Randomized study of antibodies to IFN-g and TNF-a in secondary
`progressive multiple sclerosis
`
`Multiple Sclerosis (2001) 7, 277±284
`ã 2001 Arnold All rights reserved 1352 ± 4585/01
`www.arnoldpublishers.com/journals
`
`S Skurkovich*,1, A Boiko2, I Beliaeva2, A Buglak2, T Alekseeva2, N Smirnova2, O Kulakova3,
`V Tchechonin4, O Gurova4, T Deomina2, OO Favorova3, B Skurkovich1,5 and E Gusev2
`1Advanced Biotherapy Laboratories, Rockville, MD, USA; 2Department of Neurology and Neurosurgery, Russian State
`Medical University, Moscow, Russia; 3Department of Molecular Biology and Biotechnology, Russian State Medical
`University, Moscow, Russia; 4Immunochemical Laboratory of the Russian State Scienti®c Center of Social and Legal
`Psychiatry, Moscow, Russia; 5Brown University Medical Center, Providence, Rhode Island, USA
`
`Studies of cytokines in multiple sclerosis (MS) have shown that immune mechanisms connected with disturbance of the synthesis of cytokines
`probably play critical roles in the initiation and prolongation of MS. In a double-blind, placebo-controlled trial, 45 patients with active secondary
`progressive MS were randomized to three groups of 15 patients, each receiving a short course of antibodies to IFN-g, to tumor necrosis factor
`(TNF)-a, or a placebo. After 12 months with analysis of disability (Expanded Disability Status Scale scores), accompanied by interval determinations
`of lymphocyte subpopulations, cytokine production levels, MRI, and evoked potentials, it was found that only patients who received antibodies to
`IFN-g showed statistically signi®cant improvement compared to the placebo group ± a signi®cant increase in the number of patients without
`con®rmed disability progression. This was supported by MRI data (a decrease in the number of active lesions) and systemic changes in cytokine
`status (a decrease in IL-1b, TNF-a, and IFN-g concentrations in supernatants of activated blood cells of these MS patients and an increase in TGF-
`b production). Neutralization of IFN-g could be a new approach to treating secondary progressive MS. Long-term administration of humanized
`monoclonal antibodies to IFN-g and simultaneous use of antibodies to IFN-g together with IFN-b products are planned.
`Multiple Sclerosis (2001) 7, 277±284
`Keywords: multiple sclerosis; anti-interferon-g; anti-tumor necrosis factor-a; cytokines
`
`Introduction
`
`Multiple sclerosis (MS) is a demyelinating disease of the
`central nervous system (CNS). The pathological process
`in brain tissue during MS is thought to be the result of
`chronic, mainly cell-mediated, autoimmune reactions
`directed against numerous myelin and possibly other
`endo- and exogenous antigens. Studies of cytokines in
`MS patients indicate that several immune mechanisms
`connected with the disturbance of the synthesis of these
`cytokines, particularly of interferon (IFN)-g, probably
`play critical roles in the initiation and prolongation of
`the pathological process in MS.1 For this reason, the
`cytokine system is the target for numerous immunomo-
`dulating approaches to treating MS, both in experi-
`mental models and in clinical trials. More than 20 years
`ago one of us proposed that
`the development of
`autoimmune diseases was connected with the distur-
`bance of the synthesis of certain cytokines and that the
`removal of these cytokines could be therapeutic.2 The
`®rst successful anti-cytokine therapy was performed in
`1975 in patients with severe rheumatoid arthritis.3 Later
`it was proposed to treat MS by anti-cytokine therapy.4
`Disease-modifying treatment of relapsing ± remitting MS
`with IFN-b products became a great achievement in MS
`IFN-b in treating
`care.5 Recent data on the use of
`secondary progressive MS are promising,6 but
`the
`
`*Correspondence: S Skurkovich, 802 Rollins Avenue, Rock-
`ville, MD 20852 USA
`Received 24 January 2001; accepted 16 May 2001
`
`treatment of secondary progressive MS where the
`degenerative process might be irreversible still remains
`a great problem.
`Cytokines are low-molecular weight soluble proteins
`involved in signaling between cells of the immune
`system. They are released by leukocytes as well as other
`cell types, including cells of the CNS, and modulate both
`local and systemic immune reactions.7 MS is postulated
`to be a Th1 type cell-mediated autoimmune disease.
