Articles

Oral ngolimod in primary progressive multiple sclerosis

(INFORMS): a phase 3, randomised, double-blind,
placebo-controlled trial
Fred Lublin*, David H Miller*, Mark S Freedman, Bruce A C Cree, Jerry S Wolinsky, Howard Weiner, Catherine Lubetzki, Hans-Peter Hartung,
Xavier Montalban, Bernard M J Uitdehaag, Martin Merschhemke, Bingbing Li, Norman Putzki, Fonda C Liu, Dieter A Hring, Ludwig Kappos,
on behalf of the INFORMS study investigators

Summary
Background No treatments have been approved for primary progressive multiple sclerosis. Fingolimod, an oral sphingosine
1-phosphate receptor modulator, is eective in relapse-onset multiple sclerosis, but has not been assessed in primary
progressive multiple sclerosis. We assessed the safety and ecacy of ngolimod in patients with primary progressive
multiple sclerosis.
Methods In INFORMS, a multicentre, double-blind, placebo-controlled parallel-group study, patients with primary
progressive multiple sclerosis recruited across 148 centres in 18 countries were randomly allocated (1:1) with computergenerated blocks to receive oral ngolimod or placebo for at least 36 months and a maximum of 5 years. Patients were
initially assigned to ngolimod 125 mg per day or placebo (cohort 1); however, after a protocol amendment on Nov 19,
2009, patients were switched in a masked manner to ngolimod 05 mg, whereas those on placebo continued on
matching placebo. From then onwards, patients were assigned to receive ngolimod 05 mg/day or placebo (cohort 2).
Key inclusion criteria were age 2565 years, clinical diagnosis of primary progressive multiple sclerosis, 1 year or more of
disease progression, and two of the following criteria: positive brain MRI; positive spinal cord MRI; or positive
cerebrospinal uid. Additional eligibility criteria included disease duration of 210 years and objective evidence of
disability progression in the previous 2 years. Patients and study investigators were masked to group assignment. We
used a novel primary composite endpoint based on change from baseline in Expanded Disability Status Scale (EDSS),
25 Timed-Walk Test, or Nine-Hole Peg Test to assess time to 3-month conrmed disability progression in study
participants treated for at least 3 years. All randomised patients took at least one dose of study drug. The primary ecacy
analysis included all patients in cohort 2 and those assigned to placebo in cohort 1. The safety analysis included all patients
in cohorts 1 and 2. This study is registered with ClinicalTrials.gov, number NCT00731692. The study is now closed.
Findings 970 patients were randomly assigned between Sept 3, 2008, and Aug 30, 2011 (147 to ngolimod 125 mg and
133 to placebo in cohort 1; 336 to ngolimod 05 mg and 354 to placebo in cohort 2). The ecacy analysis set (n=823)
consisted of 336 patients randomly allocated to ngolimod 05 mg and 487 to placebo. Baseline characteristics were
similar across groups and representative of a primary progressive multiple sclerosis population (48% women, mean
age 485 years [SD 84], mean EDSS 467 [SD 103], 87% free of gadolinium-enhancing lesions). By end of study,
3-month conrmed disability progression had occurred in 232 and 338 patients in the ngolimod and placebo groups,
respectively, resulting in Kaplan-Meier estimates of 772% (95% CI 71878251) of patients in the ngolimod group
versus 803% (73318725) of patients in the placebo group (risk reduction 505%; hazard ratio 095, 95% CI
080112; p=0544). Safety results were generally consistent with those of studies of ngolimod in patients with
relapse-onset multiple sclerosis. Lymphopenia occurred in 19 (6%) patients in the ngolimod group versus none in the
placebo group, bradycardia in ve (1%) versus one (<1%), and rst-degree atrioventricular block in three (1%) versus
six (1%). Serious adverse events occurred in 84 (25%) patients in the ngolimod group and 117 (24%) in the placebo
group, including macular oedema in six (2%) versus six (1%), and basal-cell carcinoma in 14 (4%) versus nine (2%).
Interpretation The anti-inammatory eects of ngolimod did not slow disease progression in primary progressive
multiple sclerosis. Therapeutic strategies for primary progressive multiple sclerosis might need dierent approaches
to those used for relapse-onset multiple sclerosis.
Funding Novartis Pharma AG.

Introduction
Multiple sclerosis is a chronic, inammatory, neurodegenerative disorder of the CNS.1 Although relapseonset multiple sclerosis (encompassing relapse-onset
and secondary progressive multiple sclerosis) is the

most frequent form, 1015% of patients with

multiple sclerosis have progressive disability from
onset with no, or very infrequent, superimposed
relapses and remissions (primary progressive multiple
sclerosis).2,3

www.thelancet.com Published online January 27, 2016 http://dx.doi.org/10.1016/S0140-6736(15)01314-8

Published Online
January 27, 2016
http://dx.doi.org/10.1016/
S0140-6736(15)01314-8
See Online/Comment
http://dx.doi.org/10.1016/
S0140-6736(16)00158-6
*Contributed equally
The INFORMS principal
investigators are listed in the
appendix
The Corinne Goldsmith
Dickinson Center for Multiple
Sclerosis, Icahn School of
Medicine at Mount Sinai,
New York, NY, USA
(Prof F Lublin MD); Queen
Square MS Centre, UCL
Institute of Neurology,
London, UK
(Prof D H Miller FMedSci);
The Ottawa Hospital Research
Institute, University of
Ottawa, Ottawa, ON, Canada
(Prof M S Freedman MD);
Multiple Sclerosis Center,
University of California
San Francisco, CA, USA
(Prof B A C Cree MD); University
of Texas Health Science Center
at Houston, Houston, TX, USA
(Prof J S Wolinsky MD); Brigham
and Womens Hospital,
Harvard Medical School,
Boston, MA, USA
(Prof H Weiner MD); University
Paris 6, Salptrire Hospital
APHP, Center of Clinical
Investigation, Paris, France
(Prof C Lubetzki MD);
Department of Neurology,
Medical Faculty,
Heinrich-Heine University,
Dsseldorf, Germany
(Prof H-P Hartung MD);
Hospital Universitari Vall
dHebron, Barcelona, Spain
(Prof X Montalban MD); VU
University Medical Center,
Amsterdam, Netherlands
(Prof B M J Uitdehaag MD);
Novartis Pharma AG, Basel,
Switzerland
(M Merschhemke MD,
N Putzki MD, D A Hring PhD);
Novartis Pharmaceuticals

Articles

Corporation, East Hanover, NJ,

USA (B Li PhD, F C Liu PharmD);
and Departments of Medicine,
Clinical Research, Biomedicine
and Biomedical Engineering,
University Hospital, University
of Basel, Switzerland
(Prof L Kappos MD)
Correspondence to:
Prof Fred D Lublin, The Corinne
Goldsmith Dickinson Center for
Multiple Sclerosis, Icahn School
of Medicine at Mount Sinai,
New York, NY 10029-6574, USA
fred.lublin@mssm.edu

