B Cell Depletion Therapy for
Michael R. Pranzatelli, MD,1*
Elizabeth D. Tate, C-FNP, MN,1
Jennifer A. Swan, BS,1Anna L. Travelstead, BS, MT,2
Jerry A. Colliver, PhD,3Steven J. Verhulst, PhD,3
Carl J. Crosley, MD,4
William D. Graf, MD,5Suja A. Joseph, MD,6
Howard M. Kelfer, MD,7and
G. Praveen Raju, MD, PhD8
1National Pediatric Myoclonus Center and Departments of
Neurology;2Flow Cytometry Facility;3Biostatistics and
Research Consulting, Southern Illinois
University School of Medicine, Springfield, Illinois, USA;
4Department of Neurology, SUNY, Syracuse, New York, USA;
5Section of Child Neurology, Children’s Mercy Hospitals and
Clinics, Kansas City, Missouri, USA;6Department of Neurol-
ogy, Medical College of Wisconsin, Milwaukee, Wisconsin,
USA;7Department of Neurology, Cook Children’s Medical
Center, Fort Worth, Texas, USA;8Department of Pediatrics,
Harvard Medical School,
Children’s Hospital Boston, Boston, Massachusetts, USA
opsoclonus-myoclonus syndrome and CSF B cell expan-
sion received rituximab, adrenocorticotropic hormone
(ACTH), and IVIg. Motor severity lessened 73% by 6 mo
and 81% at 1 yr (P < 0.0001). Opsoclonus and action
myoclonus disappeared rapidly, whereas gait ataxia and
some other motor components improved more slowly.
ACTH dose was tapered by 87%. Reduction in total CSF
B cells was profound at 6 mo (-93%). By study end, pe-
ripheral B cells returned to 53% of baseline and serum
IgM levels to 63%. Overall clinical response trailed pe-
ripheral B cell and IgM depletion, but improvement con-
tinued after their levels recovered. All but one non-ambu-
latory subject became ambulatory without additional
chemotherapy; two relapsed and remitted; four had ritux-
imab-related or possibly related adverse events; and two
had low-titer human anti-chimeric antibody. Combination
of rituximab with conventional agents as initial therapy
was effective and safe. A controlled trial with long-term
safety monitoring is indicated. ? 2009 Movement Disorder
Key words: ACTH; anti-B cell agent; dancing eyes; Kins-
bourne syndrome; neuroblastoma; paraneoplastic syndrome;
The unique combination of opsoclonus and myoclo-
nus has come to connote a paraneoplastic syndrome
and B cell pathology within the CNS.1Treatment fail-
ure, partial response, and neurological relapse compro-
mise the outcome of opsoclonus-myoclonus syndrome
(OMS).2Rituximab, the prototypic chimeric anti-B cell
monoclonal antibody (anti-CD20), has been applied to
the therapy of various autoimmune neurological disor-
ders.3We demonstrated previously that its adjunctive
use treats the characteristic CSF B cell expansion in
pediatric OMS4with clinical benefit,5–7as recent case
reports attest.8–11This study was designed to look at
the feasibility and safety of combining rituximab to-
gether with adrenocorticotropic hormone (ACTH) and
IVIg, the two most commonly used conventional thera-
pies for OMS,2as initial therapy for untreated patients.
Such an approach to gaining more complete neurologi-
cal remission and preventing relapse of OMS is new.
PATIENTS AND METHODS
This was a 1-yr, investigator-sponsored, open- and
off-label, prospective study, with video-documented
evaluations and blinded scoring of clinical efficacy.
The primary study end point was preset at 6 mo after
the final rituximab, because cerebrospinal fluid (CSF)
testing also was done then. ‘‘Relapse’’ was defined as
distinct OMS worsening or symptom reappearance last-
ing at least 72 h. Failure to walk within 6 mo or
Additional Supporting Information may be found in the online ver-
sion of this article.
*Correspondence to: Michael R. Pranzatelli, SIU-SOM, P.O. Box
19643, Springfield, Illinois 62794-9643.
E-mail: mpranzatelli@siumed. edu
Potential conflict of interest: Dr. Pranzatelli has clinical trial con-
tracts and/or research grants from Genentech/IDEC and Questcor. He
was a paid ad hoc consultant for two Genentech B cell conferences.