`Cytokines upregulating cellular immunity and pro-
`duced at least by monocytes (interleukin-1 [IL-1], tumor
`necrosis factor-a [TNF-a]) and by Th1 helpers (IL-2, IFN-
`g, lymphotoxin or TNF-b) are believed to contribute to
`propagating the pathological process and tissue injury in
`MS, while cytokines produced by Th2 cells (IL-4, IL-10)
`have been shown to suppress Th1-cells and cell-
`mediated immunity. Th3-cells can also participate in
`Th1 suppression, producing transforming growth factor-
`b (TGF-b). Human endogenous retroviruses (HERVs)
`have been detected in patients with MS,8 schizophre-
`nia,9 and certain other autoimmune diseases and may
`play a causative role in these diseases. Activation of this
`retrovirus could lead to the production of signi®cant
`amounts of IFN-g, which may directly damage central
`and peripheral nerve cells and lead to neurological
`autoimmune diseases.
`One possible mechanism for the action of IFN-b in
`MS may be through its inhibition of IFN-g-induced
`immune activation.10
`
`Hopewell EX1042
`
`1
`
`

`

`278
`
`Anti-IFN-gg and anti-TNF-aa in MS
`S Skurkovich et al
`
`the most
`Anti-cytokine therapy remains one of
`exciting directions in immunomodulation in MS.11
`We report results of one of the ®rst trials of antibodies
`to the pro-in¯ammatory cytokines TNF-a and IFN-g in
`secondary progressive MS.
`
`Methods
`
`Antibodies
`Healthy female goats were immunized with recombi-
`IFN-g or recombinant-TNF-a (Peprotech) until
`nant
`titer of the plasma reached 76104 and 86104 IU/ml
`respectively. The goats were plasmapheresed and the
`plasma was pooled. IgG was isolated and puri®ed. The
`®nal concentration of the antibody was 86104 u/ml
`and 96104 IU/ml respectively (ELISA).
`It was ali-
`quoted to 2 ml/vial. Each ml of preparation consisted
`of 48 ± 50 mg of protein.
`
`Placebo
`Commercial albumin from an American blood bank
`was used. Each ampoule of albumin consisted of 2 ml
`(50 mg of protein/ml).
`
`Patients
`A total of 45 patients with de®nite MS were enrolled in
`a randomized, double-blind, placebo-controlled, paral-
`lel group study. All subjects gave written informed
`consent to participation in the study. The study was
`approved by the institutional research ethics commit-
`tee. The inclusion criteria were: (1) clinically de®nite
`MS according to Poser criteria12
`that
`followed a
`secondary progressive course; (2) with active disease,
`i.e., showing deterioration of at least one point on
`Kurtzke's Expanded Disability Status Scale (EDSS)
`within 12 months prior to enrolment; (3) age 18 ± 50
`years old;
`(4) EDSS score between 3.0 and 7.0 at
`
`to goat protein.
`(5) negative skin test
`selection;
`Exclusion criteria included: (1) substantial cognitive
`impairment that could have compromised the patient's
`ability to give informed consent; (2) moderate or severe
`diseases other then MS; (3) primary progressive or
`relapsing ± remitting MS;
`(4) duration of secondary
`progression shorter than 1 year (to be sure of
`the
`progression) and longer than 5 years (to minimize the
`in¯uence of
`irreversible degenerative changes);
`(5)
`present or planned pregnancy; (6) receiving or having
`received in the last 12 months any immunosuppressive
`drugs, interferons or copaxone; (7) use of corticoster-
`iods in the 4 weeks before the ®rst visit; (8) previous
`exposure to any product containing goat proteins; (9)
`inability to undergo MRI scanning or immunological
`investigation; and (10) positive skin test to goat protein.
`Forty-®ve patients were randomized using tables of
`random numbers. Patients with early active secondary
`progression as a rule with relapses (i.e., with steady
`progression between relapses) were selected as the
`target for the trial (Figure 1).
`
`Treatment protocol
`The treatment of the patients, the handling of any side
`effects and adverse events, and the evaluation of
`clinical parameters were performed by quali®ed
`blinded neurologists, whenever possible by the same
`physician. Assessing physicians were different from
`those treating. Patients were randomized to three
`groups, 15 in each, using a book of random numbers.
`The ®rst group of patients received antibodies (AB) to
`TNF-a,
`the second ± AB to IFN-g; and the third ±
`placebo. The randomization was controlled by a
`statistician, and the type of
`treatment received by
`each patient was kept concealed from the physician
`and patients. All antibodies were produced and
`supplied by Advanced Biotherapy Inc.
`(USA)
`in
`
`Figure 1 Selection of patients with early secondary progressive MS
`
`Multiple Sclerosis
`
`2
`
`

`

`identical-looking vials. Albumin, packaged the same
`way as the AB, was used as a placebo and diluted in
`the same manner. Puri®ed (IgG) goat polyclonal AB
`were diluted in sterile, distilled water for injections
`and administered twice a day intramuscularly (i.m.)
