Clinical–MRI correlations in a European trial of interferon beta-1b in secondary progressive MS
Citation Manager Formats
Make Comment
See Comments

Abstract
Background: The recently completed placebo-controlled multicenter randomized trial of interferon beta-1b (Betaferon) in 718 patients with secondary progressive MS shows significant delay of disease progression and reduction of relapse rate. This study provides an opportunity to assess the level of relationship between clinical and MRI outcomes in this cohort of patients with secondary progressive MS.
Methods: Brain T2-weighted lesion volume was measured annually in all available patients, with visual analysis to identify any new or enlarging (active) T2 lesions at each annual time point. A subgroup of 125 patients had monthly gadolinium-enhanced, T1-weighted imaging at months 0 to 6 and 18 to 24. Relapses were documented and expanded disability status scale (EDSS) was measured every 3 months.
Results: For the annual MRI outcomes, a significant but modest correlation was identified between the change in T2 lesion volume from baseline to the final scan and the corresponding change from baseline in EDSS (r = 0.17, p < 0.0001). There were significant correlations between the cumulative number of active T2 lesions and 1) change in EDSS (r = 0.18, p < 0.0001) and 2) relapse rate (r = 0.24, p < 0.0001). In the subgroup of 125 patients undergoing monthly imaging, MRI lesion activity was correlated with relapse rate over months 0 to 24 (r = 0.24, p = 0.006) but not with change in EDSS.
Conclusions: These results confirm that the clinical–MRI relationships previously identified in relapsing-remitting MS still are apparent in the secondary progressive phase of the disease and support the use of MRI as a relevant outcome measure. In view of the relatively modest nature of the correlations, it seems unwise to rely on such MRI measures alone as primary efficacy variables in secondary progressive MS trials.
Three large randomized Phase III trials each have reported a significant clinical benefit with beta interferon treatment in relapsing-remitting MS (RRMS).1-5⇓⇓⇓⇓ All of these studies incorporated clinical endpoints as the primary outcome measure, but MRI protocols also were performed to supplement and extend the clinical observations. The perceived advantages of MRI measures rather than clinical endpoints in monitoring treatment efficacy include the more objective and retrievable nature of MRI data and a greater sensitivity to disease activity and progression.6,7⇓
An impressive reduction in MRI markers of activity and progression generally has been identified with beta interferon treatment in RRMS. Treatment effects on clinical outcomes have been more modest. The different degree of effect on MRI and clinical outcomes, and the significant but limited correlations between MRI and clinical measures of disease evolution demonstrated in these and other studies,2,4,8,9⇓⇓⇓ urge caution against the interpretation of MRI findings in isolation. MRI indices, therefore, have been used as secondary outcomes in Phase III treatment trials in RRMS, with clinical indices providing the primary trial endpoints.
See also page 2185
Less is known about the extent of any MRI–clinical relationship in the later, secondary progressive phase of MS (SPMS). The degree of this association is an important factor in determining the utility of MRI as an outcome in Phase III trials in SPMS. The recently reported, placebo-controlled, randomized Phase III trial of interferon beta-1b (IFNβ-1b) in SPMS incorporated both clinical and MRI outcomes to assess treatment,10 the results of which have been reported elsewhere.11,12⇓ This study provides an opportunity to establish the level of correlation between clinical and MRI measures in by far the largest cohort of SP patients to date over a planned study duration of 3 years.
The primary MRI outcome measures in this trial were 1) change in T2 lesion volume measured annually in all patients, and 2) the number of active lesions measured monthly during months 0 to 6 in a subgroup of 125 patients undergoing gadolinium (Gd)–enhanced, T1-weighted imaging. The number of active T2 lesions also was visually assessed on the annual MRI scans to provide a further supplementary endpoint. We report here in detail the relationship between these MRI outcomes and the major clinical endpoints in this study, as well as the extent of association between the different MRI outcomes.
Methods.
The study design, methodology, and main clinical and MRI results have been reported elsewhere.10-12⇓⇓ Briefly, the trial comprised a multicenter, double-masked, randomized placebo-controlled Phase III study involving 718 patients (358 in the placebo arm and 360 treated with 8 million IU IFNβ-1b subcutaneously on alternate days) with a planned study duration of 3 years. The study was terminated prematurely based on the positive results of a planned interim analysis after all patients had been in the study at least 24 months.11 Clinical assessment was performed throughout the study by trained raters who measured disability using the Kurtzke expanded disability status scale (EDSS).13 The primary outcome measure was time to confirmed progression of EDSS (an increase of at least 1.0 point [or 0.5 points if the baseline EDSS was 6.0 to 6.5], sustained for 3 months). The number of relapses over the study period and change in EDSS also were identified to provide additional clinical outcome measures.