Received 19 May 2009; Revised 27 October 2009; Accepted 9
Published online in Wiley InterScience(www.interscience. wiley.-
com). DOI: 10.1002/mds.22941
Vol. 00, No. 00, 2010, pp. 000–000
? 2009 Movement Disorder Society
respond to measures for relapse by 1 mo indicated a
need for additional immunotherapy. For comparison,
the OMS relapse frequency in patients treated only
with conventional agents is 50 to 70%.2,12
Children with OMS were recruited to the National
Pediatric Myoclonus Center and examined by the prin-
cipal investigator (M.P.). New referrals for OMS aver-
age 30 to 40 annually and about 25% are untreated.
Parents of 12 untreated children meeting inclusion and
informed consent for this Institutional Review Board
approved study (SCRIHS protocol 04-112), which was
conducted from 2004 to 2007 (IND 11,771). The de-
mographic data (means 6 SD) were as follows: age
1.9 6 0.4 yr, range 1.3 to 2.6 yr; OMS onset age 1.6
6 0.5 yr; OMS duration 0.3 6 0.2 yr; OMS score 23
6 5. The categorical subgroups included five boys and
seven girls, seven acute and five subacute cases, six
moderate and six severe cases, two neuroblastomas,
and four prior relapsers.
Subcutaneous IV ports were placed in eight toddlers
with poor venous access. The doses of rituximab,
ACTH, and IVIg were chosen to match those used in
our adjunctive trial.5ACTH and IVIg were started
before rituximab, with 3 to 7 d in between the sequen-
tial introduction of agents. Rituximab (Rituxan; South
San Francisco, CA)/Biogen IDEC (San Diego, CA)
was given IV (1 mg/mL in D5¼ NS) once weekly for
four consecutive weeks at a dose of 375 mg/m2.
Patients were premedicated with oral acetaminophen
(15 mg/kg), diphenhydramine IV (1.5 mg/kg up to 25
mg), and dexamethasone IV (0.05–0.08 mg/kg up to 1
mg).13Rituximab was infused at 20 mL for 30 min, 40
mL for 30 min, and then 60 mL/h. A 52-week protocol
for ACTH1–39, extending our previous protocol,14was
initiated. Acthar Gel (80 IU/mL; Questcor Pharmaceut-
icals, Union City, CA) was injected IM at 75 IU/m2
twice a day for 1 week, daily for 1 week, on alternate
days for 2 weeks, then slowly tapered to a final dose
of 5 IU/m2at 1 yr. IVIg was induced at 2 g/kg (di-
vided over 2 days) and maintained at 1 g/kg once a
month with acetaminophen and diphenhydramine pre-
treatments. Clinical evaluations were made about 1 mo
after IVIg. Concomitant prophylactic treatments were
trimethoprim-sulfamethoxazole, ranitidine HCl, cal-
cium with vitamin D, and a 2-g low sodium diet.
Patients received speech, occupational, and physical
Lumbar puncture was performed at baseline and 6
mo after completion of rituximab, using methods to
obtain CSF atraumatically in children.4In fresh CSF
and corresponding blood, the expression of lymphocyte
surface antigens was investigated by flow cytometry,
using a comprehensive panel of monoclonal antibodies
to adhesion proteins in combination with anti-CD3 and
anti-CD45 antibodies.4CSF-quantitative Ig was meas-
ured by Nephelometry at Specialty Labs (Santa Mon-
Blood for complete blood count, quantitative Ig,
blood lymphocyte subsets, human anti-chimeric anti-
body (HACA) was collected at baseline and at intervals.
The HACA assay, a proprietary bridging ELISA assay
of Genentech, was performed by Covance Laboratories
(Chantilly, VA), using rituximab as the capture reagent
and biotinylated rituximab and streptavidin-horseradish
peroxidase (Jackson ImmunoResearch
West Grove, PA) for detection. A calibrator curve was
prepared with proprietary goat polyclonal antibodies to
rituximab. Serum Ig was quantitated by the Tina-quant
antigen-antibody turbity assay in the clinical laboratory
(St. John’s Hospital, Springfield, IL).
Clinical outcome was rated by the co-investigator
(E.T.) from videotapes using a validated 12-item motor
evaluation scale.15Each item was scored in increasing
severity from 0 to 3.1Subscores were converted to a
total score to designate mild (0–12), moderate (13–24),
and severe (25–36) categories. The rater was blinded
to the order (pretreatment vs. treatment) in all subjects.
