UBC Faculty Research and Publications

Predictive value of early magnetic resonance imaging measures is differentially affected by the dose… Traboulsee, Anthony; Li, David K B; Cascione, Mark; Fang, Juanzhi; Dangond, Fernando; Miller, Aaron May 11, 2018

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-12883_2018_Article_1066.pdf [ 841.24kB ]
Metadata
JSON: 52383-1.0366282.json
JSON-LD: 52383-1.0366282-ld.json
RDF/XML (Pretty): 52383-1.0366282-rdf.xml
RDF/JSON: 52383-1.0366282-rdf.json
Turtle: 52383-1.0366282-turtle.txt
N-Triples: 52383-1.0366282-rdf-ntriples.txt
Original Record: 52383-1.0366282-source.json
Full Text
52383-1.0366282-fulltext.txt
Citation
52383-1.0366282.ris

Full Text

RESEARCH ARTICLE Open AccessPredictive value of early magneticresonance imaging measures isdifferentially affected by the dose ofinterferon beta-1a given subcutaneouslythree times a week: an exploratory analysisof the PRISMS studyAnthony Traboulsee1*, David K. B. Li1, Mark Cascione2, Juanzhi Fang3, Fernando Dangond4 and Aaron Miller5AbstractBackground: On-treatment magnetic resonance imaging lesions may predict long-term clinical outcomes inpatients receiving interferon β-1a. This study aimed to assess the effect of active T2 and T1 gadolinium-enhancing(Gd+) lesions on relapses and 3-month confirmed Expanded Disability Status Scale (EDSS) progression in thePRISMS clinical trial.Methods: Exploratory analyses assessed whether active T2 and T1 Gd + lesions at Month 6, or active T2 lesions atMonth 12, predicted clinical outcomes over 4 years in PRISMS.Results: Mean active T2 lesion number at Month 6 was significantly lower with interferon beta-1a givensubcutaneously (IFN β-1a SC) 44 μg and 22 μg 3×/week (tiw) than with placebo (p < 0.0001). The presence of ≥4versus 0 active T2 lesions predicted disability progression at Years 3–4 in the IFN β-1a SC 22 μg group only (p < 0.05), whereas the presence of ≥2 versus 0–1 active T2 lesions predicted disability progression in the placebo/delayed treatment (DTx) (Years 2–4; p < 0.05) and IFN β-1a SC 22 μg groups (Years 3–4; p < 0.05). Greater active T2lesion number at 6 months predicted relapses in the placebo/DTx group only (≥4 vs. 0, Years 1–4; ≥2 vs. 0–1, Years2–4; p < 0.05), and the presence of T1 Gd + lesions at 6 months predicted disability progression in the IFN β-1a SC44 μg group only (Year 1; p < 0.05). The presence of ≥2 versus 0–1 active T2 lesions at 12 months predicteddisability progression over 3 and 4 years in the IFN β-1a SC 44 μg group.Conclusion: Active T2 lesions at 6 months predicted clinical outcomes in patients receiving placebo or IFN β-1a SC22 μg, but not in those receiving IFN β-1a SC 44 μg. Active T2 lesions at 12 months may predict outcomes in thosereceiving IFN β-1a SC 44 μg and are possibly more suggestive of poor response to therapy than T2 results at 6 months.Keywords: MRI, Multiple sclerosis, T2 lesions, Gadolinium-enhancing lesions, Treatment response, Beta-interferon* Correspondence: t.traboulsee@ubc.ca1University of British Columbia, S113-2211 Wesbrook Mall, Vancouver, BC V6T1Z7, CanadaFull list of author information is available at the end of the article© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Traboulsee et al. BMC Neurology  (2018) 18:68 https://doi.org/10.1186/s12883-018-1066-8BackgroundConsiderable evidence indicates that early treatment ofrelapsing forms of multiple sclerosis (RMS) is critical inorder to delay disease progression and accumulation ofirreversible disability. Therefore, as the repertoire ofdisease-modifying drugs (DMDs) for RMS has grown, itis increasingly important for physicians who treat pa-tients with RMS to know when to consider treatmentchanges in order to derive maximum benefit from theavailable treatment options within the available windowof opportunity. There is currently no single evidence-based algorithm to guide these decisions due to the in-herent heterogeneity of RMS. In its absence, clinical andsubclinical indicators of continued disease activity arebeing evaluated for their ability to define breakthroughdisease and potential to predict long-term outcomes.Randomized clinical trials assessing DMDs for RMSrely on clinical disease endpoints: the occurrence of re-lapses and confirmed disability progression [1–3]. How-ever, modern MS clinical trial populations show lowerlevels of clinical disease activity compared with previ-ously reported cohorts [3–6]. The reduced level of dis-ease activity has led to a greater focus on alternativeoutcomes capable of acting as sensitive surrogates for re-lapses or disability progression [7–10].MRI lesions, including T2-weighted hyperintense orT1 gadolinium-enhancing (Gd+) lesions, indicate focalinflammatory activity in RMS and may be present in pa-tients who are clinically stable. These lesions have beenproposed as a surrogate marker capable of predictinglong-term treatment outcomes in patients with RMS. Al-though the pathophysiological relationship between MRIlesions and disease progression in MS is complex, MRIis an important component of MS diagnosis, and MRIlesion loads may allow treatment success to be predictedat early time points [9, 11]. Multiple studies of patientstreated with interferon (IFN) β-1a therapies have sug-gested an association between the presence of early on-treatment T2 or T1 Gd +MRI lesions and an increasedrisk of relapse or disease progression at subsequent timepoints [7, 9, 10, 12, 13]. MRI results at early time pointsmay, therefore, prove to be a useful tool for assessingthe overall efficacy of DMDs, identifying breakthroughdisease, and predicting long-term responses of individualpatients to therapy.PRISMS-2 was a 2-year randomized clinical trial de-signed to evaluate the use of IFN β-1a injected subcuta-neously (SC) three times weekly (tiw) in patients withrelapsing–remitting MS (RRMS) [3]. Both the 44- and22-μg doses of IFN β-1a SC tiw demonstrated significantimprovements on clinical and MRI measures of disease,compared with placebo [3]. A long-term follow-up study(PRISMS-4), in which patients treated with placebo inthe original PRISMS study were reassigned to either IFNβ-1a 44 or 22 μg SC tiw, assessed the efficacy of IFN β-1a SC tiw over a 4-year period [14]. Therapy with IFNβ-1a SC tiw maintained a treatment benefit after 4 yearsof treatment, and outcomes were superior in patientswho were treated with IFN β-1a SC tiw for all 4 years ofthe study, compared with those who began IFN β-1a SCtiw treatment after 2 years [14].This study investigated whether early T2 and T1 Gd +MRI lesions predicted subsequent clinical outcomes inpatients treated with IFN β-1a 44 μg SC tiw, IFN β-1a22 μg SC tiw, or placebo/delayed treatment over a 4-year period in the PRISMS clinical trial.MethodsStudy design and treatmentThe details of the PRISMS-2 study design have beenpublished previously [3]. Briefly, patients with RRMS, anExpanded Disability Status Scale (EDSS) score of ≤5, andno prior treatment with IFN β were randomized 1:1:1 toIFN β-1a 44 μg SC tiw, IFN β-1a 22 μg SC tiw, or pla-cebo. All patients had proton density (PD)/T2-weightedMRI scans twice a year. A subgroup also had monthlyT2 and T1 Gd + scans before and during the first9 months of treatment (frequent-MRI cohort).Exploratory analysesExploratory analyses assessed the number of T2 lesionsper patient at Month 6 and Month 12. The number ofactive T2 lesions was calculated as the sum of new,newly enlarging, and recurring T2 lesions on the 6-month MRI scan with reference to the baseline MRIscan. Further exploratory analyses investigated the pre-dictive value of active T2 and T1 Gd + lesions at6 months on EDSS progression (increase of ≥0.5 pointsif baseline EDSS was ≥6.0 or increase of ≥1 point if base-line EDSS was < 6.0, confirmed 3 months later) and re-lapses over Years 1 and 2 (PRISMS-2) and Years 3 and 4(PRISMS-4). The effect of active T2 lesions at 6 monthson time to EDSS progression was calculated over a 4-year period. Additional exploratory analyses evaluatedthe predictive effect of active T2 lesions at 12 months onEDSS progression and relapses over Years 1, 2, 3, and 4,as well as time to EDSS progression over 4 years.Statistical analysesBetween-treatment comparisons of mean active T2 le-sion number per patient used a negative binomial re-gression model with baseline burden of disease andtreatment as independent factors. The effect of T2 le-sions (≥4 vs. 0 and ≥ 2 vs. 0–1) on relapse and EDSSprogression was assessed in the entire cohort using a lo-gistic regression model. The logistic regression modelsexamining the predictive value of T1 Gd + lesions ana-lyzed data from the frequent-MRI cohort. Hazard ratiosTraboulsee et al. BMC Neurology  (2018) 18:68 Page 2 of 13(HRs) and p-values for between-group differences intime to first EDSS progression were calculated using aCox-proportional hazards model.ResultsIn total, 560 patients were randomized: 187 to the pla-cebo group, 184 to the IFN β-1a 44 μg SC tiw group,and 189 to the IFN β-1a 22 μg SC tiw group. Baselinecharacteristics were similar among all three treatmentgroups, as has been previously reported (Table 1) [3].The proportions of patients who experienced relapsesand sustained disability progression by Years 2 and 4 areshown in Table 2.Early T2 MRI scan results in PRISMS-2In total, among subjects with data, 147 of 186 patients(79.0%) in the placebo/delayed treatment group had oneor more active T2 lesions at 6 months, compared with100 of 185 (54.1%) and 87 of 182 (47.8%) patients in theIFN β-1a 22 and 44 μg SC tiw groups, respectively. Themean (standard deviation [SD]) number of active T2 le-sions per patient at 6 months was 4.1 (5.72) in the pla-cebo group, compared with 2.1 (4.28) in the IFN β-1a22 μg SC tiw group and 1.3 (2.34) in the IFN β-1a 44 μgSC tiw group (p < 0.0001 for each IFN β-1a group vs.placebo).Relationship between active T2 lesions at 6 months andEDSS progressionNo statistically significant differences in the proportionsof patients with EDSS progression at Years 1, 2, 3, and 4were seen between patients with ≥4 versus 0 active T2lesions at Month 6 in the IFN β-1a 44 μg SC tiw or pla-cebo/delayed treatment groups (Fig. 1a and c). However,the presence of ≥4 versus 0 active T2 lesions at Month 6predicted EDSS progression at Years 3 and 4 in the IFNβ-1a 22 μg SC tiw group (Fig. 1b). The presence of ≥2versus 0–1 active T2 lesions at 6 months predicted EDSSprogression in the placebo/delayed treatment (Years 2,3, and 4) and IFN β-1a 22 μg SC tiw (Years 3 and 4)groups, but not in the IFN β-1a 44 μg SC tiw group(Figure S1a–c shows this in more detail [see Additionalfile 1]). Notably, EDSS progression in patients treatedwith IFN β-1a 44 μg SC tiw who had ≥4 or 0 active T2lesions at Month 6 occurred at a rate similar to that inpatients treated with placebo who had no active T2 le-sions at Month 6 (Figure S2 shows this in more detail[see Additional file 2]). The predictive performance ofactive T2 lesions on EDSS progression is shown in TableS1(a) (see Additional file 3). Positive and negative pre-dictive values for EDSS progression according to activeT2 lesions at Month 6 (≥4 vs. 0) were consistently higherin the placebo/delayed treatment group than in the IFNβ-1a 44 μg SC tiw group.Time to first EDSS progression over 4 years was sig-nificantly reduced for patients with ≥4 versus 0 active T2lesions at Month 6 in the IFN β-1a 22 μg SC tiw group,but not in the IFN β-1a 44 μg SC tiw or the placebo/de-layed treatment group (Fig. 2a–c). For patients with ≥2versus 0–1 active T2 lesions, significant reductions intime to EDSS progression were seen in the placebo/de-layed treatment group (p = 0.0132), but not in eitherIFN β-1a