T h e n e w e n g l a n d j o u r n a l o f m e d i c i n e
`
`original article
`
`Natalizumab plus Interferon Beta-1a
`for Relapsing Multiple Sclerosis
`
`Richard A. Rudick, M.D., William H. Stuart, M.D., Peter A. Calabresi, M.D.,
`Christian Confavreux, M.D., Steven L. Galetta, M.D., Ernst-Wilhelm Radue, M.D.,
`Fred D. Lublin, M.D., Bianca Weinstock-Guttman, M.D., Daniel R. Wynn, M.D.,
`Frances Lynn, M.Sc., Michael A. Panzara, M.D., M.P.H.,
`and Alfred W. Sandrock, M.D., Ph.D., for the SENTINEL Investigators*
`
`A bs tr ac t
`
`Background
`Interferon beta is used to modify the course of relapsing multiple sclerosis. Despite
`interferon beta therapy, many patients have relapses. Natalizumab, an α4 integrin
`antagonist, appeared to be safe and effective alone and when added to interferon
`beta-1a in preliminary studies.
`
`Methods
`We randomly assigned 1171 patients who, despite interferon beta-1a therapy, had
`had at least one relapse during the 12-month period before randomization to receive
`continued interferon beta-1a in combination with 300 mg of natalizumab (589
`patients) or placebo (582 patients) intravenously every 4 weeks for up to 116 weeks.
`The primary end points were the rate of clinical relapse at 1 year and the cumulative
`probability of disability progression sustained for 12 weeks, as measured by the
`Expanded Disability Status Scale, at 2 years.
`
`Results
`Combination therapy resulted in a 24 percent reduction in the relative risk of sus-
`tained disability progression (hazard ratio, 0.76; 95 percent confidence interval, 0.61
`to 0.96; P = 0.02). Kaplan–Meier estimates of the cumulative probability of progres-
`sion at two years were 23 percent with combination therapy and 29 percent with
`interferon beta-1a alone. Combination therapy was associated with a lower annual-
`ized rate of relapse over a two-year period than was interferon beta-1a alone (0.34
`vs. 0.75, P<0.001) and with fewer new or enlarging lesions on T2-weighted magnetic
`resonance imaging (0.9 vs. 5.4, P<0.001). Adverse events associated with combina-
`tion therapy were anxiety, pharyngitis, sinus congestion, and peripheral edema.
`Two cases of progressive multifocal leukoencephalopathy, one of which was fatal,
`were diagnosed in natalizumab-treated patients.
`
`Conclusions
`Natalizumab added to interferon beta-1a was significantly more effective than inter-
`feron beta-1a alone in patients with relapsing multiple sclerosis. Additional re-
`search is needed to elucidate the benefits and risks of this combination treatment.
`(ClinicalTrials.gov number, NCT00030966.)
`
`From the Mellen Center for Multiple
`Sclerosis Treatment and Research,
`Cleveland Clinic Foundation, Cleveland
`(R.A.R.); the MS Center of Atlanta, Atlanta
`(W.H.S.); the Johns Hopkins Multiple
`Sclerosis Center, Baltimore (P.A.C.);
`Hôpital Neurologique, Lyon, France (C.C.);
`University of Pennsylvania School of
`Medicine, Philadelphia (S.L.G.); Univer-
`sity Hospital Basel, Basel, Switzerland
`(E.-W.R.); Mt. Sinai School of Medicine,
`New York (F.D.L.); Baird Multiple Sclerosis
`Center, State University of New York at
`Buffalo, Buffalo (B.W.-G.); Consultants in
`Neurology Multiple Sclerosis Center,
`Northbrook, Ill. (D.R.W.); and Biogen Idec,
`Cambridge, Mass. (F.L., M.A.P., A.W.S.).
`Address reprint requests to Dr. Rudick at
`the Mellen Center for Multiple Sclerosis
`Treatment and Research, Cleveland Clinic
`Foundation, 9500 Euclid Ave., Cleveland,
`OH 44195, or at rudickr@ccf.org.
`
`*The Safety and Efficacy of Natalizumab in
`Combination with Interferon Beta-1a in
`Patients with Relapsing Remitting Multi-
`ple Sclerosis (SENTINEL) Investigators
`are listed in the Supplementary Appendix,
`available with the full text of this article at
`www.nejm.org.
`
`N Engl J Med 2006;354:911-23.
`Copyright © 2006 Massachusetts Medical Society.
`
`n engl j med 354;9 www.nejm.org march 2, 2006
`
`911
`
`The New England Journal of Medicine
`Downloaded from nejm.org on October 22, 2023. For personal use only. No other uses without permission.
` Copyright © 2006 Massachusetts Medical Society. All rights reserved.
`
`Merck 2058
`TWi v Merck
`IPR2023-00049
`
`

