The Trials and Errors in MS Therapy.

The International MS Journal 2008; 15: 79–90 79
� The Trials and Errors in MS Therapy
Introduction
Treatment options in multiple sclerosis (MS) have
undoubtedly broadened over the past decade:
immunosuppressants (e.g. mitoxantrone) as well as
immunomodulatory drugs (beta-interferons [IFNβ],
glatiramer acetate [GA]) are nowadays frequently
applied. The recent approval of the monoclonal
antibody natalizumab has opened a new era of
immunoselective MS therapy. However, available
substances are only partially effective and some of
them are associated with considerable long-term
toxicity. Moreover, the definite risk-to-benefit ratio is
not yet entirely clear. Consequently, there is still
tremendous research activity to develop novel
compounds for future MS therapy.1,2
Despite considerable progress in the care of MS
patients one has to keep in mind that overall, there
are few successful agents in MS and many face a
large number of therapeutic disappointments3–5
(those investigated before 2002 were discussed by
the authors in a previous review4 and will not be
mentioned again in this article). The medical
community had to experience that rational
pathophysiological concepts, valid data from animal
models or even promising Phase I/II studies do not
guarantee final efficacy and translation into the
clinic. In addition, some compounds are associated
with unforseen adverse effects when tested in humans
making the risk-to-benefit ratio unacceptable
(Table 1). Whereas positive studies usually cause a
sensation and find their way into prestigious
journals, many well-designed but negative trials in
the past have unfortunately never been published.6
Although this publication bias has already been
The development programme for novel therapies in
multiple sclerosis (MS) has an impressive track record
which is unique in the field of neurology, and has led to the
approval of several drugs during the past decade.
However, therapeutic prosperities face numerous trials that
either failed to show efficacy or that had to be halted
because of other reasons including adverse events.
Moreover, certain treatment strategies are controversial,
both for reasons of practicability and for their true clinical
benefit. There are serious caveats that highly
immunoselective approaches such as monoclonal
antibodies can only be applied at the expense of an
increased risk of acute or long-term adverse effects. This
review focuses on the most important clinical trials on yet
unlicensed compounds in relapsing-remitting and
secondary-progressive MS which failed, were halted or
are associated with significant adverse effects since 2002.
Furthermore, we discuss the implications these experiences
have for our current view of MS pathogenesis as well as
future study design. Examples include agents that target
leukocyte differentiation molecules, co-stimulatory
molecules, adhesion molecules and chemotaxis, as well as
novel immunomodulators and anti-infective therapies.
KEY WORDS:
MULTIPLE SCLEROSIS; TRIAL; THERAPY; CAMPATH-1H; CTLA-4-IG; ANTIBIOTICS
The Trials and Errors in MS Therapy
Christoph Kleinschnitz, Sven G Meuth, Heinz Wiendl
Department of Neurology, University of Würzburg, Würzburg, Germany
Search strategy and selection criteria: Studies were identified by querying PubMed from 2002 until
April 2007 with the term “multiple sclerosis” and combining it with “therapy”, “treatment trials”, “failed
treatment trials” and “adverse events”. Studies were also identified from conference information and personal
communications of the authors. Abstracts and reports from meetings were included because failed trials are
sometimes not published in scientific journals.
Summary

reduced nowadays the situation is still unsatisfactory
since there is a lot to learn from negative results, and
critical reflection is highly important for
understanding human MS pathogenesis and
improving future clinical trial design.
This review summarizes the pharmacological
background, the experimental basis and the clinical
trial data of yet unlicensed compounds which were
ineffective, led to early trial termination for other
reasons or were associated with considerable
adverse effects when tested in human Phase I–III
studies in relapsing-remitting (RR) and secondary-
progressive (SP) MS since 2002.
