Placebo-Controlled Trial of an Oral BTK Inhibitor in Multiple Sclerosis

Xavier Montalban, M.D., Ph.D., Douglas L. Arnold, M.D., Martin S. Weber, M.D., Ivan Staikov, M.D., Ph.D., Karolina Piasecka-Stryczynska, M.D., Ph.D.,
Jonathan Willmer, M.D., Emily C. Martin, Ph.D., Fernando Dangond, M.D., Sana Syed, M.D., M.P.H., and Jerry S. Wolinsky, M.D.,
for the Evobrutinib Phase 2 Study Group*

BACKGROUND
Bruton’s tyrosine kinase (BTK) regulates the functions of B cells and myeloid cells that are implicated in the pathogenesis of multiple sclerosis. Evobrutinib is a selective oral BTK inhibitor that has been shown to inhibit B-cell activation both in vitro and in vivo.

METHODS
In this double-blind, randomized, phase 2 trial, we assigned patients with relapsing multiple sclerosis to one of five groups: placebo, evobrutinib (at a dose of 25 mg once daily, 75 mg once daily, or 75 mg twice daily), or open-label dimethyl fumarate (DMF) as a reference. The primary end point was the total (cumulative) number of gadolinium- enhancing lesions identified on T1-weighted magnetic resonance imaging at weeks 12, 16, 20, and 24. Key secondary end points included the annualized relapse rate and change from baseline in the score on the Expanded Disability Status Scale (EDSS).

RESULTS
A total of 267 patients were randomly assigned to a trial group. The mean (±SD) total number of gadolinium-enhancing lesions during weeks 12 through 24 was 3.85±5.44 in the placebo group, 4.06±8.02 in the evobrutinib 25-mg group, 1.69±4.69 in the evobrutinib 75-mg once-daily group, 1.15±3.70 in the evobrutinib 75-mg twice-daily group, and 4.78±22.05 in the DMF group. The baseline adjusted rate ratios for the total number of lesions over time as compared with placebo were 1.45 in the evo- brutinib 25-mg group (P = 0.32), 0.30 in the evobrutinib 75-mg once-daily group (P= 0.005), and 0.44 in the evobrutinib 75-mg twice-daily group (P= 0.06). The unad- justed annualized relapse rate at week 24 was 0.37 in the placebo group, 0.57 in the evobrutinib 25-mg group, 0.13 in the evobrutinib 75-mg once-daily group, 0.08 in the evobrutinib 75-mg twice-daily group, and 0.20 in the DMF group. There was no significant effect of trial group on the change from baseline in the EDSS score. Eleva- tions in liver aminotransferase values were observed with evobrutinib.

CONCLUSIONS
Patients with relapsing multiple sclerosis who received 75 mg of evobrutinib once daily had significantly fewer enhancing lesions during weeks 12 through 24 than those who received placebo. There was no significant difference with placebo for either the 25-mg once-daily or 75-mg twice-daily dose of evobrutinib, nor in the an- nualized relapse rate or disability progression at any dose. Longer and larger trials are required to determine the effect and risks of evobrutinib in patients with mul- tiple sclerosis. (Funded by EMD Serono; ClinicalTrials.gov number, NCT02975349.)From Vall d’Hebron University Hospital, Barcelona (X.M.); St. Michael’s Hospital, University of Toronto, Toronto (X.M.), and Montreal Neurological Institute and NeuroRx Research, Montreal (D.L.A.) — both in Canada; the Institute of Neuropa- thology and the Department of Neurolo- gy, University Medical Center, Göttingen, Germany (M.S.W.); the Department of Neurology, Acibadem City Clinic Tokuda Hospital, Sofia, Bulgaria (I.S.); the De- partment of Histology and Embryology, Poznan University of Medical Science, Poznan, Poland (K.P.-S.); the Global Clin- ical Development Center, EMD Serono Research and Development Institute, Bil- lerica, MA ( J.W., E.M., F.D., S.S.); and
McGovern Medical School, University of Texas Health Science Center at Houston, Houston (J.S.W.). Address reprint requests to Dr. Montalban at Vall d’Hebron Univer- sity Hospital, Passeig de la Vall d’Hebron, 119-129, 08035 Barcelona, Spain, or at St. Michael’s Hospital, 30 Bond St., 3rd Fl., Shuter Wing 3-046, Toronto, ON M5B 1W8, Canada, or at xavier.montalban@ cem-cat.org or [email protected].
*A complete list of the investigators in the Evobrutinib Phase 2 Study Group is provided in the Supplementary Appen- dix, available at NEJM.org.
This article was published on May 10, 2019, at NEJM.org.
DOI: 10.1056/NEJMoa1901981
Copyright © 2019 Massachusetts Medical Society.

T he new engl a nd jour na l of medicineultiple sclerosis is a progressive demyelinating, inflammatory, and neu- rodegenerative autoimmune disease thatresults in the formation of lesions in the protec- tive layer around nerves in the brain and spinal cord.1 Goals in the treatment of patients’ dis- abling symptoms include reducing the frequency of relapses and slowing disability progression. A change in measures of disease activity on magnetic resonance imaging (MRI) may be used as a surrogate marker of treatment response.2
The activity and interactions of B cells, T cells, and myeloid cells are involved in the immuno- pathological features of multiple sclerosis.1,3,4 Antigen-activated B cells exert effector functions through antigen presentation and the produc- tion of cytokines and antibodies. Macrophages and microglia that are abundant in multiple sclerosis lesions contribute to tissue damage and repair.5,6
Bruton’s tyrosine kinase (BTK), a member of the Tec family of kinases, transmits signals through a variety of receptors in B cells and myeloid cells, so it presents a rational target in multiple sclerosis.7-9 BTK inhibitors are currently under investigation in several types of auto- immune disease, including systemic lupus ery- thematosus, rheumatoid arthritis, and multiple sclerosis.10 Evobrutinib is a selective, covalent, oral inhibitor of BTK that blocks B-cell activation and cytokine release11 and has been shown to inhibit the activation, differentiation, and polarization of proinflammatory M1 macrophages and their release of cytokines in vitro.12 Evobrutinib has shown in vivo efficacy against experimental autoimmune encephalomyelitis (an animal mod- el of brain inflammation) regardless of B-cell activity.13,14
We selected the doses of evobrutinib that were used in this trial to cover a range of BTK- receptor binding that was anticipated to be ef- fective on the basis of animal models. The irre- versible binding of evobrutinib to BTK results in a pharmacodynamic effect that is longer than the pharmacokinetic plasma half-life of evobru- tinib. The drug has greater selectivity than first- generation BTK inhibitors15 and was designed with the expectation of having fewer off-target side effects. We report the results of a phase 2 trial comparing evobrutinib with placebo and using dimethyl fumarate (DMF) as a reference in patients with clinical and imaging evidence of
active relapsing–remitting multiple sclerosis and in those with secondary progressive multiple sclerosis with superimposed relapses.16-18