`daily for 5 consecutive days. Syringes for injection of
`the ABs were prepared to be identical in appearance
`and administered by teams separate from the preparer
`and from the evaluating physician. The safety of this
`regimen was demonstrated in a previous trial.8,13
`Among adverse effects were a ¯u-like syndrome with
`fever and headache in one patient who received AB to
`TNF-a (6.7%) and mild local
`skin in¯ammatory
`reactions (no necrosis was registered) ± in three cases
`who received AB to TNF-a (20%); in two ± AB to IFN-g
`(13.3%); and one ± placebo (6.7%). No changes in liver
`enzymes or other hematological or biochemical char-
`acteristics were seen during the short course of AB or
`after it during the observation period. The side effects
`were never a reason for halting treatment. During the
`observation period, four patients dropped out because
`they started treatments with products of IFN-b: two
`from the group who received AB to TNF-a (after 8 and
`10 months of follow-up) and two from the placebo
`group (after 7 and 8 months of follow-up). These cases
`cannot be analyzed as
`treatment
`failures
`since,
`independently of the treatment results, these patients
`at this point found money for the payment of the
`treatment previously shown to be effective in second-
`ary progressive MS. All these patients were included
`in the analyses for the previous time points (Figure 2).
`At selection, all patients underwent a medical
`examination,
`including a neurological examination.
`Visual and brainstem auditory evoked potentials and
`an MRI study were performed and blood was taken for
`immunological
`investigation. Follow-up obligatory
`visits were at 1, 3, 6, and 12 months. At each visit a
`
`279
`
`Anti-IFN-gg and anti-TNF-aa in MS
`S Skurkovich et al
`
`repeat neurological examination was done and blood
`was taken for hematological and biochemical examina-
`tion. In addition, at the 1-month visit, immunological
`and electrophysiological examinations were performed
`and at
`the 6-month visit ± MRI and immunological
`examinations as well. Every patient had both MRI
`examinations according to universal strict protocol.14
`Five-mm axial slices were obtained,
`repositioning
`based on standardized anatomic landmarks, proton
`density and T2-weighted fast spin-echo sequence
`before contrast, axial T1-weighted spin-echo sequence
`after injection of gadolinium (obtained 10 min after
`0.1 mmol/kg of intravenous (i.v.) gadolinium-DTPA).
`Scans were analyzed by the same blinded radiologists.
`The numbers of hyperintensive lesions at T2-weighted
`scans and the number of enhancing lesions at T1-
`weighted scans were measured. Studies of visual (VEP)
`and brainstem auditory evoked potentials (BAEP) were
`performed according to universally accepted proto-
`col.15,16 If the clinical status of patients worsened, they
`were able to visit out of order. In cases where an
`increased EDSS score was seen at a scheduled visit
`between months 3 and 12, patients were asked to come
`again 1 and 3 months
`later
`to con®rm EDSS
`progression. During
`the observation period, MS
`patients visited their consulting neurologist from ®ve
`to 10 times (including baseline).
`The primary outcome measure was the number of
`patients without con®rmed progression in disability as
`measured by an increase of at least 1.0 level on EDSS,
`sustained for at
`least 3 months, exclusive of any
`exacerbation. Secondary outcome measures were
`selected as:
`time without con®rmed progression,
`number of
`relapses, number of patients
`free of
`relapses, number of active lesions on T1 MRI scans,
`changes in parameters of VEP or BAEP. During the
`follow-up, patients received corticosteroids if they had
`
`Figure 2 Pro®le of the study, obligatory visits, and investigations
`
`Multiple Sclerosis
`
`3
`
`

`

`280
`
`Anti-IFN-gg and anti-TNF-aa in MS
`S Skurkovich et al
`
`relapses with severe deterioration. This treatment did
`not lead to exclusion from the study.
`
`Immunological investigation
`All subjects had their blood drawn between 09.00 and
`11.00 h. After gentle mixing with an equal volume of
`Ca2+/MG2+-free Hank's balanced salt solution (HBSS,
`Sigma), blood was centrifuged with Ficoll-Paque
`(Pharmacia) at 400 g for 40 min at room temperature.
`Buffy coats of mononuclear cells between Ficoll and
`HBSS layers, carefully transferred with sterile pipettes
`into centrifuge tubes, were spun at 800 g for 10 min at
`48C three times. After cell counting of
`lympho/
`monocytes in an RPMI-1640 medium (with addition
`of 10% fetal calf serum (Sigma), 2 mM L-glutamine,
`50 u/ml gentamicin),
`they were cultivated for 48 h
`with PHA (®nal concentration 5 mg/ml, Sigma). Super-
`natants were collected and kept at 7208C until
`analysis. Concentrations of cytokines were measured
`by a standard ELISA technique17 with kits for IFN-g
`and TGF-b (RD, USA), IL-1b, IL-2 and IL-6 (Sigma,
`USA) and TNF-a (Proteinoviy Kontur, Russia) accord-
`ing to manufacturer's guidelines. Due to technical
`
`reasons, the assay for levels of IL-1, IL-2, and IL-6
`production was done for only nine patients from the
`AB to IFN-g group and nine from the AB to TNF-a
`group, and for 10 patients from the placebo group.