SC tiw group (Figure S3 shows this in moredetail [see Additional file 4]).Relationship between active T2 lesions at 6 months andrelapsesThe presence of ≥4 versus 0 active T2 lesions at Month6 predicted relapses in the placebo/delayed treatmentgroup (Years 1–4; Fig. 3a), but not in either IFN β-1a SCtiw group (Fig. 3b and c). Similar results were seen whenpatients with ≥2 versus 0–1 active T2 lesions at Month 6were compared (Figure S4 shows this in more detail [seeAdditional file 5]). The predictive performance of activeT2 lesions at 6 months on relapse at Year 2 and Year 4is shown in Table S1(b) (see Additional file 3).Table 1 Demographic characteristics at baselinePlacebo IFN β-1a 22 μg SC tiw IFN β-1a 44 μg SC tiwNumber of activeT2 lesions at 6 monthsAll (n = 187) Number of active T2lesions at 6 monthsAll (n = 189) Number of active T2 lesions at 6 monthsAll (n = 184)0 (n = 39) ≥4 (n = 64) 0 (n = 85) ≥4 (n = 28) 0 (n = 95) ≥4 (n = 24)Age, years, mean (SD) 36.4 (7.3) 33.4 (7.8) 34.7 (7.5) 35.6 (6.5) 30.2 (7.3) 34.8 (7.0) 35.5 (8.0) 33.0 (6.7) 35.2 (7.9)Sex, female, n (%) 31 (79) 47 (73) 141 (75) 53 (62) 21 (75) 126 (67) 63 (66) 14 (58) 122 (66)Race, white, n (%) 38 (97) 63 (98) 184 (98) 84 (99) 28 (100) 188 (99) 94 (99) 24 (100) 182 (99)Time since MS onset,years, mean (SD)6.6 (5.4) 5.3 (4.3) 6.1 (4.8) 7.9 (6.45) 5.4 (4.3) 7.7 (6.1) 7.1 (5.6) 7.4 (6.9) 7.8 (6.3)Number of relapses inpast 2 years, mean (SD)2.7 (0.8) 3.3 (1.4) 3.0 (1.3) 2.9 (1.0) 3.2 (1.2) 3.0 (1.1) 2.9 (1.0) 3.5 (1.6) 3.0 (1.1)EDSS score, median 2.0 2.5 2.5 2.5 2.0 2.5 2.5 3.0 2.5EDSS Expanded Disability Status Scale, IFN β-1a interferon beta-1a, SC subcutaneously, SD standard deviation, tiw three times weeklyTraboulsee et al. BMC Neurology  (2018) 18:68 Page 3 of 13Table 2 Proportion of patients with relapses or confirmed disability progression, Year 2 (PRISMS-2) and 4 (PRISMS-4)Placebo/delayed treatment (n = 187) IFN β-1a 22 μg SC tiw (n = 189) IFN β-1a 44 μg SC tiw (n = 184)Relapse by Year 2 (%) 84.5 73.0 67.9Relapse by Year 4 (%) 90.4 82.0 78.8Confirmed EDSS progression by Year 2 (%) 39.0 31.2 27.7Confirmed EDSS progression by Year 4 (%) 48.1 45.0 39.7EDSS progression was defined as an increase of ≥0.5 points if baseline EDSS was ≥6 or increase of ≥1 point if baseline EDSS was < 6, confirmed 3 months laterEDSS Expanded Disability Status Scale, IFN β-1a interferon beta-1a, SC subcutaneously, tiw three times weekly***4YearProportion of patients withconfirmed EDSS progression (%)020406080bProportion of patients withconfirmed EDSS progression (%)020406080cYear020Proportion of patients withconfirmed EDSS progression (%)4060801 2 31 2 3 41 2 3 4YearaIFN -1a 22 µg SC tiw ≥4 T2 (n=28)IFN -1a 22 µg SC tiw 0 T2 (n=85)IFN -1a 44 µg SC tiw ≥4 T2 (n=24)IFN -1a 44 µg SC tiw 0 T2 (n=95)Placebo/delayed treatment ≥4 T2 (n=64)Placebo/delayed treatment 0 T2 (n=39)Fig. 1 Proportion progressed at each year by ≥4 versus 0 active T2 lesions at 6 months. a Placebo/delayed treatment, ≥4 versus 0 T2 lesions at6 months; b IFN β-1a 22 μg SC tiw, ≥4 versus 0 T2 lesions at 6 months; c IFN β-1a 44 μg SC tiw, ≥4 versus 0 T2 lesions at 6 months. p valuesindicate differences between patients with differing lesion loads at 6 months within the treatment group. No statistically significant differenceswere seen in the placebo/delayed treatment or IFN β-1a 44 μg SC tiw group. Values were calculated with a logistic regression model withpredictor (≥4 vs. 0 T2 lesions) as a fixed effect; number of relapses within the previous 2 years, age, baseline EDSS score, and baseline burden ofdisease were independent variables, and p values were calculated for the predictive effect of T2 lesion subgroups. *p < 0.05; **p < 0.01. EDSS:Expanded Disability Status Scale; IFN β-1a: interferon beta-1a; SC: subcutaneously; tiw: three times weeklyTraboulsee et al. BMC Neurology  (2018) 18:68 Page 4 of 13Relationship between T1 Gd + lesions at 6 months andclinical outcomes (relapses and EDSS progression)The presence of T1 Gd + lesions at 6 months did notpredict EDSS progression in the placebo/delayedtreatment group or the IFN β-1a 22 μg SC tiw groupin Years 1–4 (Fig. 4a and b). A significantly higherproportion of patients in the IFN β-1a 44 μg SC tiwgroup who had T1 Gd + lesions at 6 months experi-enced EDSS progression at Year 4 versus thosewithout T1 Gd + lesions (p < 0.05; Fig. 4c). Presum-ably due to the effectiveness of treatment, the numberof patients in the IFN β-1a 44 μg SC tiw group withT1 Gd + lesions at 6 months was small (n = 9). How-ever, when comparing patients with ≥2 T1 Gd + le-sions versus those with 0–1 T1 Gd + lesions at Month6, there was a statistically significant difference in theproportion of patients who experienced EDSS pro-gression at Year 2 in the IFN β-1a 44 μg SC tiw0.00.2Cumulative probability0.40.80.61.00 12 24 36 48Months from first dose≥4 active T2 lesionsHazard ratio (95% Cl):1.574 (0.797–3.109)p value: 0.191259.6%36.1%0 active T2 lesions6 93 18 2115 30 3327 42 4539abc0.00.2Cumulative probability0.40.80.61.00 12 24 36 48Months from first dose≥4 active T2 lesionsHazard ratio (95% Cl):2.584 (1.325–5.041)p value: 0.005465.9%37.6%0 active T2 lesions6 93 18 2115 30 3327 42 45390.00.2Cumulative probability0.40.80.61.00 12 24 36 48Months from first dose≥4 active T2 lesionsHazard ratio (95% Cl):1.087 (0.510–2.318)p value: 0.828648.0%43.4%0 active T2 lesions6 93 18 2115 30 3327 42 4539Fig. 2 Time to sustained EDSS progression by ≥4 versus 0 active T2 lesions at 6 months. a Placebo/delayed treatment group; b IFN β-1a 22 μgSC tiw group; c IFN β-1a 44 μg SC tiw group. A Cox proportional hazards model, adjusted for number of relapses in the previous 2 years, age,baseline EDSS score, and baseline burden of disease, was used to estimate hazard ratios and p values. p values are for time to first progressionover 4 