`

`T h e n e w e n g l a n d j o u r n a l o f m e d i c i n e
`
`The adhesion molecule α4 β1 integrin
`
`is a key initiator of the inflammatory cas-
`cade involved in the pathogenesis of multi-
`ple sclerosis.1-4 Natalizumab (Tysabri, Biogen Idec
`and Elan Pharmaceuticals) is the first α4 integrin
`antagonist in a new class of selective adhesion-
`molecule inhibitors for the treatment of multiple
`sclerosis. Natalizumab binds to α4 integrin on
`the surface of leukocytes, inhibiting their migra-
`tion into the brain and thereby reducing inflam-
`mation.
`Current disease-modifying therapies for relaps-
`ing–remitting multiple sclerosis (interferon beta
`and glatiramer acetate) are only partially effec-
`tive,5-8 and most patients with multiple sclerosis
`have breakthrough disease activity despite ther-
`apy with these agents. Hence, there is a need for
`additional treatment options in multiple sclero-
`sis. Natalizumab is an attractive therapy to add to
`current disease-modifying therapies in patients
`with breakthrough disease because preliminary
`efficacy9 and safety10 data have been favorable
`and because the mechanism of action of natali-
`zumab may complement those of other disease-
`modifying therapies.11-17
`The Safety and Efficacy of Natalizumab in
`Combination with Interferon Beta-1a in Patients
`with Relapsing Remitting Multiple Sclerosis
`(SENTINEL) study was a two-year, phase 3 clinical
`trial designed to determine whether natalizumab,
`when added to interferon beta-1a, has efficacy in
`addition to that associated with interferon beta-1a
`alone. The trial was also designed to confirm the
`safety of natalizumab when added to interferon
`beta-1a.
`
`Me thods
`
`Patients
`One hundred twenty-four clinical centers in Eu-
`rope and the United States enrolled 1196 patients
`beginning on January 14, 2002. All patients gave
`written informed consent. The study protocol was
`developed by the investigator advisory committee
`and the sponsors and was approved by central
`and local ethics committees, and the study was
`overseen by an independent safety-monitoring
`committee. Data were collected by the investiga-
`tors and an independent organization (PPD Inter-
`national) and were held and analyzed by Biogen
`Idec and Elan Pharmaceuticals. During the study,
`the investigator advisory committee and repre-
`
`sentatives of Biogen Idec met at least monthly to
`review and manage the study. The manuscript was
`written by Drs. Rudick and Panzara, with input
`from each of the other authors; all the authors
`vouch for the veracity and completeness of the
`data and analyses.
`Eligible patients were 18 to 55 years of age;
`had a diagnosis of relapsing–remitting multiple
`sclerosis,18 a score on the Expanded Disability
`Status Scale (EDSS) (possible scores range from
`0 to 10, with higher scores indicating more severe
`disease) between 0 and 5.0,19 and a magnetic
`resonance imaging (MRI) scan revealing lesions
`consistent with a diagnosis of multiple sclerosis;
`had received treatment with interferon beta-1a for
`at least 12 months before randomization; and
`had had at least one relapse during the 12-month
`period before randomization. Patients were in-
`eligible if they had primary progressive, second-
`ary progressive, or progressive relapsing multiple
`sclerosis20; if they had had a relapse within 50
`days before randomization; or if they had been
`treated with an approved disease-modifying ther-
`apy other than interferon beta-1a intramuscularly
`once weekly within the 12-month period before
`randomization.
`
`Study Design and Randomization
`This study was a randomized, double-blind, place-
`bo-controlled, parallel-group, phase 3 clinical trial.
`Data from 1171 of the 1196 patients enrolled were
`analyzed, because a single center with 25 patients
`was excluded before unblinding owing to irregu-
`larities in data. Patients were randomly assigned, in
`a 1:1 ratio, to receive 300 mg of natalizumab (589
`patients) or placebo (582 patients) intravenously ev-
`ery 4 weeks in addition to interferon beta-1a (Avon-
`ex, Biogen Idec) at a dose of 30 μg intramuscularly
`once weekly for up to 116 weeks. Randomization
`was stratified according to study site in blocks of
`four (two active and two placebo) with the use of a
`computer-generated schedule and a multidigit iden-
`tification number, implemented by way of an inter-
`active voice-response system. All study personnel,
`patients, sponsor personnel involved in the conduct
`of the study, and members of the investigator advi-
`sory committee were blinded to the treatment as-
`signments throughout the study.
`
`Study Procedure and End Points
`Each site designated primary and backup exam-
`ining neurologists and treating neurologists. The
`
`912
`
`n engl j med 354;9 www.nejm.org march 2, 2006
`
`The New England Journal of Medicine
`Downloaded from nejm.org on October 22, 2023. For personal use only. No other uses without permission.
` Copyright © 2006 Massachusetts Medical Society. All rights reserved.
`
`