Research in context
Evidence before this study
We searched PubMed on July 8, 2015, with no restriction on
language or publication date, using the search term primary
progressive multiple sclerosis. We identied 324 articles, of which
four were primary reports of results from randomised, blinded,
placebo-controlled clinical trials. Of these, two were large-scale,
phase 3 trials: the PROMiSe trial, which compared glatiramer
acetate with placebo, was designed as a 3-year trial but was
terminated early; and the 2-year OLYMPUS trial, which compared
rituximab with placebo. Both studies used the Expanded
Disability Status Scale as the primary outcome measure and
showed no treatment benet on disability progression or brain
volume loss (BVL), despite eects on some MRI variables.
Subsequent discussion of these studies included the eect of the
recruited patient populations, trial durations, and the sensitivity
of the endpoints used to detect treatment dierences in patients
with primary progressive multiple sclerosis.
Added value of this study
INFORMS addressed the important limitations of previous
studies in primary progressive multiple sclerosis. INFORMS was

Although part of the multiple sclerosis disease

spectrum,4 primary progressive multiple sclerosis
diers from relapse-onset multiple sclerosis in several
ways. The inammatory component, as measured by
development of gadolinium (Gd)-enhancing MRI
lesions, is less prominent in primary progressive
multiple sclerosis than in relapse-onset multiple
sclerosis.2,5 Men and women are aected equally, by
contrast with the higher frequency of relapse-onset
multiple sclerosis in women.3,6 Patients with primary
progressive multiple sclerosis are generally about
10 years older at diagnosis than those with relapsing
multiple sclerosis (mean age around 40 years for
primary progressive multiple sclerosis vs 30 years for
relapse-onset multiple sclerosis).2,6 Moreover, disability
progresses more rapidly in primary progressive
multiple sclerosis, such that the severity of disability
according to age is similar to that in relapse-onset
multiple sclerosis.7
Despite the range of eective options available for
relapse-onset multiple sclerosis, no treatment has been
shown to change the disease course in primary
progressive multiple sclerosis.8,9 Fingolimod (Novartis
Pharma AG, Basel, Switzerland) is an oral sphingosine
1-phosphate (S1P) receptor modulator that, by
downregulation of the S1P1 receptor subtype, prevents
lymphocyte egress from lymphoid tissues into
circulation. In relapse-onset multiple sclerosis,
ngolimod reduces the frequency of relapses, delays
disability progression, and reduces MRI lesion activity
versus placebo, and decreases the frequency of relapses
and active MRI lesions versus intramuscular interferon
2

designed with a novel composite endpoint to increase the

sensitivity to detect a treatment eect, and patients were
exposed to study drug for at least 3 years. Furthermore, the
study was powered to 90% for detection of a clinically
meaningful treatment eect on disability progression.
INFORMS successfully recruited a population with primary
progressive multiple sclerosis with very low inammatory
activity (few relapses and low number of gadolinium-enhancing
lesions at baseline and throughout the study).
Implications of all the available evidence
Despite the adequate design and an informative study
population with respect to worsening neurological function,
INFORMS did not show a benet for ngolimod versus placebo
in terms of disability progression or BVL. The nding that the
pharmacodynamic eects of ngolimod did not aect disability
progression or BVL, despite eects on MRI-detected
inammatory lesion activity, adds to the important discussion
of the pathophysiology of progressive phase multiple sclerosis
and will guide future research.

beta-1a.1012 Furthermore, ngolimod signicantly limits

brain volume loss (BVL) in relapse-onset multiple
sclerosis compared with both placebo and intramuscular
interferon beta-1a,12,13 a measure associated with
worsening of disability.14,15
In addition to preventing lymphocyte egress from
lymphoid tissues, ngolimod can cross the bloodbrain
barrier and bind to S1P1, S1P3, and S1P5 receptors
on neural cells.16,17 Therefore, part of ngolimods
therapeutic action might be independent of its eect on
peripheral lymphocytes. S1P1 and S1P3 receptors are
strongly expressed in multiple sclerosis lesions, and this
expression has been associated with astrogliosis, a key
pathological feature of multiple sclerosis lesions.18
In-vitro and in-vivo data suggest that ngolimod
can directly inhibit neurodegeneration. Fingolimod
treatment of isolated human astrocytes desensitised the
S1P-receptor-mediated neuroinammatory pathways,
resulting in reduced astrogliosis and neurodegeneration.18
Animal models of multiple sclerosis have shown that
ngolimod treatment can reduce neuroinammatory
disease and improve CNS tissue integrity.19,20 The ability
of ngolimod to reduce inammatory inltrates into the
CNS, coupled with the potential eects on intrinsic
mechanisms in the brain, resulting in reduced
neurodegeneration, provided the rationale for the
INFORMS trial.
Previous phase 3 studies8,9 in primary progressive
multiple sclerosis used change in the Expanded
Disability Status Scale (EDSS)21 as the primary outcome
measure. However, EDSS might not be sensitive
enough to reliably detect change in clinical status in

www.thelancet.com Published online January 27, 2016 http://dx.doi.org/10.1016/S0140-6736(15)01314-8

Articles

patients with primary progressive multiple sclerosis

during a phase 3 trial.22 In this study, we combined
two additional and validated measures of disability
progression, the 25 Timed-Walk Test (25TWT),23 and
the Nine-Hole Peg Test (9-HPT),23,24 with the EDSS into
a novel composite endpoint that could provide a
sensitive and comprehensive assessment of the change
in clinical status throughout the trial. The aim of this
study was to assess the eect of ngolimod compared
with placebo on delaying the time to sustained disability
progression, with this novel composite endpoint.

Methods
Study design
This multicentre, double-blind, placebo-controlled,
parallel-group study was done in accordance with the
International Conference on Harmonisation Guidelines
for Good Clinical Practice25 and the Declaration of
Helsinki.26 Patients with primary progressive multiple
sclerosis were recruited across 148 centres in 18 countries.
Each sites institutional review board approved the
protocol. A steering committee (for membership, see
appendix) oversaw the study.

Patients
Key eligibility criteria were age 2565 years with a clinical
diagnosis of primary progressive multiple sclerosis
according to the 2005 revised McDonald criteria.27
Patients had to have 1 year or more of disease progression
plus two of the following criteria: positive brain MRI;
positive spinal cord MRI; or positive cerebrospinal uid
(CSF). Central review to conrm that the diagnostic
criteria for primary progressive multiple sclerosis were
met was needed for all patients before randomisation.
Additional inclusion criteria included time from rst
reported symptoms of 210 years before study entry;
evidence of disability progression documented by an
increase in EDSS score of 05 points or more in the
past 2 years; objective evidence of disability measured
by EDSS score of 356; pyramidal functional system
score of 2 or more; and a 25TWT of less than 30 s.
Exclusion criteria were similar to previous relapseonset multiple sclerosis studies (for complete details,
see appendix).1012 All patients gave written informed
consent.

