Intravenous immunoglobulin therapy in multifocal motor neuropathy

8 downloads 0 Views 128KB Size Report
chronic inflammatory demyelinating ... inflammatory demyelinating polyneuropathy (CIDP) (Lewis .... therapy over a 6-month period would result in progressive.
Brain (2001), 124, 145–153

Intravenous immunoglobulin therapy in multifocal motor neuropathy A double-blind, placebo-controlled study Jean-Marc Le´ger,1,2 Be´ne´dicte Chassande,1 Lucile Musset,3 Vincent Meininger,1 Pierre Bouche2 and Nicole Baumann4 1Fe ´ de´ration

de Neurologie Mazarin, 2Service d’Explorations Fonctionnelles-Neurologie, 3Laboratoire d’Immunochimie, 4INSERM U495, Groupe Neuropathies Pe´riphe´riques Pitie´-Salpeˆtrie`re (GNPS), Hoˆpital de la Salpeˆtrie`re, Paris, France

Correspondence to: Dr J.-M. Le´ger, Fe´de´ration de Neurologie Mazarin, Hoˆpital de la Salpeˆtrie`re, 47 boulevard de l’Hoˆpital, 75651 Paris cedex 13, France E-mail: [email protected]

Summary We conducted a double-blind, placebo-controlled, study of 19 patients fulfilling eligibility criteria for multifocal motor neuropathy with persistent conduction block. They were enrolled and divided into two groups: those who had never been treated previously with intravenous immunoglobulins (IVIg) (Group 1: 10 patients) and those who presented recurrent symptoms after previously successful treatment with IVIg (Group 2: nine patients). They were randomized prospectively to receive either IVIg or placebo at a dose of 500 mg/kg/day for 5 consecutive days, once a month for 3 months. At month 4, patients found to be responders remained on the same treatment for the 3 following months, while nonresponders were switched to the alternative study drug for the 3 following months. Clinical assessment was conducted with the MRC score in 28 muscles and a selfevaluation scale (five daily motor activities scored from 0 to 5). In Group 1, nine patients completed the study, of whom initially four received IVIg and five placebo; four patients responded to IVIg (two at months 4 and 7, and a further two at month 7 after switching treatment at month 4), two patients responded to placebo at months 4

and 7, and three patients did not respond to either treatment. In Group 2, nine patients completed the study. Five patients first received IVIg and all responded at months 4 and 7. Four patients first received placebo and none responded at month 4; all were then switched to IVIg and three responded at month 7. When the 18 patients were considered together, seven out of the nine patients who received IVIg first were responders at month 4, compared with two of the nine patients who received placebo first, a difference that was statistically significant (P ⍧ 0.03). On the other hand, there was no significant difference in MRC score but a significant difference in the self-evaluation score, at month 4, between IVIg patients and placebo patients. Electrophysiological studies did not show significant differences at month 4 in motor parameters between IVIg patients and placebo patients. IgM anti-GM1 titres did not change significantly in patients treated with IVIg compared with those who received placebo, between baseline, month 4 and month 7. However, of five patients who had significantly high anti-GM1 titres (>3200) at baseline, four responded to IVIg. This trial confirms that IVIg is a promising therapeutic option for multifocal motor neuropathy.

Keywords: multifocal motor neuropathy; double-blind trial; intravenous immunoglobulin; conduction block; IgM anti-GM1 antibody Abbreviations: CB ⫽ conduction block; CI ⫽ confidence interval; CIDP ⫽ chronic inflammatory demyelinating polyneuropathy; ELISA ⫽ enzyme-linked immunosorbent assay; IVIg ⫽ intravenous immunoglobulin; MMN ⫽ multifocal motor neuropathy; MRC ⫽ Medical Research Council

Introduction Multifocal sensorimotor neuropathy with persistent conduction block was described in 1982 by Lewis and colleagues on the basis of clinical and electrophysiological criteria that were distinct from those of common chronic © Oxford University Press 2001

inflammatory demyelinating polyneuropathy (CIDP) (Lewis et al., 1982). Multifocal motor neuropathy with persistent conduction block (MMN) was first reported by Roth and colleagues (Roth et al., 1986) in Switzerland and by Parry

146

Jean-Marc Le´ ger et al.