The MRI design comprised a core annual imaging protocol applied to all 718 patients, with change in T2 brain lesion volume providing the primary MRI endpoint (this also was a secondary trial outcome measure). Image acquisition was performed locally, and MRI analysis subsequently was performed at the Central MRI Analysis Center, Institute of Neurology, London, UK. Brain lesion volume quantification was performed by a number of blinded raters using a semiautomated local thresholding technique.14,15⇓ In addition to this quantitative analysis, simple visual assessment of the serial MR images was performed to document the presence of any active (new or enlarging) T2 lesions appearing over the study duration.
A subgroup of 125 patients from 7 centers also underwent monthly Gd-enhanced, T1-weighted spin echo imaging in addition to a dual echo proton density/T2-weighted spin echo sequence, from baseline (month 0) to month 6, and then again from months 18 to 24. From these images, the number of newly and persistently active lesions was identified. The number of newly active lesions for months 1 to 6 provided a further primary MRI endpoint (and secondary trial efficacy endpoint).
Statistical analysis was performed using Goodman–Kruskal correlation coefficients, together with 95% CI (gamma coefficient in SAS FREQ procedure; SPSS, Chicago, IL). The significance of the correlation was determined using a normal approximation based on the pooled treatment mean correlation and its standard error. Cross-sectional correlations (based on the actual data each year) and longitudinal correlations (based on changes from baseline to last measurement time) were identified between T2 lesion volume and EDSS, and the extent to which baseline lesion volume determined subsequent clinical and MRI progression was determined. Analysis also was performed to assess the relationship between clinical endpoints and the number of active T2 lesions measured serially on the annual studies.
To completely explore the association between the MRI results and clinical outcome, correlation coefficients and 95% CI were provided for the actual data as assessed each year and as changes from baseline to last scan became available, where the correlations between “baseline to end” changes were considered the primary ones. p Values from statistical significance tests proving non-zero correlations also were provided, but these serve purely descriptive purposes. Because of the exploratory nature of all correlation analyses, no adjustment for multiple testing was implemented.
In the frequently scanned subgroup of 125 patients, the correlations between MRI activity over months 0 to 6 and 18 to 24 and both relapse rate and change in EDSS were calculated. Finally, the relationship between the data generated using the different MRI techniques was assessed, again using Goodman–Kruskal correlation coefficients.
A stepwise statistical modeling procedure was applied as previously described11 for the proportion of patients with confirmed progression. The base logistic model included treatment and baseline EDSS category (3.5 or less, 4.0 to 5.5, and 6.5 or more) as main factors, and in separate expanded models, each MRI variable of interest was added to this model to assess its additional impact and to evaluate whether treatment groups were homogeneous within subgroups. The MRI variables used in this context were baseline T2 lesion volume, change in T2 lesion volume during year 1 categorized according to tertiles (all patients), the presence of Gd enhancement at baseline, and the number of newly active lesions during months 1 to 6 (again categorized according to tertiles) in the frequently scanned patient subgroup.
Results.
Clinical–MRI correlations.
There were significant, albeit modest, cross-sectional correlations between EDSS and total lesion volume (TLV) demonstrated at baseline and each follow-up year (table 1), with correlation coefficients in the range of 0.09 to 0.15 for the trial population as a whole; similar magnitude correlations were found for both the treated and placebo subgroups in isolation. Significant but small correlations also were identified for the change in EDSS against change in TLV, with an r value for change over the total study period of 0.17 (p < 0.0001). Furthermore, the baseline TLV correlated minimally with subsequent change in EDSS over the study period (r = 0.07, p = 0.02 for overall trial population). A modest relationship also was identified between change in TLV and relapse rate (r = 0.146, p < 0.0001) (table 1).
Cross-sectional and longitudinal correlations between EDSS/relapse rate and TLV according to treatment group
Correlations of a similar magnitude were found for the change in EDSS vs the number of active T2 lesions at each annual time point and for the entire study duration (table 2), with an r value for change over the total study period of 0.18 (p < 0.0001). A stronger relationship was found between the relapse rate and activity on the annual T2 scans (table 2), with an r value over the study period of 0.24 for the total trial population (p < 0.0001).