`Study of the cell subpopulations was performed using
`¯ow cytometry (Becton-Dickenson) with mAB to CD3,
`CD4, CD8, CD29 and CD16 (all
`from Becton-Dick-
`enson).
`
`Statistical evaluation
`Statistical evaluation was performed using SPSS and
`EPI INFO packages and included different methods of
`parametric and non-parametric statistics.
`
`Results
`
`Baseline characteristics of all groups did not differ
`signi®cantly (Table 1). The number of patients free of
`progression was lower in the group who received AB
`to IFN-g (Figure 3). The difference between patients
`who received AB to IFN-g and placebo became
`signi®cant 9 months after the short treatment course
`(12/15 vs 3/13, w2=6.92, P=0.008 with Yates correc-
`
`Table 1 Baseline clinical-demographic characteristics of MS patients included in the study
`
`Characteristics (number of patients)
`
`Age (years): mean+s.e. (range)
`
`Age of onset (years): mean+s.e. (range)
`
`Gender (w/m)
`Disease duration (years): mean+s.e. (range)
`
`Duration of secondary progression (years):
`mean+s.e. (range)
`EDSS: mean+s.e. (range)
`
`sFS: mean+s.e. (range)
`
`AB, antibodies; s.e., standard error
`
`All
`(n=45)
`
`AB to TNF-a
`(n=15)
`
`AB to IFN-g
`(n=15)
`
`38.54+1.62
`(22 ± 54)
`27.71+1.72
`(15 ± 44)
`23/22
`11.31+1.18
`(4 ± 30)
`2.65+0.17
`(1 ± 5)
`4.18+0.19
`(3.0 ± 7.0)
`13.97+0.67
`(8 ± 21)
`
`38.33+2.46
`(22 ± 51)
`27.93+2.40
`(15 ± 44)
`9/6
`10.67+1.84
`(4 ± 30)
`2.83+0.37
`(1 ± 5)
`3.97+0.29
`(3.0 ± 6.0)
`14.13+1.09
`(8 ± 21)
`
`38.93+1.95
`(25 ± 49)
`27.47+1.69
`(20 ± 41)
`7/8
`11.71+1.73
`(4 ± 24)
`2.46+0.29
`(1 ± 5)
`4.47+0.30
`(3.0 ± 6.5)
`15.33+0.68
`(10 ± 20)
`
`Placebo
`(n=15)
`
`38.80+2.23
`(24 ± 54)
`27.53+2.11
`(16 ± 42)
`7/8
`11.27+1.65
`(4 ± 27)
`2.53+0.27
`(1 ± 5)
`4.13+0.43
`(3.0 ± 7.0)
`12.80+0.87
`(8 ± 18)
`
`Figure 3 Number of patients without con®rmed EDSS progression for 12 months
`
`Multiple Sclerosis
`
`4
`
`

`

`tion), and still remained signi®cant 12 months later (7/
`15 vs 1/13, P=0.038 in Fisher two-sided test), while the
`number of such patients in the subgroup who received
`AB to IFN-g decreased from the 9- to 12-month time
`points. No signi®cant difference with patients who
`received AB to TNF-a were seen. The difference
`between groups was con®rmed by Kaplan-Meier
`analyses, showing signi®cantly less probability for
`these patients to progress in disability during the
`observation period. The time without progression in
`comparison to the placebo group (5.93+0.94 months)
`was signi®cantly longer (9.0+1.06 months, P=0.042)
`for the AB to IFN-g group, but not for the AB to TNF-a
`group (7.33+1.05 months, P40.05). It is important to
`notice that after 3 months, the group of patients who
`received AB to IFN-g were divided into two sub-
`groups ± the majority had a less active MS course and a
`stable or better EDSS, while there were several patients
`who still had active disease with an increase in EDSS
`score. Altogether, eight relapses during 12 months
`were registered in the placebo group (six had to be
`treated with corticosteriods), seven in the AB to TNF-a
`group (®ve were treated with corticosteroids) and ®ve
`in the AB to IFN-g group (all were treated with
`corticosteroids). Thus, while there was a clear decrease
`in these numbers for the AB to IFN-g group, neither
`the number of relapses, nor the number of corticoster-
`oid courses differed signi®cantly between groups. The
`difference in the number of patients free of relapses
`during the 12-month period also did not
`reach
`statistical signi®cance (for the 12-month time point,
`10/15 in the AB to IFN-g group and 6/13 in the
`placebo group, P40.05).