The level of significance was P < 0.05. Time-course
data were analyzed on the Statistical Analysis System
by one-way analysis of variance (ANOVA) with
repeated measures, and follow-up comparisons of
means were made by the least square means procedure.
Bonferroni corrections were made for multiple compar-
isons. CSF data were analyzed by paired t tests, and
correlation analysis by Pearson correlations.
OMS Motor Severity
Treatment reduced the total score (ANOVA, P <
0.0001; Fig. 1A). At the 6-mo evaluation, all subjects
2M.R. PRANZATELLI ET AL.
Movement Disorders, Vol. 00, No. 00, 2010
had improved by ‡6 scale points, 8 by ‡12 points, 7
by ‡18 points, and 1 by ‡ 24 points. Opsoclonus
and action myoclonus disappeared rapidly in parallel
with B cell depletion, but some other motor compo-
nents, such as gait ataxia, improved more slowly
(Fig. 1B). Clinical improvement continued even as
blood B cells and serum IgM began to recover (Fig.
1D). Treatment had functional impact. Seven children
were not ambulating independently at the initial visit,
but 6 mo after completion of rituximab, only one
was not walking. He received additional immunother-
apy and was ambulatory at 1 yr. ACTH dose was
tapered steadily, a total decrease of 87% (P 5
0.0001) by 1 yr. The subsequent time points differed
significantly from the first.
At baseline, the frequency of total B cells was 4.4
6 1.6% (normal ? 1%). Six mo after the final rituxi-
mab, there was significant reduction in total B cell fre-
quency (Fig. 1C) and 83% reduction in the CSF-to-
blood B cell ratio (P 5 0.001). CSF B cells were
undetectable in five of 11 children. CD191
CD201B cells (Fig. 1E) were correlated (r 5 0.87,
P < 0.0001).
Treatment with rituximab depleted both relative and
absolute blood B cell pools. Total blood B cells
(CD32CD191) plummeted to 0% by 1 mo after the
FIG. 1. (A) Clinical efficacy (reduction in motor severity) and effect on immunological measures. Data are total score means 6 SEM. Asterisks
indicate significant differences compared with the score at initial evaluation. The initiation of rituximab is denoted by time 5 0, and time 5 2
represents the 1-mo evaluation after completion of rituximab. Compared with baseline, mean total score decreased by 43% at 1 mo, 59% at 3 mo,
73% at 6 mo, and 81% at 1 yr (P < 0.0001). (B) Mean subscores for key motor components of OMS Scale (scale items 1, 7, and 11). Upper ex-
tremity action myoclonus was scored as reaches target with no jerks (0), minimal jerks (1), moderate to severe jerks (2), or unable to reach target
due to jerks (3). (C) Dot plots of CD191B cell frequency in CSF. For pediatric neurological controls of mean age (6SD) 9.3 6 1.2 yr (n 5 16),
the median CSF B cell frequency was 0.71% (National Pediatric Myoclonus Center database). (D) The relation between changes in total score,
blood B cells, and serum IgM concentration is shown. IgM was still below the reference range in five of nine subjects at 1 yr.
3 RITUXIMAB IN OMS
Movement Disorders, Vol. 00, No. 00, 2010
last rituximab infusion, corresponding to a reduction in
total B cell counts from 1,377 (1,001–1,753, 95% CI)
to 1.7 (0.2–3.1) per mm3(P 5 0.0002). Six mo after
the last rituximab infusion, the reduction in B cell fre-
quency (261%) was still significant (P < 0.0001).
Serum IgM concentration declined rapidly and
returned to 63% of the pretreatment values by 12 mo
(P < 0.0001). Serum IgG and IgA levels did not
change significantly. Baseline CSF IgG concentrations
were normal at 0.79 mg/dL (0.7–0.9, 95% CI) and not
reduced by rituximab. Six of 11 subjects initially had
detectable CSF IgM; by 6 mo, only one had detectable
HACA was measured in eight children. Two chil-
dren (25%) developed HACA antibodies, one at 6 mo
and the other at 12 mo (Table 1). The antibody con-
centrations were low (<100 ng/mL) and showed no
clear relationship to efficacy or safety.