Correlations between EDSS/relapse rate and number of active lesions on annual PD/T2 scans
In the frequent MRI subgroup, the total number of newly active lesions correlated significantly with the relapse rate (r = 0.24, p = 0.006) but not with change in EDSS over the first 24 months of the study (table 3).
Correlations between monthly MRI activity in frequently scanned subgroup (n = 125) and clinical/MRI indices
The regression model confirmed the predictive value of baseline lesion volume for subsequent confirmed clinical progression (p < 0.005), with a higher proportion of patients progressing in subgroups with larger baseline TLV (50.9% in placebo patients with TLV less than 14.5 cm3, 53.8% with TLV 14.4 to 31.9 cm3, and 56.5% in placebo patients with TLV greater than 31.8 cm3). However, the model showed no significant predictive value of lesion volume change in the first study year. In the frequent MRI cohort, neither the presence of Gd enhancement at baseline nor new lesion activity in the first 6 months predicted subsequent clinical progression.
Post hoc stratification also was performed to assess the effect of relapse status on TLV (table 4). Treatment effects were highly significant, irrespective of relapse status before and during the study. Whereas there was a trend for a greater increase in TLV in the groups with ongoing relapses, none of these differences were significant. The effect of gender on the change in TLV also was assessed (table 4). Treatment effects were significant, irrespective of gender, and there were no significant differences in the change in TLV according to gender in either the placebo or treatment groups.
Post hoc subgroup analysis for percentage change in TLV over study duration according to relapse status and sex
Correlations between different MRI outcomes.
In the annual MRI analyses, the change in TLV correlated significantly with T2 lesion activity (table 5) over the 3 years (r = 0.36, p < 0.0001). The year-on-year correlations also were significant for all patients combined and for both treatment arms considered separately. The baseline TLV was modestly predictive of the cumulative number of new or enlarging PD/T2 lesions over the study period (r = 0.16, p < 0.0001); the strength of the correlation was similar in the placebo (r = 0.18) and IFNβ-1b (r = 0.16) groups (table 6).
Correlations between change in TLV and new/enlarging T2 lesions
Predictive value of baseline TLV for subsequent MRI progression
The relationship between baseline TLV and absolute change in TLV was more complex and differed according to treatment group (table 6). In the placebo group, there was a positive correlation (r = 0.18, p < 0.0001), and in the IFNβ-1b group, a negative correlation (r = −0.16, p < 0.001) over the whole study period. In a year-by-year analysis, the negative correlation in the IFNβ-1b group was most clearly evident during the first year of treatment (when the r value was −0.22, p < 0.0001).
In the frequent MRI subgroup, the number of newly active lesions during months 0 to 6 was significantly correlated with the number of new or enlarging T2 lesions that appeared on the annual scan (table 3) at month 12 (r = 0.65, p < 0.0001). The number of new active lesions from months 18 to 24 correlated strongly with the number of new or enlarging PD/T2 lesions appearing on the annual scan at month 24 (r = 0.70, p < 0.0001).
Discussion.
This is the largest cohort of SPMS patients in which clinical–MRI correlations have been described. This study demonstrates the existence of several modest correlations between MRI and clinical markers of disease progression in SPMS. Such a correlation was found between changes in T2 brain lesion volume and EDSS over the whole study period and was equally apparent in both the placebo and IFNβ-1b groups (table 1). The extent of these correlations are of a similar order to those observed in recent large Phase III studies of patients with earlier RRMS.2,4⇓ The magnitude of the treatment effect on MRI and clinical outcomes is quantitatively different, with 38.9% of the treated group demonstrating confirmed progression in EDSS despite stabilization of TLV and a reduction in new lesion activity of 57.3%.11,12⇓ The major factors likely to contribute to this difference include the lack of specificity of standard T2-weighted imaging to more destructive pathologic elements (demyelination and axonal loss), the impact of spinal cord disease, and the well-known limitations of the EDSS in terms of its reliability and responsiveness. The importance of this latter factor is supported by the fact that the year-by-year cross-sectional correlations between T2 lesion volume and EDSS improved with each year of follow-up (at baseline: r = 0.09, p = 0.002; at 3-year follow-up: r = 0.15, p < 0.0001). The EDSS range at entry was relatively narrow—from 3 to 6.5—but increased during follow-up as some patients accrued more disability. As noted previously,16 a greater range of disability, as indicated by a widened range of EDSS measurements, provides a better opportunity for appreciating the MRI–EDSS relationship.