Randomisation and masking

Patients were randomly assigned (1:1) with computergenerated blocks to receive either ngolimod or placebo.
The randomisation sequence was automatically generated
by a validated system under the responsibility of Novartis
Drug Supply Management. The randomisation occurred
in blocks of four within centre in a 1:1 ratio to ngolimod
or placebo. Allocation was concealed through the use of
blinded code-break cards with removable, scratch-o
cover for the whole double-blind treatment period. All
randomised drug assignments remained masked to

patients, investigator sta, people performing the

assessments, and data analysts for the whole double-blind
treatment period (at least 36 months and up to 5 years).
Treatment codes were accessible only to members of the
data and safety monitoring board (DSMB), which was
independent of Novartis and not otherwise involved in
the study. We achieved masking by use of identical
packaging and identical capsule colour and size for
treatment and placebo.
Heart rate reduction is a known pharmacological eect
of ngolimod that can potentially unmask study
participants. To maintain masking, an independent rst
dose administrator monitored pulse rate after the rst
dose of study drug. Employees of the funder who were
independent of the study team monitored rst dose
safety and were masked to study allocation.
Patients were initially randomly allocated (1:1) to receive
ngolimod 125 mg/day or placebo. After the decision to
select the 05 mg dose of ngolimod for submission to
regulatory authorities and to discontinue development of
ngolimod 125 mg, the protocol was amended on
Nov 19, 2009, and we stopped allocation to the 125 mg
cohort of patients. Patients who had been assigned to
125 mg ngolimod were labelled as cohort 1 and were
switched in a masked manner to ngolimod 05 mg;
those on placebo continued on matching placebo. From
this timepoint onwards, patients were randomly allocated
to a new cohort (cohort 2), which included patients who
were recruited after the protocol amendment, and were
randomly allocated (1:1) to receive ngolimod 05 mg/day
or placebo.

See Online for appendix

Procedures
Patients received either oral ngolimod or placebo
once daily for 36 months or up to a maximum
treatment duration of 5 years. Clinical assessments
(EDSS, 25TWT, and 9-HPT) were done at screening,
at randomisation (baseline), and at study visits,
including safety assessments at 2 weeks and 1 month,
2 months, 3 months, 6 months, 9 months, and
12 months during the rst year after randomisation
and then every 3 months until month 36.
An independent, specially trained and certied
evaluating physician did ecacy assessments
(Neurostatus training and documentation DVD). MRI
scans were analysed at a central MRI centre by trained
sta masked to study group assignment. EDSS
assessments were scheduled at screening, baseline,
and then every 3 months during the double-blind
treatment period, at the end of treatment, and at the
3-month follow-up visit. EDSS assessment by the
independent physician was needed at an unscheduled
visit to conrm occurance of a multiple sclerosis
relapse. The 25TWT and 9-HPT were assessed at
baseline and every 3 months throughout the
double-blind treatment period, at the end of treatment,
and at the 3-month follow-up visit.

www.thelancet.com Published online January 27, 2016 http://dx.doi.org/10.1016/S0140-6736(15)01314-8

For more on Neurostatus see

http://www.neurostatus.net

Articles

1520 patients assessed for eligibility

550 ineligible
120 inclusion criteria not met
294 exclusion criteria met
34 withdrew consent
7 intercurrent medical events
95 other reasons

970 randomised

280 randomised in cohort 1

147 assigned ngolimod 125 mg

68 discontinued treatment
6 abnormal laboratory values
4 abnormal test procedure results
2 administrative issues
25 had adverse event
2 died
1 lost to follow-up
4 protocol deviation
12 withdrew consent
1 no longer needed treatment
11 unsatisfactory therapeutic eects

79 completed study
1 o treatment

147 included in safety analysis

690 randomised in cohort 2

133 assigned placebo

336 assigned ngolimod 05 mg

52 discontinued treatment
1 abnormal laboratory values
2 abnormal test procedure results
4 administrative issues
11 had adverse event
2 died
2 protocol deviation
10 withdrew consent
20 unsatisfactory therapeutic
eects

116 discontinued treatment

19 abnormal laboratory values
3 abnormal test procedure results
2 administrative issues
28 had adverse event
1 died
3 lost to follow-up
5 protocol deviation
32 withdrew consent
23 unsatisfactory therapeutic
eects

354 assigned placebo

118 discontinued treatment

4 abnormal laboratory values
3 abnormal test procedure results
2 administrative issues
18 had adverse event
3 lost to follow- up
6 protocol deviation
36 withdrew consent
2 no longer needed treatment
44 unsatisfactory therapeutic
eects

81 completed study
0 o treatment

220 completed study

9 o treatment

236 completed study

2 o treatment

133 included in ecacy and safety analysis

336 included in ecacy and safety analysis

354 included in ecacy and safety analysis

Figure 1: Study prole

Outcomes
The primary objective was to assess the eect of
ngolimod versus placebo on delaying the time to
conrmed disability progression (CDP), which was
centrally assessed by the study statistical and
programming team. CDP was dened as the rst
occurrence of a progression according to at least one of
the following three criteria: increase from baseline EDSS
score by 1 point in patients with baseline EDSS score of
50 or lower, or by 05 points in patients with baseline
EDSS score of 55 or higher; increase of at least 20%
from baseline in the 25TWT; or increase of at least 20%
from baseline in time taken to complete the 9-HPT.
Progression in at least one of the three components had
to be conrmed for the same component (or components)
at least 3 months later at a scheduled visit.
Protocol-dened key secondary objectives were: eect of
ngolimod 05 mg versus placebo on delaying the time to
3-month CDP as measured by EDSS, and eect of
ngolimod 05 mg versus placebo on percent brain
volume change (PBVC) as a measure of BVL. PBVC was
measured using SIENA applied to T1-weighted images.28
Other secondary objectives were to assess the eect of
05 mg ngolimod versus placebo on: time to 3-month
CDP by 25TWT and 9-HPT; MRI parameters
4

(inammatory disease activity as indicated by presence of

Gd-positive lesions and disease burden as indicated by T2
and T1 hypointense lesion evolution over time); and
patient-reported outcomes using the Patient Reported
Indices in Multiple Sclerosis (PRIMUS), EuroQoL
(EQ-5D), Unidimensional Fatigue Impact Scale (UFIS),
and Multiple Sclerosis Walking Scale (MSWS-12). We also
obtained data for the pharmacokineticpharmacodynamic
relation for ngolimod.
Specications of the adverse event monitoring procedure,
as dened in the study protocol, are detailed in the
appendix, which also provides other details of the methods,
the list of members of the independent DSMB, and the
members of the independent diagnosis adjudication board.

