MRI Outcomes with Cladribine Tablets for MS (Comi Et Al)

MRI outcomes with cladribine tablets for MS (Comi et al)

Online Resource 1 – Text and Tables

MRI outcomes with cladribine tablets for multiple sclerosis in the CLARITY study

Journal of Neurology

Giancarlo Comi, MD, Stuart D. Cook, MD, Gavin Giovannoni MBBCh, PhD, Kottil Rammohan, MD, Peter Rieckmann, MD, Per Soelberg Sørensen, MD, DMSc, Patrick Vermersch, MD, PhD, Anthony C. Hamlett, PhD, Vissia Viglietta, MD, PhD, and Steven J. Greenberg, MD, on behalf of the CLARITY Study Group

Address correspondence to:

Professor Giancarlo Comi, Department of Neurology and Institute of Experimental Neurology, Università Vita-Salute San Raffaele, via Olgettina 48, 20132, Milan, Italy

Telephone: +39 02 2643 2990 or +39 02 2643 2881

Fax: 39 02 2643 3746 or +39 02 2643 2277 or +39 02 2643 3085

Email:

Methodology: MRI procedures

Prior to screening their first patient, each trial site sent a test scan (without gadolinium) to assess image quality and shipment procedures, to evaluate the accessibility of the electronic data carrier, and to assess the ability to correctly reposition patients for comparable brain images. Scans were assessed by a centralized imaging facility (WorldCare Clinical Inc., Cambridge, MA, USA). Upon final approval of this test scan, sites were allowed to begin screening patients. Throughout the trial, each patient was scanned using the same machine, procedure, patient set-up and selection of scan parameters; a record of the slice positions used at baseline scanning was kept to enable accurate slice position reproduction at sequential visits. Patients underwent T1 axial, axial FLAIR and fast spin echo T2 axial scans, then T1 axial post-Gd scans (conducted 7 minutes after Gd injection), each for 25–27 contiguous slices of 5 mm thickness, with repetition times (TR) of 450, 10,000, 6,025 and 45 msec, and echo times (TE) of 60, 120, 102 and 60 sec, respectively. Parallel slices were performed with the slice angle determined by a line joining the most inferioranterior and inferiorposterior border of the corpus callosum (the genu and the splenium).

New and prior scans were reviewed synchronously to identify new T1 Gd+ lesions. The presence of true Gd enhancement was confirmed by identifying the lack of intrinsic T1 hyperintensity on the T1 pre-Gd sequence. New and enlarging T2-bright lesions were also identified by synchronous review of scans, and were recorded regardless of associated enhancement to provide the ‘active T2 lesion’ count. Those that were also newly T1 Gd+ (‘overlapping’ lesions) were not included in the count of CU lesions. Lesion volumes were determined from marked lesion borders, avoiding normal structures and excluding ‘dirty white matter’ that was present within deep and periventricular white matter. Lesions of diameter >5 mm were defined as enlarging lesions if they had either an increase in diameter of at least 100% or an increase in size on at least two consecutive slices. Lesions of diameter <5 mm were defined as enlarging lesions if they had both an increase in diameter of at least 100% and an increase in size on at least two consecutive slices. A conservative approach was used to avoid false-positive recording of new lesions, particularly in the posterior cranial fossa, in view of the potential for a high degree of artifact in this region. On T2 scans, a lesion was considered new rather than enlarging if it was contiguous with a pre-existing lesion, but connected to it by an area of relatively lower signal. Tapering between regions, but with homogeneous signal intensity, was interpreted as enlargement.

All assessments were evaluated for accuracy and consistency by neuroradiologist reviewers who were trained in the use of CheshireTM (a validated, commercially-available image-processing software suite; Hayden Image. Processing Group, MA, USA), and blinded to the study treatment, assignment, patient, site, and site/investigator assessment. The same neuroradiologist reviewed all time points for a given patient. All neuroradiologists had independently reviewed a single common set of eight representative MRI cases to ensure consistency in the assessment of lesion volume. Retraining was undertaken if results fell below an acceptable threshold of inter- or intra-reviewer consistency (90%). All neuroradiologists underwent quarterly testing throughout the trial, with a total of 60 variability-testing training cases over the course of the study.