Failed Trials of Novel
Immunomodulators:
Phosphodiesterase Inhibitors,
Hydrolytic Enzymes
Phosphodiesterase Inhibitors
Phosphodiesterase (PDE) inhibitors are anti-
inflammatory by modulating intracellular cyclic
adenosine (cAMP) and cyclic guanosine (cGMP)
monophosphate concentrations.7 Moreover, they are
able to alter the local cytokine milieu, T-cell priming
and co-stimulatory signalling thus skewing the
immune balance towards Th2-driven responses.8
Treatment with PDE inhibitors has proven to be
effective in several experimental autoimmune
encephalomyelitis (EAE) models.9–12 PDE inhibitors
can be applied orally.
The non-specific PDE inhibitor ibudilast induced a
favourable immune shift in certain T-cell subtypes
and natural killer (NK) cells in human MS.13 However,
an open-label, cross-over study of 18 patients testing
the PDE-4 specific inhibitor rolipram was recently
terminated due to lack of clinical efficacy. Another
Phase I/II clinical trial on rolipram in MS was recently
discontinued after enrolment of only eight patients as
an increase in the average number of new and total
Gadolinium (Gd)-enhancing lesions was observed in
MRI.14 Clinical outcome measures remained stable or
improved. Despite aggravated MRI activity, rolipram
seems immunologically active in vivo in the expected
manner since T-cell proliferation and the production of
Th1 cytokines were inhibited.
The concept of PDE inhibition can be regarded as a
genuine example of immunomodulation. However, data
from animal experiments are partially contradictory
and net effects in the human system are difficult to
predict. While the use of first-generation PDE
inhibitors was significantly limited by concomitant
nausea and emesis, second-generation compounds
have largely been improved concerning their side-
effect profile. The anti-inflammatory properties of
PDE inhibitors are thought to result from changes
in the levels of intracellular cAMP and cGMP.
However, cAMP is probably a key player in
regulating T-cell–mediated immune suppression.15,16
Therefore, down-regulation of cAMP could in parallel
impede regulatory T-cell function, thus counteracting
the anti-inflammatory effects of PDE. These competing
effects might explain the dissociation between
immunological, clinical and MRI outcome measures
observed in the above mentioned Phase I/II study.
Hydrolytic Enzymes
Hydrolytic enzymes (HE), e.g. phlogenzym consisting
of bromelain, trypsin and the anti-oxidant rutosid,
have a long tradition as potential therapy for various
diseases and are popular among laymen. Proponents
of HE argue that these substances increase the
specific hydrolytic activity on putatively ingested
(auto)antigens in the serum.17–19 Oral administration
of phlogenzym nearly completely protected from EAE
which was associated with a shift towards an anti-
inflammatory Th2 cytokine profile and increased T-
cell activation thresholds.20
So far only one, but large, clinical trial was
performed to test the safety and efficacy of oral HE
in MS.21 This randomized, double-blind, placebo-
controlled study included 301 patients suffering from
RRMS. The study drug was well tolerated but failed
to show any treatment effect on clinical or MRI
parameters.21
HE belong to a group of “popular” MS drugs that
are accompanied by considerable non-scientific
media hype. Unsubstantiated hopes are raised in
patients because of single case observations, but
fundamental proof of efficacy is mostly lacking.
Nonetheless, their use among MS patients is common
partly due to uncritical prescription by physicians.
This well-designed study contributes valid negative
data and underlines that the administration of HE
cannot be recommended in MS.