A factor that might determine subsequent progression in disability is the extent of pre-existing irreversible disease; areas of chronic persistent demyelination might be an unsuitable environment for maintaining axonal viability and lead eventually to axonal degeneration, even in the absence of an ongoing active pathologic process. However, in this study, there was only a minimal overall correlation between baseline TLV and EDSS change over the next 3 years (r = 0.07, p = 0.02), implying that TLV is of little value in predicting of subsequent clinical progression in SPMS. The situation is different in the earliest clinical stages of the disease, where T2-weighted imaging is more predictive of clinical evolution.17 This discrepancy supports the concept that pathologic changes other than those defined by T2-weighted imaging become increasingly important contributors to clinical progression later in the disease course.
Gd enhancement has been correlated in MS and experimental allergic encephalomyelitis with the presence of active inflammation.18,19⇓ Enhancement, however, does not quantify demyelination and axonal loss, the substrates of disability in MS. The lack of correlation between monthly activity (mainly enhancing lesions) and EDSS change over 2 years, therefore, is not surprising and is consistent with the observations of a recent, large meta-analysis.20 The possibility that the amount of inflammatory activity predicts disability over a longer follow-up period is not excluded, however: one small study reports a relationship between MRI activity during a 6-month period of monthly scanning and progression in disability after 5 years in a small cohort of patients with SPMS.21
When enhancing inflammatory lesions occur in clinically eloquent locations, they are likely to be expressed clinically,22,23⇓ and the significant relationship between enhancing activity and relapse rate found in this and other studies20 is not surprising. The moderate extent of the correlation partly reflects the fact that most enhancing lesions occur in clinically silent brain regions but also may reflect pathologic heterogeneity of enhancing lesions; that is, purely inflammatory, nondemyelinating lesions may not produce a clinical deficit, whereas inflammation and demyelination, although producing an indistinguishable enhancing lesion, is more likely to cause conduction block and clinical deficit.
Significant correlations also were identified between relapse rate and annual T2 lesion activity for all annual time points and over the whole study period (table 2). This is expected, given that new T2 lesions usually represent new areas of inflammation and demyelination. The results suggest a valuable role of new T2 lesions as a marker of disease progression. Treatment effect was demonstrated with simple visual assessment of serial T2-weighted images, and the clinical–MRI correlations for this outcome are at least as strong as T2 lesion volume quantification.
The correlation of TLV and new PD/T2 lesion activity in the placebo group was significant but modest (r = 0.36 over the whole study period). This reflects the fact that the net change in TLV represents a composite of both an increase from new lesion formation and a decrease from lesion resolution. Furthermore, the impact of new lesions on change in TLV is not linear over time, with activity shortly before the exit scan having a greater impact on any net change in TLV than earlier activity.24 A recent study confirms that the net change in T2 lesion volume substantially underestimates volume of new T2 lesions that have developed, as a result of resolution in other areas of T2 hyperintensity,25 and clinical correlation of new lesions differ from those of net changes.
There was a strong correlation between annual T2 and monthly enhanced lesion activity during the first 2 years. In year 1, T2 activity was compared with enhancing activity from months 1 to 6, whereas in year 2, it was compared with monthly activity from months 19 to 24, that is, the period immediately preceding the annual PD/T2 study. The correlation was slightly stronger in the second period (r = 0.7 versus 0.65 in the first period). This may reflect the fact that the most recent enhancing activity on monthly scans is likely to be the most visible on a follow-up PD/T2-weighted image obtained at a longer interval; a similar relationship between the timing of monthly enhancing activity measures and changes in TLV over a period of approximately 1 year has been reported.24
When the relationship between baseline TLV and subsequent TLV change was studied, an interesting difference was found between the placebo and treated groups (table 6). In the placebo group, there was a positive correlation between baseline TLV and both change in TLV over the next 3 years (r = 0.18) and the cumulative number of new or enlarging lesions (r = 0.18). Patients with a larger TLV at study entry probably had accumulated abnormality more rapidly than those with a lower TLV before study entry and continued to do so over the duration of the study. In contrast, in the IFNβ-1b group, there was a negative correlation between baseline TLV and subsequent TLV change (r = −0.16, p < 0.0001). Patients with higher TLV should, on average, have a higher volume of inflammatory or edematous lesions with the potential to resolve, and this probably accounts for the greater decrease in TLV, which was fully apparent at the first year of follow-up.