methods

Trial Design
From March 2017 through July 2018, we con- ducted this randomized, placebo-controlled, phase 2 trial at 56 centers in Europe and Russia (see the Supplementary Appendix, available with the full text of this article at NEJM.org). After a 4-week screening period, patients were randomly assigned to one of five groups, in a 1:1:1:1:1 ratio, by an interactive Web-response system to receive oral evobrutinib (at a dose of 25 mg once daily, 75 mg once daily, or 75 mg twice daily), placebo, or open-label DMF (at a dose of 120 mg twice daily for the first week and 240 mg twice daily thereafter) (Fig. S1 in the Supplementary Appen- dix). After 24 weeks of the placebo-controlled phase of the trial, patients in the placebo group were switched to receive 25 mg of evobrutinib once daily for a further 24-week blinded exten- sion phase; patients who were receiving evobru- tinib or DMF continued to be given the same dose. There was a 4-week safety follow-up period after the end of trial. A temporary interruption or discontinuation of treatment was required in patients who had any of several prespecified laboratory abnormalities. (Details are provided in the Supplementary Appendix.)
Trial visits were scheduled every 4 weeks (±3 days) after screening to assess MRI results, starting at week 12; clinical disease activity, in- cluding relapse assessment, which was also evalu- ated at unscheduled visits for suspected new or worsening neurologic symptoms including re- lapse; the score on the Expanded Disability Sta- tus Scale (EDSS) at weeks 12, 24, 36, and 48 only; and safety during the 52-week period. The treating investigator at each trial site was un- aware of trial-group assignments with the excep- tion of open-label DMF. The assessing neurolo- gists and independent central MRI readers were unaware of all trial-group assignments.

Trial Oversight
The trial was conducted in accordance with the principles of the Declaration of Helsinki, the guidelines for Good Clinical Practice of the In- ternational Conference on Harmonisation, andtrial of an oral btK inhibitor in multiple sclerosisadditional local regulations. The trial was de- signed by the sponsor (EMD Serono) with input from a steering committee. The sponsor was also involvein the collection, analysis, and in- terpretation of the data and the writing of the manuscript. The sponsor provided the evobruti- nib, DMF, and placebo that were used in the trial; it paid for data collection and analysis by IQVIA (a clinical research organization) and for professional writing assistance. An independent data monitoring committee monitored safety. All the authors vouch for adherence of the trial to the protocol (available at NEJM.org), the ac- curacy and completeness of data reporting, and complete reporting of adverse events. All the authors approved the final version of the manu- script and were involved in the decision to sub- mit the manuscript for publication. Confiden- tiality agreements were in place between the authors and the sponsorpatientsPatients were eligible for treatment if they were between the ages of 18 and 65 years, had relaps- ing–remitting multiple sclerosis or secondary progressive multiple sclerosis with superimposed relapses,17,18 and had a score of no more than 6 on the EDSS (which ranges from 0 [no disability] to 10 [death]); all the patients provided written informed consent (Fig. S1 in the Supplementary Appendix). Patients were offered and encouraged to undertake treatment with available therapies for multiple sclerosis, and they consented to par- ticipate in the trial with full awareness of alterna- tive treatment options.

The diagnosis of secondary progressive mul- tiple sclerosis was based on investigator judg- ment; requirements for the determination of re- lapse were stipulated in the inclusion criteria. Key exclusion criteria were progressive multiple sclerosis, either primary or secondary with no superimposed relapses; a disease duration of more than 15 years with an EDSS score of 2 or less; and exposure to DMF within 6 months before randomization. (Details regarding inclu- sion and exclusion criteria are provided in the Supplementary Appendix.)End points
The primary end point was the total (cumulative) number of gadolinium-enhancing lesions identi- fied on T1-weighted MRI at weeks 12, 16, 20,
and 24.19 Key secondary end points were the an- nualized relapse rate, based on qualified relapses; qualified relapse-free status; change from base- line in the EDSS score at week 24; and safety. A qualified relapse was defined as new, worsen- ing, or recurrent neurologic symptoms attribut- ed to multiple sclerosis that lasted for at least 24 hours without fever, infection, or adverse reaction to a prescribed medication and that was pre- ceded by a stable or improving neurologic status of at least 30 days. A qualified relapse was ac- companied by new clinical signs, such as chang- es in the neurologic examination or an increase in the EDSS score.

All outcome measures that were based on relapse data used qualified relapses. The 13 ad- ditional secondary end points related to the ac- tive treatment groups in the trial are listed in the Supplementary Appendix. Investigation of the effects of evobrutinib, as compared with placebo, on health-related quality of life was an explor- atory end point. Serious adverse events were defined as described in the Supplementary Ap- pendix and did not include relapse as an adverse event.