Statistical analysis
The study was powered to 90% to detect a reduction
between ngolimod and placebo in the time to 3-month
CDP based on the composite endpoint in a log-rank test
at a two-sided signicance level of 5%: a sample size of
654 patients (327 in each group) was needed to detect a
3-year event rate reduction in the ngolimod 05 mg
group of 25% (50% event rate on placebo vs 375% on
ngolimod). The sample size calculations incorporated
an assumed dropout rate of 10% per year, based on

www.thelancet.com Published online January 27, 2016 http://dx.doi.org/10.1016/S0140-6736(15)01314-8

Articles

Fingolimod
05 mg
(n=336)

Placebo
(n=487)

Total
(n=823)

Demographics

Fingolimod
05 mg
(n=336)

Placebo
(n=487)

Total
(n=823)

2879
(1645)

2865
(1462)

(Continued from previous column)

Sex

9-HPT score (s)

Men

173
(51%)

252
(52%)

425
(52%)

Mean

2844
(1147)

Women

163
(49%)

235
(48%)

398
(48%)

Median

2526
2533
(22532999) (21983058)

2528
(1392183)

Median

49
(425550)

49
(420540)

49
(2465)

Treatment naive

272
(81%)

372
(76%)

644
(78%)

Mean

485
(86)

485
(83)

485
(84)

Any interferon beta

36
(11%)

66
(14%)

102
(12%)

Natalizumab

3
(1%)

2
(<1%)

5
(1%)

Glatiramer acetate

26
(8%)

33
(7%)

59
(7%)

19
(6%)

36
(7%)

55
(7%)

Age (years)

History of DMT use

Age distribution (years)

1830

6
(2%)

4
(1%)

10
(1%)

3140

60
(18%)

90
(18%)

150
(18%)

4150

127
(38%)

194
(40%)

321
(39%)

>50

143
(43%)

199
(41%)

342
(42%)

Race
White

324
(96%)

467
(96%)

791
(96%)

Black

7
(2%)

13
(2%)

Asian

4
(1%)

4
(<1%)

Other

5
(1%)

10
(2%)

15
(2%)

(1%)

Clinical characteristics
Disease duration since diagnosis (years)
Mean

280
(26)

291
(23)

287
(24)

Median

198
(097414)

235
(109431)

214
(01201)

Disease duration since onset of symptoms (years)

Mean

58
(25)

59
(24)

58
(24)

54
(36756)

57
(3978)

56
(120)

Mean

470
(103)

466
(103)

467
(103)

Median*

450
(4060)

450
(4060)

450
(2065)

Mean

905
(561)

909
(762)

908
(687)

Median

723
(5551018)

690
(550960)

705
(311177)

Median
EDSS score

25TWT score (s)

(Table 1 continues in next column)

results from studies of ngolimod in relapse-onset

multiple sclerosis.1012
The primary ecacy analysis population included all
patients randomly allocated to receive ngolimod
05 mg or placebo (cohort 2) and all patients randomly

Other multiple
sclerosis medicines
MRI characteristics

Number of Gd-enhancing T1 lesions

n

336

484

820

Mean

03
(110)

03
(103)

03
(106)

Median

0
(00)

Free of Gd290
enhancing T1 lesions (86%)

0
(00)

0
(014)

423
(87%)

713
(87%)

Total volume of T2 lesions (mm3)

n

336

485

821

Mean

94427
(10 1797)

10 0382
(13 0309)

97945
(11 9435)

Median

61095
(19410
13 1370)

52710
(20470
12 8170)

57050
(440
91 9640)

Normalised brain volume (cm3)

n

335

483

818

Mean

14909
(865)

14917
(849)

14914
(855)

Median

14910
(14310
15500)

14980
(14330
15530)

14930
(12060
17250)

Data are n (%), median (range), or mean (SD). EDSS=Expanded Disability Status
Scale. 25TWT score=25 Timed-Walk Test. 9-HPT=Nine-Hole Peg Test.
DMT=disease-modifying therapy. Gd=gadolinium. *Inclusion criteria of EDSS 356
and 25TWT <30 s had to be met either at screening or baseline and, therefore, could
also have been outside this range on either visit. Additionally, a small number of
protocol deviations (one in the ngolimod 05 mg group and three in the placebo
group) resulted in inclusion of patients outside these criteria. Dominant hand.

Table 1: Baseline characteristics of patients included in primary ecacy

analysis by study group

allocated to placebo in cohort 1. Patients had to have

taken at least one dose of study drug to be included
(modied intention-to-treat principle). We decided to
use the patients in the placebo group from cohort 1 in
the primary analysis population after a masked review of
baseline data showed no dierences between patients in
cohort 1 and cohort 2. Cohort 1 and cohort 2 were also

www.thelancet.com Published online January 27, 2016 http://dx.doi.org/10.1016/S0140-6736(15)01314-8

Articles

Fingolimod 05 mg (n=336)

Placebo (n=487)

n (%)

n (%)

KM estimate (95% CI)*

KM estimate (95% CI)

Risk
Hazard ratio
reduction (95% CI)*

p value

3-month CDP (composite)

232 (69%)

772% (71878251)

338 (69%)

803% (73318725)

505%

095 (080 112)

3-month CDP (EDSS)

154 (46%)

543% (47166145)

240 (49%)

587% (53306418)

1199%

088 (072108)

0544
0217

3-month CDP (9-HPT)

84 (25%)

336% (26114108)

133 (27%)

413% (32105055)

694%

093 (071122)

0607

3-month CDP (25TWT)

184 (55%)

629% (57106862)

276 (57%)

700% (61787821)

559%

094 (078114)

0546

KM=Kaplan-Meier. CDP=conrmed disease progression. EDSS=Expanded Disability Status Scale. 9-HPT=9-Hole Peg Test. 25TWT=25 Timed-Walk Test. *Time to event using a
Cox regression model adjusted for treatment, region, baseline EDSS, baseline 25TWT, baseline 9-HPT for the composite endpoint (baseline EDSS, baseline 9-HPT, and baseline
25TWT for these respective parameters), and age. Kaplan-Meier estimate of probability of having the specied disability progression.

Table 2: Ecacy analysis for primary endpoint and components

A
Fingolimod 05 mg
Placebo

100
Patients with 3-month CDP
by composite endpoint (%)

90
80
70
60
50
40
30
20

Cox regession HR 095 (95% CI 080112); p=0544

Log-rank test p=0689

10
0

Number at risk
Fingolimod 05 mg 336
Placebo 487

26

52

78

104

130

156

182

208

234

260

286

312

239
359

177
248

134
175

105
139

86
109

57
80

33
48

14
28

0
5

0
2

0
1

0
0

Patients with 3-month CDP by EDSS (%)

B
100
90
80
70

Role of the funding source

60
50
40
30
20

Cox regession HR 088 (95% CI 072108); p=0217

Log-rank test p=0315

10
0

Number at risk
Fingolimod 05 mg 336
Placebo 487

26

52

78

104

130

281
411

236
337

199
266

167
227

146
197

156
182
Study week
106
146

58
93

208

234

260

286

312

21
46

0
15

0
2

0
1

0
0

Figure 2: Time to 3-month conrmed disability progression

(A) Primary composite endpoint. (B) EDSS (key secondary endpoint). HR=hazard ratio. CDP=conrmed disability
progression. EDSS=Expanded Disability Status Scale.

analysed separately (appendix), although the number of

patients in cohort 1 would not have provided enough
statistical power to detect any potential treatment eect.
All ecacy results, unless otherwise specied, are based
on the primary ecacy analysis set. The safety analysis
population included all patients from cohorts 1 and 2.
For the primary analysis, we used a Cox proportional
hazards model to test for dierences between ngolimod
and placebo in the time to 3-month CDP based on the
6

composite endpoint, with region, age, baseline EDSS,

baseline 25TWT, and baseline 9-HPT as covariates.
The key secondary analysis of time to 3-month CDP
based on EDSS was done as for the primary ecacy
endpoint, but we only analysed disability progression
based on the EDSS; baseline 25TWT and baseline
9-HPT were not included as covariates in the Cox
proportional hazards model. We analysed PBVC with a
random coecients model, including treatment and
region as xed eects, and time, number of
Gd-enhancing lesions at baseline, baseline T2 lesion
volume, and normalised brain volume at baseline as
continuous covariates. To control the type I error rate,
we tested the primary and key secondary ecacy
hypotheses in sequential order in a hierarchical
step-down procedure. We used SAS versions 9.3 and 9.4
for all statistical analyses. This study was overseen by an
independent DSMB. This study is registered with
ClinicalTrials.gov, number NCT00731692.