Methodology: statistical procedures

As specified in the statistical analysis plan, because both of the cladribine tablets cumulative dose groups were significant for the primary outcome (relapse rate over the 96-week study), MRI parameters were tested at the 0.05 level in the order of T1, T2, and CU for cladribine tablets 5.25 mg/kg versus placebo, followed by T1, T2, and CU for cladribine tablets 3.5 mg/kg versus placebo.

Missing baseline data for the efficacy MRI parameters (lesion number and volume) were imputed according to the pre-specified statistical analysis plan, based on the median baseline measurement for all patients across all treatment groups. For patients with no post-baseline MRI scans, data were imputed as the median measurement from all patients with post-baseline MRI scans at that time point. Missing data on lesion-free status were imputed using the proportion of patients with known status at the end of 96 weeks across all three treatment groups, multiplied by the number of patients with missing lesion status in each treatment group; this value for each treatment group was rounded to the nearest integer and patients with missing lesion status were randomly assigned to lesion-free status or to not lesion-free status.

Results: MRI outcomes across serial time points

Data from serial time points through the study (Weeks 24, 48 and 96) allowed examination of the course of MRI events over time. The number of T1 Gd+ lesions in the cladribine tablets 3.5 and 5.25 mg/kg groups were similar to that of the placebo group at baseline (Table 1), but were reduced at the first assessment point (Week 24) and at each subsequent time point, with mean numbers ranging from 0.06–0.11 and 0.04–0.07 versus 0.38–0.97, respectively (see Online Resource 2; Figure 1). Similar improvements were seen in active T2 and CU lesions, emerging at the first assessment at 24 weeks, and sustained for the rest of the study. Mean lesion numbers in the cladribine tablets 3.5 and 5.25 mg/kg versus placebo groups ranged from 0.22–0.45 and 0.15–0.33 versus 0.87–1.59 for active T2 lesions; and from 0.29–0.38 and 0.16–0.49 versus 0.97–1.91 for CU lesions, respectively.

Results: MRI outcomes across patient subgroups

There was a tendency for younger patients with lower disease duration at baseline to have higher MRI activity at Week 96 and for a lower proportion to be free from MRI activity, in all three treatment groups. However no trends emerged, with cladribine tablets treatment demonstrating comparable efficacy, in patient subgroups stratified by baseline EDSS or prior DMD treatment status (Tables 1–3).

Table 1. MRI disposition: patients contributing scans at each assessment during the 96-week study

Placebo / Cladribine tablets
3.5 mg/kg / Cladribine tablets
5.25 mg/kg
Patients contributing scans for each assessment, n (%) / (n = 437) / (n = 433) / (n = 456)
Week 24
T1 Gd+ lesions / 408 (93.4) / 410 (94.7) / 428 (93.9)
T2 lesions / 408 (93.4) / 410 (94.7) / 428 (93.9)
CU lesions / 408 (93.4) / 411 (94.7) / 428 (93.9)
Week 48
T1 Gd+ lesions / 384 (87.9) / 399 (92.1) / 417 (91.4)
T2 lesions / 384 (87.9) / 398 (91.9) / 417 (91.4)
CU lesions / 384 (87.9) / 399 (91.9) / 417 (91.4)
Week 96
T1 Gd+ lesions / 365 (83.5) / 379 (87.5) / 390 (85.5)
T2 lesions / 366 (83.7) / 379 (87.5) / 392 (86.0)
CU lesions / 366 (83.7) / 379 (87.5) / 392 (86.0)

Table2. MRI outcomes at 96 weeks stratified by baseline relapse category in the 12 months prior to study

