-33.
29. Pranzatelli MR, Tate ED, McGee NR, et al. Key role of CXCL13/CXCR5
axis for cerebrospinal fluid B cell recruitment in pediatric OMS.
J Neuroimmunol. 2012;243:81-88.
30. Pranzatelli MR, Tate ED, McGee NR, Travelstead AL, Verhulst SJ,
Ransohoff M. Expression of CXCR3 and its ligands CXCL9, -10, -11 in
pediatric OMS. Clin Exp Immunol. 2013;172:427-436.
31. Pranzatelli MR, Tate ED, McGee NR, Ransohoff RM, Travelstead A.
CCR4 agonists MDC/CCL22 and TARC/CCL17 are elevated in pedi-
atric OMS sera: rapid down-regulation of CCL22 by ACTH or corti-
costeroids. J Clin Immunol. 2013;33:817-825.
32. Havrdova E, Galetta S, Stefoski D, Comi G. Freedom from disease
activity. Neurology. 2010;74(Suppl 3):S3-S7.
56 M.R. Pranzatelli, E.D. Tate / Pediatric Neurology 73 (2017) 48e56
33. Nixon R, Bergvall N, Tomic D, Sfikas N, Cutter G, Giovannoni G. No
evidence of disease activity: indirect comparisons of oral therapies
for the treatment of relapsing-remitting multiple sclerosis. Adv Ther.
2014;31:1134-1154.
34. Pranzatelli MR, Tate ED, Galvan I, Wheeler A. Controlled pilot study
of piracetam for pediatric opsoclonus-myoclonus. Clin Neuro-
pharmacol. 2001;24:352-357.
35. Pranzatelli MR, Travelstead AL, Tate ED, Allison TJ, Verhulst SJ. CSF
B-cell expansion in opsoclonus-myoclonus syndrome: a biomarker
of disease activity. Mov Disord. 2004;19:770-777.
36. Tate ED, Allison TJ, Pranzatelli MR, et al. Neuroepidemiologic trends
in 105 cases of pediatric opsoclonusdmyoclonus syndrome. J Pediatr
Oncol Nurs. 2005;22:8-19.
37. Zhou Y, Ling EA, Dheen ST. Dexamethasone suppresses monocyte
chemoattractant protein-1 production via mitogen activated pro-
tein kinase phosphate-1 dependent inhibition of Jun N-terminal
kinase and P38 mitogen-activated protein kinase in activated rat
microglia. J Neurochem. 2007;102:667-678.
38. Huo Y, Rangarajan P, Ling EA, Dheen ST. Dexamethasone inhibits
the Nox-dependent MAPK pathways in activated microglia. BMC
Neurosci. 2011;12:49.
39. Previti ML, Zhang W, Van Nostrand WE. Dexamethasone diminishes
the pro-inflammatory and cytotoxic effects of amyloid beta-protein in
cerebrovascular smooth muscle cells. J Neuroinflammation. 2006;3:18.
40. Lutgendorf MA, Ippolito DL, Mesngon MT, et al. Effect of
dexamethasone administered with magnesium sulfate on
inflammation-mediated degradation of the blood-brain barrier
using an in vitro model. Reprod Sci. 2014;21:483-491.
41. Hill JE, Makky K, Shrestha L, Hillard CJ, Gasser PJ. Natural and syn-
thetic corticosteroids inhibit uptake-2 mediated transport in CNS
neurons. Physiol Behav. 2011;104:306-311.
42. Ramesh G, Martinez AN, Martin DS, Philipp MT. Effects of dexa-
methasone and meloxicam on Borrelia burgdorferi-induced
inflammation in glial and neuronal cells of the central nervous
system. J Neuroinflammation. 2017;14:28.
43. San Miguel JF, Weisel KC, Song KW, et al. Impact of prior
treatment and depth of response on survival in MM-003, a
randomized phase 3 study comparing pomalidomide plus low-
dose dexamethasone versus high-dose dexamethasone in
relapsed/refractory multiple myeloma. Haematologica. 2015;100:
1334-1339.
44. Doubek M, Brychtova Y, Panovska A, et al. Ofatumumab added to
dexamethasone in patients with relapsed or refractory chronic
lymphocytic leukemia: results from a phase II study. Am J Hematol.
2015;90:417-421.
45. Din B, Wang X, Shi Y, Li Y. Long-term effect of high-dose dexa-
methasone with or without low-dose dexamethasone maintenance
in untreated immune thrombocytopenia. Acta Haematol. 2015;133:
124-128.
46. Yan Z, Li Z, Zhang H, et al. [Efficacy of high-dose dexamethasone
plus low-dose rituximab as a second-line treatment in 65 patients
with primary immune thrombocytopenia]. Zhonghua Xue Ye Xue Za
Zhi. 2015;36:206-209.
47. Gettig J, Cummings JP, Matuszewski K. H.p. Acthar gel and cosyn-
tropin review: clinical and financial implications. P T. 2009;34:
250-257.