Statistical Analysis
We estimated that the enrollment of 44 patients in each group would provide a power of 85% to detect a 90% lower total number of gadolinium- enhancing lesions in each evobrutinib group than in the placebo group, using the Wilcoxon rank-sum test at a two-sided significance level of 5% and assuming a negative binomial distribu- tion for the total number of lesions in each group. We based our assumptions regarding the mean total number of lesions (estimated to be

5.5 lesions in the placebo group) on four MRI evaluations and the negative binomial distribu- tion on results from recent phase 2 studies in- volving patients with multiple sclerosis.19,20 On the basis of an assumed annual dropout rate of 12%, we set a target enrollment of 50 patients per group.
The primary efficacy analysis at week 24 was based on the modified intention-to-treat analy- sis population, which consisted of the patients who had undergone randomization and who had undergone MRI at baseline and at least once after baseline. We analyzed the total number of gadolinium-enhancing lesions using negative binomial regression, with offset given by the log

T he new engl a nd jour na l of medicinenumber of available scans and adjustment for lesion activity at baseline. This analysis resulted in a comparison of each evobrutinib group with placebo on the basis of the estimated lesion rate ratio. In the lesion analyses, data regarding scans that had been performed within 3 weeks after the receipt of high-dose glucocorticoids were considered to be missing, but the total le- sion count was not imputed unless all available scans were affected. Other lesion-count end points were analyzed similarly.
The relative change in the annualized relapse rate was evaluated with a negative binomial model for the number of relapses, with offset given by the log years on trial and adjustment for baseline relapse activity. This analysis resulted in a comparison of each evobrutinib group with placebo on the basis of the estimated relapse rate ratio. Data for patients who discontinued a trial agent early were considered in the total number of relapse events, and these patients were followed until discontinuation.

Relapse-free status was analyzed by means of logistic regression, after adjustment for baseline values; patients who withdrew from the trial without relapse before week 24 were not consid- ered to be free of relapse. The change from baseline in the EDSS score was analyzed by means of the stratified Wilcoxon rank-sum test, with adjustment for the baseline EDSS score. Missing values for the change from baseline in the EDSS score at week 24 were imputed with the use of the median value among patients in the same trial group and baseline covariates. We used a multistage testing algorithm to handle multiple comparisons of the three evobrutinib dose groups on the basis of the evaluation of four efficacy end points (primary and key sec- ondary end points). The family-wise error rate for the null hypotheses associated with dose- group comparisons for a given end point was controlled by the truncated Hochberg procedure for the primary end point and the first two key secondary end points (annualized relapse rate and relapse-free status) and by the standard Hochberg procedure for the third key secondary end point (change from baseline in the EDSS score). The analysis of additional secondary end points was not adjusted for multiple compari- sons, so these outcomes are reported as point estimates and 95% confidence intervals without P values. The statistical analysis plan, including

Figure 1 (facing page). Enrollment in Placebo-Controlled Phase and Extension Phase.
After 24 weeks of receiving the assigned trial agent, patients in the placebo group were switched to receive 25 mg of evobrutinib once daily (QD) for a further 24- week blinded extension phase; patients who were re- ceiving evobrutinib and dimethyl fumarate continued to be given the same dose. The percentages of patients who completed or discontinued the extension phase are based on the original number of patients who under- went randomization to each trial group.

the handling of missing data, is provided in the Supplementary Appendix.
Within-group analyses of 48-week data were based on a subgroup of patients in the modified intention-to-treat population, including all the patients who entered the extension portion of the trial and underwent at least one MRI assess- ment after week 24. Efficacy end points were compared between the two time points (week 24 and 48) or trial phases (weeks 0 to 24 and weeks 25 to 48) with the use of the Hodges–Lehmann estimate of shift in location and the Wilcoxon signed-rank test. There were no statistical com- parisons between evobrutinib and placebo at week 48, since the placebo group was switched to active treatment for the extension period at 24 weeks.
The safety analysis population consisted of all the patients who had received at least one dose of evobrutinib, placebo, or DMF. Descrip- tive statistics were used to evaluate safety and secondary end points, including all end points related to the DMF reference group. No statisti- cal comparisons of DMF with evobrutinib or placebo were prespecified or conducted.

results

patientsOf the 267 patients who underwent randomiza- tion, 261 were included in the modified inten- tion-to-treat analysis, after the exclusion of 6 patients owing to a lack of post-baseline MRI assessments. Of the 244 patients (91%) who completed 24 weeks of the trial, 229 were in- cluded in the extension analysis after the exclu- sion of 15 patients who had not undergone MRI assessment after week 24; of the 267 patients who had undergone randomization, 227 (85%) completed 48 weeks of treatment (Fig. 1).

24-Week Placebo-Controlled Phase

54 Were assigned to receive dimethyl fumarate
54 Were assigned to receive evobrutinib (75 mg twice daily)
53 Were assigned to receive evobrutinib (75 mg once daily)
Open-label reference group
54 Were assigned to receive placebo
52 Were assigned to receive evobrutinib (25 mg once daily)

49 (91%) 5 (9%) 47 (90%) 5 (10%) 48 (91%) 5 (9%) 48 (89%) 6 (11%) 52 (96%) 2 (4%)
Completed Discontinued Completed Discontinued Completed Discontinued Completed Discontinued Completed Discontinued
treatment treatment treatment treatment treatment treatment treatment treatment treatment treatment

4 Had adverse event
1 Was lost to follow-up
1 Had adverse event 3 Withdrew consent 1 Had other reason
2 Had adverse event 3 Withdrew consent

6 Had adverse event

2 Had adverse event

trial of an oral btK inhibitor in multiple sclerosis
5
n engl j med nejm.org
The New England Journal of Medicine
Downloaded from nejm.org at SPRINGFIELD COLL on May 10, 2019. For personal use only. No other uses without permission.
Copyright © 2019 Massachusetts Medical Society. All rights reserved.

24-Week Blinded-Extension Phase

244 Were included in a blinded extension

49 Were assigned to receive evobrutinib (25 mg once daily)
47 Were assigned to receive evobrutinib (25 mg once daily)
48 Were assigned to receive evobrutinib (75 mg once daily)
48 Were assigned to receive evobrutinib (75 mg twice daily)
52 Were assigned to receive dimethyl fumarate

42 (78%) 7 (13%) 43 (83%) 4 (8%) 44 (83%) 4 (8%) 46 (85%) 2 (4%) 52 (96%) 0
Completed Discontinued Completed Discontinued Completed Discontinued Completed Discontinued Completed Discontinued
treatment treatment treatment treatment treatment treatment treatment treatment treatment treatment
1 Had adverse event 5 Withdrew consent 1 Had other reason 1 Had adverse event 1 Had lack of efficacy 2 Withdrew consent 2 Had adverse event 1 Withdrew consent 1 Had other reason

1 Had adverse event 1 Withdrew consent
1 Was lost to follow-up

T he new engl a nd jour na l of medicine
6
n engl j med nejm.org
The New England Journal of Medicine
Downloaded from nejm.org at SPRINGFIELD COLL on May 10, 2019. For personal use only. No other uses without permission.
Copyright © 2019 Massachusetts Medical Society. All rights reserved.