MRI outcome
Baseline category / Placebo
(n = 437) / Cladribine tablets 3.5 mg/kg
(n = 433) / Cladribine tablets 5.25 mg/kg
(n = 456)
LSM ± SE / LSM ± SE / Treatment difference± SEa
Relative
reductionb % / LSM ± SE / Treatment difference ± SEa
Relative
reductionb %
T1 Gd+ lesions/patient/scan
≤1 relapses / 0.73 ± 0.06 / 0.09 ± 0.06 / –0.65 ± 0.08
89.0* / 0.10 ± 0.06 / –0.64 ± 0.08
87.7*
2 relapses / 1.09 ± 0.10 / 0.06 ± 0.10 / –1.03 ± 0.14
94.5* / 0.00 ± 0.10 / –1.09 ± 0.14
100.0*
≥3 relapses / 1.70 ± 0.23 / 0.21 ± 0.21 / –1.49 ± 0.31
87.6* / 0.07 ± 0.24 / –1.63 ± 0.33
95.9*
Active T2 lesions/patient/scan
≤1 relapses / 1.24 ± 0.07 / 0.33 ± 0.07 / –0.91 ± 0.10
73.4* / 0.30 ± 0.07 / –0.94 ± 0.10
75.8*
2 relapses / 1.55 ± 0.12 / 0.30 ± 0.13 / –1.25 ± 0.17
80.6* / 0.25 ± 0.12 / –1.29 ± 0.17
83.2*
≥3 relapses / 2.71 ± 0.28 / 0.65 ± 0.26 / –2.07 ± 0.38
76.4* / 0.25 ± 0.30 / –2.47 ± 0.41
91.1*
CU lesions/patient/scan
≤1 relapses / 1.47 ± 0.09 / 0.37 ± 0.09 / –1.10 ± 0.12
74.8* / 0.35 ± 0.08 / –1.11 ± 0.12
75.5*
2 relapses / 1.95 ± 0.15 / 0.33 ± 0.15 / –1.62 ± 0.21
83.1* / 0.26 ± 0.14 / –1.69 ± 0.20
86.7*
≥3 relapses / 3.00 ± 0.33 / 0.70 ± 0.30 / –2.30 ± 0.45
76.7* / 0.30 ± 0.36 / –2.70 ± 0.49
90.0*

Data shown for number of new lesions (not observed on baseline scans).

*All p < 0.001, estimated using a parametric ANCOVA model on ranked data with fixed effects for treatment group, baseline category, treatment group by baseline category interaction and baseline number of T1 Gd+ lesions as a covariate; CU = combined unique; Gd+ = gadolinium-enhancing; LSM = Least Squares Mean; SE = standard error.

aTreatment difference shown as point estimate for cladribine–placebo estimated using a parametric ANCOVA model on raw data with fixed effects for treatment group, baseline category, treatment group by baseline category interaction and baseline number of T1 Gd+ lesions as a covariate.

bCalculated as the ratio of the difference of LSM (placebo – cladribine) relative to the placebo LSM.

Table3. MRI outcomes at 96 weeks stratified by baseline T1 Gd+ lesion category

MRI outcome
Baseline category / Placebo
(n = 437) / Cladribine tablets 3.5 mg/kg
(n = 433) / Cladribine tablets 5.25 mg/kg
(n = 456)
LSM ± SE / LSM ± SE / Treatment difference± SEa
Relative
reductionb % / LSM ± SE / Treatment difference ± SEa
Relative
reductionb %
T1 Gd+ lesions/patient/scan
0 lesions / 0.48 ± 0.08 / 0.07 ± 0.06 / –0.41 ± 0.08
85.4* / 0.04 ± 0.06 / –0.44 ± 0.06
91.7*
≥1 lesion(s) / 1.79 ± 0.09 / 0.12 ± 0.09 / –1.67 ± 0.12
93.3* / 0.13 ± 0.09 / –1.65 ± 0.012
92.2*
Active T2 lesions/patient/scan
0 lesions / 0.97 ± 0.07 / 0.26 ± 0.07 / –0.70 ± 0.10
72.2* / 0.25 ± 0.07 / –0.71 ± 0.10
73.2*
≥1 lesion(s) / 2.38 ± 0.11 / 0.52 ± 0.11 / –1.86 ± 0.15
78.2* / 0.36 ± 0.11 / –2.02 ± 0.15
84.9*
CU lesions/patient/scan
0 lesions / 1.13 ± 0.09 / 0.30 ± 0.09 / –0.83 ± 0.12
73.5* / 0.28 ± 0.09 / –0.85 ± 0.12
75.2*
≥1 lesion(s) / 2.91 ± 0.14 / 0.56 ± 0.13 / –2.35 ± 0.18
80.8* / 0.44 ± 0.13 / –2.47 ± 0.18
84.9*

Data shown for number of new lesions (not observed on baseline scans).