`

`natalizumab and interferon beta-1a in multiple sclerosis
`
`examining neurologists performed the EDSS and
`neurologic examinations but were otherwise not
`involved in the patients’ medical care. The treat-
`ing neurologists were responsible for all patient
`care, including the management of adverse events
`and relapses of multiple sclerosis.
`Clinical visits every 12 weeks included assess-
`ment of relapses, EDSS evaluation, blood chemi-
`cal and hematologic tests, assessment of any ad-
`verse events, and immunogenicity studies. Patients
`were also seen by a treating neurologist during
`unscheduled visits within 72 hours after the de-
`velopment of new symptoms so that they could be
`assessed for possible relapses or adverse events.
`If a relapse was suspected, the patient was evalu-
`ated by the examining neurologist. Relapses were
`defined as the development of new or recurrent
`neurologic symptoms not associated with fever
`or infection, lasting at least 24 hours, and accom-
`panied by new, objective neurologic findings. At
`the discretion of the treating neurologist, relapses
`were treated with intravenous methylprednisolone
`at a dose of 1000 mg per day for three or five days.
`Patients who had disability progression that was
`sustained for 12 weeks were asked to provide con-
`sent to continue study participation and were giv-
`en the option of adding an available multiple scle-
`rosis treatment as rescue medication, according to
`protocol, while continuing to receive the study
`drug. Patients who discontinued the study drug
`were strongly encouraged to remain in the study
`for follow-up assessments, and all patients who
`continued to participate in the study were evalu-
`ated (according to the intention-to-treat principle).
`Proton-density, T2-weighted MRI scans and
`gadolinium-enhanced T1-weighted MRI scans of
`the brain were obtained at baseline and at weeks
`52 and 104. Forty contiguous, 3-mm-thick axial
`slices were acquired. MRI analyses were per-
`formed centrally at the MS-MRI Evaluation Cen-
`ter (Basel, Switzerland) by blinded raters. The
`scans were checked for artifacts, compliance with
`scanning requirements, and repositioning.
`The primary efficacy end point was the rate of
`clinical relapse at one year. Secondary end points
`at one year were the number of new or enlarging
`T2-hyperintense lesions, the number of gadolin-
`ium-enhancing lesions, and the proportion of pa-
`tients free of relapse. The primary efficacy end
`point at two years was the cumulative probabil-
`ity of sustained disability progression, defined as
`an increase by at least 1.0 point in the EDSS score
`
`from a baseline score of at least 1.0 or an increase
`by at least 1.5 points in the EDSS score from a
`baseline score of 0, sustained for 12 weeks; pro-
`gression could not be confirmed during a relapse.
`Secondary end points at two years were the rate
`of clinical relapse, the volume of T2-hyperintense
`lesions, the number of new T1-hypointense le-
`sions, and disability as measured by the Multiple
`Sclerosis Functional Composite.21 This report
`presents data pertaining to primary end points
`and key secondary efficacy end points, as well as
`safety data. Results pertaining to additional sec-
`ondary end points and tertiary end points are not
`included in this report.
`Binding antibodies against natalizumab were
`assessed with use of an enzyme-linked immuno-
`sorbent assay. Positive samples (0.5 μg per mil-
`liliter) were further tested in a flow-cytometry
`assay to determine whether these antibodies inter-
`fered with the binding of natalizumab to α4 in-
`tegrin.
`
`Statistical Analysis
`The sample size was estimated, on the basis of
`data from previous trials of natalizumab9 and
`interferon beta-1a,6 with the use of two-sided
`tests with an experiment-wise alpha of 0.05. The
`annualized rate of relapse among patients receiv-
`ing combination therapy at one year was predict-
`ed to be 0.6, as compared with 0.9 among pa-
`tients receiving interferon beta-1a alone. For the
`annualized relapse rate, the likelihood-ratio test
`was used to determine the sample size with half
`the patients receiving active drug and half receiv-
`ing placebo. With an assumed dropout rate of 17
`percent, rounding, a type I error rate of 2.5 per-
`cent, and a type II error rate of 90 percent, the
`number of patients needed was estimated to be
`1200. To power the study for the two-year end
`point of disability progression, we assumed a
`progression rate of 34.9 percent at the end of two
`years in the group assigned to interferon beta-1a
`alone and a progression rate of 22.7 percent at
`the end of two years (a 35 percent improvement)
`in the combination-therapy group. Simulations
`of the log-rank test were run with 60 percent of
`the accrual in the first 24 weeks and the remain-
`der in the next 24 weeks. With an assumed drop-
`out rate of 20 percent, the sample size of 1200
`provided at least 92 percent power with a Bonfer-
`roni adjustment for multiple end points and with
`the type I error rate maintained at 5 percent.
`
`n engl j med 354;9 www.nejm.org march 2, 2006
`
`913
`
`The New England Journal of Medicine
`Downloaded from nejm.org on October 22, 2023. For personal use only. No other uses without permission.
` Copyright © 2006 Massachusetts Medical Society. All rights reserved.
`
`