`The positive trend in the whole group who
`received AB to IFN-g was con®rmed by MRI (Table
`2). At baseline, all groups had similar MRI scans
`characteristics. Six months later the total number of
`T2-lesions was also similar, while the signi®cant
`reduction in the numbers of MRI active lesions
`(P50.05) was seen only in the group who received
`AB to IFN-g. There were no signi®cant changes in
`parameters of VEP or BAEP in all groups, either in
`
`281
`
`Anti-IFN-gg and anti-TNF-aa in MS
`S Skurkovich et al
`
`comparison to placebo or to baseline level (data not
`shown).
`Results of the immunological studies are shown in
`Table 3. At baseline all patient groups had similar
`lymphocyte sub-population ratios and levels of cyto-
`kine production. The exceptions were an elevated
`CD4/CD8 ratio and percentage of CD29+-cells in the
`AB to IFN-g group and a percentage of CD16+-cells in
`the AB to TNF-a group. One month later a signi®cant
`decrease in levels of TNF-a, IFN-g, and IL-1b produc-
`tion as well as an increase in TGF-b production was
`seen in the group who received AB to IFN-g. Patients
`in this group also had a decrease in the CD4/CD8 ratio
`compared to baseline and an increased percentage of
`CD16+-cells compared to the placebo group (insignif-
`icant changes in cell subpopulations were excluded
`from the table). Neither patients in the AB to TNF-a
`group, nor patients who received a placebo had
`signi®cant
`changes
`in cytokine production. All
`changes in the group who received AB to IFN-g
`became less prominent 6 months later.
`
`Discussion
`
`This is one of the ®rst attempts to use antibodies to
`cytokines for the treatment of MS. In this preliminary
`study we have shown that patients who received
`antibodies to IFN-g had a signi®cant decrease in the
`production of several pro-in¯ammatory cytokines, and
`this was associated with a better course of the MS
`and with less activity on MRI 6 months after
`treatment. It is interesting that this effect was found
`only in patients who received antibodies to IFN-g, but
`not antibodies to TNF-a. Positive clinical changes at 6
`months were supported by MRI data showing a
`decrease in the number of active enhancing lesions.
`MRI has been suggested as a sensitive measure of the
`pathological process in MS. Serial MRI studies have
`shown that
`frequent asymptomatic disease activity
`occurs as evidenced by a much higher number of
`active lesions as revealed by MRI scans than clinical
`relapses.18,19
`
`Table 2 Characteristics of T2 and enhancing lesions on MRI in different treatment groups at baseline and 6 months later
`
`Characteristic
`
`AB to TNF-a
`(n=15)
`
`AB to IFN-g
`(n=15)
`
`Baseline
`Number of T2-lesions: mean+s.e. (range)
`
`Number of enhancing lesions: mean+s.e. (range)
`
`Six months later
`Number of T2-lesions: mean+s.e. (range)
`
`Number of enhancing lesions: mean+s.e. (range)
`
`Patients without new enhancing lesions
`
`16.27+2.07
`(7 ± 30)
`2.20+0.46
`(0 ± 6)
`
`17.07+2.29
`(6 ± 36)
`2.0+0.51
`(0 ± 6)
`8 (53.3%)
`
`18.93+3.02
`(6 ± 40)
`2.47+0.56
`(0 ± 7)
`
`19.20+2.69
`(5 ± 40)
`1.27+0.33*
`(0 ± 4)**
`13 (86.7%)*
`
`*Difference from placebo group with P50.05; **Difference from baseline with P50.05
`
`Placebo
`(n=15)
`
`14.73+2.51
`(4 ± 32)
`1.93+0.51
`(0 ± 8)
`
`16.60+3.04
`(3 ± 38)
`2.33+0.60
`(0 ± 9)
`6 (40%)
`
`Multiple Sclerosis
`
`5
`
`

`

`282
`
`Anti-IFN-gg and anti-TNF-aa in MS
`S Skurkovich et al
`