CU = combined unique; Gd+ = gadolinium-enhancing; LSM = Least Squares Mean; SE = standard error.

bCalculated as the ratio of the difference of LSM (placebo – cladribine) relative to the placebo LSM.

Table4. MRI outcomes at 96 weeks stratified by baseline T2 lesion category

Data shown for number of new lesions (not observed on baseline scans).

CU = combined unique; Gd+ = gadolinium-enhancing; LSM = Least Squares Mean; SE = standard error.

bCalculated as the ratio of the difference of LSM (placebo – cladribine) relative to the placebo LSM.

cp value estimated using a parametric ANCOVA model on ranked data with fixed effects for treatment group, baseline category, treatment group by baseline category interaction and baseline number of T1 Gd+ lesions as a covariate.

Table5. MRI outcomes at 96 weeks stratified by baseline T2 lesion volume category

MRI outcome
Baseline category
(relapse or lesion no.) / Placebo
(n = 437) / Cladribine tablets 3.5 mg/kg
(n = 433) / Cladribine tablets 5.25 mg/kg
(n = 456)
LSM ± SE / LSM ± SE / Treatment difference± SEa
Relative
reductionb % / LSM ± SE / Treatment difference ± SEa
Relative
reductionb %
T1 Gd+ lesions/patient/scan
≤5 mL / 0.36 ± 0.10 / 0.13 ± 0.09 / –0.23 ± 0.13
63.9* / 0.09 ± 0.09 / –0.27 ± 0.13
75.0*
>5 mL / 1.05 ± 0.06 / 0.07 ± 0.06 / –0.99 ± 0.08
94.3* / 0.06 ± 0.06 / –0.99 ± 0.08
94.3*
Active T2 lesions/patient/scan
≤5 mL / 1.08 ± 0.12 / 0.42 ± 0.11 / –0.65 ± 0.16
60.2* / 0.28 ± 0.11 / –0.79 ± 0.16
73.1*
>5 mL / 1.50 ± 0.07 / 0.30 ± 0.08 / –1.19 ± 0.10
79.3* / 0.28 ± 0.07 / –1.21 ± 0.10
80.7*
CU lesions/patient/scan
≤5 mL / 1.14 ± 0.14 / 0.46 ± 0.13 / –0.68 ± 0.19
59.6* / 0.33 ± 0.13 / –0.81 ± 0.19
71.1*
>5 mL / 1.85 ± 0.09 / 0.34 ± 0.09 / –1.51 ± 0.12
81.6* / 0.33 ± 0.09 / –1.52 ± 0.12
82.2*

Data shown for number of new lesions (not observed on baseline scans).

CU = combined unique; Gd+ = gadolinium-enhancing; LSM = Least Squares Mean; SE = standard error.

bCalculated as the ratio of the difference of LSM (placebo – cladribine) relative to the placebo LSM.

Table6. MRI outcomes at 96 weeks stratified by additional baseline characteristics