`

`T h e n e w e n g l a n d j o u r n a l o f m e d i c i n e
`
`The baseline characteristics of the patients
`were analyzed with the use of a t-test, with the
`exceptions of sex, race, and diagnosis of multi-
`ple sclerosis (based on the McDonald criteria18),
`which were analyzed with the use of a chi-
`square test. The time to the onset of disability
`progression sustained for 12 weeks was used to
`determine the cumulative probability of disabil-
`ity progression estimated by the Kaplan–Meier
`method. The Cox proportional-hazards model,
`adjusted for the baseline EDSS score, was used
`to compare the Kaplan–Meier curves. The an-
`nualized relapse rate was calculated by Poisson
`regression and adjusted for the number of re-
`lapses in the year before randomization; data
`pertaining to relapses that occurred after rescue
`treatment was initiated (per protocol) were cen-
`sored. Additional baseline factors were tested for
`inclusion in each of the models: EDSS score (≤3.5
`or >3.5), gadolinium-enhancing lesions (present
`or absent), the number of T2-hyperintense lesions
`(<9 or ≥9), and age (<40 or ≥40 years).22-24
`Each covariate was tested in the model for sta-
`tistical significance by a backward-selection pro-
`cedure, and only statistically significant covari-
`ates (P≤0.10) were included in the final models.
`No additional covariates were included in the
`analysis of disability progression. Three addi-
`tional covariates (baseline EDSS score, the pres-
`ence or absence of gadolinium-enhancing lesions
`at baseline, and age) were included in the analy-
`sis of relapse rate.
`A sensitivity analysis of disability progression
`(based on the change in EDSS score) sustained for
`24 weeks was also conducted. For the annualized
`rate of relapse, sensitivity analyses were performed
`with and without censoring, as well as with and
`without adjustment for significant covariates.
`The unadjusted rate of relapse was calculated as
`the total number of relapses divided by the total
`number of subject-years of follow-up in each
`treatment group. The Hochberg procedure25 for
`multiple comparisons was used in the analysis
`of the two primary end points; hence, the sig-
`nificance level was set such that if the higher of
`the P values for the analyses of these end points
`was less than or equal to 0.05, then both end
`points were considered to be statistically signifi-
`cant; otherwise, the lower of the P values was
`tested at a significance level of 0.025.
`Secondary efficacy end points were rank-
`
`ordered, and a closed testing procedure was used
`such that if statistical significance was not achieved
`for a given end point, then end points of a lower
`rank were considered not statistically significant.
`Secondary efficacy end points were analyzed by
`logistic regression with a term for treatment group
`and with their respective baseline values as covari-
`ates; missing values were imputed by using the
`mean in the study population.
`Adverse events were analyzed with use of the
`chi-square test, and serious adverse events were
`analyzed with use of Fisher’s exact test. Poisson
`regression was used to calculate the difference
`between the rates of infection in each treatment
`group.
`All analyses followed the intention-to-treat prin-
`ciple. All reported P values are two-tailed. The date
`on which the database was locked for the two-year
`analyses was May 31, 2005, and as a result there
`were 2528 patient-years of observation and 1222
`patient-years of exposure to natalizumab.
`
`R esults
`
`Patients
`SENTINEL was stopped approximately one month
`early, on February 28, 2005, because of two re-
`ports of progressive multifocal leukoencephalopa-
`thy (PML). Of the 1171 patients, a total of 1003
`(86 percent) completed the 120-week study; 168
`patients (14 percent overall; 12 percent of the
`group assigned to interferon beta-1a plus natali-
`zumab and 16 percent of the group assigned to
`interferon beta-1a alone) withdrew from the study
`(Fig. 1). Sixty-four patients discontinued the study
`drug but completed follow-up (5 percent overall;
`5 percent of the combination-therapy group and
`6 percent of the group assigned to interferon
`beta-1a alone). There were no significant differ-
`ences in demographic or disease-related charac-
`teristics at baseline between the two treatment
`groups, with the exception of the duration of dis-
`ease (median, seven years in the combination-ther-
`apy group and eight years in the group assigned
`to interferon beta-1a alone; P = 0.02) (Table 1).
`The SENTINEL data represent 28 percent of
`the placebo-controlled experience with natalizu-
`mab (in terms of patient-years of exposure) in
`both multiple sclerosis and Crohn’s disease and
`44 percent of the overall experience in multiple
`sclerosis.
`
`914
`
`n engl j med 354;9 www.nejm.org march 2, 2006
`
`The New England Journal of Medicine
`Downloaded from nejm.org on October 22, 2023. For personal use only. No other uses without permission.
` Copyright © 2006 Massachusetts Medical Society. All rights reserved.
`
`