80 The International MS Journal 2008; 15: 79–90
The Trials and Errors in MS Therapy �

Failed Trials on Anti-Infective
Therapies: Virostatics, Antibiotics
Infectious particles have been thought to represent
“environmental triggers or causes” for MS for a long
time. In particular, certain viruses and bacteria have
been linked to disease development or maintenance
and conceptually their eradication could cure or
prevent MS.22
Virostatics
Viral infections (e.g. human herpesvirus 6 [HHV-6],
Epstein-Barr-Virus [EBV]) trigger approximately 25%
of MS relapses and often precede disease onset.23–27
Consequently, several virostatics have been tested in
clinical trials. In a double-blind, placebo-controlled
study 60 RRMS patients received aciclovir (800
mg/day) acting against herpes viruses, or placebo
over 2 years. In summary, there was a statistically
non-significant trend towards fewer relapses (~34%
reduction) in individuals on antiviral medication but
clear proof of efficacy was lacking.28 Moreover,
valaciclovir, another virostatic agent addressing the
family of herpes viruses, was tested in a Phase II
randomized, double-blind, placebo-controlled study
in 70 patients using new active MRI lesions defined
as new Gd-enhancing lesions, new non-enhancing
T2 lesions not observed previously, enlarged non-
enhancing lesions or recurrent lesions as paraclinical
read-out parameter. The study period comprised
24 weeks.29 While valaciclovir had no influence on
active MRI lesions in the whole cohort, a subgroup
analysis suggested a beneficial effect in patients with
very active disease. Both aciclovir and valaciclovir
were well tolerated in MS even during continuous
long-term application.
Antibiotics
The obligate intracellular bacterium Chlamydia
pneumoniae (CP) was proposed to be the long
searched pathogen that triggers MS in genetically
susceptible individuals30 and a previous report
suggested that relapses in MS patients are linked to
CP infections.31 Moreover, intrathecal antibody
production against CP has been described in MS
patients as part of the polyspecific immune
response.32 However, serologically defined (but
clinically asymptomatic) CP carrier status is common
The International MS Journal 2008; 15: 79–90 81
� The Trials and Errors in MS Therapy
Key Points
• A significant number of trials in MS therapy have
failed in the past
• Negative trials are often not published
• This publication bias can negatively influence the
understanding of MS pathophysiology and the
design of future trials
• The definition of a "failed" trial is complex and
can be related to lack of efficacy, inadmissible
adverse effects or premature trial termination,
e.g. due to financial considerations
• There are many lessons to learn from negative
trials, including the influence on our
understanding of MS-pathophysiology, strengths
and weaknesses of trial designs, the relevance of
patient selection or disease heterogeneity
among healthy persons as well as MS patients and
individuals suffering from other neurological diseases
making causal interpretations difficult. Moreover,
epidemiological studies evaluating the correlation
between CP infections and the development of MS
have been inconclusive so far.33,34 To further address
the role of CP in MS, the US National Multiple
Sclerosis Society funded a proof-of-principle trial on
the use of rifampicin (300 mg twice daily) and
azithromycin (500 mg every other day) versus
placebo over 6 months in eight female RRMS patients.
Both substances show profound activity against CP.
The bacterium was eradicated from the cerebrospinal
fluid (CSF) in three out of four women that had
received the antibiotics. Interestingly, the number and
volume of Gd-enhancing MRI lesions tended to
increase only in the placebo group during the
observation period. The investigators concluded that
these results provide a rationale for a larger study on
the safety and efficacy of antibiotic treatment in MS.35
Although a large number of epidemiological
observations suggest a correlation of viral and
bacterial infections with the development and
maintenance of MS, most clinical trials on anti-infective
therapies were negative or inconclusive at best. The
fact that persistent infections by common pathogens
are very frequent among healthy individuals, as well