Table 1. Demographic and Disease Characteristics of the Patients at Baseline (Modified Intention-to-Treat Population).*

Characteristic
Placebo (N = 53) Evobrutinib, 25 mg QD
(N = 50) Evobrutinib, 75 mg QD
(N = 51) Evobrutinib, 75 mg BID
(N = 53) Dimethyl Fumarate (N = 54)
All Patients (N = 261)
Age — yr 41.6±10.8 42.4 ±9.4 42.9±10.1 42.2±11.5 42.8±11.7 42.4±10.7
Female sex — no. (%) 39 (74) 32 (64) 35 (69) 36 (68) 39 (72) 181 (69)
Type of multiple sclerosis — no. (%)
Relapsing–remitting disease 47 (89) 42 (84) 43 (84) 47 (89) 49 (91) 228 (87)
Secondary progressive disease 6 (11) 8 (16) 8 (16) 6 (11) 5 (9) 33 (13)
Median time since disease onset (range) — yr 7.5
(0.1–39.4) 8.4
(0.2–26.4) 11.4
(0.4–24.6) 10.1
(0.2–39.4) 7.3
(0.3–32.5) 8.4
(0.1–39.4)
Relapse in the previous 2 yr — no. (%)
1 relapse 26 (49) 27 (54) 18 (35) 25 (47) 20 (37) 116 (44)
≥2 relapses 27 (51) 23 (46) 33 (65) 28 (53) 34 (63) 145 (56)
Score on Expanded Disability Status Scale†
Mean 3.2±1.7 3.3±1.5 3.5±1.4 3.4±1.6 3.0±1.7 3.3±1.6
Median (range) 3.0
(0.0–6.0) 3.0
(0.0–6.0) 3.5
(1.5–6.0) 3.0
(1.0–6.0) 2.5
(0.0–6.0) 3.0
(0.0–6.0)
Gadolinium-enhancing lesions on T1-weighted MRI
Patients with lesions — no. (%) 24 (45) 19 (38) 18 (35) 23 (43) 19 (35) 103 (39)
Mean no. 1.19±1.91 0.92±2.02 1.65±5.44 1.72±3.40 2.20±6.79 1.54±4.37
Median no. (range) 0 (0–9) 0 (0–10) 0 (0–38) 0 (0–19) 0 (0–46) 0 (0–46)
Mean volume of lesions on T2-weighted MRI — cm3 15.89±12.63 13.79±11.67 14.03±12.23 19.02±13.54 18.84±17.67 16.37±13.85
* Plus–minus values are means ±SD. Patients were included in the modified intention-to-treat analysis if they had undergone randomization and MRI assessment at baseline and at least once after baseline. All the patients in this trial were white (98% were non-Hispanic or Latino; 99% of the patients were from Eastern Europe and 1% from Western Europe), as recorded by the investigators. BID denotes twice daily, MRI magnetic resonance imaging, and QD once daily.† Scores on the Expanded Disability Status Scale (EDSS) range from 0 (no disability) to 10 (death).trial of an oral btK inhibitor in multiple sclerosis

All the patients were white, with a mean age of 42 years; 69% were women. A total of 87% had relapsing–remitting multiple sclerosis; the remainder had secondary progressive multiple sclerosis with superimposed relapses (Table 1). The mean (±SD) number of gadolinium-enhanc- ing lesions at baseline was 1.54±4.37, and the mean EDSS score at baseline was 3.3±1.6. The patients’ disease characteristics at baseline were similar across the trial groups.

Efficacy Outcomes
The mean total number of gadolinium-enhanc- ing lesions at weeks 12 through 24 (the primary end point) was 3.85±5.44 in the placebo group, 4.06±8.02 in the evobrutinib 25-mg group, 1.69±4.69 in the evobrutinib 75-mg once-daily group, 1.15±3.70 in the evobrutinib 75-mg twice- daily group, and 4.78±22.05 in the DMF group (Table 2). The baseline adjusted rate ratios for the total number of lesions over time, as com- pared with placebo, were 1.45 in the evobrutinib 25-mg group (95% confidence interval [CI], 0.72 to 2.91; P = 0.32 after adjustment for multiple comparisons), 0.30 in the evobrutinib 75-mg once-daily group (95% CI, 0.14 to 0.63; P = 0.005), and 0.44 in the evobrutinib 75-mg twice-daily group (95% CI, 0.21 to 0.93; P = 0.06). (P values that were adjusted for multiple comparisons and those that were not adjusted are provided in Table 2 and Fig. 2A, and in Table S1 in the Supplementary Appendix.)

The distribution of the gadolinium-enhanc- ing lesions at week 24 in the DMF group was influenced by a patient with 46 lesions at base- line and a total of 160 lesions over three post- baseline MRI evaluations. For within-group com- parisons at all doses, there was no evidence of difference in effect of evobrutinib on the num- ber of gadolinium-enhancing lesions between weeks 24 and 48 (Table S2 in the Supplementary Appendix).

At 24 weeks, 9 relapses had occurred in the placebo group, 13 in the evobrutinib 25-mg group, 3 in the evobrutinib 75-mg once-daily group, 2 in the evobrutinib 75-mg twice-daily group, and 5 in the DMF group. As compared with placebo at week 24, there was no signifi- cant effect on the annualized relapse rate in the evobrutinib groups, with an unadjusted annual- ized relapse rate of 0.37 (95% CI, 0.17 to 0.70) in
the placebo group, 0.57 (95% CI, 0.30 to 0.97) in
the evobrutinib 25-mg group, 0.13 (95% CI, 0.03 to 0.38) in the evobrutinib 75-mg once-daily group, and 0.08 (95% CI, 0.01 to 0.30) in the evobrutinib 75-mg twice-daily group (Table 2 and Fig. 2B). The unadjusted annualized relapse rate over 48 weeks was 0.37 (95% CI, 0.21 to 0.59) in the placebo–evobrutinib 25-mg group,0.52 (95% CI, 0.33 to 0.78) in the evobrutinib 25-mg group, 0.25 (95% CI, 0.12 to 0.44) in the evobrutinib 75-mg once-daily group, 0.11 (95% CI, 0.04 to 0.25) in the evobrutinib 75-mg twice- daily group, and 0.14 (95% CI, 0.06 to 0.29) in the DMF group (Fig. 2C).
The percentage of patients who were relapse- free at week 24 was 77% in the placebo group, 74% in the evobrutinib 25-mg group, 88% in the evobrutinib 75-mg once-daily group, 87% in the evobrutinib 75-mg twice-daily group, and 89% in the DMF group (Table 2); the percentages were similar at 48 weeks (Fig. S2 in the Supple- mentary Appendix). The median change from baseline in the EDSS score at weeks 24 and 48 was zero in all five groups (Table 2, and Table S2 in the Supplementary Appendix).

