Introduction
Multiple sclerosis (MS) is the most common cause of chronic neurological disability in young adults [
1,
2]. Over the past 25 years, increased understanding of MS and availability of disease-modifying therapies (DMTs) has improved options to manage the disease [
1]. Typically MS lasts several decades [
2‐
4]. Prolonged follow-up of well characterized patients provides critical information on the implications of early treatment intervention.
The Betaferon®/Betaseron® in Newly Emerging MS for Initial Treatment (BENEFIT) trial investigated the effects of interferon beta-1b (IFNB-1b) treatment at or shortly after clinically isolated syndrome (CIS)—the first neurological episode suggestive of MS—on clinical and MRI outcomes [
5‐
8]. The BENEFIT trial has been recognized as the most comprehensive follow-up study of patients who received early intervention with IFNB-1b, and has contributed to the broad consensus to offer patients early treatment to optimize disease management [
1,
9]. Early study results demonstrated that initiating interferon beta-1b at CIS, compared with those who started treatment after a first relapse or after two years was associated with improved clinical and MRI outcomes beyond the 2-year double-blind randomized trial and through the 5-year rater-blinded study period.
Additional follow-up with a prospective, cross-sectional assessment at 11 years post baseline showed that early treatment with IFNB-1b remained beneficial up to 11 years after randomization. Patients who started IFNB-1b at CIS retained a lower risk for conversion to clinically definite multiple sclerosis (CDMS) and a lower annualized relapse rate (ARR) compared with those who started IFNB-1b after a short delay (mean 1.5 years) [
5]. Disability remained low and stable in both treatment groups at the 11-year assessment.
This study aims to further extend the long-term information on this cohort and describe the course of this well-characterized inception cohort by an additional prospectively planned assessment 15 years after randomization. Data is reported from patient visits at Year 15 and from integrated analyses of the full BENEFIT cohort.
Methods
Patients with CIS and ≥ 2 brain MRI lesions suggestive of MS were randomly assigned in a 5:3 ratio to receive IFNB-1b (“early treatment”) or placebo. [
8] After conversion to CDMS (defined as a second clinical attack or confirmed disease worsening), patients were offered open-label treatment with IFNB-1b without disclosing the initial randomization. At 2 years, all patients (including those who had not converted to CDMS) were offered to continue taking interferon beta-1b. The “delayed treatment” group included patients who were initially randomized to placebo. IFNB-1b was administered at a dose of 250 µg subcutaneously, every other day.
Prospective assessments (blinded as for the initial randomization) continued for 5 years following randomization. At 15 years, all initially randomized patients were asked to participate in a comprehensive clinical and MRI reassessment. Patients could be evaluated at their original center or, if their original study center did not participate in BENEFIT 15, at a different, participating center. Clinical outcomes included assessment of conversion to CDMS, progression to secondary progressive multiple sclerosis (SPMS), ARR, Expanded Disability Status Scale (EDSS), Paced Auditory Serial Addition Task – 3 s (PASAT-3), and employment status. CDMS was reached if a new neurological event (relapse) occurred, ie, the appearance of new neurological abnormality or reappearance of a neurological abnormality, separated by at least 30 days from onset of a preceding clinical demyelinating event, or by sustained worsening of > = 1.5 points on the EDSS and a total EDSS of > = 2.5. HRQoL was assessed by EuroQol 5-Dimension (EQ-5D), including the Visual Analog Scale (VAS) and Functional Assessment of Multiple Sclerosis (FAMS – total score and Trial Outcome Index (TOI)). Standardized questions on employment situation and resource utilization were asked. If a visit at the study center was not possible, patients were offered a structured telephone evaluation that also included a validated patient reported EDSS assessment tool for which an overall high correlation with EDSS as determined by physical examination had been demonstrated (Pearson’s correlation coefficient -0.95), across all functional scores and degrees of disability.[
10,
11].
Investigators collected MRI data at study sites according to the BENEFIT MRI standardized protocol. Scans were analyzed at a central reading site (Institutes of Neurology and Healthcare Engineering, UCL, London, UK). Trained readers manually identified and quantified lesions using a local-intensity thresholding technique. Neuroradiological assessments included number of new T2 lesions, T2 lesion volume, T1 lesion volume, normalized brain volume, normalized thalamic volume, cortical thickness, and mean upper cervical cord area [
5]
.