The funder participated in the design and conduct of the

study; data collection, management, analysis, and
interpretation; and preparation, review, and approval of
the paper. All authors had full access to all the data. All
authors, including those employed by the funder, were
involved in manuscript preparation and had control over
the content, for which they take full responsibility and
have given nal approval for submission and publication.

Results
Enrolment was between Sept 3, 2008, and Aug 30, 2011,
and the study was completed on Dec 18, 2014. We
recruited patients at 148 centres in 18 countries
(see appendix for a list of the centres and principal
investigators). We screened 1520 patients, with
970 randomly allocated (280 patients were randomly
allocated [1:1] to ngolimod 125 mg or placebo in
cohort 1 and 690 patients were randomly allocated [1:1]
to ngolimod 05 mg or placebo in cohort 2).
The study prole is summarised in gure 1. Table 1
presents the baseline characteristics of the ecacy
analysis set, which were similar across treatment groups

www.thelancet.com Published online January 27, 2016 http://dx.doi.org/10.1016/S0140-6736(15)01314-8

Articles

Fingolimod 05 mg (n=336)

Placebo (n=487)

Rate
reduction

Adjusted mean
(95% CI) or n (%)

Adjusted mean
(95% CI) or n (%)

Percentage change in brain volume to

month 36

293

149%
(164 to 135)

421

153%
(165 to 141)

Number of new or newly enlarging

T2 lesions to month 36

298

013 per year

(010 to 018)

431

050 per year

(040 to 061)

73%

Number of patients free of new or newly

enlarging T2 lesions to end of study

298

238
(80%)

431

260
(60%)

Number of Gd-enhancing T1 lesions at

month 36

223

005 per scan

(002 to 009)

320

021 per scan

(015 to 030)

Number of patients free of Gd-enhancing

lesions to end of study

299

260
(87%)

432

335
(78%)

Number of new T1 hypointense lesions at

month 36

298

009 per year

(006 to 013)

433

024 per year

(018 to 031)

Number (%) of patients free of new

T1 hypointense lesions to end of study

298

243 (82%)

433

312 (72%)

78%

62%

Point estimate
(95% CI)

p value

AMD 004%
(015 to 023)*

0673

RR 0267
<00001
(0185 to 0386)
OR 279
(195 to 400)

<00001

RR 0217
<00001
(0102 to 0463)
OR 215
(139 to 333)

00006

RR 0375
<00001
(0240 to 0587)
OR 175
(121 to 253)

0003

N=total number of patients included in the analysis. n=number with outcome. AMD=adjusted mean dierence. OR=odds ratio. RR=rate ratio. *Obtained from tting a
random coecients model with treatment and region as xed eects and time, baseline number of gadolinium (Gd)-enhancing T1 lesions, baseline T2 volume, and
baseline normalised brain volume as continuous covariates. Obtained from tting a negative binomial regression model adjusted for treatment, region, baseline number
of Gd-enhanced T1 lesions, and age. Obtained from tting a logistic regression model adjusted for treatment, region, baseline number of Gd-enhanced T1 lesions, and
age. Obtained from tting a logistic regression model adjusted for treatment, region, and age. Obtained from tting a negative binomial regression model adjusted for
treatment, region, and age.

Table 3: Ecacy analysis for secondary endpoints and components

(baseline characteristics for the whole INFORMS patient

population [cohorts 1 and 2] are shown in the appendix).
The composite primary ecacy endpoint was not met
(table 2, gure 2): ngolimod showed no dierence
compared with placebo (hazard ratio 095, 95% CI
080112; p=0544) in the time to 3-month CDP. We
found no dierence in the time to 3-month CDP based
on EDSS alone (table 2, gure 2). Ecacy by cohort is
presented in the appendix; we found no signicant
treatment eect in either cohort.
We did a similar analysis for 6-month CDP and the
conclusions were consistent with that of the primary
endpoint (data not shown). Subgroup analyses by sex,
age (40 years vs 4155 years vs >55 years), EDSS at
baseline (5 vs >5), Gd lesion status at baseline, and age
plus Gd status at baseline (50 years and Gd+ 1 vs
>50 years and Gd+ <1) yielded similar results and did not
alter the conclusion of no ecacy of ngolimod in
patients with primary progressive multiple sclerosis
(data not shown).
PBVC was similar in patients treated with ngolimod
05 mg and placebo (table 3). However, ngolimod
reduced the number of new or newly enlarging T2
lesions by 73% (rate ratio 0267, 95% CI 01850386;
p<00001), of Gd-enhancing T1 lesions by 78% (0217,
01020463; p<00001), and of new T1 hypointense
lesions by 62% (0375, 02400587; p<00001; table 3).
Most patients remained free of new or newly enlarging
T2 lesions to the end of the study in both treatment
groups (mean number of new or newly enlarging T2
lesions was 05 [95% CI 040061] per year in the

placebo group vs 013 [010 018] per year in the

ngolimod group; table 3). Conrmed relapses were
reported for six (2%) of 336 patients in the ngolimod
group and 41 (8%) of 487 patients in the placebo group.
We identied no signicant dierences between
ngolimod 05 mg and placebo in change from baseline
to month 36 for any of the patient-reported outcome
measures (data not shown).
Fingolimod and ngolimod-phosphate blood concentrations were dose-proportional in patients who
received 125 mg at the beginning of the study and
were switched to 05 mg, and were similar in both
cohorts after the protocol amendment. Drug exposure
was similar to that previously documented in patients
with relapse-onset multiple sclerosis receiving 05 mg
ngolimod per day.
The incidence of adverse events in the ecacy
analysis set was generally similar between groups
(table 4; a summary of adverse events for the whole
safety analysis set is shown in the appendix), and the
overall adverse event prole was similar to that of
studies of ngolimod in relapse-onset multiple
sclerosis.10,12 Adverse events were mild or moderate in
severity in 249 (74%) of 336 receiving ngolimod and
366 (75%) of 487 receiving placebo.
Adverse events that led to discontinuation of the study
drug (including abnormal laboratory test results) were
more common with ngolimod than with placebo
(table 4). Serious adverse events were reported for 25%
of patients receiving ngolimod and 24% of those
receiving placebo.

www.thelancet.com Published online January 27, 2016 http://dx.doi.org/10.1016/S0140-6736(15)01314-8

Articles

Fingolimod
05 mg (n=336)

Placebo
(n=487)

324 (96%)

463 (95%)

All events
At least one adverse event
Any adverse event leading to discontinuation of study drug*

52 (15%)

36 (7%)

Any serious adverse event

84 (25%)

117 (24%)

Abnormal laboratory value leading to discontinuation of study drug

27 (8%)

6 (1%)