At 24 weeks, the secondary outcome of the lesion rate ratio for new or enlarging lesions on T2-weighted MRI was 0.42 (95% CI, 0.20 to 0.87) in the evobrutinib 75-mg twice-daily group, in- dicating a better response than in the other evobrutinib dose groups. The results of this and other imaging and clinical secondary end points are provided in Table S2 and the results of quality-of-life analyses are provided in Table S3 in the Supplementary Appendix.

Safety Outcomes
Adverse events that occurred during the 52-week safety period are listed in Table 3, and Table S4 in the Supplementary Appendix. The highest rate of serious adverse events (7%) occurred in pa- tients treated with evobrutinib 75 mg twice daily. Grade 3 or 4 adverse events were most frequent among patients in the evobrutinib 75-mg once- daily group (13%), the evobrutinib 75-mg twice- daily group (15%), and the DMF group (13%). The most commonly observed adverse events of any grade that were associated with evobrutinib were nasopharyngitis and increases in levels of alanine aminotransferase (ALT), aspartate amino- transferase (AST), and lipase. Patients with ele- vations in aminotransferase levels were asymp- tomatic, the elevations were reversible, and no

MRI and Clinical Outcomes at 24 Weeks (Modified Intention-to-Treat Population).*

Outcome
Placebo (N = 53) Evobrutinib, 25 mg QD
(N = 50) Evobrutinib, 75 mg QD
(N = 51) Evobrutinib, 75 mg BID
(N = 53) Dimethyl Fumarate (N = 54)
Cumulative no. of gadolinium-enhancing lesions on T1-weighted MRI at 12, 16, 20, and 24 wk†
Mean 3.85±5.44 4.06±8.02 1.69±4.69 1.15±3.70 4.78±22.05
Median 1 1 0 0 0
Range 0–20 0–38 0–27 0–25 0–160‡
Interquartile range 0–6 0–2 0–1 0–1 0–1
Lesion rate ratio (95% CI)§ NA 1.45
(0.72 to 2.91) 0.30
(0.14 to 0.63) 0.44
(0.21 to 0.93) NA
Adjusted P value vs. placebo NA 0.32 0.005 0.06 NA
Unadjusted P value vs. placebo NA 0.29 0.002 0.03 NA
Relapse at 24 wk¶
No. of relapses 9 13 3 2 5
Unadjusted annualized relapse rate (95% CI)‖ 0.37
(0.17 to 0.70) 0.57
(0.30 to 0.97) 0.13
(0.03 to 0.38) 0.08
(0.01 to 0.30) 0.20
(0.06 to 0.47)
Relapse rate ratio (95% CI)** NA 1.66
(0.67 to 4.09) 0.31
(0.08 to 1.20) 0.23
(0.05 to 1.09) NA
Relapse-free status at wk 24¶
Patients with no relapse (95% CI) — % 77
(64 to 88) 74
(60 to 85) 88
(76 to 96) 87
(75 to 95) 89
(77 to 96)
Odds ratio for no relapse (95% CI)†† NA 0.75
(0.29 to 1.95) 2.79
(0.92 to 8.41) 2.08
(0.72 to 5.99) NA
Change from baseline in EDSS score at 24 wk¶‡‡
Median 0.0 0.0 0.0 0.0 0.0
Range –1.0 to 1.0 –2.5 to 2.5 –4.5 to 0.5 –0.5 to 1.0 –1.0 to 1.0
* Plus–minus values are means ±SD. CI denotes confidence interval, and NA not applicable.
† This outcome is the primary end point.

‡ The distribution of the total number of gadolinium-enhancing lesions on T1-weighted MRI at week 24 in the dimethyl fumarate group was influenced by one patient who had 46 lesions at baseline and a total of 160 lesions over three post-baseline scans. In a post hoc analysis that excluded data for this patient, the dimethyl fumarate group had improved outcomes.
§ The lesion rate ratio was estimated on the basis of a negative binomial model for the total number of gadolinium-enhancing lesions on T1-weighted MRI summed over the available scans through week 24. The model includes factors for trial group and a covariate for the presence or absence of gadolinium-enhancing lesions at baseline, with offset equal to the log number of available scans. The lesion rate ratio, which compares the number of lesions per scan (i.e., lesion rate) between groups, equals one minus the relative reduction in le- sion rate. Data regarding scans that were performed within 3 weeks after the receipt of high-dose glucocorticoids were considered to be missing.
¶ This outcome was one of the key secondary end points.
‖ The unadjusted annualized relapse rate was defined as the number of relapses among patients divided by the number of patient-years of follow-up. For patients who discontinued the trial early, all relapses and follow-up through the safety follow-up visit were included.
** The relapse rate ratio was estimated on the basis of a negative binomial model for relapse count over 24 weeks. The model includes factors for trial group and a covariate for baseline relapse activity (≤1 or >1 relapse in 2 years before trial entry), with offset equal to the log years of follow-up. The relapse rate ratio equals one minus the relative reduction in the annualized relapse rate.
†† The relapse-free odds ratio was estimated on the basis of a logistic model that included factors for trial group and a covariate for baseline relapse activity. Patients who discontinued the trial before week 24 without relapse were not considered to be relapse-free. The calculation of the 95% confidence interval for the proportion of patients is based on the Clopper–Pearson exact method, and the calculation of the odds ratio is based on the Thomas algorithm.
Placebo Evobrutinib, Evobrutinib, Evobrutinib, Dimethyl 25 mg QD 75 mg QD 75 mg BID fumarate

