MRI outcomes in a placebo-controlled trial of natalizumab in relapsing MS

In multiple sclerosis (MS), focal inflammatory demyelinating lesions are thought to occur when activated T lymphocytes cross the blood-brain barrier (BBB).¹ The adhesion molecule α4β1-integrin (VLA-4), which is expressed on all circulating white blood cells except neutrophils, plays a role in migration of mononuclear white blood cells across the BBB.^2,3 Areas of BBB breakdown are visualized as foci of gadolinium (Gd) enhancement on T1-weighted MRI scans and occur in new lesions in relapse-onset MS.^4–6 Pathologically, Gd-enhancing lesions exhibit perivascular inflammation with myelin breakdown.^7,8

Natalizumab is a humanized monoclonal antibody that blocks the interaction of α4-integrin on mononuclear white blood cells with vascular cell adhesion molecule 1 (VCAM-1) on endothelial cells² and prevents migration of white blood cells into the brain parenchyma in experimental allergic encephalomyelitis.² In a 6-month, placebo-controlled trial, natalizumab reduced the number of new Gd-enhancing lesions by 90% in relapsing MS.⁹

A placebo-controlled, randomized, double-blind, phase III clinical trial (the AFFIRM study) investigated the efficacy and safety of natalizumab in patients with relapsing MS. Natalizumab significantly reduced the annualized relapse rate by 68% and sustained progression of disability by 42% over 2 years compared with placebo.¹⁰ Summary details of the counts of Gd-enhancing lesions and new or enlarging T2 lesions have been previously reported.¹⁰ We now present a comprehensive report of MRI lesion activity and volume outcomes from the AFFIRM study. Patients underwent annual brain MRI scans over 2 years, and both total lesion volumes and number of new lesions were assessed at each follow-up. Change in brain volume was also investigated.

METHODS

Details of the design and methodology of the AFFIRM study have been published.¹⁰ Briefly, 942 patients from 99 clinical centers in Europe, North America, Australia, and New Zealand were enrolled. Men and women 18 to 50 years of age were included in the study if they had a diagnosis of relapsing MS,¹¹ an Expanded Disability Status Scale (EDSS) score between 0.0 and 5.0,¹² an MRI scan demonstrating lesions consistent with MS, and at least one medically documented relapse within the 12 months prior to randomization. Patients were randomly assigned (2:1) to receive natalizumab (Tysabri, Biogen Idec, Inc. and Elan Pharmaceuticals, Inc.) 300 mg or placebo by IV infusion every 4 weeks for up to 116 weeks. All study personnel, patients, and sponsor personnel involved in the study conduct were blinded to treatment assignments.

RESULTS

Overall population.

Scans analyzed.

Depending on the MRI measure, analyzable scans were not available for 4 to 5% patients at year 1 and 8 to 9% of patients at year 2 (tables 2 and 3). The main reason for missing data (>80%) was the scan not being performed because the patient withdrew from the study; in the remainder (<20%), although the patient was still in the study, the scan was either not performed, had not been received at the Central MRI Analysis Center, or had been received but was of inadequate quality for analysis.

View this table:

• View inline
• View popup
• Download powerpoint

Table 2 Number of Gd-enhancing, new or enlarging T2-hyperintense, and new T1-hypointense lesions detected during the trial (intent-to-treat population, N = 942)

View this table:

• View inline
• View popup
• Download powerpoint

Table 3 MRI lesion volumes (mm³) (intent-to-treat population, N = 942)

Number of lesions.

The number of Gd-enhancing lesions, new or enlarging T2-hyperintense lesions, and new T1-hypointense lesions over the 2 years of treatment are shown in table 2 (the data on Gd-enhancing and new or enlarging T2-hyperintense lesions is as previously reported¹⁰). The mean number of Gd-enhancing lesions was reduced by 92% in the natalizumab group vs placebo group at both year 1 (0.1 vs 1.3; p < 0.001) and year 2 (0.1 vs 1.2; p < 0.001). Ninety-seven percent of natalizumab patients had no Gd-enhancing lesions on the year 2 MRI scan vs 72% of placebo patients.