Death

1 (<1%)

2 (<1%)

Most common adverse events (5% in any group)

135 (28%)

Nasopharyngitis

78 (23%)

Headache

56 (17%)

77 (16%)

Urinary tract infection

50 (15%)

79 (16%)

Fall

47 (14%)

94 (19%)

Discussion

Hypertension

43 (13%)

28 (6%)

Alanine aminotransferase increased

39 (12%)

9 (2%)

In patients with primary progressive multiple sclerosis,

ngolimod treatment did not decrease the risk of
progression of disability versus placebo. We found no
dierence between ngolimod and placebo on BVL. In
both treatment groups, BVL occurred at roughly 05%
per year, and most patients progressed in their disability
status, irrespective of treatment. The eects of ngolimod
on MRI measures of lesion activity and clinical relapses
(although very few in number) were consistent with
results from placebo-controlled studies in relapse-onset
multiple sclerosis.11,12
INFORMS is the rst large, prospective multiple
sclerosis study that used a composite of three major
domains of multiple-sclerosis-associated disability as the
primary endpoint. This composite endpoint yielded more
progression events than its components, of which the
highest event rate was recorded for 25TWT. The outcomes
of all measures included were consistent, supporting the
potential usefulness of the composite endpoint in future
studies of primary progressive multiple sclerosis.
Despite inclusion of patients with an older age (42% of
patients were older than 50 years; table 1) and more
advanced disability than previous trials, safety and
tolerability results for ngolimod 05 mg in INFORMS
were in line with those of studies in relapse-onset
multiple sclerosis, including the numbers of adverse
events of special interest such as cardiac conduction at
rst dose and infections.
Based on the event rates of both the composite primary
disability endpoint and its components, including the
traditional EDSS endpoint, the study was adequately
powered to detect a treatment eect. We recruited a large,
well dened worsening population of patients with
primary progressive multiple sclerosis who progressed
substantially during the course of the study. In the placebo
group, around 80% of patients had CDP according to the
primary composite endpoint, and 59% according to the
EDSS endpoint. Event rates according to the EDSS were
higher than those reported in the placebo groups of two
other large phase 3 trials.8,9 In PROMiSe,9 in which the
active treatment was glatiramer acetate, the placebo group
event rate was 452%, whereas in the OLYMPUS
rituximab trial,8 the placebo rate was 385%. More than

Back pain

37 (11%)

75 (15%)

Upper respiratory tract infection

37 (11%)

58 (12%)

Glutamyl transferase increased

31 (9%)

3 (1%)

Arthralgia

30 (9%)

49 (10%)

Constipation

29 (9%)

36 (7%)

Inuenza

29 (9%)

43 (9%)

Cough

28 (8%)

34 (7%)

Fatigue

25 (7%)

44 (9%)

Nausea

21 (6%)

19 (4%)

Pain in extremity

21 (6%)

36 (7%)

Dizziness

19 (6%)

29 (6%)

Lymphopenia

19 (6%)

Pyrexia

18 (5%)

21 (4%)

Upper abdominal pain

17 (5%)

12 (2%)

Melanocytic naevus

16 (5%)

31 (6%)

Depression

15 (4%)

39 (8%)

Insomnia

12 (4%)

29 (6%)

Adverse events of special interest

Bradycardia

5 (1%)

1 (<1%)

Sinus bradycardia

Atrioventricular block rst degree

3 (1%)

6 (1%)

Atrioventricular block second degree

1 (<1%)

Myocardial infarction

1 (<1%)

Myocardial ischaemia

1 (<1%)

Angina pectoris

1 (<1%)

3 (1%)

Hypertensive crisis

1 (<1%)

Secondary hypertension

1 (<1%)

Hypotension

2 (1%)

5 (1%)

Syncope or presyncope

7 (2%)

9 (2%)

Macular oedema

6 (2%)

6 (1%)

Cystoid macular oedema

1 (<1%)

Bronchitis

1 (<1%)

16 (5%)

21 (4%)

Cystitis or bacterial cystitis

9 (3%)

19 (4%)

Tinea versicolor

6 (2%)

8 (2%)

Pneumonia or bronchopneumonia

6 (2%)

8 (2%)

Meningitis

1 (<1%)

Systemic mycosis

1 (<1%)

Pulmonary sepsis

1 (<1%)

(Table 4 continues on next page)

Five deaths occurred during the study: two (1%) of

147 in the group originally allocated to receive ngolimod
125 mg (single cases of respiratory tract infection and
aspiration pneumonia); one (<1%) of 336 in the
ngolimod 05 mg group in the ecacy analysis set
(metastatic lung cancer); and two (<1%) of 487 in the
combined placebo group (single cases of convulsion and
pulmonary embolism; table 4).
Adverse events of special interest, such as cardiac
conduction abnormalities, macular oedema, infections,
and neoplasms, were in line with previous results from
studies of ngolimod in relapse-onset multiple sclerosis.

www.thelancet.com Published online January 27, 2016 http://dx.doi.org/10.1016/S0140-6736(15)01314-8

Articles

30% of the CDP events by the primary outcome occurred

within the rst 26 weeks after treatment allocation. In line
with most previous cohorts of patients with primary
progressive multiple sclerosis,8,9 the INFORMS population
showed a low rate of active inammatory MRI lesions
(<15% positive for Gd-enhancing lesions at baseline) and
a moderate T2 lesion burden. Together, these data suggest
that the inclusion criteria of INFORMS succeeded in
recruiting a characteristic progressing population.4
The rate of discontinuations in INFORMS was in line
with that of previous studies in patients with primary
progressive multiple sclerosis,8,9 which were of shorter
duration, and is also consistent with ngolimod studies
in patients with relapse-onset multiple sclerosis, in
which an approximate dropout rate of 10% per year was
reported.1012 We factored this rate of attrition into the
sample size calculations to ensure an ability to detect a
treatment dierence at the end of the study.
As with relapse-onset multiple sclerosis, the rate of
BVL in patients with primary progressive multiple
sclerosis substantially exceeded that observed in
individuals without multiple sclerosis.29 In relapse-onset
multiple sclerosis, ngolimod showed a consistent and
robust eect on BVL,12,13 which we did not see in this
study. In relapse-onset multiple sclerosis, the eect of
ngolimod on BVL is associated with a reduction in
inammatory Gd-enhancing and new or enlarging T2
lesions,15,30 but has also been attributed to actions
independent of its anti-inammatory eects. In primary
progressive multiple sclerosis, the eect of ngolimod
on lesion activity (Gd-enhancing and new or enlarging
T2 lesions) was consistent with that seen in relapseonset multiple sclerosis, although fewer patients with
primary progressive multiple sclerosis had lesion
activity (Gd-enhancing or new or newly enlarging T2
lesions) than did patients with relapse-onset multiple
sclerosis. These observations suggest that, although
ngolimod has an eect on inammatory disease
activity, it had little eect on the process that leads to
BVL and disability progression in primary progressive
multiple sclerosis. Similarly, in the PROMiSe and
OLYMPUS trials, treatment had no eect on BVL
despite an eect on Gd-enhancing activity and T2 lesion
burden at some timepoints.8,9
Unlike relapse-onset multiple sclerosis, the
inammatory inltrate in the CNS is not only less
prominent in primary progressive multiple sclerosis, but
also diers in cellular composition.2,5 The pathophysiological mechanisms that drive BVL probably dier,
at least partly, in these two phenotypes, and
neurodegenerative processes might have a more
prominent role in primary progressive multiple sclerosis
than in relapse-onset multiple sclerosis.
The decision to end development of the 125 mg dose
of ngolimod based on the better benetrisk prole of
the 05 mg dose in relapse-onset multiple sclerosis
restricted the ability to fully assess the ecacy of