No. at Risk
Lesion rate ratio (95% CI) 1.45 (0.72–2.91) 0.30 (0.14–0.63) 0.44 (0.21–0.93)
Adjusted P value vs. placebo 0.32 0.005 0.06
Unadjusted P value vs. placebo 0.29 0.002 0.03
B Relapse Rate Ratio and Annualized Relapse Rate at Wk 24
Relapse Rate Ratio Annualized Relapse Rate
Trial Group (95% CI) (95% CI)
Placebo 0.37 (0.17–0.70)
Evobrutinib
25 mg QD 1.66 (0.67–4.09) 0.57 (0.30–0.97)
75 mg QD 0.31 (0.08–1.20) 0.13 (0.03–0.38)
75 mg BID 0.23 (0.05–1.09) 0.08 (0.01–0.30)
Dimethyl fumarate 0.20 (0.06–0.47)

0 1 2 3 4 5

Favors Favors
Evobrutinib Placebo
C Annualized Relapse Rate at Wk 24 and 48
Placebo Evobrutinib, 25 mg QD Evobrutinib, 75 mg QD Evobrutinib, 75 mg BID Placebo–evobrutinib, 25 mg QD (switch) Dimethyl fumarate
0.57
0.6 (0.30–0.97) 0.52
(0.33–0.78)
0.5
0.37 0.37
0.4 (0.17–0.70) (0.21–0.59)
0.25
0.3 0.20 (0.12–0.44)
0.2 0.13 (0.06–0.47) 0.11 0.14
(0.03–0.38) 0.08 (0.04–0.25) (0.06–0.29)
0.1 (0.01–0.30)
0.0

Wk 0–24 Wk 0–48
Figure 2. Total Number of Lesions at Weeks 12 through 24, Relapse Measures at 24 Weeks, and Annualized Relapse Ratio at 24 Weeks and 48 Weeks.
Panel A shows the total number of gadolinium-enhancing lesions identified on T1-weighted magnetic resonance imaging at weeks 12, 16, 20, and 24 (primary end point) in the five trial groups. The T bars denote the standard devi- ation. The vertical dashed line separates the open-label dimethyl fumarate group from the groups in the placebo- controlled analysis. Panel B shows the relapse rate ratio and annualized relapse rate at 24 weeks (a key secondary end point). The relapse rate ratio is based on a negative binomial model for the relapse count that was adjusted for patients’ relapse activity at baseline. The annualized relapse rate was not adjusted for baseline relapse activity. The relapse rate ratio for the three evobrutinib groups was calculated as a comparison with placebo and did not include the dimethyl fumarate reference group. Panel C shows the annualized relapse rate at weeks 24 and 48. All the analy- ses were performed in the modified intention-to-treat population. BID denotes twice daily, and QD once daily.

The new engl a nd jour na l of medicine

Table 3. Adverse Events at 52 Weeks (Safety Population).
Placebo–
Evobrutinib, Evobrutinib, Evobrutinib, Evobrutinib, Dimethyl
25 mg QD* 25 mg QD 75 mg QD 75 mg BID Fumarate
Adverse Event (N = 54) (N = 52) (N = 53) (N = 54) (N = 54)
number of patients (percent)
Any adverse event 30 (56) 28 (54) 35 (66) 34 (63) 35 (65)
Any grade 3 or 4 adverse event† 6 (11) 1 (2) 7 (13) 8 (15) 7 (13)
Serious adverse event‡ 2 (4) 2 (4) 2 (4) 4 (7) 2 (4)
Adverse event leading to discontinuation 5 (9) 3 (6) 6 (11) 7 (13) 2 (4)
Adverse event deemed by investigator to be related to trial agent 14 (26) 10 (19) 15 (28) 18 (33) 26 (48)
Infection 16 (30) 17 (33) 10 (19) 12 (22) 12 (22)
Neoplasm§ 2 (4) 0 0 0 1 (2)
Most common adverse events¶
Nausea 0 2 (4) 0 1 (2) 3 (6)
Diarrhea 2 (4) 1 (2) 0 0 4 (7)
Nasopharyngitis 5 (9) 9 (17) 3 (6) 7 (13) 2 (4)
Upper respiratory tract infection 2 (4) 1 (2) 1 (2) 1 (2) 3 (6)
Urinary tract infection 5 (9) 2 (4) 1 (2) 0 2 (4)
Increase in alanine aminotransferase 4 (7) 3 (6) 6 (11) 5 (9) 3 (6)
Increase in aspartate aminotransferase 1 (2) 1 (2) 2 (4) 4 (7) 2 (4)
Increase in lipase 5 (9) 2 (4) 5 (9) 5 (9) 3 (6)
Increase in creatinine 1 (2) 0 3 (6) 3 (6) 1 (2)
Low lymphocyte count 0 0 0 1 (2) 5 (9)
Arthralgia 1 (2) 2 (4) 3 (6) 0 4 (7)
Headache 2 (4) 3 (6) 2 (4) 1 (2) 1 (2)
Flushing 0 0 0 0 12 (22)