`

`natalizumab and interferon beta-1a in multiple sclerosis
`
`Randomly assigned
`(N=1196)
`
`Excluded because of irregularities in data at one center
`(N=25)
`
`Interferon beta-1a plus natalizumab
`(N=589)
`
`Interferon beta-1a alone
`(N=582)
`
`Withdrew from study
`(N=73)
`Patient’s request
`(N=25 [4%])
`Adverse event
`(N=17 [3%])
`Noncompliance
`(N=6 [1%])
`Loss to follow-up
`(N=4 [<1%])
`Death
`(N=0 [0%])
`Other
`(N=21 [4%])
`
`Discontinued study drug
`but completed follow-up
`(N=29)
`
`Withdrew from study
`(N=95)
`Patient’s request
`(N=45 [8%])
`Adverse event
`(N=14 [2%])
`Noncompliance
`(N=5 [<1%])
`Loss to follow-up
`(N=5 [<1%])
`Death
`(N=2 [<1%])
`Other
`(N=24 [4%])
`
`Discontinued study drug
`but completed follow-up
`(N=35)
`
`Completed 2 yr of follow-up
`(N=516)
`
`Completed 2 yr of follow-up
`(N=487)
`
`Figure 1. Disposition of Patients during the Two-Year SENTINEL Study.
`
`Efficacy
`
`Kaplan–Meier estimates of the cumulative proba-
`bility of sustained disability progression at 2 years
`were 23 percent with combination therapy and 29
`percent with interferon beta-1a alone (Fig. 2 and
`Table 2). Combination therapy resulted in a 24 per-
`cent decrease in the risk of sustained disability
`progression (hazard ratio, 0.76; 95 percent confi-
`dence interval, 0.61 to 0.96; P = 0.02). In the sen-
`sitivity analysis of the risk of disability progres-
`sion sustained for 24 weeks, estimates of the
`cumulative probability of progression by 2 years
`were 15 percent for combination therapy and 18
`percent for interferon beta-1a alone (representing
`an 18 percent re duction with combination thera-
`py); however, this difference was not statistically
`significant (P = 0.17).
`Combination therapy reduced the annualized
`rate of relapse at one year, which was 0.82 with
`interferon beta-1a alone, to 0.38 (P<0.001) — a
`
`54 percent reduction (Table 2). This difference
`was maintained at two years, at which time the
`rate was 0.75 with interferon beta-1a alone and
`0.34 with combination therapy (a 55 percent re-
`duction with combination therapy, P<0.001). Sub-
`group analyses (according to relapse history,
`EDSS score, age, sex, the presence or absence of
`gadolinium-enhancing lesions, and the number
`of T2-hyperintense lesions) and a sensitivity analy-
`sis of relapse rate showed consistent results. The
`proportion of patients who were relapse-free at
`two years was 54 percent in the combination-
`therapy group, as compared with 32 percent in
`the group assigned to interferon beta-1a alone
`(P<0.001). The risk of relapse was 50 percent
`lower with combination therapy (hazard ratio,
`0.50; 95 percent confidence interval, 0.43 to 0.59;
`P<0.001).
`The number of new or enlarging T2-hyperin-
`tense lesions over the two-year period was re-
`
`n engl j med 354;9 www.nejm.org march 2, 2006
`
`915
`
`The New England Journal of Medicine
`Downloaded from nejm.org on October 22, 2023. For personal use only. No other uses without permission.
` Copyright © 2006 Massachusetts Medical Society. All rights reserved.
`
`