Compared with placebo, natalizumab reduced the mean number of new or enlarging T2-hyperintense lesions by 83% (11.0 vs 1.9; p < 0.001) over 2 years (table 2). Fifty-seven percent of patients in the natalizumab group developed no new or enlarging T2-hyperintense lesions over the 2-year treatment period compared with 15% of placebo patients. In addition, the mean number of new T2-hyperintense lesions (p < 0.001) was reduced by 82% and the mean number of enlarging T2-hyperintense lesions (p < 0.001) was reduced by 88% over the 2-year study period in the natalizumab arm (table 2). However, there was more than 10 times the number of new lesions compared with enlarging lesions in each study arm.

On unenhanced scans, the mean number of all new T1-hypointense lesions over 2 years was 1.1 in the natalizumab group and 4.6 in the placebo group, representing a 76% reduction in new T1-hypointense lesions with natalizumab (p < 0.001). The effect was observed over the first year and continued in the second year of treatment (74% and 83% reductions; p < 0.001). A higher percentage of natalizumab-treated patients (63%) had no new T1-hypointense lesions compared with placebo-treated patients (27%), and substantially fewer natalizumab-treated patients (11%) had three or more new T1-hypointense lesions compared with placebo-treated patients (44%). On Gd-enhanced scans, the mean number of nonenhancing new T1-hypointense lesions over 2 years was 1.0 in the natalizumab group and 3.8 in the placebo group, representing a 74% reduction in new nonenhancing T1-hypointense lesions with natalizumab (p < 0.001, table 2).

Sensitivity analyses of data on the number of lesions showed results that were consistent with the primary analyses (data not shown).

Lesion volumes.

Gd-enhancing, T2-hyperintense, and T1-hypointense lesion volumes are shown in table 3. Natalizumab reduced Gd-enhancing lesion volume compared with placebo at both year 1 (means 21 vs 207 mm³; p < 0.001) and year 2 (means 32 vs 192 mm³; p < 0.001), representing 90% and 83% reductions with natalizumab treatment at the 1- and 2-year time points. Compared with baseline, the mean Gd-enhancing lesion volumes were lower at both the year 1 and 2 time points in both the natalizumab- and placebo-treated arms; this decrease was greater at both year 1 (−343 mm³ vs −126 mm³; p < 0.001) and year 2 (−332 mm³ vs −141 mm³; p < 0.001) for the natalizumab-treated group.

Mean and median T2 lesion volumes are shown in figure 1. Over the first year of treatment, T2-hyperintense lesion volumes decreased in the natalizumab group and increased in the placebo group (p = 0.016). (Note also that in the placebo group in the first year, whereas the median T2 volume per se increases [figure 1 and table 3], the median difference in T2 volumes, year 1 minus baseline, decreases [table 3]; this is possible because there is no mathematical relationship between these two measures.) During the second year of treatment, T2-hyperintense lesion volume continued to increase in the placebo group and remained stable in the natalizumab group (p < 0.001). Overall, the mean T2 lesion volume was lower in natalizumab-treated patients (14,722 [95% CI: 13,238 to 16,206]) compared with placebo-treated patients (17,853 [95% CI: 15,414 to 20,292]) (p < 0.001) after 2 years. Over 2 years, natalizumab also reduced the mean change (−905 mm³ vs 2,891 mm³; p < 0.001) and median percentage change (−9.4% vs 8.8%; p < 0.001) in T2 lesion volume compared with placebo. Mean percentage of T2 volume changes (table 3) also show a marked difference between the treatment arms, although the actual mean percentage of changes is larger than median percentage of changes: the reason for this is that the mean percentage of changes are skewed by very large percentage increases in individual patients, especially among those who started with small lesion volumes.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 Mean ± SE (top) and median (bottom) T2-hyperintense lesion volumes at baseline, year 1, and year 2

Mean and median T1-hypointense lesion volumes are shown in figure 2. T1-hypointense lesion volumes decreased over the first year in both treatment groups; however, the reduction in lesion volume was greater in the natalizumab group compared with the placebo group (p = 0.004). During year 2, T1-hypointense lesion volume increased in the placebo group and remained the same in the natalizumab group (p < 0.001). Natalizumab also reduced the mean change (−1,508 mm³ vs 548 mm³; p < 0.001) and median percentage of change (−23.5% vs −1.5%; p < 0.001) in T1-hypointense lesion volume compared with placebo over 2 years. Mean percentage of T1-hypointense volume changes (table 3) also show clear differences between treatment arms, although as for T2 volume measures, the actual mean percentage of change is much larger than the median percentage of change.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2 Mean ± SE (top) and median (bottom) T1-hypointense lesion volumes at baseline, year 1, and year 2