* After 24 weeks, patients in the placebo group were switched to receive 25 mg of evobrutinib once daily for a further 24-week blinded exten- sion period. There was a 4-week safety follow-up period after the end of trial.
† No deaths occurred during the trial.
‡ Serious adverse events (each of which occurred in 1 patient) included pneumonia, lung neoplasm, and peripheral embolism in the placebo– evobrutinib 25-mg group; toxic hepatitis (asymptomatic elevation in aminotransferase levels and no elevation in bilirubin level) and overdose of the trial agent in the evobrutinib 25-mg group; traffic accident and spontaneous abortion in the evobrutinib 75-mg once-daily group; toxic hepatitis (as previously defined), elevation in aminotransferase levels, epilepsy, and restless leg syndrome in the evobrutinib 75-mg twice- daily group; and Lyme disease and gastric cancer in the dimethyl fumarate group.
§ In the placebo–evobrutinib 25-mg group, 1 patient had a lung neoplasm before switching to evobrutinib after week 24, and 1 had a skin papilloma after switching to evobrutinib; in the dimethyl fumarate group, gastric cancer was diagnosed in 1 patient.
¶This list includes the most common adverse events that were reported in at least 5% of the patients in any group. A list of additional adverse events is provided in Table S4 in the Supplementary Appendix.cases fell within the criteria of Hy’s law for drug-induced liver injury, as defined by the Food and Drug Administration.21 Shifts in laboratory values from normal (grade 0) at baseline to ele- vated ALT levels and from any baseline grade to highest grade over 52 weeks in all the trial groups (including patients who switched from placebo to evobrutinib) are shown in Table S5 in
the Supplementary Appendix. All elevations oc- curred within the first 24 weeks of the trial.
The percentage of patients who had infec- tions was higher in the evobrutinib 25-mg group and the placebo–evobrutinib 25-mg group (30 to 33%) than in other groups (Table 3). Single cases of infection occurred in the evobrutinib 25-mg group (viral infection, 2%) and the evotrial of an oral btK inhibitor in multiple sclerosisbrutinib 75-mg once-daily group (oral herpes and viral respiratory tract infection, 2% each).
Over the 24-week placebo-controlled phase of the trial, the percentage of patients who had a shift from a normal lymphocyte count to a de- creased count (grade 1 lymphopenia) was simi- lar in the placebo group (6%) and the evobruti- nib groups (4% in the evobrutinib 25-mg group, 4% in the evobrutinib 75-mg once-daily group, and 6% in the evobrutinib 75-mg twice-daily group); 2% of the patients in the evobrutinib 75-mg twice-daily group had a shift from nor- mal to grade 2 lymphopenia. The percentage of patients with a shift from normal to a decreased lymphocyte count in the DMF group was 19.6% to grade 1 and 13.7% to grade 2. Findings over 52 weeks were similar. Data regarding shifts in the lymphocyte count from normal and from baseline are provided in Table S6 in the Supple- mentary Appendix.
Trial discontinuation owing to an adverse event associated with a trial agent was most frequent in the evobrutinib 75-mg once-daily and twice-daily groups (11% and 13%, respec- tively) (Fig. 1). Most of the discontinuations were due to protocol-mandated withdrawals for eleva- tions in levels of ALT, AST, and lipase in evobru- tinib-treated patients (Table S7 in the Supple- mentary Appendix). There were no deaths during the trial.

discussion

Biologic pathways involving activated B cells and myeloid cells play a role in multiple sclerosis,1,3-6 and BTK inhibition may alter these pathways. After 24 weeks of treatment, the primary end point of the total number of gadolinium- enhancing lesions on T1-weighted MRI, measured at weeks 12 through 24, was significantly lower among patients in the evobrutinib 75-mg once- daily group than in the placebo group. However, the difference with placebo was not significant in the groups that received either a lower dose or a higher dose of evobrutinib, as assessed by the lesion rate ratio adjusted for baseline lesion ac- tivity. Also, there was no significant between- group difference with placebo in the annualized relapse rate (a secondary outcome) at any evo- brutinib dose. The trial was not designed to compare evobrutinib with DMF, but the percent- age of patients in the DMF group who had not had a relapse at 24 weeks was 89%, as comparedwith 74%, 88%, and 87% in the three evobruti- nib dose groups.

Evobrutinib 75-mg once-daily and twice-daily doses were associated with higher rates of ad- verse events, including grade 3 events, than the evobrutinib 25-mg dose or placebo. Higher evo- brutinib doses were associated with a higher frequency of elevations in ALT, AST, or lipase levels than in the other trial groups at 52 weeks. Since most discontinuations from evobrutinib were caused by hepatobiliary disorders and changes in liver aminotransferase levels, the adoption of hepatic risk-mitigation strategies and stopping rules in future clinical trials may be appropriate.
This trial has several limitations. As com- pared with other trials involving patients with multiple sclerosis,19,20,22-32 the population was old- er, the disease duration was longer, and fewer patients had had relapses within 2 years before baseline. These observations may have been re- lated to the inclusion of patients with secondary progressive multiple sclerosis with superim- posed relapses and a relatively high baseline EDSS score at trial entry. In addition, only white patients were enrolled, which may limit the gen- eralizability of the findings to other popula- tions. One patient in the DMF group had very large numbers of gadolinium-enhancing lesions at baseline and over 24 weeks, which may have influenced the results for the number of lesions in that group. In addition, this phase 2 trial was powered for MRI disease activity as an outcome measure; nevertheless, the confidence intervals for differences between evobrutinib and placebo in secondary end points did not support a signifi- cant effect for evobrutinib on other outcomes.

In conclusion, in patients with relapsing mul- tiple sclerosis, the inhibition of BTK with evo- brutinib at a dose of 75 mg once daily, but not at doses of 75 mg twice daily or 25 mg once daily, reduced the total number of enhancing MRI lesions, as compared with placebo, at weeks 12 through 24. Treatment with evobrutinib at any dose had no effect on the annualized relapse rate or disability progression and was associated with elevations in liver aminotransferase levels.

A data sharing statement provided by the authors is available with the full text of this article at NEJM.org.
Supported by EMD Serono Research and Development In- stitute.
Disclosure forms provided by the authors are available with the full text of this article at NEJM.org.
We thank the patients and their families for their participa- tion in this trial.
nejm.org 11