years. CI: confidence interval; EDSS: Expanded Disability Status Scale; IFN β-1a: interferon beta-1a; SC: subcutaneously; tiw: threetimes weeklyTraboulsee et al. BMC Neurology  (2018) 18:68 Page 5 of 13group (p = 0.0375). Notably, the number of patientsin the IFN β-1a 44 μg SC tiw group with ≥2 T1 Gd+ lesions at 6 months was very small (n = 3).The presence of T1 Gd + lesions at 6 months did notpredict relapses in any of the treatment groups (Fig. 5a–c). Similar results were seen when comparing patientswith ≥2 T1 Gd + lesions versus those with 0–1 T1 Gd +lesions at Month 6 (data not shown).Relationship between active T2 lesions at 12 months andEDSS progressionNo statistically significant differences in the proportionsof patients with EDSS progression at Years 1, 2, 3, and 4were seen between patients with ≥4 versus 0 active T2lesions at Month 12 in any of the treatment groups.However, numerically greater proportions of patientswith ≥4 versus 0 lesions exhibited EDSS progression20406080100Proportion of patientswith relapses (%)abc020406080100Proportion of patientswith relapses (%)020406080100Proportion of patientswith relapses (%)Placebo/delayed treatment ≥4 T2 (n=64)Placebo/delayed treatment 0 T2 (n=39)YearYearIFN -1a 44 µg SC tiw ≥4 T2 (n=24)IFN -1a 44 µg SC tiw 0 T2 (n=95)1 2 3 41 2 3 41 2 3 4YearIFN -1a 22 µg SC tiw ≥4 T2 (n=28)IFN -1a 22 µg SC tiw 0 T2 (n=85)0* * * *Fig. 3 Proportion relapsed at each year: ≥4 versus 0 active T2 lesions at 6 months a Placebo/delayed treatment group; b IFN β-1a 22 μg SC tiwgroup; c IFN β-1a 44 μg SC tiw group. p values indicate differences between patients with differing lesion loads at 6 months within the treatmentgroup. No statistically significant differences were seen in the placebo/delayed treatment or IFN β-1a 44 μg SC tiw group. Values were calculatedwith a logistic regression model; number of relapses within previous 2 years, age, baseline EDSS score, and baseline burden of disease wereindependent variables, and p values were calculated for the predictive effect of T2 lesion subgroups. *p < 0.05. EDSS: Expanded Disability StatusScale; IFN β-1a: interferon beta-1a; SC: subcutaneously; tiw: three times weeklyTraboulsee et al. BMC Neurology  (2018) 18:68 Page 6 of 13over 4 years in each group (Fig. 6). Within the IFN β-1a44 μg SC tiw group, the numeric difference in propor-tion with progression between those with ≥4 versus 0 le-sions grew more pronounced with each year. Thepredictive value of active T2 lesions at Month 12 forEDSS progression is shown in Table S1(a) [see Add-itional file 3]. The presence of ≥2 versus 0–1 active T2lesions at Month 12 predicted EDSS progression in theIFN β-1a 44 μg SC tiw group over Years 3 and 4 (FigureS5 shows this in more detail [see Additional file 6]).The presence of ≥4 versus 0 active T2 lesions at12 months was associated with numeric trends to-ward reduced time to first EDSS progression in eachtreatment group (HR: 1.215 [95% confidence interval020Proportion of patients withconfirmed EDSS progression (%)406080aProportion of patients withconfirmed EDSS progression (%)020406080bProportion of patients withconfirmed EDSS progression (%)Year020406080cPlacebo/delayed treatment absence Gd+ (n=32)Placebo/delayed treatment presence Gd+ (n=30)YearIFN -1a 22 µg SC tiw absence Gd+ (n=52)IFN -1a 22 µg SC tiw presence Gd+ (n=11)1 2 3 41 2 3 41 2 3 4YearIFN -1a 44 µg SC tiw absence Gd+ (n=57)IFN -1a 44 µg SC tiw presence Gd+ (n=9)*Fig. 4 Proportion with EDSS progression at each year by T1 Gd + lesions at 6 months. a Placebo/delayed treatment group; b IFN β-1a 22 μg SCtiw group; c IFN β-1a 44 μg SC tiw group. There was a significant difference in the number of patients who had T1 Gd + lesions at 6 months andwho experienced confirmed disability progression at 4 years in the IFN β-1a 44 μg SC tiw group, albeit in a small number of patients (n = 9). Nostatistically significant differences were seen in the placebo/delayed treatment or IFN β-1a 22 μg SC tiw groups. Values were based on logisticregression model adjusting for number of relapses within the previous 2 years, age, baseline EDSS score, and baseline T1 Gd + lesions. p valueswere calculated for the predictive effect of T1 Gd + lesions. *p < 0.05. EDSS: Expanded Disability Status Scale; Gd+: gadolinium-enhancing; IFNβ-1a: interferon beta-1a; SC: subcutaneously; tiw: three times weeklyTraboulsee et al. BMC Neurology  (2018) 18:68 Page 7 of 13(CI): 0.634, 2.327], 1.358 [0.707, 2.608], and 1.817 [0.851, 3.879] for placebo/delayed treatment, IFN β-1a22 μg SC tiw, and IFN β-1a 44 μg SC tiw groups, re-spectively [p > 0.05]). For patients with ≥2 versus 0–1active T2 lesions, significant reductions in time toEDSS progression were seen in the IFN β-1a 44 μgSC tiw group (HR: 1.970 [95% CI: 1.150, 3.376]; p =0.0136) but not in the placebo/delayed treatmentgroup (p = 0.5565) or IFN β-1a 22 μg SC tiw group(p = 0.5684).Relationship between active T2 lesions at 12 months andrelapsesSimilar to the results seen with lesions at Month 6, thepresence of ≥4 versus 0 active T2 lesions at Month 12predicted relapses in the placebo/delayed treatmentgroup (Years 1–4; Fig. 7), but not in either IFN β-1a SCtiw group. Again, results similar to the effect of lesionsat 6 months were seen when patients with ≥2 versus 0–1active T2 lesions at Month 12 were compared (Figure S6shows this in more detail [see Additional file 7]).04020Proportion of patients withrelapses (%)6080100YearaYearProportion of patients withrelapses (%)bProportion of patients withrelapses (%)c040206080100040206080100Placebo/delayed treatment absence Gd  (n 32)Placebo/delayed treatment presence Gd  (n 30)IFN -1a 22 µg SC tiw absence Gd  (n 52)IFN -1a 22 µg SC tiw presence Gd  (n 11)IFN -1a 44 µg SC tiw absence Gd  (n 57)IFN -1a 44 µg SC tiw presence Gd  (n 9)Year1 2 3 41 2 3 41 2 3 4Fig. 5 Proportion