Fingolimod
(n=336)

Placebo
(n=487)

(Continued from previous page)

Urosepsis

Serratia sepsis

Herpes zoster

10 (3%)

2 (<1%)
1 (<1%)
9 (2%)

Herpes zoster meningomyelitis

1 (<1%)

Herpes zoster neurological

1 (<1%)

Herpes zoster oticus or ophthalmic

1 (<1%)

Hepatocellular injury

2 (1%)

Hepatic function abnormal

1 (<1%)

1 (<1%)

Hyperbilirubinaemia

1 (<1%)

Basal cell carcinoma

14 (4%)

9 (2%)

Squamous cell carcinoma of skin

6 (2%)

1 (<1%)

Malignant melanoma (including in situ)

1 (<1%)

Breast cancer

1 (<1%)

Invasive lobular breast carcinoma

1 (<1%)

Non-Hodgkin lymphoma

1 (<1%)

Lung neoplasm, malignant

1 (<1%)

Ovarian cancer

1 (<1%)

Prostate cancer

1 (<1%)

Dyspnoea

14 (4%)

1 (<1%)
16 (3%)

Exertional dyspnoea

Convulsion

2 (1%)

5 (1%)
2 (<1%)

Epilepsy

1 (<1%)

Status epilepticus

1 (<1%)

Investigations
Blood cholesterol increased

15 (4%)

16 (3%)

Blood triglycerides increased

9 (3%)

9 (2%)

Low density lipoprotein increased

7 (2%)

3 (1%)

Weight increased

5 (1%)

1 (<1%)

Carbon monoxide diusing capacity decreased

7 (2%)

8 (2%)

Data are n (%). *Any adverse event leading to discontinuation of the study drug includes events occurring in patients
whose primary or secondary reason for discontinuing the study drug was an adverse event (including abnormal
laboratory ndings). Two additional deaths occurred in the ngolimod group of cohort 1 (appendix).

Table 4: Adverse events in ecacy population by study group

ngolimod at the higher dose. However, our study was

adequately powered for assessment of the 05 mg dose of
ngolimod versus placebo. Additionally, as presented in
the appendix, we identied no sign of a potential
treatment eect in the cohort originally assigned to
receive ngolimod 125 mg.
The results of INFORMS suggest that anti-inammatory
strategies applied at present in relapse-onset multiple
sclerosis are unlikely to be benecial in primary
progressive multiple sclerosis, and that novel approaches
might be needed to treat patients with primary
progressive multiple sclerosis.
Contributors
All authors contributed to data interpretation, wrote the rst draft, and
reviewed and edited subsequent drafts. FL, DHM, MSF, BACC, JSW,
HW, CL, H-PH, XM, BMJU, MM, BL, NP, DAH, and LK were members
of the Steering Committee and contributed to study design. DAH and
BL were the study statisticians.

www.thelancet.com Published online January 27, 2016 http://dx.doi.org/10.1016/S0140-6736(15)01314-8

Articles

Declaration of interests
FL reports personal fees from Bayer Healthcare, Actelion, Acorda,
Questcor/Malinckrodt, Roche/Genentech, Medimmune,
Osmotica, Xenoport, Receptos, Forward Pharma, BBB Technologies,
Akros, TG Therapeutics, AbbVie, and EMD Serono, grants and
personal fees from Novartis, Biogen Idec, Teva Neuroscience,
Sano/Genzyme, and Celgene, and grants from Transparency Life
Sciences. DHM reports grants and personal fees from Novartis and
Biogen Idec, personal fees from Misubishi Pharma Europe,
BayerSchering, and Chugai, and grants from Apitope. MSF reports
personal fees from Genzyme, Merck Serono, Bayer, Biogen Idec, Teva
Canada Innovation, Opexa, Novartis, and Chugai. BACC reports
personal fees from Biogen Idec, EMD Serono, MedImmune, Novartis,
Teva Neurosciences, Genzyme/Sano-Aventis, and AbbVie, and grants
from Homann-La Roche. JSW reports personal fees from Novartis
Pharmaceuticals, Sano-Aventis, Teva Pharmaceuticals, Genzyme,
Eli Lilly, Actelion, Homann-LaRoche, AbbVie, Athersys Inc, Xenoport,
EMD Serono, Alkermes, Forward Pharma A/S, Consortium MS
Centers, University of Kansas, Wayne State University, and ACTRIMS,
grants from US National Institutes of Health, Sano-Aventis, and US
National Institutes of Health/NINDS, Genzyme, and royalties for our
licensed monoclonal antibodies through University of Texas Health
Science Center at Houston from Millipore (Chemicon Intl) Corp. CL
reports personal fees from Biogen, Roche, Novartis, and Genzyme, and
personal fees for scientic collaboration from Vertex. H-PH reports
personal fees from Biogen Idec, Genzyme, Teva, Roche, Novartis, and
MedImmune. XM reports personal fees from Actelion, Almirall, Bayer,
Biogen, Genzyme, Merck, Novartis, Octapharma, Receptos, Roche,
Sano, Teva, and Trophos. BMJU reports personal fees from Novartis,
Biogen Idec, Genzyme, Merck Serono, Novartis, Roche, and Teva. MM,
BL, NP, FCL, and DAH are employees of Novartis. LKs institution
(University Hospital Basel) received in the past 3 years and used
exclusively for research support: steering committee, advisory board
and consultancy fees (Actelion, Addex, Bayer Health Care, Biogen,
Biotica, Genzyme, Lilly, Merck, Mitsubishi, Novartis, Ono Pharma,
Pzer, Receptos, Sano-Aventis, Santhera, Siemens, Teva, UCB,
Xenoport); speaker fees (Bayer Health Care, Biogen, Merck, Novartis,
Sano-Aventis, Teva); support of educational activities (Bayer Health
Care, Biogen, CSL Behring, Genzyme, Merck, Novartis, Sano, Teva);
royalties (Neurostatus Systems GmbH); and grants (Bayer Health Care,
Biogen, Merck, Novartis, Roche, Swiss MS Society, the Swiss National
Research Foundation, the European Union, Roche Research
Foundations). HW declares no competing interests.
Acknowledgments
Novartis Pharma AG (Basel, Switzerland) supported this
study. Central MRI analysis was done at the Queen Square MS Centre,
UCL Institute of Neurology (London, UK). Members of the central MRI
analysis team were: David G MacManus, Tarek A Yousry,
Claudia A M Wheeler-Kingshott, zgr Yaldizli, Jon Stutters,
Catherine M Dalton, Virginia Santana, Almudena Garcia-Gomez,
Carolina Crespo, David H Miller. The Queen Square MS Centre is
supported by the UK MS Society and the UCL-UCLH joint Biomedical
Research Centre. We thank the patients who participated in the study;
the study-site personnel; Ana de Vera, Goeril Karlsson and the Novartis
clinical team; and Katy Demery (Novartis) and Paul Coyle (Western Edge
Medical Communications Ltd) for editorial assistance.
References
1
Noseworthy JH, Lucchinetti C, Rodriguez M, Weinshenker BG.
Multiple sclerosis. N Engl J Med 2000; 343: 93852.
2
Antel J, Antel S, Caramanos Z, Arnold DL, Kuhlmann T. Primary
progressive multiple sclerosis: part of the MS disease spectrum or
separate disease entity? Acta Neuropathol 2012; 123: 62738.
3
Koch M, Kingwell E, Rieckmann P, Tremlett H. The natural history
of primary progressive multiple sclerosis. Neurology 2009;
73: 19962002.
4
Lublin FD, Reingold SC, Cohen JA, et al. Dening the clinical
course of multiple sclerosis: the 2013 revisions. Neurology 2014;
83: 27886.
5
Mahad DH, Trapp BD, Lassmann H. Pathological mechanisms in
progressive multiple sclerosis. Lancet Neurol 2015; 14: 18393.