To assess the extent of change in T1-hypointense lesion volume relative to change in total (T2) lesion volume, the T1/T2 lesion volume ratio (LVR) was investigated. At baseline, the mean T1/T2 LVR was 0.343 (median 0.349; SD = 0.189; interquartile range 0.19 to 0.46) in the placebo arm and 0.328 (median 0.319; SD = 0.187; interquartile range 0.18 to 0.46) in the natalizumab arm. At 2-year follow up, the mean T1/T2 LVR was 0.311 (median 0.295; SD = 0.239; interquartile range 0.18 to 0.40) in the placebo arm and 0.270 (median 0.256; SD = 0.166; interquartile range 0.15 to 0.38) in the natalizumab arm (p = 0.002; comparison between the treated and placebo groups, Friedman's ANCOVA [ranked data], adjusted for baseline T1/T2 LVR). The changes in T1/T2 LVR from baseline to 2 years were i) placebo arm: mean −0.03, median −0.032, interquartile range −0.11 to 0.03; ii) natalizumab arm: mean −0.058, median −0.048, interquartile range −0.12 to 0.1 (comparison between placebo and natalizumab arms p = 0.002, Friedman's ANCOVA [ranked data], adjusted for baseline T1/T2 LVR). Thus, T1/T2 LVR decreased in both arms but more so in natalizumab-treated patients.

Sensitivity analyses of lesion volume data showed results that were consistent with the primary analyses (data not shown).

Brain parenchymal fraction.

Over 2 years of treatment, the mean percentage of reductions in BPF were similar between the natalizumab and placebo groups (0.80 and 0.82%; p = 0.822). During the first year of treatment, both groups exhibited a decrease in BPF; however, there was a greater reduction in the natalizumab group compared with the placebo group (0.56% vs 0.40%; p = 0.002). During the second year of treatment, a reduction (equivalent to that observed during the first year) was again observed in the placebo group (0.43%), whereas the reduction in the natalizumab group decreased to 0.24%; there was a difference between the natalizumab and placebo groups during the second year of treatment (p = 0.004).

DISCUSSION

MRI findings from the AFFIRM study confirm and extend those observed in the earlier 6-month proof-of-concept trial⁹; natalizumab was associated with a marked and sustained decrease in the number of Gd-enhancing lesions, which are indicative of BBB breakdown in MS. In addition, the accumulation of new lesions seen either as areas of T2 hyperintensity or T1 hypointensity (the latter representing about one-third of the former) was significantly reduced by natalizumab over 2 years of treatment. The 92% reduction in Gd-enhancing lesions emphasizes the profound effect of natalizumab in preventing new focal lesions that manifest BBB leakage and is consistent with its mechanism of action as an α4-integrin antagonist.

Several observations can be made that are relevant to planning and interpretation of MRI scans in phase III clinical trials of a similar size and duration. First, when annual scans are performed, the number of new T2-hyperintense lesions considerably exceeds (by 4–5 times) the number of Gd-enhancing lesions, and hence is a much more sensitive outcome measure. The T2 measure visualizes new T2-hyperintense lesions that may have appeared at any time during the year because the majority of such lesions become persistent areas of T2 hyperintensity. In contrast, Gd enhancement is a transient feature of new lesions that typically lasts for only a few weeks and therefore depicts only active lesions at the time of MRI.

Second, new T2-hyperintense lesions are much more common (>10 times) than enlarging lesions. Although this partly reflects the use of conservative criteria for enlargement that had been previously developed by an expert group,¹⁴ we considered that such an approach was necessary because it is difficult to be certain of enlargement during analysis of scans, as the appearance of lesions is susceptible to change with minor differences in repositioning. For future studies, it may be considered more efficient to report new T2 lesions only and not expend extra effort in detecting (uncertainly) a few enlarging lesions unless there is a hypothesis that therapy will have a discordant effect on new vs enlarging T2-hyperintense lesions; genuine biologic enlargement of lesions, even if not appreciated visually, should also be reflected in the measure of total lesion volume.