relapsed at each year by T1 Gd + lesions at 6 months. a Placebo/delayed treatment group; b IFN β-1a 22 μg SC tiw group;c IFN β-1a 44 μg SC tiw group. No statistically significant differences were seen in any of the treatment groups. Values were based on logisticregression model adjusting for number of relapses within the previous 2 years, age, baseline EDSS score, and baseline T1 Gd + lesions. p valueswere calculated for the predictive effect of T1 Gd + lesions. EDSS: Expanded Disability Status Scale; Gd+: gadolinium-enhancing; IFN β-1a:interferon beta-1a; SC: subcutaneously; tiw: three times weeklyTraboulsee et al. BMC Neurology  (2018) 18:68 Page 8 of 13DiscussionThese exploratory analyses demonstrate that active T2lesions at Month 6 were not predictive of longer termclinical disease activity in the IFN β-1a 44 μg SC tiwgroup in PRISMS, and were predictive in the placebo/delayed treatment (EDSS progression and relapses) andIFN β-1a 22 μg SC tiw groups (EDSS progression only).Our data both challenge and extend the previous studiesthat linked the presence of active T2 lesions tosuboptimal treatment responses and clinical outcomes[7, 10, 13, 15]. The results of such previous studies ledto suggestions that active lesions identified during treat-ment with IFN β-1a or other DMDs signify treatmentfailure and should trigger consideration of treatmentchanges [7, 15]. Our findings instead indicate that early(6 months) active T2 lesions are not predictive of long-term outcomes with high-dose IFN SC tiw, while T1 Gd+ lesions at 6 months may be relevant to future clinical020Proportion of patients withconfirmed EDSS progression (%)406080aProportion of patients withconfirmed EDSS progression (%)020406080bProportion of patients withconfirmed EDSS progression (%)Year020406080cPlacebo/delayed treatment, 4 active T2 lesions (n 62)Placebo/delayed treatment, 0 active T2 lesions (n 47)YearIFN -1a 22 µg SC tiw 4 T2 (n 28) IFN -1a 22 µg SC tiw 0 T2 (n 89) 1 2 3 41 2 3 41 2 3 4YearIFN -1a 44 µg SC tiw 4 T2 (n 16)IFN -1a 44 µg SC tiw 0 T2 (n 115)Fig. 6 Proportion progressed at each year by ≥4 versus 0 active T2 lesions at 12 months. a Placebo/delayed treatment, ≥4 versus 0 T2 lesions at12 months; b IFN β-1a 22 μg SC tiw, ≥4 versus 0 T2 lesions at 12 months; c IFN β-1a 44 μg SC tiw, ≥4 versus 0 T2 lesions at 12 months. Nostatistically significant differences were seen in any treatment group. Values were calculated with a logistic regression model with predictor (≥4vs 0 T2 lesions) as a fixed effect; number of relapses within the previous 2 years, age, baseline EDSS score, and baseline burden of disease wereindependent variables, and p-values were calculated for the predictive effect of T2 lesion subgroups. EDSS: Expanded Disability Status Scale; IFNβ-1a: interferon beta-1a; SC: subcutaneously; tiw: three times weeklyTraboulsee et al. BMC Neurology  (2018) 18:68 Page 9 of 13outcomes. Additionally, the presence of continued T2activity after 1 year of treatment suggests that responseto treatment is lacking.Our findings provide new data suggesting that the pre-dictive strength of early on-treatment T2 lesions may bemodified by treatment factors. In this historical cohortof patients with RMS, active T2 lesions at 6 months werepredictive of clinical outcomes in the placebo/delayedtreatment and IFN β-1a 22 μg SC tiw groups, but not inthe IFN β-1a 44 μg SC tiw group, suggesting that thepredictive strength of early on-treatment MRI lesionsmay depend on IFN β-1a SC tiw dose. Notably, EDSSprogression in the IFN β-1a 44 μg SC tiw group, irre-spective of 6-month T2 lesion number, occurred at a rate1 2 3 41 2 3 41 2 3 4Placebo/delayed treatment 4 T2 (n=62)Placebo/delayed treatment 0 T2 (n=47)** *** **aIFN -1a 22 µg SC tiw 4 T2 (n=28)IFN -1a 22 µg SC tiw 0 T2 (n=89)bIFN -1a 44 µg SC tiw 4 T2 (n=16)IFN -1a 44 µg SC tiw 0 T2 (n=115)c04020Proportion of patients withrelapses (%)608004020Proportion of patients withrelapses (%)608010004020Proportion of patients withrelapses (%)6080100YearYearYear100Fig. 7 Proportion relapsed at each year: ≥4 versus 0 active T2 lesions at 12 months. a Placebo/delayed treatment group; b IFN β-1a 22 μg SCtiw group; c IFN β-1a 44 μg SC tiw group. p-values indicate differences between patients with differing lesion loads at 12 months within thetreatment group. No statistically significant differences were seen in the IFN β-1a SC tiw groups. Values were calculated with a logistic regressionmodel; number of relapses within previous 2 years, age, baseline EDSS score, and baseline burden of disease were independent variables, andp-values were calculated for the predictive effect of T2-lesion subgroups. *p < 0.05; **p < 0.01. EDSS: Expanded Disability Status Scale; IFN β-1a:interferon beta-1a; SC: subcutaneously; tiw: three times weeklyTraboulsee et al. BMC Neurology  (2018) 18:68 Page 10 of 13similar to that in patients in the placebo/delayed treat-ment group (especially before the treatment switch toIFN β-1a 44 μg SC tiw at the end of Year 2) who had noT2 lesions. Indeed, no trend toward greater likelihood ofEDSS progression or relapse was seen in the IFN β-1aSC tiw group, even when patients with ≥4 active T2 le-sions at 6 months were compared with patients with noT2 lesions.Importantly, we also show that on-treatment T1 Gd +lesions at 6 months may be predictive of future disabilityprogression. We propose that the active T2 lesions at6 months are indicative of accumulated disease activityover the entire period of Months 0–6 when the treat-ment might not yet be fully effective, while a true lack oftreatment response is evident by the T1 Gd + lesion ac-tivity at Month 6.MAGNIMS Consensus Guidelines consider that usinga baseline MRI scan as the reference point for new activ-ity after treatment initiation carries the possibility ofmisinterpretation, as new activity