10

7
8

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

Tremlett H, Paty D, Devonshire V. The natural history of primary

progressive MS in British Columbia, Canada. Neurology 2005;
65: 191923.
Confavreux C, Vukusic S. Age at disability milestones in multiple
sclerosis. Brain 2006; 129: 595605.
Hawker K, OConnor P, Freedman MS, et al. Rituximab in patients
with primary progressive multiple sclerosis: results of a randomized
double-blind placebo-controlled multicenter trial. Ann Neurol 2009;
66: 46071.
Wolinsky JS, Narayana PA, OConnor P, et al. Glatiramer acetate in
primary progressive multiple sclerosis: results of a multinational,
multicenter, double-blind, placebo-controlled trial. Ann Neurol 2007;
61: 1424.
Cohen JA, Barkhof F, Comi G, et al. Oral ngolimod or
intramuscular interferon for relapsing multiple sclerosis.
N Engl J Med 2010; 362: 40215.
Calabresi PA, Radue EW, Goodin D, et al. Safety and ecacy of
ngolimod in patients with relapsing-remitting multiple sclerosis
(FREEDOMS II): a double-blind, randomised, placebo-controlled,
phase 3 trial. Lancet Neurol 2014; 13: 54556.
Kappos L, Radue EW, OConnor P, et al. A placebo-controlled trial
of oral ngolimod in relapsing multiple sclerosis. N Engl J Med
2010; 362: 387401.
Barkhof F, de Jong R, Skas N, et al. The inuence of patient
demographics, disease characteristics and treatment on brain
volume loss in Trial Assessing Injectable Interferon vs FTY720 Oral
in Relapsing-Remitting Multiple Sclerosis
(TRANSFORMS), a phase 3 study of ngolimod in multiple
sclerosis. Mult Scler 2014; 20: 170413.
Jacobsen C, Hagemeier J, Myhr KM, et al. Brain atrophy and
disability progression in multiple sclerosis patients: a 10-year
follow-up study. J Neurol Neurosurg Psychiatry 2014; 85: 110915.
Radue EW, Barkhof F, Kappos L, et al. Correlation between brain
volume loss and clinical and MRI outcomes in multiple sclerosis.
Neurology 2015; 84: 78493.
Miron VE, Ludwin SK, Darlington PJ, et al. Fingolimod (FTY720)
enhances remyelination following demyelination of organotypic
cerebellar slices. Am J Pathol 2010; 176: 268294.
Meno-Tetang GM, Li H, Mis S, et al. Physiologically based
pharmacokinetic modeling of FTY720 (2-amino-2[2-(-4-octylphenyl)
ethyl]propane-1,3-diol hydrochloride) in rats after oral and
intravenous doses. Drug Metab Dispos 2006; 34: 148087.
Colombo E, Di Dario M, Capitolo E, et al. Fingolimod may support
neuroprotection via blockade of astrocyte nitric oxide. Ann Neurol
2014; 76: 32537.
Choi JW, Gardell SE, Herr DR, et al. FTY720 (ngolimod) ecacy in
an animal model of multiple sclerosis requires astrocyte
sphingosine 1-phosphate receptor 1 (S1P1) modulation.
Proc Natl Acad Sci USA 2011; 108: 75156.
Foster CA, Mechtcheriakova D, Storch MK, et al. FTY720 rescue
therapy in the dark agouti rat model of experimental autoimmune
encephalomyelitis: expression of central nervous system genes and
reversal of blood-brain-barrier damage. Brain Pathol 2009; 19: 25466.
Kurtzke JF. Rating neurologic impairment in multiple sclerosis:
an expanded disability status scale (EDSS). Neurology 1983;
33: 144452.
Hartung HP, Aktas O. Bleak prospects for primary progressive
multiple sclerosis therapy: downs and downs, but a glimmer of
hope. Ann Neurol 2009; 66: 42932.
Schwid SR, Goodman AD, McDermott MP, Bever CF, Cook SD.
Quantitative functional measures in MS: what is a reliable change?
Neurology 2002; 58: 129496.
Mathiowetz V, Volland G, Kashman N, Weber K. Adult norms for
the Box and Block Test of manual dexterity. Am J Occup Ther 1985;
39: 38691.
International Conference on Harmonisation of Technical
Requirements for Registration of Pharmaceuticals for Human Use.
ICH harmonised tripartite guideline. Guideline for good clinical
practice: E6(R1). Geneva: International Conference on Harmonisation
of Technical Requirements for Registration of Pharmaceuticals for
Human Use, June 10, 1996. http://www.ich.org/products/guidelines/
ecacy/ecacy-single/article/good-clinical-practice.html (accessed
Oct 5, 2014).

www.thelancet.com Published online January 27, 2016 http://dx.doi.org/10.1016/S0140-6736(15)01314-8

Articles

26

27

28

World Medical Association. Declaration of Helsinki: ethical

principles for medical research involving human subjects.
http://www.wma.net/en/30publications/10policies/b3/index.html
(accessed Oct 5, 2014).
Polman CH, Reingold SC, Edan G, et al. Diagnostic criteria for
multiple sclerosis: 2005 revisions to the McDonald Criteria.
Ann Neurol 2005; 58: 84046.
Smith SM, Zhang Y, Jenkinson M, et al. Accurate, robust, and
automated longitudinal and cross-sectional brain change analysis.
NeuroImage 2002; 17: 47989.

29

30

Hedman AM, van Haren NE, Schnack HG, Kahn RS,

Hulsho Pol HE. Human brain changes across the life span:
a review of 56 longitudinal magnetic resonance imaging studies.
Hum Brain Mapp 2012; 33: 19872002.
Sormani M, De Stefano N, Francis G, et al. Fingolimod eect on
brain volume loss independently contributes to its eect on
disability. Mult Scler 2015; 21: 91624.

www.thelancet.com Published online January 27, 2016 http://dx.doi.org/10.1016/S0140-6736(15)01314-8

11