Third, the number of new T1-hypointense lesions is a feasible and sensitive measure for the evaluation of a treatment effect. As expected, T1-hypointense lesions are a feature of about 30% of new T2-hyperintense lesions. The total count of new T1-hypointense lesions on the precontrast scans included Gd-enhancing lesions that would have been potentially acute and reversible; therefore, additional analysis was performed on the postcontrast scans. This revealed that a large majority of the new T1-hypointense lesions were nonenhancing, hence more likely to reflect areas of persistent hypointensity that have been associated with a greater degree of tissue matrix damage including axonal loss when compared with isointense lesions.¹⁷ This subset of lesions was reduced in natalizumab-treated patients, indicating the potential of natalizumab to reduce the accumulation of axonal loss arising from new lesions in relapsing-remitting MS.

The Gd-enhancing lesion volume measures mirrored the number of lesions, and an equally clear-cut difference was observed between treatment groups. Although a clear difference in T2-hyperintense and T1-hypointense lesion volume evolution was also seen in favor of the natalizumab-treated group, the longitudinal pattern of change was nonlinear in both groups and warrants more detailed consideration.

During the first year, there was only a minimal increase in T2-hyperintense lesion volume in the placebo-treated group. This result indicates that resolution of preexisting lesions (i.e., present at baseline) almost matched the increase due to new lesions. During the second year, there was a clear increase in T2-hyperintense lesion volume in the placebo group, indicating that the accumulation of new lesions outweighed the resolution of preexisting lesions. Because the number of new T2-hyperintense lesions was similar in both years, the difference in volume likely reflects that there were more subacute inflammatory lesions at study entry that had potential to resolve than was the case later in the study. This idea is supported by the Gd-enhancing volumes in the placebo group being approximately 50% higher at baseline than at year 1 and year 2 and by the observation that relapse rate during the year prior to study entry was 1.3, whereas in the first year on study it had reduced to 0.7.¹⁰ Another possible mechanism for a larger increase in T2 volume in year 2 despite no increase in the number of new or enlarging T2 lesions is that there had been enlargement of areas of T2 abnormality that could not be detected visually, i.e., due to coalescence or expansion of preexisting lesions. Reliable visual detection of areas of enlarging T2 abnormality is difficult and conservative criteria were employed for their detection.¹⁴

The relative stability of T2-hyperintense lesion volume in placebo-treated patients in the first year of the study emphasizes the crucial importance of having a parallel control group to distinguish the effects of spontaneous regression in disease activity from genuine treatment effects. In the natalizumab-treated group, there was a clear reduction in T2-hyperintense lesion volume in year 1, with little change in year 2. This indicates that resolution of preexisting lesions outweighed the development of new lesions (which were very few for patients on treatment) in year 1, whereas in year 2, when there were few lesions either resolving or appearing, total volumes were predictably stable.

There was a small drop in T1-hypointense lesion volume in the placebo-treated group during year 1 and a somewhat larger increase during year 2. The mechanisms underlying this evolution are likely to be similar to those seen for T2 volumes, but the sharper initial drop in T1-hypointense lesion volume (and the decrease in T1/T2 LVR over 2 years) may reflect a relatively high proportion of subacute lesions at study entry that had resolution of edema or remyelination, both of which will favor evolution from T1 hypointensity to T1 isointensity.¹⁸ The large drop in T1-hypointense volume in the natalizumab-treated group during year 1 fits with partial resolution of preexisting lesions, in addition to there being very few new lesions; an additional factor may be that new lesions in natalizumab-treated patients are also less likely to evolve to regions of T1 hypointensity.¹⁹ Such mechanisms may also account for the greater decrease in T1/T2 LVR seen in the natalizumab arm. As expected, with few new lesions in the second year and a stable preexisting lesion load at year 1, the T1-hypointense lesion volumes were relatively stable in the natalizumab arm during year 2.

Natalizumab significantly reduced the number of new MRI lesions in all the subgroups studied except for those that had fewer than nine T2 lesions at study entry. However, the latter group was small, and neither its placebo nor active treatment group exhibited much clinical or MRI activity during the study. This finding suggests that patients with relapsing MS who have few MRI lesions are likely to experience a more quiescent disease course during clinical trials. There may be a case for requiring a minimum number of lesions as a criterion for entry into future studies of relapsing MS.