The MRI results of this study support the clinical efficacy endpoint measures.11 Highly significant and powerful treatment effects were identified for both qualitative and quantitative analysis of the annual MRI studies.12 The results also indicated significant clinical–MRI correlations in SPMS, extending previous observations on RRMS1,2⇓ to the later stages of the disease course. The value of simple visual assessment of T2-weighted MR studies is evident, suggesting a greater role for this outcome measure in future Phase III trials. However, the modest overall nature of the clinical–MRI correlations suggests that it would be unwise to rely on measurement based on T2-weighted or Gd-enhanced lesions alone as the primary efficacy variables in SPMS trials. Measures of MR markers of the more neurodegenerative aspects of MS disease (e.g., atrophy, T1 hypointensity) are likely to be valuable in evaluating therapies at this stage of disease26,27⇓
Appendix
The European Study Group on Interferon Beta-1b in Secondary Progressive MS (MRI).
MRI investigators (principle investigators in bold print): Aberdeen, Scotland:R. Knight, J.E.C. Hern, O.J. Robb; Amsterdam, the Netherlands:C. Polman, J. Valk, F. Barkhof, J.H. van Waesberghe, T. Schweigmann; Barcelona, Spain:J. Montalbán, A. Rovira, S. Pedraza; Basel, Switzerland:L. Kappos, E.W. Radü; Belfast, Northern Ireland:S. Hawkins, K.E. Bell, C.S. McKinstry;Berlin, Germany:H. Altenkirch, K. Baum, K.M. Einhäupl, P. Marx, R. Lehmann; Berlin, Germany (Schering AG): C. Christel; Birmingham, England:, D. Francis, E.B. Rolfe; Bordeaux, France:B. Brochet, V. Dousset; Cardiff, Wales:C.M. Wiles, S.F.S. Halpin, M.D. Hourihan; Dublin, Ireland:M. Hutchinson, D. McErlaine; Düsseldorf, Germany:G. Stoll, T. Kahn; Erfurt, Germany:H.W. Kölmel, R. Kachel; Florence, Italy:L. Amaducci (d. 1998), L. Massacesi, C. Fonda; Göttingen, Germany:S. Poser, A. Riegel, B. Welskop; Groningen, the Netherlands:J. Minderhoud, J. De Keyser, H. van Woerden, T. de Jong; Helsinki, Finland:J. Wikström, O. Salonen; Huddinge, Sweden:S. Fredrikson, B. Isberg; Leuven, Belgium: G. Wilms, P. Demaerel; London, England:D. Miller;Lyons, France:C. Confavreux, J.-C. Froment; Masku, Finland:M. Panelius, P. Sonninen, H. Oivanen, J. Ruutiainen; Melsbroek, Belgium:M. D’Hooghe;Milan, Italy:G. Comi, M. Filippi, M. Rovaris; Munich, Germany:R. Hohlfeld, T.A. Yousry, C. Becker, F. Stadie, P. Eppmann; Newcastle, England:N. Cartlidge, A. Coulthard, P. English; Osnabrück, Germany:P. Haller, A.W. Frank; Paris, France:O. Lyon-Caen, E. Cabanis, M.-T. Iba-Zizen; Rennes, France:G. Edan, M. Carsin, Y. Rolland; Rome, Italy:C. Fieschi, S. Bastianello, E. Giugni; Sheffield, England:S.J.L. Howell, T.J. Hodgson, C.A.J. Romanowski; Toulouse, France:M. Clanet, I. Berry, D. Ibarrolla, O. Martin; Vienna, Austria:H. Kollegger, L. Deecke, S. Trattnig; Würzburg, Germany:R. Gold, H.-P. Hartung, D. Hahn, W. Kenn, and T. Pabst.
Queen Square MRI Central Evaluation Unit (London, UK): G.J. Barker, P. Brex, J. Cook, A Fletcher, C. Fogg, D. Galetti, H. Gallagher, M. Gawne-Cain, B. Gomez-Anson, S. Gregory, E. Gunn, T. Holmes, L. Livingstone, M. Lowis, D.G. MacManus, R. Maunder, B. McNulty, C. Middleditch, D.H. Miller, P.D. Molyneux, I.F. Moseley, C. Noctor, J. O’Riordan, T. Pearce, P. Robinson, A. Stepney, V. Stevenson, P. Tofts, N. Tubridy, L. Wang, and I. Walsh.