`Table 3 Immunological parameters in different treatment groups at baseline and 6 months later (insigni®cant changes in
`subpopulation levels were excluded)
`
`Variables
`
`AB to TNFa (n=15)
`
`AB to IFN-g (n=15)
`
`Placebo (n=15)
`
`Baseline
`CD3+ cells (%): mean+s.e. (range)
`CD4+ cells (%): mean+s.e. (range)
`CD8+ cells (%): mean+s.e. (range)
`CD4/CD8 ratio: mean+s.e. (range)
`CD29+ cells (%): mean+s.e. (range)
`CD16+ cells (%): mean+s.e. (range)
`TNF-a (pg/ml): mean+s.e. (range)
`TGF-b (pg/ml): mean+s.e. (range)
`IFN-g (pg/ml): mean+s.e. (range)
`IL1-b (ng/ml): mean+s.e. (range)
`IL2 (ng/ml): mean+s.e. (range)
`IL6 (ng/ml): mean+s.e. (range)
`One month
`CD4/CD8 ratio: mean+s.e. (range)
`CD16+ cells (%): mean+s.e. (range)
`TNF-a (pg/ml): mean+s.e. (range)
`TGF-b (pg/ml): mean+s.e. (range)
`IFN-g (pg/ml): mean+s.e. (range)
`IL1-b (ng/ml): mean+s.e. (range)
`IL2 (ng/ml): mean+s.e. (range)
`IL6 (ng/ml): mean+s.e. (range)
`Six months
`CD16+ cells (%): mean+s.e. (range)
`TNF-a (pg/ml): mean+s.e. (range)
`TGF-b (pg/ml): mean+s.e. (range)
`IFN-g (pg/ml): mean+s.e. (range)
`IL1-b (ng/ml): mean+s.e. (range)
`IL2 (ng/ml): mean+s.e. (range)
`IL6 (ng/ml): mean+s.e. (range)
`
`67.80+1.23 (59 ± 76)
`70.27+2.25 (55 ± 85)
`41.80+1.0 (36 ± 48)
`40.40+1.31 (33 ± 49)
`24.0+1.10 (18 ± 32)
`26.80+1.47 (20 ± 38)
`1.81+0.11 (1.1 ± 2.7)*
`1.57+0.06 (1.1 ± 2.1)
`7.07+0.42 (4 ± 10)*
`7.27+0.66 (4 ± 12)
`8.40+0.89 (4 ± 17)
`8.40+0.86 (3 ± 14)
`1109.4+106.7 (357 ± 1812) 1244.6+139.6 (345 ± 2490)
`488.33+53.98 (207 ± 879)
`402.47+51.39 (120 ± 750)
`674.53+145.5 (112 ± 2351)
`710.8+127.37 (72 ± 1497)
`1.80+0.04 (1.63 ± 2.0)
`1.75+0.04 (1.55 ± 1.90)
`21.13+0.57 (18.49 ± 23.64) 21.68+0.71 (19.7 ± 26.7)
`14.45+1.03 (9.82 ± 19.0)
`14.87+0.89 (10.3 ± 19.1)
`
`68.27+1.24 (60 ± 78)
`38.60+0.95 (32 ± 45)
`24.67+1.58 (15 ± 38)
`1.62+0.10 (0.9 ± 2.1)
`8.53+0.79 (5 ± 16)
`7.73+0.54 (5 ± 12)
`1105.3+256.1 (182 ± 3800)
`456.47+64.63 (120 ± 939)
`642.6+229.26 (90 ± 3514)
`1.81+0.04 (1.67 ± 2.11)
`21.37+0.47 (19.34 ± 24.12)
`15.48+0.54 (12.3 ± 18.11)
`
`1.64+0.12 (1.1 ± 2.9)
`1.62+0.05 (1.3 ± 2.0)**
`1.66+0.09 (0.9 ± 2.0)
`6.87+0.38 (5 ± 10)
`9.6+1.06 (5 ± 19)
`7.73+0.61 (5 ± 14)
`875.6+93.1* (209 ± 1670)** 1285.5+143.0 (364 ± 2000)
`1206.3+137.8 (216 ± 2160)
`574.13+71.4* (137 ± 1015)** 407.47+86.2 (112 ± 1230)
`506.2+77.89 (99 ± 1180)
`772.6+123.95 (150 ± 1992) 499.3+98.98* (58 ± 1250)** 816.2+206.52 (84 ± 2766)
`1.8+0.03 (1.64 ± 1.98)
`1.65+0.03* (1.51 ± 1.84)**
`1.84+0.04 (1.66 ± 2.06)
`21.59+0.58 (18.0 ± 24.1)
`22.09+0.67 (20.31 ± 26.8)
`21.22+0.56 (18.45 ± 24.0)
`14.29+0.92 (9.91 ± 18.0)
`15.81+0.91 (10.1 ± 18.9)
`14.97+0.6 (12.3 ± 18.3)
`
`9.67+0.70 (6 ± 14)*
`8.53+0.74 (5 ± 16)
`1161.3+82.39 (568 ± 1870) 1147.7+161.2 (457 ± 2810)
`541.13+76.69 (125 ± 1366) 476.88+65.97 (59 ± 952)
`757.33+101.2 (54 ± 1560)
`605.67+88.7 (120 ± 1860)
`1.77+0.04 (1.49 ± 1.92)
`1.75+0.04 (1.6 ± 1.92)
`21.5+0.46 (18.93 ± 24.2)
`21.79+0.60 (19.5 ± 24.3)
`14.40+0.70 (11.59 ± 17.4)
`15.0+0.99 (10.89 ± 19.8)
`
`7.47+0.35 (5 ± 10)
`1322.4+120.3 (780 ± 2270)
`528.53+91.51 (156 ± 1148)
`851.3+121.46 (334 ± 2060)
`1.82+0.03 (1.68 ± 1.89)
`21.75+0.50 (18.99 ± 23.89)
`14.96+0.47 (13.11 ± 17.83)
`
`*Difference from placebo group with P50.05; **Difference from baseline with P50.05
`
`In our pilot study, the primary and two secondary
`outcome measures reached signi®cant difference with
`placebo even with a small total number of patients.