Higher levels of MRI activity in the placebo group during the trial were associated with higher levels of disease activity at baseline, most notably a higher relapse rate during the year prior to study entry and the presence of Gd-enhancing lesions at baseline. Similar observations were made in subgroups of the previous smaller phase II study.²⁰ This observation suggests that subgroups selected for having more frequent recent relapses or Gd-enhancing lesions will be more sensitive when evaluating MRI lesion activity measures in treatment trials since they are likely to have more on-study activity. In a similar vein, a recent analysis of placebo data from multiple trials showed that prior relapse rate is correlated with the likelihood of Gd enhancement at a single time point.²¹

Whereas the BPF, a measure of brain tissue volume that is normalized to intracranial volume, was decreased more in natalizumab-treated patient during year 1, the rate of brain atrophy was significantly less in natalizumab-treated patients during year 2. A plausible explanation for an initial reduction in brain volume in natalizumab-treated patients is the profound decrease of inflammatory brain lesions with resolution of associated edema. An early reduction in brain volume has been associated with other treatments that have an anti-inflammatory effect: high-dose IV methylprednisolone²² and interferon beta.²³ The smaller amount of brain atrophy in year 2 may reflect a longer-term benefit of natalizumab in preventing new lesion formation and the consequent axonal transaction (and resulting wallerian degeneration) that occurs in such lesions.²⁴

ACKNOWLEDGMENT

The authors thank Nancy Bormann for assistance with preparation of the manuscript and Nisha Patel for assistance with statistical analysis. The MS MRI group at the Institute of Neurology is also supported by the MS Society of Great Britain and Northern Ireland.

APPENDIX

*Principal investigators.

MRI Central Reading Center, Institute of Neurology, University College London, London, UK: D. Miller*, T. Yousry, K. Miszkiel, G. Barker, D. MacManus, D. Soon, K. Fernando, C. Webb, K. Hunter, L. Hearsum, C. Middleditch, T. Pepple, C. Miller, Z. Stockbridge, J. Goldsmith, H. Proctor, P. Bartlett, A. Cannon, H. Murphy, A. Rajkumar, J. Lewis, A. John, T. Alfaro-Vidal, K. Wilder-Ahmed, V. Santana, T. Solano, S. Zalita. MRI Brain Atrophy Analysis Center, Cleveland Clinic, Cleveland, OH: E. Fisher*, R. Rudick.

In addition to the authors, the following investigators participated in the AFFIRM study.