before treatment wasinitiated or before drug therapy became effective couldbe mistakenly identified as representing suboptimal re-sponse [16]. For instance, in PRISMS-2, the period be-tween baseline and 6 months would include the first8 weeks of administration, when the dose of IFN β-1aSC tiw was being titrated up to the full assigned dose.Although reduction in MRI activity in the subgroupundergoing monthly scanning was observed as early as2 months after N β-1a SC tiw treatment began [17], inrecognition of the possibility that the full treatment ef-fect may not be evident by 6 months, we also examinedas a predictor the new T2 activity at 12 months (withreference to the 6-month scan). The comparison of ≥4versus 0 T2 lesions at 12 months was not shown to sig-nificantly predict future progression. However, numerictrends suggest that lesions at 12 months may be a moreuseful means of identifying nonresponders to IFN β-1aSC 44 μg than lesions at 6 months; indeed, comparing≥2 versus 0–1 active T2 lesions at 12 months predictedfuture progression in this group. Taken together, theseobservations suggest that T2 activity over the first6 months of IFN β-1a SC 44 μg treatment may not indi-cate treatment failure, while activity after a full year issuggestive of nonresponse. Conversely, as the compari-son of ≥2 vs 0–1 T2 lesions at 6 months predicted futureprogression in both the placebo/delayed treatment andIFN β-1a SC 22 μg groups, the IFN β-1a SC 22 μg dose,while beneficial to many patients, shares some similaritywith placebo in that results as early as 6 months may besuggestive of future progression.Previous studies evaluating treatment response accord-ing to MRI results in patients receiving IFN β-1a haveassessed differing treatment regimens and utilized vary-ing on-treatment MRI time points (6 months to 2 yearsafter study baseline), assessment types (T1 Gd + or T2lesions), and thresholds for numbers of lesions [8, 10,12, 18, 19]. As such, the inconsistent results seen be-tween studies suggest that the predictive value of earlyMRI lesions may depend on IFN β dose, the timing andtype of MRI scans obtained, and the overall study proto-col. Importantly, a recent study of accrual of long-termdisability in a cohort of patients with RMS showed thatemergence of new/enlarging T2 lesions in the first2 years on study did not predict clinical worsening (asmeasured by EDSS, the Timed-25 Foot Walk, the 9-Hole Peg Test, or the Paced Serial Auditory AdditionTest-3) over 10 years, again suggesting that radiologicalactivity alone may not be predictive of long-term out-comes [20].As the number of effective DMDs available for RMStreatment grows, some experts have advocated a “zerotolerance” approach to subclinical disease activity, sug-gesting that even a single MRI lesion is evidence oftreatment failure and should trigger a change in therapy[21]. However, the results presented in this analysis sug-gest the need for a more nuanced approach to the prog-nostic value of on-treatment MRI activity. Thepredictive nature of T2 lesions is strongly influenced bywhether patients are on high-dose, high-frequency treat-ment at the time when MRI scans are performed, andthe presence of a small number of active T2 lesions onearly MRI scans should not automatically warrant treat-ment changes. On the other hand, patients who con-tinue to have T1 Gd + lesions, or active T2 lesions after ayear of therapy, may be true non-responders of IFN β-1aSC tiw treatment and might benefit from alternativetherapies.Limitations of this study include the exploratory na-ture of the analyses and the primary focus on T2 lesions.T1 Gd + lesions were assessed only over 9 months andin just a subset of patients. Data from other studies havesuggested that the presence of early T1 Gd + lesions onMRI scans may also be predictive of subsequent clinicaloutcomes [18].ConclusionsThe results of this study suggest that although early(6 months) on-treatment active T2 lesions may predictlong-term clinical activity in patients receiving placeboor low-dose IFN β-1a SC tiw, this association is not seenin patients treated with high-dose IFN β-1a SC tiw(44 μg). On-treatment T1 Gd + lesions at Month 6, how-ever, may predict subsequent clinical outcomes with IFNβ-1a 44 μg SC tiw. Notably, very few patients on treat-ment have T1 Gd + lesions, indicating that the vast ma-jority of patients respond well to the high-dose IFN β-1aSC tiw treatment. Emergence of active T2 lesions after6 months of IFN β-1a 44 μg SC tiw treatment in theTraboulsee et al. BMC Neurology  (2018) 18:68 Page 11 of 13absence of T1 Gd + lesions should not be cause to con-sider treatment changes. However, active T2 lesions at12 months and T1 Gd + lesions at 6 months may because to consider nonresponse to treatment. Early MRIresults can therefore be successfully integrated into thetreatment algorithms that are used to decide whetherpatients are responding to treatment.Additional filesAdditional file 1: Figure S1. Proportion with EDSS progression at eachyear by ≥2 versus 0–1 active T2 lesions at 6 months. (a) Placebo/delayedtreatment group; (b) IFN β-1a 22 μg SC tiw group; (c) IFN β-1a 44 μg SCtiw group. (PDF 258 kb)Additional file 2: Figure S2. Proportion with EDSS progression at eachyear in the placebo/delayed treatment and IFN β-1a 44 μg SC tiw groupsby ≥4 versus 0 active T2 lesions at 6 months. (PDF 189 kb)Additional file 3: Table S1. (a) Performance of T2 lesions on predictingEDSS progression at Year 2 and Year 4. (b) Performance of T2 lesions onpredicting relapse at Year 2 and Year 4. (DOCX 15 kb)Additional file 4: Figure S3. Time to sustained EDSS progression by ≥2versus 0–1 active T2 lesions at 6 months. (a) Placebo/delayed treatmentgroup; (b) IFN β-1a 22 μg SC tiw group; (c) IFN β-1a 44 μg SC tiw group.