MRI outcome
Baseline category / Placebo
(n = 437) / Cladribine tablets 3.5 mg/kg
(n = 433) / Cladribine tablets 5.25 mg/kg
(n = 456)
LSM ± SE / LSM ± SE / Treatment difference± SEa
Relative
reductionb% / LSM ± SE / Treatment difference ±SEa
Relative
reductionb%
Analyzed by age category at study entry
T1 Gd+ lesions/patient/scan
≤40 years / 1.08 ± 0.07 / 0.10 ± 0.06 / –0.98 ± 0.09
90.7 / 0.07 ± 0.07 / –1.01 ± 0.09
93.5
>40 years / 0.59 ± 0.08 / 0.07 ± 0.08 / –0.52 ± 0.11
88.1 / 0.07 ± 0.07 / –0.52 ± 0.10
88.1
Active T2 lesions/patient/scan
≤40 years / 1.85 ± 0.08 / 0.40 ± 0.11 / –1.40 ± 0.11
75.7 / 0.33 ± 0.08 / –1.51 ± 0.11
81.6
>40 years / 0.78 ± 0.09 / 0.19 ± 0.10 / –0.58 ± 0.13
74.4 / 0.23 ± 0.09 / –0.55 ± 0.13
70.5
CU lesions/patient/scan
≤40 years / 2.17 ± 0.10 / 0.48 ± 0.09 / –1.69 ± 0.13
77.9 / 0.36 ± 0.10 / –1.81 ± 0.13
83.4
>40 years / 0.99 ± 0.11 / 0.23 ± 0.11 / –0.76 ± 0.15
76.8 / 0.28 ± 0.10 / –0.70 ± 0.15
70.7
Analyzed by prior DMD therapy in the 12 months prior to study
T1 Gd+ lesions/patient/scan
DMD-naïve / 0.78 ± 0.06 / 0.08 ± 0.06 / –0.70 ± 0.08
89.7 / 0.07 ± 0.06 / –0.72 ± 0.08
92.3
DMD-treated / 1.05 ± 0.09 / 0.11 ± 0.10 / –0.94 ± 0.13
89.5 / 0.08 ± 0.09 / –0.97 ± 0.12
92.4
Active T2 lesions/patient/scan
DMD-naïve / 1.30 ± 0.08 / 0.31 ± 0.07 / –0.99 ± 0.10
76.2 / 0.28 ± 0.07 / –1.01 ± 0.11
77.7
DMD-treated / 1.57 ± 0.11 / 0.42 ± 0.12 / –1.14 ± 0.16
72.6 / 0.28 ± 0.11 / –1.29 ± 0.15
82.2
CU lesions/patient/scan
DMD-naïve / 1.54 ± 0.09 / 0.35 ± 0.09 / –1.19 ± 0.12
77.3 / 0.33 ± 0.09 / –1.21 ± 0.12
78.6
DMD-treated / 1.92 ± 0.13 / 0.47 ± 0.14 / –1.45 ± 0.19
75.5 / 0.33 ± 0.13 / –1.59 ± 0.18
82.8
Analyzed by baseline EDSS score
T1 Gd+ lesions/patient/scan
EDSS ≤3 / 0.79 ± 0.06 / 0.10 ± 0.06 / –0.69 ± 0.09
87.3 / 0.08 ± 0.07 / –0.72 ± 0.09
91.1
EDSS ≥3.5 / 0.98 ± 0.08 / 0.06 ± 0.08 / –0.92 ± 0.11
93.9 / 0.08 ± 0.07 / –0.92 ± 0.11
93.9
Active T2 lesions/patient/scan
EDSS ≤3 / 1.38 ± 0.08 / 0.38 ± 0.08 / –1.00 ± 0.11
72.5 / 0.30 ± 0.08 / –1.08 ± 0.11
78.3
EDSS ≥3.5 / 1.39 ± 0.10 / 0.27 ± 0.10 / –1.12 ± 0.14
80.6 / 0.26 ± 0.09 / –1.13 ± 0.13
81.3
CU lesions/patient/scan
EDSS ≤3 / 1.62 ± 0.09 / 0.42 ± 0.09 / –1.20 ± 0.13
74.1 / 0.34 ± 0.10 / –1.28 ± 0.14
79.0
EDSS ≥3.5 / 1.73 ± 0.12 / 0.31 ± 0.12 / –1.41 ± 0.17
81.5 / 0.32 ± 0.11 / –1.41 ± 0.16
81.5
Analyzed by disease duration at baseline
T1 Gd+ lesions/patient/scan
<3 years / 0.93 ± 0.11 / 0.12 ± 0.09 / –0.81 ± 0.14
87.1 / 0.05 ± 0.10 / –0.87 ± 0.15
93.5
3–10 years / 0.98 ± 0.07 / 0.09 ± 0.08 / –0.89 ± 0.11
90.8 / 0.06 ± 0.07 / –0.92 ± 0.11
93.9
>10 years / 0.68 ± 0.09 / 0.04 ± 0.08 / –0.64 ± 0.13
94.1 / 0.09 ± 0.09 / –0.59 ± 0.12
86.8
Active T2 lesions/patient/scan
<3 years / 1.66 ± 0.13 / 0.39 ± 0.11 / –1.26 ± 0.17
75.9 / 0.35 ± 0.13 / –1.31 ± 0.18
78.9
3–10 years / 1.58 ± 0.09 / 0.37 ± 0.09 / –1.21 ± 0.13
76.6 / 0.28 ± 0.09 / –1.31 ± 0.13
82.9
>10 years / 0.94 ± 0.11 / 0.24 ± 0.12 / –0.70 ± 0.16
74.5 / 0.25 ± 0.10 / –0.68 ± 0.15
72.3
CU lesions/patient/scan
<3 years / 1.90 ± 0.15 / 0.44 ± 0.14 / –1.46 ± 0.21
76.8 / 0.39± 0.15 / –1.51 ± 0.22
79.5
3–10 years / 1.90 ± 0.11 / 0.40 ± 0.11 / –1.50 ± 0.16
78.9 / 0.31 ± 0.11 / –1.59 ± 0.15
83.7
>10 years / 1.17 ± 0.13 / 0.28 ± 0.14 / –0.90 ± 0.19
76.9 / 0.31 ± 0.12 / –0.87 ± 0.18
74.4