Australia. Austin Health, Melbourne: R. Macdonell*, A. Hughes, I. Taylor, Y.C. Lee, H. Ma; Royal Melbourne Hospital, Melbourne: J. King*, T. Kilpatrick, H. Butzkueven, M. Marriott; University of Sydney, Sydney: J. Pollard*, P. Spring, J. Spies, M. Barnett. Belgium. Algemeen Ziekenhuis St. Jan, Brugge: I. Dehaene*, L. Vanopdenbosch, M. D'Hooghe, M. Van Zandijcke, O. Derijck; C.H.U. de Charleroi, Charleroi: P. Seeldrayers*, J. Jacquy, T. Piette, C. De Cock; Limburgs Universitair Centrum, Diependeek: R. Medaer*, P. Soors, E. Vanroose, L. Vanderhoven; Nationaal MS Centrum, Melsbroek: M. D'Hooghe*, G. Nagels, B. Dubois, M.-C. Deville, R. D'Haene. Canada. CHVO Hospital de Hull, Gatineau, Quebec: F. Jacques*, D. Hallé, S. Gagnon, E. Likavcan; Dalhousie MS Research Unit, QEII, Halifax, Nova Scotia: T.J. Murray*, V. Bhan, R. MacKelvey, C.E. Maxner; SCO. Elisabeth Bruyere Health Center, Ottawa, Ontario: S. Christie*, R. Giaccone, D.A. Guzman; Health Services Centre, Winnipeg, Manitoba: M. Melanson*, F. Esfahani, A.J. Gomori, M.H. Nagaria; Hospital Charles Lemoyne, Greenfield Park, Quebec: F. Grand'Maison*, L. Berger, Z. Nasreddine, M. Duplessis; Kingston General Hospital, Kingston, Ontario: D. Brunet*, M. Melanson, A. Jackson, G. Pari; St. Michael's Hospital, Toronto, Ontario: P. O'Connor*, T. Gray, M. Hohol, P. Marchetti; Sunnybrook and Women's College Health Sciences Center, Toronto, Ontario: L. Lee*, B. Murray, J. Sahlas, J. Perry; Vancouver Coastal Health Authority, UBC Hospital, Vancouver, British Columbia: V. Devonshire*, J. Hooge, S. Hashimoto, J. Oger, P. Smyth; University Campus London Health Sciences Center, London, Ontario: G. Rice*, M. Kremenchutzky. Czech Republic. Faculty Hospital Brno Bohunice, Brno: P. Stourac*, Z. Kadanka, Y. Benesova, I. Niedermayerova; Faculty Hospital Motol, Prague: E. Meluzinova*, P. Marusic, M, Bojar, K. Zarubova, E. Houzvicková, J. Piková; Faculty Hospital Hradec Kralove, Hradec Kralove: R. Talab*, G. Warberzinek, M. Valis, M. Talábová; Faculty Hospital Olomouc, Olomouc: B. Muchova*, K. Urbánek, Z. Kettnerova, J. Mares, P. Otruba; Faculty Hospital of Ostrava Poruba, Ostrava: O. Zapletalová*, P. Hradilek, D. Dolezil, I. Woznicova, R. Höfer; Faculty Hospital of Plzen, Plzen: Z. Ambler*, J. Fiedler, J. Sucha, V. Matousek; Faculty Hospital St. Anne, Brno: I. Rektor*, M. Dufek, R. Mikulik, J. Mastik, I. Tyrlikova; General Teaching Hospital, Prague: E. Havrdová*, D. Horakova, H. Kalistová, M. Táblová; Regional Hospital of Pardubice, Pardubice: E. Ehler*, A. Novotná, P. Geier. Denmark. Rigshospitalet, Copenhagen: P. Soelberg-Sorensen*, M. Ravnborg, B. Petersen, M. Blinkenberg. Finland: Helsingin yliopistollinen keskussaitaala, Helsinki: M. Färkkilä*, H. Harno, M. Kallela, O. Häppölä; Tampere University Hospital, Tampere: I. Elovaara*, H. Kuusisto, M. Ukkonen, J. Peltola, J. Palmio. France. Hôpital D'adultes de la Timone, Marseille: J. Pelletier*, L. Feuillet, L. Suchet, A. Dalecky, D. Tammam; Hôpital Pontchaillou, Rennes: G. Edan*, E. Le Page, M. Mérienne, J. Yaouanq; Hôpital Purpan, Toulouse: M. Clanet*, C. Mekies, C. Azais-Vuillemin, A. Senard, G. Lau; Hôpital Universitaires, Strasbourg: G. Steinmetz*, J. Warter*, V. Wolff, M. Fleury, C. Tranchant. Germany: Neurologische Klinik des Klinikums Offenbach, Offenbach: E. Stark*, U. Buckpesch-Heberer, K.