Analyses were performed for the BENEFIT 15 cohort. Integrated analyses were carried out using the full BENEFIT cohort, allowing for time to event calculations. Outcomes included time to conversion to CDMS, time to first relapse, time to recurrent relapse, number of patients with confirmed and sustained 1-point EDSS progression, and number of patients with confirmed 2.5-point EDSS progression. The number of patients with diagnosis of MS according to the McDonald 2001 and 2010 criteria at fifteen years after occurrence of CIS was determined in patients who had an MRI performed at Year 15. Patients who developed CDMS or McDonald MS by Year 11 of follow-up were also included in this analysis.
Safety was also assessed at Year 15. An adverse event (AE) was defined as any untoward medical occurrence (ie, any unfavorable and unintended sign, symptom, or disease temporally associated with the use of a medicinal product, whether or not considered related to the medicinal product, also covering laboratory findings or results of other diagnostic procedures considered clinically relevant). Conditions were documented that started or deteriorated after signing of informed consent.
Statistical procedures
Statistical modeling was used to estimate treatment effects, and to explore the relationship between target variables and treatment. The study was exploratory in nature, with conversion to CDMS and/or SPMS, relapse rate, EDSS, PASAT-3, resource utilization, and employment/retirement status defined as variables of primary interest. All variables were analyzed descriptively with appropriate statistical methods. Particular outcomes (time to CDMS, time to first relapse, time to use of an ambulatory device, and time to dependence on an ambulatory device) were evaluated using Kaplan–Meier (KM) methods, log-rank tests and proportional hazards regression for time-to-event outcomes and a generalized linear regression model for ARR, with steroid use during the first event (yes or no); multifocal or monofocal onset of disease, number of T2 lesions at screening (2 to 4, 5 to 8, or ≥ 9); number of gadolinium-enhancing (Gd +) lesions at screening; age; and sex included as the set of covariates in the proportional hazards regression model and the generalized linear regression model. For PASAT-3 distribution over time a parametric longitudinal linear mixed model (with baseline PASAT-3 as the covariate) was established. The statistical evaluation was performed using software package SAS® (Statistical Analysis System) release 9.2 (SAS Institute Inc., Cary, North Carolina, United States).
Standard protocol approvals, registrations, and patient consents
The institutional review boards of participating institutions approved the protocol for the study. Patients provided informed consent at enrollment of the trial. The BENEFIT 15 trial is listed on clinicaltrials.gov under NCT03269175.
Discussion
BENEFIT 15 is a unique study cohort with 15-year follow-up data from randomization that compared early vs delayed treatment with IFNB-1b in patients presenting with CIS. Considering the time elapsed since the initial randomization, the number of patients enrolled in BENEFIT 15 is a respectable proportion—55.8%—of randomized patients, thus providing a unique opportunity to determine long-term outcomes of early treatment with IFNB-1b. The possibility that patients with a more severe disease course dropped out before this 15 year follow-up study cannot entirely be ruled out although such selective drop out may also have occurred with asymptomatic participants. However, the 15-year follow-up population did not appear to have a selection bias, as baseline characteristics were very similar to those of the original BENEFIT population, with a T2 lesion load at baseline that reflected substantial disease burden. Patients from sites that participated in this 15-year assessment were invited irrespective of their current medication, and great effort was made to reach more severely disabled patients with reduced mobility. In those cases where a clinic visit was not feasible, remote standardized telephone assessments were conducted. It is therefore remarkable that the conversion rates to CDMS after 15 years from initial randomization still favour the early treatment group (67.6% vs 73.5% with delayed treatment, by KM estimates). The overall relatively low EDSS and high PASAT-3 scores suggest limited disease progression by Year 15. EDSS scores compared favorably with natural history populations with shorter duration of follow-up [
12,
13]. This, along with the low SPMS conversion rate after 15 years of follow up in both the early treatment and delayed but still early treatment groups, supports the overall beneficial effect of early initiation of DMTs. With an average of 1.5 years until the delayed treatment groups received treatment, both groups initiated treatment relatively early.
Patients in this study generally reported good HRQoL across both measures. Population surveys have demonstrated a relationship between age and HRQoL measures, and the mean EQ-5D VAS in the BENEFIT 15 cohort was comparable to that of EQ-5D VAS norm in an international general population for its respective age group [
14].