82 The International MS Journal 2008; 15: 79–90
The Trials and Errors in MS Therapy �
Table 1: Treatment trials in MS that have failed, been halted prematurely or associated with adverse effects
Agent
(Assumed)
Mechanism
of action
Characteristics/
Trial design
Disease
course
Outcome
MRI
Outcome
Clinical/side-effects
Further/
Ongoing trials Comments
Anti-CD52
(Alemtuzumab,
Campath-1)
Depletion of
pathogenetic
relevant
T-cells
Dose-comparison study;
(12 mg/day and
24 mg/day iv),
head-to-head design
versus IFNB-1a 44 µg sc,
Phase II
Open-label study;
36 patients; 5 doses
of 20 mg on
5 consecutive days
RRMS
SPMS
Positive
Positive
Positive
1 fatal case of ITP;
autoimmune
hyperthyroidism
Negative
Continued disease
progression in
32/36 patients
Finished
Phase III study
planned
Finished
High anti-inflam-
matory potential,
considerable adverse
effect profile (ITP,
hyperthyroidism)
Co-stimulatory molecules
CTLA-4-Ig
(abatacept,
RG2077)
Negative
regulator of
T-cell function;
effects on
CD4+ and
CD25+
regulatory
cells
Pilot study; 16 patients;
single infusions (2, 10,
20 or 35 mg/kg) or
multi-dose of 10 mg/kg
330 patients; double-
blind, placebo-controlled,
multicentre, Phase II trial
(2, 10 mg/kg infusions
on Days 1, 15 and 29,
and than every 4 weeks
until Day 197)
RRMS
RRMS
–
Accumulation of
inflammatory
MRI activity
(low-dose verum
group); fewer
new Gd-T1
enhancing lesions
in 10 mg/kg
group
No major
adverse effects
Accumulation of
relapses in
low-dose verum
group; less
relapses in
10 mg/kg group
Finished
Study was
prematurely
halted
Reason for worse
outcome in the verum
group probable
randomization
failure; the clinical
efficacy in MS
remains unclear
Anti-
CD40L
(anti-
CD154,
IDEC-131)
Antibody
interacting
with the co-
stimulatory
pathway
CD40-CD40L
Pilot study (IDEC-131);
15 patients
46 patients; double-blind,
placebo-controlled,
Phase II trial; (15 mg/kg)
iv for 5 weeks and then
every month for 3 months
RRMS
RRMS
Positive
–
No relapses for at
least 6 months
MS study halted,
because 3 patients
developed
thrombembolism in
a study of Crohn´s
disease
Finished
Stopped,
although all
patients had
pre-existing risk
factors for
thromboembolism
Cross-reactivity with
platelet CD40L
possible
Anti-adhesion molecules
Anti-LFA1
(CD11/
CD18,
Hu23F2G)
LFA1/ICAM-
antagonist,
inhibition of
cell adhesion
between
leukocytes
and vascular
endothelial
cells
Open Phase I study
(24 patients) and a
subsequent Phase II
study (169 patients)
with humanized
anti-LFA1 (Hu23F2G)
SPMS,
RRMS
Negative Negative Finished Proof of efficacy
lacking
Chemotaxis
CCR1
antagonist
(BX-471)
Chemokine
receptor
antagonist,
reduces
transmigration
of autoreactive
T-cells to the
CNS
600 mg orally 3 times
per day, Phase II
RRMS Negative Negative Finished; trials
on other CCR
antagonists
(CCR2 – 2
Phase I and 1
Phase II;
CCR5 – Phase II
and III trial in
HIV) planned/
ongoing
Complexity of the
cytokine system,
determination of the
most promising CCR
difficult
Novel immunomodulators
Phosphodiesterase
inhibitors
(ibudilast,
rolipram)
Down-
regulation of
inflammatory
responses by
changing levels
of cAMP and
cGMP; shifting
the cytokine
milieu to
Th2-driven
responses
Open-label, cross-over
study; 18 patients
(rolipram)
RRMS,
SPMS
– Negative Terminated
due to lack
of clinical
efficacy;
dose-
dependent
side-effects,
e.g. nausea
and emesis
Regarded as true
immunomodulatory
treatment; first
generation of PDE
inhibitors with
problems concerning
side-effects; proof of
efficacy lacking

The International MS Journal 2008; 15: 79–90 83
� The Trials and Errors in MS Therapy
as MS patients significantly limits any statements on
causality. Nevertheless, EBV is still the candidate that
best fulfils the requirements of an “environmental
trigger” for MS.25–27,36,37 Recent data suggest that
repeated or chronic infections with EBV and even non-
pathogenic viruses (e.g. TT virus) can induce the
expansion of T-cells specific for conserved viral
peptides which are identical (and thus cross-reactive)
to tissue-derived antigens (molecular mimicry). Against
this background, it is doubtful that elimination of
viruses after MS onset would still be beneficial since
during this stage of the disease adaptive immune
responses might promote central nervous system (CNS)
inflammation independently from the initial trigger.