-H. Henn, T. Skoberne; Neurologische Klinik der Ruhr-Universität Bochum am St. Josef-Hospital, Bochum: S. Schimrigk*, K. Hellwig, N. Brune; Universitatsklinikum Hamburg-Eppendorf, Hamburg: C. Weiller*, J. Gbadamosi, J. Röther, C. Heesen, C. Buhmann. Greece. District General Hospital G. Gennimatas, Athens: C. Karageorgiou*, D. Korakaki, Dr. Giannoulis, S. Tsiara; Pammakaristos General Hospital, Athens: T. Thomaides*, I. Thomopoulos, H. Papageorgiou, F. Armakola. Hungary: Jahn Ferenc Hospital, Budapest: S. Komoly*, C. Rózsa*, J. Matolcsi, G. Szabó, B. Molnár, G. Lovas; Josa Andras Hospital, Nyiregyhaza: P. Dioszeghy*, P. Szulics, Z. Magyar, J. Incze, J. Farkas; Kenézy Gyula Hospital, Debrecen: B. Clemens*, J. Kánya, Zs. Valicskó, E. Bense, Zs. Nagy; National Institute of Psychiatry and Neurology, Budapest: G. Geréby*, J. Perényi, Zs. Simon, M. Szapper, L. Gedeon; Petz Aladár County Hospital, Györ: A. Csanyi*, G. Rum, S. Lipóth, A. Szegedi, L. Jávor, I. Nagy, I. Adám; Semmelweis University, Budapest: I. Szirmai*, M. Simó, C. Ertsey, I. Amrein, A. Kamondi; St. Imre Hospital, Budapest: P. Harcos*, E. Dobos, B. Szabó, V. Balas; Szent György Hospital, Székesfehérvár: A. Guseo*, E. Fodor, E. Jófejü, K. Eizler; University of Debrecen, Debrecen: L. Csiba*, T. Csépány, E. Pallagi, D. Bereczki; Uzsoki Hospital, Budapest: G. Jakab*, M. Juhász, B. Szabó, I. Mayer, G. Katona. Ireland. St. Vincent's University Hospital, Dublin: M. Hutchinson*, J. O'Dwyer, K. O'Rourke. The Netherlands. Amphia Ziekenhuis, Breda: E.A.C.M. Sanders*, J.F. Rijk-van Andel, M.A.M. Bomhof, P. van Erven; Erasmus MC, Academic Hospital Rotterdam, Rotterdam: R.Q. Hintzen*, I. Hoppenbrouwers, R.F. Neuteboom, D. Zemel, P.A. van Doorn, B.C. Jacobs; Jeroen Bosch Aiekenhuis, 's-Hertogenbosch: E.Th.L. Van Munster*, J.P. ter Bruggen, R. Bernsen; Stichting MS Centrum, Nijmegen: P.J.H. Jongen*, E.A.A. de Smet, H.F.H. Tacken; VUMC, De Boelelaan, Amsterdam: C. Polman*, J. Zwemmer, J. Nielsen, N. Kalkers, J. Kragt, B. Jasperse. New Zealand. Auckland Hospital, Auckland: E. Willoughby*, N.E. Anderson, A. Barber; Christchurch Hospital, Christchurch: T. Anderson*, P.J. Parkin, J. Fink, S. Avery, D. Mason. Poland. Katedra i Klinika Neurologii AM, SPCSK, Warsaw: H. Kwiecinski*, B. Zakrzewska-Pniewska, A. Kaminska, A. Podlecka, M. Nojszewska; II Klinika Neurologii, Instytut Psychiatrii I Neurologii, Warsaw: A. Czlonkowska*, J. Zaborski, W. Wicha, J. Kruszewska-Ozimowska, L. Darda-Ledzion; Katedra I Klinika Neurologii, Lodz: K. Selmaj*, A. Mochecka-Thoelke, J. Pentela-Nowicka, A. Walczak, M. Stasiolek; Katedra I Klinika Neurologii, Lublin: Z. Stelmasiak*, H. Bartosik- Psujek, K. Mitosek-Szewczyk, E. Belniak, U. Chyrchel; Katedra I Klinika Neurologii, Slaskiej AM, Katowice-Ligota: A. Wajgt*, M. Maciejowski, L. Strzyzewska- Lubos, L. Lubos, E. Matusik; Klinika Neurologii, 10 WSK, Bydgoszcz: Z. Maciejek*, A. Niezgodzinska-Maciejek, D. Sobczynska, T. Slotala, S. Wawrzyniak; Niezalezny Zespol Opieki Zdrowotnej-“Kendron,” Bialystok: J. Kochanowicz*, K. Kuczynski, R. Zimnoch, M. Pryszmont; Samodzielny Publiczny Szpital Kliniczny, Oddzial Kliniczny Neurologii, Bialystok: W. Drozdowski*, E. Baniukiewicz, A. Kulakowska, H. Borowik, M. Lewonowska; Szpital Uniwersytecki w Krakowie, Oddzial Kliniczny Neurologii, Kraków: A. Szczudlik*, T. Róg, E. Gryz-Kurek, J. Pankiewicz, J. Furgal, A. Kimkowicz; Wojewódzki Szpital Specjalistczny IM M. Kopernika, Gdansk: W. Fryze*, T. Wierbicki, L. Michalak, J. Kowalewska, J. Swiatkiewicz. Sweden. Karolinska University Hospital, Huddinge, Huddinge: J. Hillert*, E. Åkesson, S. Fredrikson, P. Diener; Karolinska University Hospital, Stockholm, Solna Stockholm: T. Olsson*, E. Wallström, F. Piehl, L. Hopia, L. Brundin, M. Marta, M. Andersson; MS Centrum, Sahlgrenaks University Hospital, Mölndal: J. Lycke*, B. Runmarker, C. Malmeström, P. Vaghfeldt, B. Skoog. Switzerland. Centre Hospitalier Universitaire Vaudios, Lausanne: M. Schluep*, J. Bogousslavskyr, R. Du Pasquier; Universitätsspital Basel, Neurologisch-Neurochirurgische Poliklinik, Basel: L. Kappos*, L. Achtnichts, J. Kuhle, C. Buitrago-Telez, R. Schläger, Y. Naegelin. Turkey. University of Istanbul, Istanbul: M. Eraksoy*, N. Bebek, G. Akman Demir, B. Topcuoglu, M. Kurtuncu; Istanbul University, Istanbul: A. Siva*, S. Saip, A. Altintas, A. Kiyat. United Kingdom. Guy's, Kings & St. Thomas School of Medicine, London: M. Sharief*, M. Kasti; Institute of Neurology, London: G. Giovannoni*, E.T. Lim, W. Rashid; King's College Hospital, London: E. Silber*, G. Saldanha; North Staffordshire Royal Infirmary, Stoke on Trent: C. Hawkins*, G. Mamutse, J. Woolmore; Oldchurch Hospital, Essex: C. Hawkes*, L. Findley, R. DaSilva, H. Gunasekara; Radcliffe Infirmary, Oxford: J. Palace*, Z. Cader, E. Littleton, G. Burke; Royal Hallamshire Hospital, Sheffield: B. Sharrack*, O. Suliman, S. Klaffke; Royal London Hospital, London: M. Swash*, H. Dhillon; Royal Victoria Infirmary, Newcastle: D. Bates*, M. Westwood, P. Nichol; St. George's Hospital, London: D. Barnes*, D. Wren, N. Stoy; University Hospital of Wales, Cardiff: N. Robertson*, T. Pickersgill, O. Pearson, C. Lawthom; Walton Centre for Neurology and Neurosurgery, Liverpool: C. Young*, R. Mills, B. Lecky. United States. Clinical and Magnetic Resonance Research Center, Albuquerque, NM: C. Ford*, J. Katzman, G. Rosenberg; East Bay Region Associates in Neurology, Berkeley, CA: J. Cooper*, B. Wrubel, B. Richardson; Landon Center on Aging, Kansas City, KS: S. Lynch*, L. Ridings, A. McVey, W. Nowack; Lehigh Valley Hospital, Allentown, PA: A. Rae-Grant*, G.A. Mackin, J.E. Castaldo, L.J. Spikol; Mayo Clinic Scottsdale, Scottsdale, AZ: J. Carter*, D. Wingerchuk, R. Caselli, D. Dodick; MeritCare Neuroscience Clinic, Fargo, ND: S. Scarberry*, R. Bailly, K. Garnaas, B. Haake; Michigan Institute for Neurological Disorders, Farmington Hills, MI: H. Rossman*, M. Belkin, W.D. Boudouris, R.P. Pierce; Oregon Health and Sciences University, Portland, OR: M. Mass*, V. Yadav, D. Bourdette, R.H. Whitham; Texas Neurology, PA, Dallas, TX: J.T. Phillips*, D. Heitzman, A. Martin, C.F. Greenfield; University of California Davis Medical Center, Sacramento, CA: M. Agius*, D.P. Richman, N. Vijayan, V.L. Wheelock; University of Chicago Medical Center, Chicago, IL: A. Reder*, B. Arnason, A. Noronha, R. Balabanov, A. Ray; University of Miami School of Medicine, Miami, FL: W. Sheremata*, S. Delgado, B. Shebert, J. Maldonado; University of Washington, Seattle, WA: J. Bowen*, G.A. Garden, B.J. Distad; Yale Center for MS Treatment and Research, New Haven, CT: M. Carrithers*, M. Rizzo, T. Vollmer, J. Preiningerova, J. Guarnaccia, A. Lo, G.B. Richardson.