(PDF 229 kb)Additional file 5: Figure S4. Proportion relapsed at each year by ≥2versus 0–1 active T2 lesions at 6 months. (a) Placebo/delayed treatmentgroup; (b) IFN β-1a 22 μg SC tiw group; (c) IFN β-1a 44 μg SC tiw group.(PDF 269 kb)Additional file 6: Figure S5. Proportion progressed at each year by ≥2versus 0–1 active T2 lesions at 12 months. (a) Placebo/delayed treatment,≥2 versus 0–1 T2 lesions at 12 months; (b) IFN β-1a 22 μg SC tiw, ≥2 ver-sus 0–1 T2 lesions at 12 months; (c) IFN β-1a 44 μg SC tiw, ≥2 versus 0–1 T2 lesions at 12 months. (PDF 183 kb)Additional file 7: Figure S6. Proportion relapsed at each year by ≥2versus 0–1 active T2 lesions at 12 months. (a) Placebo/delayed treatmentgroup; (b) IFN β-1a 22 μg SC tiw group; (c) IFN β-1a 44 μg SC tiw group.(PDF 273 kb)AbbreviationsCI: Confidence interval; DMD: Disease-modifying drug; EDSS: ExpandedDisability Status Scale; Gd + : Gadolinium-enhancing; HR: Hazard ratio;IFN: Interferon; MRI: Magnetic resonance imaging; MS: Multiple sclerosis;PD: Proton density; RMS: Relapsing forms of multiple sclerosis;RRMS: Relapsing–remitting multiple sclerosis; SC: Subcutaneously;SD: Standard deviation; tiw: Three times weekly; Y: YearAcknowledgementsThe authors thank Matthew Thomas, DPhil, of Caudex, Oxford, UK (supportedby EMD Serono, Inc., Rockland, MA, USA (a business of Merck KGaA,Darmstadt, Germany) and Pfizer Inc., New York, NY, USA) for editorialassistance in drafting the manuscript, collating the comments of authors,and assembling tables and figures. The authors also thank the University ofBritish Columbia MS/MRI Research Group for analysis and review of the MRIscan data.FundingThis study and analyses were supported by EMD Serono, Inc., Rockland, MA,USA (a business of Merck KGaA, Darmstadt, Germany) and Pfizer Inc., NewYork, NY, USA. The sponsor planned the post hoc analyses reported here incooperation with the authors. Analysis and interpretation was performed bythe authors. The authors were involved in all stages of development andfinalization of the manuscript and received editorial assistance from anindependent medical-writing-services agency paid by EMD Serono, Inc.,Rockland, MA, USA (a business of Merck KGaA, Darmstadt, Germany) andPfizer Inc., New York, NY, USA.Availability of data and materialsThe data that support the findings of this study are available from EMDSerono, Inc., Rockland, MA, USA (a business of Merck KGaA, Darmstadt,Germany) but restrictions apply to the availability of these data. Theserestrictions and application process are detailed in EMD Serono’s ResponsibleData Sharing Policy, available at http://www.emdserono.com/en/research/commitment_to_responsible_clinical_trial_data_sharing/commitment_to_responsible_clinical_trial_data_sharing.html.Authors’ contributionsDKBL contributed collection of original data, analysis and interpretation ofdata, and revision of the manuscript for important intellectual content. JFcontributed acquisition and interpretation of post hoc analysis data andrevision of the manuscript for important intellectual content. AT, MC, FD, andAM contributed to the analysis and interpretation of the data, and to thewriting of the manuscript and/or its revision for important intellectualcontent. AT, DKBL, MC, JF, FD, and AM read and approved the finalmanuscript.Ethics approval and consent to participateEthics approval was obtained from appropriate Institutional EthicsCommittees/Institutional Review Boards: Royal Melbourne Hospital,Melbourne, Victoria, Australia; Ethics Review Committee, Central Sydney AreaHealth Service, Camperdown, Australia; Health Sciences Centre, University ofWestern Ontario, London, Ontario, Canada; Ottawa General Hospital, Ottawa,Ontario, Canada; Office of Research Services, Clinical Screening Committeefor Research Involving Human Subjects, University of British Columbia,Vancouver, British Columbia, Canada; Forskningsetikkommittén I Lund/Malmö, Lund University Hospital, Lund, Sweden; Independent Review Board,Amsterdam, Netherlands; Newcastle & North Tyneside Health Authorities,Newcastle upon Tyne, UK; Ethical Committee, University Hospital,Nottingham, UK; United Medical and Dental Schools, Division ofPharmacological Sciences, Department of Clinical Pharmacology, St.Thomas’s Hospital, London, UK; Central Oxford Research Ethics Committee,Headington, Oxford, UK; Commissie voor Medische Ethiek/KlinischOnderzoek, UZ Leuven, Leuven, Belgium; Limburgs Universitair Centrum,Diepenbeek, Belgium; Commission d’Ethique Hospito-Facultaire, Universitécatholique de Louvain, Louvain-la-Neuve, Belgium; Helsinki UniversityHospitals Ethical Committee, Helsinki, Finland; Joint Commission on Ethics ofthe Turku University Central Hospital, Turku University Central Hospital, Turku,Finland; Ethik-Kommission de Medizinischen Fakultät der Universität Würz-burg, Würzburg University, Würzburg, Germany; Ethik-Kommission desDepartementes für Innere Medizin, Kantonsspital Basel, Basel, Switzerland;Academisch Zienkenhuis Vrije Univeriteit Commissie Voor Medische Ethiek/Klinisch Onderzoek, Leuven, Belgium; Medisch Ethische Commissie,Academisch Ziekenhuis Rotterdam, Rotterdam, Netherlands; St. George’sHealthcare NHS Trust, St George’s Hospital, London, UK; and Commissiond’Ethique du Département de Médicine, Hôpitaux Universitaires de Genève,Geneva, Switzerland. All patients gave written informed consent.Competing interestsA Traboulsee has acted as a consultant for Biogen, Genzyme, Roche, andTeva, and is Principal Investigator on clinical trials for Biogen, Chugai,Genzyme, and Roche.D Li is the Director of the University of British Columbia MS/MRI ResearchGroup, which has been contracted to perform central analysis of MRI scansfor therapeutic trials with Genzyme, Hoffmann-La Roche, Merck Serono,Nuron, Perspectives, and Sanofi-Aventis. He has acted as a consultant toVertex Pharmaceuticals; has served on scientific advisory