Relapse and Other Therapies
During the study, two subjects (25%) relapsed, one
with a prior history (Table 1). No consistent relapse
profile was identified. A partial responder had a com-
plete response to additional immunotherapy given out-
side the study protocol.
The frequency of relapse or need for treatment of
side effects was low (Table 1). Adverse events caused
no lasting difficulties and were proportionate among
the three immunotherapies.
Three-pronged initial immunotherapy is based on the
strategy that delivering adequate treatment at the onset
of OMS with agents working by various mechanisms
will induce faster neurological remission, lower the
relapse rate, and proffer the best long-term prognosis.
It was rapidly acting, very efficacious, and feasible in
OMS, with only slightly more side effects than with
conventional agents alone. B cell targeting resulted in
remarkable CSF B cell depletion. The clinical response
was sustained in most patients and better than antici-
pated for ACTH and IVIg alone, based on a survey
study of 105 cases.2In a disorder notorious for relapse,
the 17% relapse rate in this study is one-third to one-
quarter lower than reported for conventional agents
alone2,12and merits long-term follow-up.
Despite the small sample size that characterizes
studies of rare orphan diseases, this response has impli-
cations for OMS in adults and other paraneoplastic
syndromes in which CSF B cells are expanded.16It is
pertinent to movement disorder specialists, especially
those dealing with children. The patient who responded
to additional immunotherapy emphasizes how retrieva-
ble children with OMS can be.
Open-label methodology leaves concerns of placebo
effects. However, the authors have data on a negative
double-blind trial in OMS with the same scale,15sug-
TABLE 1. Relapse, partial-responders, adverse events,
First 6 months
Second 6 months
Because not ambulatoryb
Treatment emergent (SAE)c
Other treatment emergentd
Bacterial infection (SAE)e
aOne responded to ACTH increase by 10 IU/m2, a dose doubling,
the other to treatment of infection alone. Both were girls and had pe-
ripheral B cell repletion. One was severe at baseline; the other had a
prior tumor. Neither was HACA positive.
bReceived cyclophosphamide and became ambulatory.
cCough on first infusion. Rituximab was stopped and then resumed
dReduced diastolic blood pressure (asymptomatic) was managed
by stopping rituximab for 30 min or giving fluids while continuing.
eE. coli urinary tract infection 5 mo after rituximab despite pent-
fExcessive weight gain or facial puffiness.
gAsymptomatic, occurring on high-dose end of ACTH schedule.
hFever, headache, nausea, vomiting, or malaise.
iConcentrations (ng/mL): 8.4, 20. One with HACA at 6 months
was seronegative at 12 months.
adverse event; IVIg, intravenous immunoglobulins; HACA, human
AE, adverse event; SAE, serious
4M.R. PRANZATELLI ET AL.
Movement Disorders, Vol. 00, No. 00, 2010
gesting that our observations will withstand a double-
blind study. Also, not all immunotherapy for OMS is
The safety profile of combination therapy was good.
Each agent caused adverse events, but they were re-
versible. Occurrence of bacterial infection, although
low, emphasizes the need for taking prophylactic steps.
The risk of more serious infection was perhaps also
offset by preserved serum IgG levels from long-term
plasma cells (they lack CD20). The frequency of
HACA in our study was typical, and titers were not
high. These results are conducive to encouraging a fur-
ther, controlled, larger cohort study with long-term
LEGENDS TO THE VIDEO
Segment 1. Representative before-and-after video-
tapes were excerpted to illustrate response in four
patients. Treatment status is indicated as ‘‘Preri-
tuximab’’ (labeled in white letters and displayed on the
left of each frame) or ‘‘Postrituximab’’ (labeled in yel-
low letters on the right). Opsoclonus was elicited during
tracking of a small object held above the frame. To
evoke action myoclonus with or without dysmetria,
children were instructed to bring a drinking cup to their
mouth, stack small wooden blocks into a straight tower,
and put a paperclip into a small bottle with arms out-
stretched. Ataxia was demonstrated on testing of station
Acknowledgments: We thank Ronald McDonald House
(Springfield, IL), Miracle Flights for Kids (Green Valley,
NV), Air Charity Network (Addison, TX), the Baylis Day
Surgery Staff (Springfield, IL), and participating families.