`

`T h e n e w e n g l a n d j o u r n a l o f m e d i c i n e
`
`Table 1. Baseline Characteristics of the Patients.*
`
`Characteristic
`
`Age — yr
`
`Mean ±SD
`
`Range
`
`Sex — no. of patients (%)
`
`Male
`
`Female
`
`Race — no. of patients (%)†
`
`White
`
`Other
`
`Duration of disease — yr
`
`Median
`
`Range
`
`No. of relapses in previous 1 yr — no. of patients (%)
`
`0
`
`1
`
`2
`
`Interferon Beta-1a
` plus Natalizumab
`(N = 589)
`
`Interferon
`Beta-1a Alone
`(N = 582)
`
`Total
`(N = 1171)
`
`38.8±7.7
`
`18–55
`
`147 (25)
`
`442 (75)
`
`550 (93)
`
`39 (7)
`
`7.0‡
`
`1–34
`
`0
`
`390 (66)
`
`153 (26)
`
`44 (7)
`
`39.1±7.6
`
`19–55
`
`38.9±7.7
`
`18–55
`
`162 (28)
`
`420 (72)
`
`542 (93)
`
`40 (7)
`
`8.0
`
`1–34
`
`1 (<1)
`
`357 (61)
`
`174 (30)
`
`50 (9)
`
`309 (26)
`
`862 (74)
`
`1092 (93)
`
`79 (7)
`
`7.0
`
`1–34
`
`1 (<1)
`
`747 (64)
`
`327 (28)
`
`94 (8)
`
`≥3
`
`Missing data
`
`No. of relapses in previous 1 yr
`
`Mean ±SD
`
`Range
`
`EDSS score — no. of patients (%)§
`
`0
`
`1.0–1.5
`
`2.0–2.5
`
`3.0–3.5
`
`4.0–4.5
`
`5.0
`
`≥5.5
`
`EDSS score
`
`Mean ±SD
`
`Range
`
`duced from 5.4 with interferon beta-1a alone to
`0.9 with combination therapy (P<0.001), repre-
`senting an 83 percent reduction with combina-
`tion therapy (Table 2). The mean number of
`gadolinium-enhancing lesions at two years was
`0.9 with interferon beta-1a alone and 0.1 with
`combination therapy, representing an 89 percent
`reduction (P<0.001).
`
`2 (<1)
`
`0
`
`2 (<1)
`
`1.44±0.75
`
`1–7
`
`1.49±0.72
`
`1.47±0.73
`
`0–5
`
`0–7
`
`24 (4)
`
`145 (25)
`
`214 (36)
`
`125 (21)
`
`68 (12)
`
`12 (2)
`
`1 (<1)
`
`2.4±1.1
`
`0–6.0
`
`Safety
`
`19 (3)
`
`143 (25)
`
`203 (35)
`
`126 (22)
`
`72 (12)
`
`16 (3)
`
`3 (<1)
`
`2.5±1.1
`
`0–5.5
`
`43 (4)
`
`288 (25)
`
`417 (36)
`
`251 (21)
`
`140 (12)
`
`28 (2)
`
`4 (<1)
`
`2.4±1.1
`
` 0–6.0
`
`At least one adverse event was reported by 584
`patients assigned to receive interferon beta-1a plus
`natalizumab (>99 percent) and 578 assigned to re-
`ceive interferon beta-1a alone (>99 percent). Ad-
`verse events significantly associated with combi-
`nation therapy were anxiety, pharyngitis, sinus
`congestion, and peripheral edema (Table 3). The
`
`916
`
`n engl j med 354;9 www.nejm.org march 2, 2006
`
`The New England Journal of Medicine
`Downloaded from nejm.org on October 22, 2023. For personal use only. No other uses without permission.
` Copyright © 2006 Massachusetts Medical Society. All rights reserved.
`
`