Employment findings of the patients enrolled in the BENEFIT study at Year 15 reflect patients’ ability to accomplish daily professional tasks. These tasks often can be limited by MS symptoms and the disabilities inherent to the disease. The Global MS Employment Report demonstrated that 43% of unemployed patients with MS discontinued work within 3 years of diagnosis, and 70% after 10 years of diagnosis [
15]. In fact, results from a prospective observational study showed that, in patients with MS, accumulation of disability and increase in relapse at 5 years following first episode of central nervous system demyelination were each associated with a decline in employment trajectory [
16]. In the context of BENEFIT 15, the majority of patients (66.3%) were employed at Year 15 and had not taken days off from work in the past 12 months because of MS, regardless of whether they were in the early treatment or the delayed treatment arm of the initial BENEFIT cohort. This result supports the value of IFNB-1b treatment initiation at an early stage, especially when considering that the burden of MS can be substantial, throughout adult life.[
17] In 2017, a survey of 4590 working-age patients with MS in Germany (mean (SD) age: 51.8 (11.0) years; mean (SD) age at diagnosis: 36.3 (10.6) years) revealed that 51% were employed or self-employed [
18]. Despite their patient population having a shorter average disease duration, the findings from our study compare favorably with these results, with 66.3% of patients in our study being employed at Year 15.
Another survey of 1727 patients with MS in New Zealand, with a similar age and disease duration as our patient population, found that more than half (54%) were not working [
19], while the study population in BENEFIT 15 including patients who either attended study centers or were evaluated via telephone interview had a non-employment rate of only 32.2%. This although great effort was taken to include all types of patients with varying disability, as patients with higher disability may be prone to non-employment. HRQoL data from the BENEFIT 15 study also compared favorably with population-based values and with data from other MS cohorts [
20‐
22].
No new safety signals were detected with IFNB-1b long-term treatment in the current cross-sectional study. Clinical outcomes from the 15-year BENEFIT trial follow-up further support initiating DMT at or shortly after CIS. In the follow up of a placebo controlled CIS study comparing early interferon beta intramuscular treatment 10 years after randomization with delayed treatment Kinkel et al. reported that 18.4% had an EDSS ≥ 3.0.[
23] In a study by Chung et al. conducted in a British cohort of untreated patients 30 years after CIS only 42% had remained ambulatory (EDSS scores of ≤ 3.5) and 34% developed SPMS [
22].
With a follow-up of 15 years, the BENEFIT-15 cohort offers the longest duration of follow-up in a randomized population treated at or shortly after CIS, thus suggesting that many patients who start treatment with IFNB-1b at an early disease stage can retain clinical and social wellbeing in the future.
Acknowledgements
The authors thank the patients and the BENEFIT 15 investigators for their continuing contributions to the study. VMLY&R provided writing support that included medical editing, English-language assistance, formatting of text and figures, and coordinating reviews, which was funded by Bayer AG. BENEFIT and BENEFIT 15 Study group. Principal investigators: Austria – S Strasser-Fuchs (Graz), T Berger (Innsbruck), K Vass (Vienna)*; Belgium – C Sindic, V van Pesch (Brussels), B Dubois (Leuven), D Dive, V Delvaux (Liège), J Debruyne, G Laureys (Ghent); Canada – L Metz (Calgary), G Rice (London), P Duquette, Y Lapierre (Montreal), M Freedman (Ottawa), A Traboulsee (Vancouver), P O’Connor (Toronto)*; Czech Republic – P Štourač (Brno), R Taláb, M Valis (Hradec Kralove); O Zapletalová (Ostrava), I Kovářová, E Medová (Prague), J Fiedler (Plzen); Denmark – J Frederiksen (Glostrup); France – B Brochet (Bordeaux), T Moreau (Dijon), P Vermersch (Lille), J Pelletier (Marseille), G Edan (Rennes), M Clanet (Toulouse), P Clavelou (Clermont Ferrand), C Lebrun-Frenay (Nice), O Gout (Paris); Finland M Kallela, J Erälinna (Helsinki), T Pirttilä, (Kuopio), J Ruutiainen (Turku), K Koivisto (Seinäjoki), M Reunanen, AM Keskinarkaus (Oulu), I Elovaara, M Sumelahti (Tampere); Germany – A Villringer, H Altenkirch, S Brockhaus, F Paul (Berlin), K Wessel (Braunschweig), H-P Hartung, W Steinke (Düsseldorf), H Kölmel (Erfurt), P Oschmann, M Berghoff (Giessen), R Diem, M Weber (Göttingen), A Dressel (Greifswald), F Hoffmann (Halle/Saale), K Baum (Hennigsdorf), S Jung (Homburg/Saar), H Felicitas Petereit, D Reske (Cologne), M Sailer (Magdeburg), J Köhler, F Zipp (Mainz), N Sommer, B Tackenberg (Marburg), R Hohlfeld, T Kuempfel (Munich), K-H Henn (Offenbach), A Steinbrecher (Regensburg), H Tumani (Ulm), R Gold, P Rieckmann, (Würzburg); Hungary – R Komoly, G Gács, G Jakab, P Salacz (Budapest), L Csiba, T Csepany (Debrecen), L Vécsei (Szeged); Israel – A Miller (Haifa), D Karussis, P Petrou (Jerusalem), J Chapman (Tel-Hashomer); Italy – A Ghezzi, M Zaffaroni (Gallarate), G Comi, V Martinelli (Milan), P Gallo (Padua), V Cosi, R Bergamaschi (Pavia), L Durelli, M Clerico (Turin); Netherlands – B Anten (Sittard), L Visser (Tilburg); Norway – K-M Myhr (Bergen); Poland – A Szczudlik (Kraków), K Selmaj, M Stasiolek (Łódź), Z Stelmasiak, H Bartosik-Psujek (Lublin), R Podemski, M Ejma (Wrocław), Z Maciejek, S Wawrzyniak (Bydgoszcz); Portugal – L Cunha, L Sousa (Coimbra); Slovenia – S Sega-Jazbec (Ljubljana); Spain – X Montalbán, T Arbizu, A Saiz, S Martinez Yelamos (Barcelona), J Bárcena (Barakaldo), R Arroyo, J Matias-Guiu (Madrid), O Fernández, P Serrano (Málaga), G Izquierdo (Seville), B Casanova (Valencia); Sweden – J Lycke, (Mölndal); Switzerland – L Kappos, J. Kuhle (Basel), A Chan (Bern), K Beer, (St Gallen); UK – R Coleman (Aberdeen), J Chataway (London), J O’Riordan (Dundee), S Howell (Sheffield). *This contributor has passed away.
Declarations
Ethical statement
L Kappos received no personal compensation. His institutions (University Hospital Basel/Foundation Clinical Neuroimmunology and Neuroscience Basel) have received the following exclusively for research support: steering committee, advisory board and consultancy fees (AbbVie, Actelion, Auriga Vision AG, Bayer HealthCare, Biogen, Celgene, Dörries Frank-Molnia & Pohlman, Eli Lilly, EMD Serono, Genentech, Genzyme, Glaxo Smith Kline, Janssen, Merck, Minoryx, Novartis, Roche, Sanofi, Santhera, Senda Biosciences, Shionogi and Wellmera AG); speaker fees (Bristol Myers Squibb, Celgene, Janssen, Merck, Novartis, and Roche); support for educational activities (Biogen, Desitin, Novartis, Sanofi and Teva); license fees for Neurostatus products; and grants (European Union, Innosuisse, Novartis, Roche, Swiss MS Society and Swiss National Research Foundation). F Barkhof has received compensation for consultancy from Bayer, Biogen Idec, Merck, Novartis, Roche, Combinostics, and IXICO, and has received research support from the Dutch MS Research Foundation, NIHR biomedical research center at UCLH and the EU (FP7 and IMI). MS Freedman has received compensation from Alexion, Atara Biotherapeutics, Bayer HealthCare, Beigene, BMS/Celgene, EMD Inc./EMD Serono and Merck Serono, Hoffman La-Roche, Janssen (J&J), Novartis, Pendopharm, Sanofi-Genzyme for consulting services. He also participates in a Sanofi-Genzyme and EMD Serono-sponsored speakers bureaus. G Edan has received honoraria for consulting from Biogen, Merck, Novartis, Sanofi, Roche, and LFB, and has received personal compensation for serving on the BENEFIT scientific advisory board and for speaking from Bayer AG. His institution has also received research support from Novartis, Sanofi, Merck, Biogen, Roche, and Teva. H-P Hartung has received honoraria for consulting and speaking at symposia from Bayer AG, Biogen Idec, BMS Celgene, GeNeuro, Merck, Novartis, Roche, Sanofi, and TG Therapeutics, with approval by the rector of Heinrich-Heine University. X Montalbán has received speaking honoraria and travel expenses for participation in scientific meetings and has been a steering committee member of clinical trials or participated in advisory boards of clinical trials in the past years with Actelion, Bayer, Biogen, Celgene, Genzyme, Merck, Novartis, Roche, Sanofi-Genzyme, Teva Pharmaceutical, Excemed, MSIF and NMSS. R Koelbach is an employee of Ingress-Health HWM GmbH (a Cytel company) and was an employee of Parexel International GmbH at the time of this study. DG MacManus has nothing to disclose. EM Wicklein is a salaried employee of Bayer AG.