`These positive clinical results of a short course of AB
`to IFN-g may be related to the patient selection criteria,
`whereby only active early progressive patients were
`included. It means that within the group of secondary
`progressive MS patients, a subgroup of patients
`sensitive to anti-cytokine and other immunomodula-
`tive treatment are still present. The high percentage of
`such patients among secondary progressive cases
`might be one of the reasons for positive results in
`the IFN-b-1b trial in Europe.5 The clear de®nition of
`this subgroup, close to relapsing 7 remitting or remit-
`is not easy and additional
`ting ± relapsing MS,20
`clinical and MRI markers could be useful.
`In our
`study the short duration of the progressive stage, the
`presence of
`relapses during progression, and the
`numerous active lesions on T1-scans appeared to be
`suitable for such purpose, but a large further study is
`necessary to con®rm this.
`Changes in cytokine production may foreshadow
`changes in MS course. This was shown both for
`relapsing ± remitting and secondary progressive MS.21 ±
`IFN-g and TNF-a in the
`23 The concentration of
`
`supernatants of activated cells and the levels of these
`cytokine messenger RNA in cells were proposed as
`markers of active in¯ammation and demyelination.20
`In our study the decrease in IFN-g and TNF-a
`production 1 month after the short course of AB to
`IFN-g were accompanied by an increase in TGF-b and
`a decrease in IL-1b production.
`IL-1 is the main
`cytokine triggering the process of immune activation.
`High production of this cytokine in active MS has also
`been found.24 TGF-b is an anti-in¯ammatory cytokine.
`Its production increased during the regression of
`exacerbations of MS.25,26 Changes in all these cytokine
`levels re¯ect the systemic change in immunoregula-
`tion. The patients with active secondary progressive
`MS may have mostly unfavorable changes in cytokine
`status27 and may positively react even to a short course
`of AB to IFN-g.
`The future of anti-cytokine therapy in MS has been
`widely discussed. It is generally known that admin-
`IFN-g provokes MS attacks.28 This has
`istration of
`been attributed to a direct stimulation of lymphocyte
`proliferation, which leads
`to in¯ammation and
`demyelination. On endothelial cells,
`for example,
`IFN-g can increase lymphocyte transmigration through
`the blood±brain barrier and MHC II-expression at the
`
`Multiple Sclerosis
`
`6
`
`

`

`sites of active demyelination in chronic active MS
`plaques.29 All
`the new strategies for
`treating MS
`include reducing IFN-g activity in different ways,
`the most accepted one by administration of IFN-b.
`The strategy of IFN-g neutralization is less popular in
`part because of the theoretical risk of reactivation of
`potential virus infections. But it can be argued that
`neutralization may only remove hyperproduced IFN-g,
`not stopping its production in the cells. Furthermore,
`IFN-g removal still leaves other antiviral interferons.
`In some patients with active MS, this strategy may be
`reasonable, as was proposed for the blocking of TNF-
`a. However, we did not ®nd any changes in disease
`course in patients given antibodies to TNF-a.
`In
`preliminary testing we also found that patients given
`a combination of antibodies to IFN-g and to TNF-a
`than anti-IFN-g alone. Lack of
`did not do better
`improvement from anti-TNF-a therapy agrees with the
`of TNF-a modulating
`results
`of
`several
`trials
`agents,30,31 though such an approach was effective in
`treating the MS animal model of experimental allergic
`encephalitis (EAE).32 At the same time, we did not see
`a dramatic increase in disease activity after use of
`anti-TNF-a antibodies as in the Lenercept MS Study,28
`nor any severe deterioration in MRI as in Van Oosten
`et al.27 The majority of our patients continued to
`progress at the rate they had before. This could be
`explained by: (a) low activity of antibodies used or (b)
`speci®c selection of patients. TNF-a is one of the most
`pleiotropic cytokines and acts in a complex way. Its
`real role in MS is still unknown. A very severe EAE
`was induced in mice lacking TNF following immuni-
`zation with myelin oligodendrocyte
`glycoprotein
`(MOG).33 This is not surprising given the importance
`of this cytokine in apoptosis of in¯ammatory cells
`and their migration to the site of in¯ammation in the
`brain. TNF may inhibit IL-12 production, triggering
`production of IFN-g.34
`the current results should be
`The limitations of
`noted. We used a very short course of polyclonal
`antibodies. The course was well tolerated, but further
`use of these antibodies was limited due to the risk of
`allergic reactions and serum sickness. The true
`magnitude of clinical and MRI changes could not be
`determined given the short duration of treatment and
`relatively short follow-up. Levels of cytokine produc-
`tion varied greatly in each group of patients. Finally,
`there was a subgroup of patients who did not respond
`to the antibodies to IFN-g and still maintained a very
`active course of the disease.