`

`natalizumab and interferon beta-1a in multiple sclerosis
`
`Table 1. (Continued.)
`
`Characteristic
`
`Interferon Beta-1a
`plus Natalizumab
`(N = 589)
`
`Interferon
`Beta-1a Alone
`(N = 582)
`
`Total
`(N = 1171)
`
`No. of gadolinium-enhancing lesions — no. of patients (%)
`0
`1
`2
`
`3
`≥4
`
`Missing data
`No. of gadolinium-enhancing lesions
`Mean ±SD
`
`Range
`No. of T2-hyperintense lesions — no. (%)
`<9
`≥9
`Missing data
`Duration of interferon beta-1a therapy before study — mo
`Mean ±SE
`Median
`
`Range
`
`392 (67)
`98 (17)
`31 (5)
`
`20 (3)
`43 (7)
`
`5 (<1)
`
`0.9±2.5
`
`0–24
`
`67 (11)
`519 (88)
`3 (<1)
`
`33.6±0.7
`29.0
`
`10–88
`
`374 (64)
`105 (18)
`32 (5)
`
`26 (4)
`42 (7)
`
`3 (<1)
`
`0.9±1.9
`
`0–16
`
`52 (9)
`528 (91)
`2 (<1)
`
`35.4±0.7
`32.0
`
`11–99
`
`766 (65)
`203 (17)
`63 (5)
`
`46 (4)
`85 (7)
`
`8 (<1)
`
`0.9±2.2
`
`0–24
`
`119 (10)
`1047 (89)
`5 (<1)
`
`34.5±0.5
`31.0
`
`10–99
`
`* Percentages may not total 100, because of rounding.
`† Race was determined at the time of enrollment by the treating investigator.
`‡ P≤0.02 for the comparison with the group assigned to interferon beta-1a alone.
`§ EDSS denotes the Expanded Disability Status Scale (possible scores range from 0 to 10, with higher scores indicating
`more severe disease).
`
`worst adverse events associated with combination
`therapy were mild in 10 percent of the patients,
`moderate in 54 percent, and severe in 35 percent;
`the respective percentages for interferon beta-1a
`alone were 5 percent, 57 percent, and 37 percent.
`Serious adverse events were observed in 18 per-
`cent of the patients assigned to combination ther-
`apy and 21 percent of those assigned to interferon
`beta-1a alone (P = 0.23). The most common seri-
`ous adverse event was a relapse of multiple scle-
`rosis, which occurred in 5 percent of the patients
`in the combination-therapy group and 9 percent of
`those in the interferon beta-1a group (P = 0.002).
`One of the serious adverse events reported was
`PML, which occurred in a patient who had received
`29 doses of natalizumab. A second patient received
`a diagnosis of PML after her completion of the
`two-year study and after she had received 37 doses
`of natalizumab. The details of these cases of PML
`have been reported previously.26,27 Two of the pa-
`tients assigned to interferon beta-1a alone died:
`
`one was a 47-year-old woman with a history of
`sinus arrhythmia and heart murmur, and the
`other was a 23-year-old woman with a history of
`headache, pain, and use of prescribed methadone
`who died during sleep.
`Depression was assessed every six months with
`use of the Beck Depression Inventory II.28
`There were no differences between the treatment
`groups in Beck Depression Inventory II scores
`during the study (data not shown).
`The incidence of infection was 83 percent in
`the combination-therapy group and 81 percent in
`the group assigned to interferon beta-1a alone;
`infections occurred at a rate of 1 per patient-year
`in each group. When the data pertaining to infec-
`tion were reanalyzed to include multiple occur-
`rences, the rate increased in each group, as ex-
`pected. However, there remained no significant
`difference between the groups, with infection
`rates of 1.54 per patient-year with combination
`therapy and 1.53 per patient-year with interferon
`
`n engl j med 354;9 www.nejm.org march 2, 2006
`
`917
`
`The New England Journal of Medicine
`Downloaded from nejm.org on October 22, 2023. For personal use only. No other uses without permission.
` Copyright © 2006 Massachusetts Medical Society. All rights reserved.
`
`