boards for Novartis,Nuron, and Roche; has served on a data and safety advisory board for Opexa;and has received research funding from the Canadian Institute of HealthResearch and Multiple Sclerosis Society of Canada.M Cascione has received funding/honoraria for research, consultation, andspeakers bureau participation from Acorda, Bayer HealthCare, Biogen, EMDSerono, Inc., Genentech, Genzyme/Sanofi, Novartis, Pfizer, Roche, and TevaPharmaceuticals.J Fang was an employee of EMD Serono, Inc., Rockland, MA, USA (a businessof Merck KGaA, Darmstadt, Germany) at the time of writing.F Dangond is an employee of EMD Serono, Inc., Billerica, MA, USA (abusiness of Merck KGaA, Darmstadt, Germany).Traboulsee et al. BMC Neurology  (2018) 18:68 Page 12 of 13A Miller has received research support from Biogen, Sanofi-Genzyme, Mal-linckrodt (Questcor), Novartis, and Roche/Genentech. He has acted as a con-sultant for Accordant Health Services (Caremark), Acorda Therapeutics,Alkermes, Biogen, EMD Serono, Sanofi Genzyme, GlaxoSmithKline, Mallinck-rodt (Questcor), Novartis, and Roche/Genentech. He has served on thespeakers bureau for Biogen, Genentech, and Sanofi Genzyme for unbrandeddisease awareness programs only.Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.Author details1University of British Columbia, S113-2211 Wesbrook Mall, Vancouver, BC V6T1Z7, Canada. 2Tampa Neurology Associates, South Tampa Multiple SclerosisCenter, 2919 W. Swann Avenue, Suite 401, South Tampa, FL 33609, USA.3EMD Serono, Inc., One Technology Place, Rockland, MA 02370, USA. 4EMDSerono, Inc., 45A Middlesex Turnpike, Billerica, MA 01821, USA. 5Mount SinaiHospital, 5 East 98th Street, 1st Floor, New York, NY 10029, USA.Received: 21 March 2017 Accepted: 1 May 2018References1. Cohen JA, Barkhof F, Comi G, Hartung HP, Khatri BO, Montalban X, et al.Oral fingolimod or intramuscular interferon for relapsing multiple sclerosis.N Engl J Med. 2010;362:402–15.2. Coles AJ, Twyman CL, Arnold DL, Cohen JA, Confavreux C, Fox EJ, et al.Alemtuzumab for patients with relapsing multiple sclerosis after disease-modifying therapy: a randomised controlled phase 3 trial. Lancet. 2012;380:1829–39.3. PRISMS Study Group. Randomised double-blind placebo-controlled study ofinterferon beta-1a in relapsing/remitting multiple sclerosis. Lancet. 1998;352:1498–504.4. Uitdehaag BM, Barkhof F, Coyle PK, Gardner JD, Jeffery DR, Mikol DD. Thechanging face of multiple sclerosis clinical trial populations. Curr Med ResOpin. 2011;27:1529–37.5. Fox RJ, Miller DH, Phillips JT, Hutchinson M, Havrdova E, Kita M, et al.Placebo-controlled phase 3 study of oral BG-12 or glatiramer in multiplesclerosis. N Engl J Med. 2012;367:1087–97.6. Gold R, Kappos L, Arnold DL, Bar-Or A, Giovannoni G, Selmaj K, et al.Placebo-controlled phase 3 study of oral BG-12 for relapsing multiplesclerosis. N Engl J Med. 2012;367:1098–107.7. Prosperini L, Gallo V, Petsas N, Borriello G, Pozzilli C, One-year MRI. Scanpredicts clinical response to interferon beta in multiple sclerosis. Eur JNeurol. 2009;16:1202–9.8. Rio J, Comabella M, Montalban X. Predicting responders to therapies formultiple sclerosis. Nat Rev Neurol. 2009;5:553–60.9. Sormani MP, Bruzzi P. MRI lesions as a surrogate for relapses in multiplesclerosis: a meta-analysis of randomised trials. Lancet Neurol. 2013;12:669–76.10. Rudick RA, Lee JC, Simon J, Ransohoff RM, Fisher E. Defining interferon betaresponse status in multiple sclerosis patients. Ann Neurol. 2004;56:548–55.11. Polman CH, Reingold SC, Banwell B, Clanet M, Cohen JA, Filippi M, et al.Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonaldcriteria. Ann Neurol. 2011;69:292–302.12. Rio J, Rovira A, Tintoré M, Huerga E, Nos C, Tellez N, et al. Relationshipbetween MRI lesion activity and response to IFN-beta in relapsing-remittingmultiple sclerosis patients. Mult Scler. 2008;14:479–84.13. Durelli L, Barbero P, Bergui M, Versino E, Bassano MA, Verdun E, et al. MRIactivity and neutralizing antibody as predictors of response to IFN-betatreatment in MS. J Neurol Neurosurg Psychiatry. 2008;79:646–51.14. PRISMS Study Group, University of British Columbia MS/MRI Analysis Group.PRISMS-4: long-term efficacy of interferon-beta-1a in relapsing MS.Neurology. 2001;56:1628–36.15. Dobson R, Rudick RA, Turner B, Schmierer K, Giovannoni G. Assessingtreatment response to interferon-beta: is there a role for MRI? Neurology.2014;82:248–54.16. Wattjes MP, Rovira A, Miller D, Yousry TA, Sormani MP, de Stefano MP, et al.Evidence-based guidelines: MAGNIMS consensus guidelines on the use ofMRI in multiple sclerosis–establishing disease prognosis and monitoringpatients. Nat Rev Neurol. 2015;11:597–606.17. Li DK, Paty DW, UBC MS/MRI Analysis Research Group, PRISMS Study Group.Magnetic resonance imaging results of the PRISMS trial: a randomized,double-blind, placebo-controlled study of interferon-beta1a in relapsing-remitting multiple sclerosis. Ann Neurol. 1999;46:197–206.18. Bermel RA, You X, Foulds P, Hyde R, Simon JH, Fisher E, et al. Predictors oflong-term outcome in multiple sclerosis patients treated with interferonbeta. Ann Neurol. 2013;73:95–103.19. Sormani MP, Rio J, Tintore M, Signori A, Li D, Cornelisse P, et al. Scoringtreatment response in patients with relapsing multiple sclerosis. Mult Scler.2013;19:605–12.20. Cree BA, Gourraud PA, Oksenberg JR, Bevan C, Crabtree-Hartman E, GelfandJM, et al. Long-term evolution of multiple sclerosis disability in thetreatment era. Ann Neurol. 2016;80:499–510.21. Giovannoni G, Turner B, Gnanapavan S, Offiah C, Schmierer K, Marta M. Is ittime to target no evident disease activity (NEDA) in multiple sclerosis?Multiple Sclerosis Related Disorders. 2015;4:329–33.Traboulsee et al. BMC Neurology  (2018) 18:68 Page 13 of 13

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0366282/manifest

Comment

Related Items