Moreover, this approach is limited by the fact that
agents which are capable of completely eradicating
persisting (herpes) viruses do not exist. Available data
on the true efficacy of antibiotics in MS are likewise
scarce and a causal relation between bacterial
infections and MS is still unproven. The reported
positive effects of certain tetracyclines such as
minocyclin38–40 are most likely not related to their
antimicrobial properties per se but due to other
mechanisms like inhibition of matrix metalloproteinases
and reduced leukocyte transmigration.41
Targeting the Adhesion Molecule
LFA-1: Anti-LFA-1 Monoclonal
Antibody (Hu23F2G)
One critical step in the pathophysiology of MS is
recruitment of autoreactive leukocytes from the
periphery to the CNS. This process is regulated
by cell adhesion molecules expressed on the surface
of infiltrating cells and their respective counterparts
on the endothelium. It therefore appears plausible
that specific antibodies directed against these
structures can potently reduce leukocyte migration
across the blood–brain barrier.
Certainly, the most exciting development in MS
therapy during the last few years is the approval of the
humanized monoclonal antibody (mAb) natalizumab
which targets α4β1-integrin (VLA-4) on the surface of
encephalitogenic leukocytes.42 Numerous excellent
articles have dealt with the specific risks and benefits of
Table 1 (continued)
Agent
(Assumed)
Mechanism
of action
Characteristics/
Trial design
Disease
course
Outcome
MRI
Outcome
Clinical/side-effects
Further/
Ongoing trials Comments
Hydrolytic
enzymes
Increase the
specific
hydrolytic
activity on
putative
ingested (auto)
antigens in the
serum
Oral drug (90 mg
bromelain + 48 mg
trypsin + 100 mg
rutosid); Phase III
RRMS Negative Negative Finished Use in clinical
practice obsolete
Anti-infectious therapies
Antiviral agents
(aciclovir,
valaciclovir)
Eradication
of potential
infectious
MS triggers
Double-blind, placebo-
controlled trial;
aciclovir (800 mg)
for 2 years;
60 patients
Phase II, valaciclovir
for 24 weeks,
randomized, double-
blind, placebo-
controlled; 70 patients
RRMS
RRMS
Not used
Marginal
beneficial effects
in patients with
high disease
activity on MRI
Not significant
Tendency towards
fewer exacerbations
Negative
Finished
Finished
Casual role of
viruses in MS
unproven. Efficacy
of eradication
questionable due
to molecular
mimicry/antigen
spreading
Antibiotics
(rifampicin,
azithromycin)
Eradication
of potential
infectious
MS triggers
Pilot study of rifampicin
(300 mg twice daily)
and azithromycin
(500 mg every
other day) over
6 months; 8 patients
RRMS Positive Negative Finished Causal role of
bacteria in MS
unproven;
positive effects most
likely not directly
related to
antimicrobial
properties
cAMP: cyclic adenosine monophosphate; cGMP: cyclic guanosine monophosphate; CCR-1: chemokine receptor 1; HIV: human immunodeficiency virus;
ICAM: intracellular adhesion molecule 1; IFNB: interferon beta; ITP: idiopathic thrombocytopenic purpura; iv: intravenously; MRI: magnetic resonance imaging;
PDE: phosphodiesterase; RRMS: relapsing-remitting multiple sclerosis; sc: subcutaneously; SPMS: secondary-progressive multiple sclerosis.