`This study has demonstrated that even a short
`course of IFN-g antibodies may bring changes in the
`cytokine pro®le associated with positive changes in
`MS course and MRI ®ndings. Future trials with
`chimeric, humanized, or all-human monoclonal anti-
`bodies are needed. They could determine more
`accurately the magnitude and duration of the clinical
`and MRI effect of antibodies to IFN-g on MS. The next
`step may be to test the use of IFN-b combined with
`anti-IFN-g, both of which act to block the action of
`IFN-g, for a possible stronger, more lasting effect.
`
`Anti-IFN-gg and anti-TNF-aa in MS
`S Skurkovich et al
`
`References
`
`283
`
`1 Link H. (1998) The cytokine storm in multiple sclerosis.
`Multiple Sclerosis 4: 12 ± 15.
`2 Skurkovich SV, Klinova EG, Eremkina EI, Levina NV.
`(1974) Immunosuppressive effect of an anti-interferon
`serum. Nature 22: 551 ± 552.
`3 Skurkovich SV, Eremkina EI. (1975) Probable role of
`interferon in allergy. Ann Allergy 35: 356 ± 360.
`4 Skurkovich SV et al. (1977) Lymphocytes' cytotoxicity
`towards cells of human lymphoblastoid lines in patients
`with rheumatoid arthritis and systemic lupus erythema-
`tosus. Ann Allergy 39: 344 ± 350.
`5 Chof¯on M. (2000) Recombinant human interferon beta
`in relapsing-remitting multiple sclerosis: a review of the
`major clinical trials. Eur J Neurol 7: 369 ± 380.
`6 Kappos L, and the European Study Group on interferon
`beta-1b in secondary progressive MS. (1998) Placebo
`controlled multicentre randomised trial of interferon
`beta-1b in treatment of secondary progressive multiple
`sclerosis. Lancet 352: 1498 ± 1504.
`7 Cavallo MG, Pozzilli P, Thorpe R. (1994) Cytokines and
`autoimmunity. Clin Exp Immunol 96: 1 ± 7.
`8 Christensen T, Sorensen PD, Hansen HJ, Moller-Larsen
`A. (2001) Endogenous retroviruses in multiple sclerosis.
`Ugeskr Laeger 163: 297 ± 301.
`9 Karlsson H et al. (2001) Retroviral RNA identi®ed in the
`cerebrospinal ¯uids and brains of
`individuals with
`schizophrenia. Proc Natl Acad Sci USA 98: 4634 ± 4639.
`10 Revel M et al. (1995) Antagonism of interferon beta on
`interferon gamma: inhibition of signal transduction in
`vitro and reduction of serum levels in multiple sclerosis
`patients. Multiple Sclerosis 1 (Suppl): S5 ± S11.
`11 Gusev EI et al. (1997) Anticytokine therapy of secondary
`progressive MS. Nevrologicheskiy Vestnik 29: 55 ± 58.
`12 Poser CM et al.
`(1983) New diagnostic criteria for
`multiple sclerosis: guidelines for research protocols.
`Ann Neurol 13: 227 ± 231.
`13 Deomina T et al. (1997) A three-month pilot trial of
`antibodies to cytokines (Interferon-alpha,
`Interferon-
`gamma, TNF-alpha) in multiple sclerosis. Eur J Neurol
`4 (Suppl.1): S62 ± S63.
`14 Miller DH et al. (1991) Magnetic resonance imaging in
`monitoring the treatment of multiple sclerosis: concerted
`action guidelines. J Neurol Neurosurg Psychiatry 54:
`683 ± 688.
`15 Comi G et al. (1989) Correlation between multimodal
`evoked potentials and magnetic resonance imaging in
`multiple sclerosis. J Neurol 236: 4 ± 8.
`16 Giesser BS et al. (1987) Trimodal evoked potentials
`compared with magnetic resonance imaging in the
`diagnosis of multiple sclerosis. Arch Neurol 44: 281 ±
`284.
`17 Gusev EI, Demina TL, Boiko AN, Pinegin BV. (1994)
`Prolonged dynamical clinico-immunological observa-
`tion of 85 patients with de®nite multiple sclerosis. J
`Neurol 8: 500 ± 510.
`18 Stone LA et al.
`(1995) Blood-barrier disruption on
`contrast-enhanced MRI in patients with mild relapsing-
`remitting multiple sclerosis. Neurology 45: 1122 ± 1126.
`19 Khoury SJ, Weiner HL. (1998) Multiple sclerosis. What
`have we learned from magnetic resonance i