Data shown for number of new lesions (not observed on baseline scans).

For all treatment differences, p values were <0.001, estimated using a parametric ANCOVA model on ranked data with fixed effects for treatment group, baseline category, treatment group by baseline category interaction and baseline number of T1 Gd+ lesions as a covariate.

CU = combined unique; Gd+ = gadolinium-enhancing; LSM = Least Squares Mean; SE = standard error.

aPoint estimate of the treatment difference (cladribine–placebo) estimated using a parametric ANCOVA model on raw data with fixed effects for treatment group, baseline category, treatment group by baseline category interaction and baseline number of T1 Gd+ lesions as a covariate.

bCalculated as the ratio of the difference of LSM (placebo – cladribine) relative to the placebo LSM.

Table 7. MRI-free status at 96 weeks stratified by additional baseline disease criteria

MRI outcome
Baseline category / Placebo
(n = 437) / Cladribine tablets 3.5 mg/kg
(n = 433) / Cladribine tablets 5.25 mg/kg
(n = 456)
Lesion-free,
n (%) / Lesion-free,
n (%) / OR (95% CI)a / Lesion-free,
n (%) / OR (95% CI)a
Analyzed by age category at study entry
T1 Gd+ lesion-free status
≤40 years / 93 (37.5) / 215 (82.7) / 7.96
(5.28, 12.02) / 216 (87.8) / 12.00
(7.57, 19.02)
>40 years / 118 (62.4) / 161 (93.1) / 8.07
(4.19, 15.56) / 199 (94.8) / 10.89
(5.54, 21.37)
Active T2 lesion-free status
≤40 years / 48 (19.4) / 140 (53.8) / 4.86
(3.26, 7.24) / 143 (58.1) / 5.78
(3.86, 8.67)
>40 years / 76 (40.2) / 127 (73.4) / 4.10
(2.63, 6.41) / 142 (67.6) / 3.10
(2.06, 4.68)
CU lesion-free status
≤40 years / 39 (15.7) / 135 (51.9) / 5.79
(3.80, 8.81) / 138 (56.1) / 6.85
(4.48, 10.47)
>40 years / 75 (39.7) / 123 (71.1) / 3.74
(2.41, 5.80) / 139 (66.2) / 2.98
(1.98, 4.48)
Analyzed by prior DMD prior to study
T1 Gd+ lesion-free status
DMD-naïve / 150 (50.8) / 280 (87.5) / 6.77
(4.53, 10.12) / 282 (91.3) / 10.10
(6.40, 15.93)
DMD-treated / 61 (43.0) / 96 (85.0) / 7.50
(4.06, 13.85) / 133 (90.5) / 12.61
(6.63, 24.00)
Active T2 lesion-free status
DMD-naïve / 87 (29.5) / 206 (64.4) / 4.32
(3.08, 6.06) / 192 (62.1) / 3.92
(2.79, 5.51)
DMD-treated / 37 (26.1) / 61 (54.0) / 3.33
(1.97, 5.64) / 93 (63.3) / 4.89
(2.96, 8.08)
CU lesion-free status
DMD-naïve / 79 (26.8) / 198 (61.9) / 4.44
(3.15, 6.25) / 186 (60.2) / 4.13
(2.93, 5.83)
DMD-treated / 35 (24.6) / 60 (53.1) / 3.46
(2.03, 5.89) / 91 (61.9) / 4.97
(2.99, 8.24)
Analyzed by baseline EDSS score
T1 Gd+ lesion-free status
EDSS ≤3 / 116 (44.1) / 234 (86.0) / 7.80
(5.13, 11.88) / 229 (91.2) / 13.19
(8.00, 21.76)
EDSS ≥3.5 / 95 (54.6) / 142 (88.2) / 6.21
(3.54, 10.93) / 186 (90.7) / 8.14
(4.66, 14.23)
Active T2 lesion-free status
EDSS ≤3 / 67 (25.5) / 160 (58.8) / 4.18