References
1. Reich DS, Lucchinetti CF, Calabresi PA. Multiple sclerosis. N Engl J Med 2018; 378:169-80.
2. Giovannoni G. Disease-modifying treatments for early and advanced multi- ple sclerosis: a new Evobrutinib treatment paradigm. Curr Opin Neurol 2018;31:233-43.
3. Lehmann-Horn K, Kinzel S, Weber MS. Deciphering the role of B cells in multiple sclerosis-towards specific targeting of pathogenic function. Int J Mol Sci 2017; 18(10):E2048.
4. Fletcher JM, Lalor SJ, Sweeney CM, Tubridy N, Mills KH. T cells in multiple sclerosis and experimental autoimmune encephalomyelitis. Clin Exp Immunol 2010;162:1-11.
5. Rawji KS, Mishra MK, Yong VW. Re- generative capacity of macrophages for remyelination. Front Cell Dev Biol 2016;4: 47.
6. Vogel DY, Vereyken EJ, Glim JE, et al. Macrophages in inflammatory multiple sclerosis lesions have an intermediate ac- tivation status. J Neuroinflammation 2013; 10:35.
7. Hartkamp LM, Fine JS, van Es IE, et al. Btk inhibition suppresses agonist-induced human macrophage activation and inflam- matory gene expression in RA synovial tissue explants. Ann Rheum Dis 2015;74: 1603-11.
8. Corneth OBJ, Verstappen GMP, Paulis- sen SMJ, et al. Enhanced Bruton’s tyrosine kinase activity in peripheral blood B lym- phocytes from patients with autoimmune disease. Arthritis Rheumatol 2017;69: 1313-24.
9. Ní Gabhann J, Hams E, Smith S, et al. Btk regulates macrophage polarization in response to lipopolysaccharide. PLoS One 2014;9(1):e85834.
10. Akinleye A, Chen Y, Mukhi N, Song Y, Liu D. Ibrutinib and novel BTK inhibitors in clinical development. J Hematol Oncol 2013;6:59.
11. Haselmayer P, Camps M, Liu-Bujalski L, et al. Efficacy and pharmacodynamic modeling of the BTK inhibitor evobruti- nib in autoimmune disease models. J Im- munol 2019;202:2888-906.
12. Alankus YB, Grenningloh R, Hasel- mayer P, Bender A, Bruttger J. BTK inhibi-

tion prevents inflammatory macrophage differentiation: a potential role in MS. Mult Scler 2018;24:Suppl 2:264. abstract.
13. Torke S, Grenningloh R, Boschert U, Weber M. Inhibition of Bruton’s tyrosine kinase selectively prevents antigen-acti- vation of B cells and ameliorates B cell- mediated experimental autoimmune en- cephalomyelitis. Mult Scler 2018;24:Suppl 2:
276. abstract.
14. Boschert U, Crandall T, Pereira A, et al. T cell mediated experimental CNS auto- immunity induced by PLP in SJL mice is modulated by evobrutinib (M2951) a novel Bruton’s tyrosine kinase inhibitor. Mult Scler 2017;23:Suppl 3:327. abstract.
15. Crawford JJ, Johnson AR, Misner DL, et al. Discovery of GDC-0853: a potent, selective, and noncovalent Bruton’s tyro- sine kinase inhibitor in early clinical de- velopment. J Med Chem 2018;61:2227-45.
16. McDonald WI, Compston A, Edan G, et al. Recommended diagnostic criteria for multiple sclerosis: guidelines from the International Panel on the diagnosis of multiple sclerosis. Ann Neurol 2001;50: 121-7.
17. Polman CH, Reingold SC, Banwell B, et al. Diagnostic criteria for multiple scle- rosis: 2010 revisions to the McDonald cri- teria. Ann Neurol 2011;69:292-302.
18. Lublin FD, Reingold SC, Cohen JA, et al. Defining the clinical course of multiple sclerosis: the 2013 revisions. Neurology 2014;83:278-86.
19. Kappos L, Li D, Calabresi PA, et al. Ocrelizumab in relapsing-remitting multi- ple sclerosis: a phase 2, randomised, placebo-controlled, multicentre trial. Lan- cet 2011;378:1779-87.
20. Hauser SL, Waubant E, Arnold DL, et al. B-cell depletion with rituximab in relaps- ing–remitting multiple sclerosis. N Engl J Med 2008;358:676-88.
21. Department of Health and Human Services. Guidance for industry — drug- induced liver injury: premarketing clini- cal evaluation. Silver Spring, MD: Food and Drug Administration, July 2009 (https:// 21. www.fda.gov/downloads/guidances/ UCM174090.pdf).
22. Kappos L, Gold R, Miller DH, et al. Efficacy and safety of oral fumarate in
patients with relapsing-remitting multi- ple sclerosis: a multicentre, randomised, double-blind, placebo-controlled phase IIb study. Lancet 2008;372:1463-72.
23. Cohen JA, Arnold DL, Comi G, et al. Safety and efficacy of the selective sphin- gosine 1-phosphate receptor modulator ozanimod in relapsing multiple sclerosis (RADIANCE): a randomised, placebo-con- trolled, phase 2 trial. Lancet Neurol 2016; 15:373-81.
24. Olsson T, Boster A, Fernández Ó, et al. Oral ponesimod in relapsing-remitting multiple sclerosis: a randomised phase II trial. J Neurol Neurosurg Psychiatry 2014; 85:1198-208.
25. Sorensen PS, Lisby S, Grove R, et al. Safety and efficacy of ofatumumab in relapsing-remitting multiple sclerosis: a phase 2 study. Neurology 2014;82:573-81.
26. Selmaj K, Li DK, Hartung HP, et al. Siponimod for patients with relapsing- remitting multiple sclerosis (BOLD): an adaptive, dose-ranging, randomised, phase 2 study. Lancet Neurol 2013;12:756-67.
27. The CAMMS223 Trial Investigators. Alemtuzumab vs. interferon beta-1a in early multiple sclerosis. N Engl J Med 2008; 359:1786-801.
28. Comi G, Pulizzi A, Rovaris M, et al. Effect of laquinimod on MRI-monitored disease activity in patients with relapsing- remitting multiple sclerosis: a multicentre, randomised, double-blind, placebo-con- trolled phase IIb study. Lancet 2008;371: 2085-92.
29. O’Connor PW, Li D, Freedman MS, et al. A phase II study of the safety and efficacy of teriflunomide in multiple sclerosis with relapses. Neurology 2006;66:894-900.
30. Hauser SL, Bar-Or A, Comi G, et al. Ocrelizumab versus interferon beta-1a in relapsing multiple sclerosis. N Engl J Med 2017;376:221-34.
31. Fox RJ, Miller DH, Phillips JT, et al. Placebo-controlled phase 3 study of oral BG-12 or glatiramer in multiple sclerosis. N Engl J Med 2012;367:1087-97.
32. Gold R, Kappos L, Arnold DL, et al. Placebo-controlled phase 3 study of oral BG-12 for relapsing multiple sclerosis. N Engl J Med 2012;367:1098-107.
Copyright © 2019 Massachusetts Medical Society.