Writing committee: D.H. Miller, P.D. Molyneux, G.J. Barker, D.G. MacManus, I.F. Moseley, K. Wagner, and L. Kappos.
Steering committee: Amsterdam, the Netherlands: C.H. Polman; Basel, Switzerland: L. Kappos; Berlin, Germany (Schering AG): F. Dahlke, M. Ghazi, K. Wagner; London, England: A.J. Thompson; Rome, Italy: C. Pozzilli.
Independent advisory board: Bethesda, MD: H. McFarland (Chairman); Vancouver, Canada: J. Petkau (Statistical Advisor); Rennes, France: O. Sabouraud; Würzburg, Germany: K. Toyka. Data entry and statistical analyses: Parexel, Berlin, Germany.
Acknowledgments
Supported by Schering AG, Berlin, Germany.
Footnotes
- Received October 30, 2000.
- Accepted September 12, 2001.
References
- ↵
Paty DW, Li DKB, the UBC MS/MRI Study Group, the IFNB Multiple Sclerosis Study Group. Interferon beta-1b is effective in relapsing remitting MS: MRI analysis results of a multicenter, randomized, double blind, placebo controlled trial. Neurology . 1993; 43: 662–667.
- ↵
IFNB Multiple Sclerosis Study Group, University of British Columbia MS/MRI Analysis Group. Interferon beta-1b in the treatment of multiple sclerosis: final outcome of the randomized, controlled trial. Neurology . 1995; 45: 1277–1285.
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
Polman C, Dahlke F, Thompson AJ, et al. Interferon beta-1b in secondary progressive multiple sclerosis: outline of the clinical trial. Mult Scler . 1995; 1: S51–54.
- ↵
- ↵
- ↵
Kurtzke JF. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology . 1983; 33: 1444–1452.
- ↵
- ↵
Molyneux PD, Fletcher A, Gunn B, et al. Precision and reliability for measurement of change in MRI lesion volume in multiple sclerosis: a comparison of two computer assisted techniques. J Neurol Neurosurg Psychiatry . 1998; 65: 42–47.
- ↵
Gawne-Cain ML, O’Riordan JI, Coles A, et al. MRI lesion volume measurement in MS and its correlation with disability: a comparison of fast FLAIR and spin echo sequences. J Neurol Neurosurg Psychiatry . 1998; 64: 197–203.
- ↵
O’Riordan JI, Thompson AJ, Kingsley DP, et al. The prognostic value of brain MRI in clinically isolated syndromes of the CNS: a 10-year follow-up. Brain . 1998; 121: 495–503.
- ↵
Hawkins CP, Munro PM, MacKenzie F, et al. Duration and selectivity of blood–brain barrier breakdown in chronic relapsing experimental allergic encephalomyelitis studied by gadolinium-DTPA and protein markers. Brain . 1990; 113: 365–378.
- ↵
- ↵
Kappos L, Moeri D, Radue EW, et al. Predictive value of gadolinium-enhanced magnetic resonance imaging for relapse rate and changes in disability or impairment in multiple sclerosis: a meta-analysis. Gadolinium MRI Meta-analysis Group. Lancet . 353: 964–969
- ↵
- ↵
Thorpe JW, Kidd D, Moseley et al. Serial gadolinium-enhanced MRI of the brain and spinal cord in early relapsing remitting multiple sclerosis. Neurology . 1996; 46: 373–378.
- ↵
Youl BD, Turano G, Miller DH, et al. The pathophysiology of acute optic neuritis: an association of gadolinium leakage with clinical and electrophysiological deficits. Brain . 1991; 114: 2437–2450.
- ↵
- ↵
Lee MA, Smith S, Palace J, Matthews PM. Defining multiple sclerosis disease activity using MRI T2-weighted difference imaging. Brain . 1998; 121: 2095–2102.
- ↵
Molyneux PD, Kappos L, Polman C, et al. The effect of interferon beta-1b treatment on MRI measures of cerebral atrophy in secondary progressive multiple sclerosis. Brain . 2000; 123: 2256–2263.
- ↵
Barkhof F, van Waesberghe JHTM, Filippi M, et al. T1 hypointense lesions in secondary progressive MS: effect of interferon beta-1b. Brain . 2001; 124: 1396–1402.
Disputes & Debates: Rapid online correspondence
REQUIREMENTS
If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.