This study was supported by grants from Genentech, South
San Francisco, CA, and Biogen IDEC, San Diego, CA, Ques-
tcor Pharmaceuticals, Union City, CA, Thrasher Research
Fund, Chicago Institute of Neurosurgery and Neuroresearch
Foundation, and the Spastic Paralysis Research Foundation
(Illinois-Eastern Iowa District, Kiwanis International) (to
M.R.P.). The results were presented as a poster at the 36th
Annual Meeting of the Child Neurology Society, Quebec
City, Quebec, Canada, October 10–13, 2007.
Author Roles: M.P. and E.T. designed and organized the
study; all authors contributed to its execution. E.T., M.P.,
and J.S. collected data. J.S. videotaped the patients and made
the movies, which were edited by E.T. and M.P. Statistical
analysis was performed by J.S, J.C., and S.V. The manuscript
was drafted by M.P. and reviewed and critiqued by all
1. Pranzatelli MR, Travelstead AL, Tate ED, Allison TJ, Verhulst
SJ. CSF B cell expansion in opsoclonus-myoclonus syndrome: a
biomarker of disease activity. Mov Dis 2004;19:770–777.
2. Tate ED, Allison TJ, Pranzatelli MR, Verhulst SJ. Neuroepide-
miologic trends in 105 US cases of pediatric opsoclonus-myoclo-
nus syndrome. J Pediatr Oncol Nurs 2005;22:8–19.
3. Dalakas MC. B cells as therapeutic targets in autoimmune neuro-
logical disorders. Nat Clin Pract Neurol 2008;4:557–567.
4. Pranzatelli MR, Travelstead AL, Tate ED, et al. B- and T-cell
markers in opsoclonus-myoclonus syndrome: immunophenotyp-
ing of CSF lymphocytes. Neurology 2004;62:1526–1532.
5. Pranzatelli MR, Tate ED, Travelstead AL, et al. Rituximab (anti-
CD20) adjunctive therapy for opsoclonus-myoclonus syndrome.
J Pediatr Hematol Oncol 2006;28:585–593.
6. Pranzatelli MR, Tate ED, Travelstead AL, Verhulst SJ. CSF B
cell over-expansion in paraneoplastic opsoclonus-myoclonus:
effect of rituximab, an anti-B cell monoclonal antibody. Neurol-
ogy 2003;60(suppl 1):A395.
7. Pranzatelli MR, Tate ED, Travelstead AL, Longee D. Immuno-
logic and clinical responses to rituximab in a child with opsoclo-
nus-myoclonus. Pediatrics 2005;115:e115–e119.
8. Tersak JM, Safier RA, Schor NF. Rituximab (anti-CD20) in the
treatment of refractory neuroblastoma-associated opsoclonus-
myoclonus syndrome. Ann Neurol 2005;58(suppl 9):S111.
9. Bell J, Moran C, Blatt J. Response to rituximab in a child with
neuroblastoma and opsoclonus-myoclonus. Pediatr Blood Cancer
10. Leen WG, Weemaes CM, Verbeek MM, Willemsen MA, Rotte-
veel JJ. Rituximab and intravenous immunoglobulins for relaps-
ing postinfectious opsoclonus-myoclonus syndrome. Pediatr Neu-
11. Corapcioglu F, Mutlu H, Kara B, et al. Response to rituximab
and prednisolone for opsoclonus-myoclonus-ataxia syndrome in a
child with ganglioneuroblastoma. Pediatr Hematol Oncol 2008;
12. Pranzatelli MR. The immunopharmacology of the opsoclonus-
myoclonus syndrome. Clin Neuropharmacol 1996;19:1–47.
13. Guilino LB, Bussel JB, Neufeld EJ. Treatment with rituximab in
benign and malignant hematologic disorders in children. J Pediatr
14. Pranzatelli MR, Huang Y, Tate E, et al. Monoaminergic effects
of high dose corticotropin in corticotropin-responsive pediatric
opsoclonus-myoclonus. Mov Disord 1998;13:522–528.
15. Pranzatelli MR, Tate ED, Galvan I, Wheeler A. Controlled pilot
study of piracetam for pediatric opsoclonus-myoclonus. Clin
16. de Graaf M, de Beukelaar J, Bergsma J, et al. B and T cell
imbalances in CSF of patients with Hu-antibody associated PNS.
J Neuroimmunol 2008;195:164–170.
5 RITUXIMAB IN OMS
Movement Disorders, Vol. 00, No. 00, 2010