`

`T h e n e w e n g l a n d j o u r n a l o f m e d i c i n e
`
`P=0.02
`
`Interferon beta-1a
`alone (29%)
`
`Interferon beta-1a plus
`natalizumab (23%)
`
`0
`
`12
`
`24
`
`36
`
`48
`
`60
`Weeks
`
`72
`
`84
`
`96
`
`108
`
`120
`
`0.5
`
`0.4
`
`0.3
`
`0.2
`
`0.1
`
`0.0
`
`Sustained Progression
`
`Proportion of Patients with
`
`No. at Risk
`Interferon beta-1a alone
`Interferon beta-1a plus natalizumab
`
`582
`589
`
`550
`569
`
`517
`543
`
`493
`520
`
`461
`494
`
`441
`479
`
`415
`459
`
`396
`438
`
`367
`421
`
`347
`399
`
`343
`395
`
`Figure 2. Kaplan–Meier Plots of the Time to Sustained Disability Progression.
`The hazard ratio for sustained progression in the combination-therapy group as compared with the group given
`interferon beta-1a alone was 0.76 (95 percent confidence interval, 0.61 to 0.96).
`
`beta-1a alone (P = 0.95). Common infections were
`nasopharyngitis (39 percent vs. 35 percent); uri-
`nary tract infection, not otherwise specified (18
`percent vs. 19 percent); sinusitis, not otherwise
`specified (18 percent vs. 15 percent); upper respi-
`ratory tract infection, not otherwise specified (17
`percent vs. 18 percent); and influenza (17 percent
`vs. 15 percent). Serious infections occurred in 2.7
`percent and 2.9 percent of the patients assigned
`to combination therapy and interferon beta-1a
`alone, respectively. There were no cases of tuber-
`culosis. The incidence of cancer was 1 percent in
`the combination-therapy group and 2 percent in
`the group assigned to interferon beta-1a alone.
`Infusion reactions, defined as any event occur-
`ring within two hours after the start of an infu-
`sion, occurred in 24 percent of the patients in the
`combination-therapy group and 20 percent of
`those in the group assigned to interferon beta-1a
`alone (P = 0.11). The most common infusion reac-
`tion was headache. Most reactions were treated
`symptomatically and did not result in discontin-
`uation of the study drug. Hypersensitivity reac-
`tions included all events reported on the basis of
`clinical judgment as hypersensitivity, an allergic
`reaction, an anaphylactic or anaphylactoid reac-
`tion, urticaria, or hives by the investigator and
`were categorized according to severity. Eleven
`
`patients assigned to combination therapy (1.9
`percent) had a hypersensitivity reaction; 8 of the
`11 hypersensitivity reactions were isolated cases
`of urticaria (2 of which were severe). In addition,
`two patients assigned to interferon beta-1a alone
`(0.3 percent) had hypersensitivity reactions asso-
`ciated with the infusion of placebo; both were
`cases of mild urticaria. There was no cardiopul-
`monary compromise associated with any event.
`Natalizumab was discontinued, and the episodes
`resolved without sequelae.
`Eight percent of the patients in the combina-
`tion-therapy group and 7 percent of those in the
`group assigned to interferon beta-1a alone discon-
`tinued the study drug because of an adverse event.
`Three percent and 2 percent, respectively, with-
`drew from the study because of an adverse event.
`Natalizumab-treated patients had increases in
`lymphocytes, monocytes, eosinophils, and baso-
`phils — changes consistent with the drug’s known
`pharmacodynamic effects and the presence of
`α4 integrins on these cell types. Increases in nu-
`cleated red cells also were seen transiently in a
`small number of patients. These laboratory chang-
`es were not associated with any clinical manifes-
`tations and were reversible, with values returning
`to baseline within 16 weeks after the last dose.
`Elevations in neutrophils were not observed. No
`
`918
`
`n engl j med 354;9 www.nejm.org march 2, 2006
`
`The New England Journal of Medicine
`Downloaded from nejm.org on October 22, 2023. For personal use only. No other uses without permission.
` Copyright © 2006 Massachusetts Medical Society. All rights reserved.
`
`

`

`natalizumab and interferon beta-1a in multiple sclerosis
`
`0.001
`
`0.001
`
`0.001
`0.001
`0.001
`
`0, 43
`
`0
`
`0.9±3.2
`
`47 (8)
`33 (6)
`67 (12)
`435 (75)
`
`0, 64
`
`3
`
`5.4±8.7
`
`292 (50)
`59 (10)
`55 (9)
`176 (30)
`
`0, 12
`
`0
`
`0.1±0.6
`
`4 (<1)
`4 (<1)
`13 (2)
`568 (96)
`
`0, 27
`
`0
`
`0.9±2.1
`
`80 (14)
`39 (7)
`76 (13)
`394 (67)
`
`0.75
`0.70
`0.75 (0.67–0.84)
`
`0.33
`0.31
`0.34 (0.29–0.39)
`
`96 (16)
`105 (18)
`164 (28)
`217 (37)
`
`31 (5)
`41 (7)
`158 (27)
`359 (61)
`
`0.001
`
`0.75 (0.67–0.84)
`
`0.34 (0.29–0.39)
`
`0.02‡
`
` Value
`
`P
`
`29
`
`23
`
`Beta-1a Alone
`Interferon
`
`(N = 582)
`
`2 Yr
`
`plus Natalizumab
`Interferon Beta-1a
`
`(N = 589)
`
`<0.001
`
`<0.001
`
`<0.001
`<0.001
`
` Value
`
`P
`
`0, 43
`
`0
`
`0.8±2.5
`
`0, 28
`1.0
`2.4±4.1
`
`—
`
`—
`
`—
`
`—
`
`(N = 582)
`
`Interferon
`
`† Sustained disability progression was defined as an increase by at least 1.0 point in the Expanded Disability Status Scale (EDSS) score from a baseline sco