(2.89, 6.04) / 155 (61.8) / 4.72
(3.24, 6.88)
EDSS ≥3.5 / 57 (32.8) / 107 (66.5) / 4.07
(2.58, 6.41) / 130 (63.4) / 3.56
(2.33, 5.44)
CU lesion-free status
EDSS ≤3 / 58 (22.1) / 156 (57.4) / 4.75
(3.26, 6.94) / 150 (59.8) / 5.25
(3.57, 7.72)
EDSS ≥3.5 / 56 (32.2) / 102 (63.4) / 3.64
(2.32, 5.72) / 127 (62.0) / 3.43
(2.24, 5.25)
Analyzed by disease duration at baseline
T1 Gd+ lesion-free status
<3 years / 40 (40.8) / 105 (83.3) / 7.25
(3.91, 13.45) / 93 (92.1) / 16.86
(7.37, 38.54)
3–10 years / 86 (44.3) / 165 (89.2) / 10.36
(6.02, 17.84) / 173 (83.3) / 9.45
(5.62, 15.87)
>10 years / 85 (58.6) / 106 (86.9) / 4.68
(2.51, 8.70) / 149 (93.7) / 10.52
(5.12, 21.62)
Active T2 lesion-free status
<3 years / 16 (16.3) / 74 (58.7) / 7.29
(3.84, 13.87) / 56 (55.4) / 6.38
(3.28, 12.39)
3–10 years / 50 (25.8) / 109 (58.9) / 4.13
(2.67, 6.38) / 119 (60.7) / 4.45
(2.89, 6.85)
>10 years / 58 (40.0) / 84 (68.9) / 3.32
(2.00, 5.51) / 110 (69.2) / 3.37
(2.10, 5.40)
CU lesion-free status
<3 years / 15 (15.3) / 72 (57.1) / 7.38
(3.84, 14.18) / 54 (53.5) / 6.36
(3.24, 12.48)
3–10 years / 44 (22.7) / 109 (58.9) / 4.89
(3.13, 7.64) / 116 (59.2) / 4.94
(3.18, 7.68)
>10 years / 55 (37.9) / 77 (63.1) / 2.80
(1.70, 4.61) / 107 (67.3) / 3.37
(2.10, 5.40)

Data shown for number/proportion of patients free from T1 Gd+, active T2 or CU lesions throughout the 96-week study.

CI = confidence interval; CU = combined unique; DMD = disease-modifying drug; EDSS = Expanded Disability Status Scale; Gd+ = gadolinium-enhancing; MRI = magnetic resonance imaging; OR = odds ratio.

aOR and associated 95% CI values were estimated using a logistic regression model with fixed effects for treatment group, baseline category, and their interaction; all OR and 95% CI values were associated with a p value of < 0.001, calculated based on Wald Chi-square test from analysis of endpoint using a logistic regression model with fixed effects for treatment group, baseline category, and their interaction.

Figure caption

Fig. 1Key MRI outcomes over serial time points during the 96-week study, shown as mean (± SD) number of lesions by treatment group (intent-to-treat population), for (A) T1 gadolinium-enhancing lesions, (B) active T2 lesions, (C) combined unique lesions and (D) T1 hypointense lesions

aMean (± SD) number of lesions per patient per scan over the entire 96-week study.

bDefined as new (post-baseline) T2 non-enhancing or enlarging lesions.

cDefined as new T1 Gd-enhancing and/or new T2 non-enhancing or enlarging lesions (without double counting).

p < 0.001 for all comparisons of active treatment versus placebo (nonparametric ANCOVA model on ranked data with fixed effects for treatment group and region and baseline T1 Gd+ lesion as a covariate)

SD = standard deviation.