and Clarke (Parry and Clarke, 1988) in the USA. Patients with MMN have a stereotyped clinical picture (Pestronk et al., 1988; Van den Bergh et al., 1989; Sadiq et al.,1990; Lange et al., 1992; Bouche et al., 1995; Le´ ger, 1995; NobileOrazio, 1996; Le Forestier et al., 1997) comprising a chronic asymmetrical motor syndrome, usually starting in a distal upper limb and remaining prominently in the arms. Age at onset is generally between 20 and 75 years, and men are more frequently affected than women. The motor deficit involves individual motor nerves, and is frequently accompanied by cramps and fasciculations. Motor findings include asymmetrical, predominantly distal weakness and various degrees of atrophy. Some patients report paraesthesias, but sensory signs are usually absent or clinically insignificant. In regions with normal strength, tendon reflexes are often preserved. In areas of weakness, reflexes may initially be normal but can become reduced with progression of the disease. Electrodiagnostic studies may reveal conduction block (CB) at any level of the motor nerves. The reduction in the compound action potential between proximal and distal stimulation may be as much as 80%, and is usually accompanied by slowing of motor nerve conduction velocities, restricted to corresponding segments of nerves. The diagnostic value of CB is greater when the change is focal and is in a distal nerve segment other than the regions of entrapment. Motor nerve conduction velocities and distal latencies are often little changed in regions without CB. Sensory studies in these patients are normal. IgM anti-GM1 antibodies are found in 30–80% of cases (Pestronk et al., 1988; Willison, 1994; Bouche et al., 1995; Le Forestier et al., 1997). The rare pathological studies that have been reported suggest a demyelinating process (Kaji et al., 1993; Bouche et al., 1995). The course of the disease is slowly progressive, and there may be involvement of other motor nerves in the upper and then the lower limbs, leading to varying degrees of motor deterioration. Recent reports have stated that high doses of intravenous immunoglobulins (IVIg) can be an effective treatment in patients with MMN in 60–80% of cases, but most of these analyses were retrospective, uncontrolled and non-blinded (Kaji et al., 1992; Chaudhry et al., 1993; Nobile-Orazio et al., 1993; Comi et al., 1994; Le´ ger et al., 1994). The only prospective, double-blind, placebo-controlled studies (Azulay et al., 1994; Van den Berg et al., 1995) were conducted in a small group of patients and with one or two IVIg/placebo infusions. The results of other studies have tended to show that, after several months, the efficacy of IVIg is no longer maintained, even when the infusions are repeated at the same doses (Azulay et al., 1997; Meucci et al., 1997; Van den Berg et al., 1998). In addition, the electrophysiological outcome of CB and the predictive relevance of anti-GM1 antibodies in responding patients need to be better defined. At present there is no therapeutic alternative to IVIg therapy, and it seemed important to know whether repetition of IVIg therapy over a 6-month period would result in progressive

improvement of the neuropathy. Should its effectiveness be proven, IVIg could be recommended as the treatment of choice. IVIgs have been shown to have a beneficial effect in other immune-mediated polyneuropathies, including Guillain– Barre´ syndrome and CIDP (van der Me´ che´ and van Doorn, 1997; Le´ ger, 1999), suggesting that a study of IVIg in MMN would be useful in order to improve our understanding of the effects of IVIg treatment on clinical symptoms, the functional and electrophysiological deficit, and immunological status. The aim of the present study was to ascertain the efficacy of IVIg in MMN in a randomized double-blind, placebo-controlled study. The secondary objectives were to determine changes in electrophysiology and IgM anti-GM1 antibody titres.

Patients and methods Patients Patients were selected on the clinical and electrophysiological diagnostic criteria of MMN used elsewhere (Le´ ger et al., 1994; Bouche et al., 1995). Nineteen patients were enrolled between December 1995 and October 1997 and were divided into two groups: those who had never been treated with IVIg (Group 1) and those who presented recurrent symptoms after previously successful treatment with IVIg (Group 2: patients who were responders to IVIg in open treatment, according to criteria defined below). For newly diagnosed patients, the neuropathy was continuously progressive for ⬎2 months; cases with a previously established diagnosis of MMN were static or showed recent deterioration. Patients had not received immunosuppressors or IVIg during the 2 and 3 months, respectively, before inclusion. Patients were excluded if they had a severe concurrent medical condition that might cause neuropathy or interfere with the treatment, were pregnant or were less than 18 years of age. Prior to the study, our institution’s ethics committee approved the protocol and all patients gave written informed consent.

Study design We conducted a double-blind, randomized, placebo-controlled study. The patients were randomly assigned to receive either IVIg (Endobulin, 5% lyophilized; Baxter, Vienna, Austria) at 500 mg/kg/day for 5 consecutive days, or placebo (1% human albumin, lyophilized) once a month for 3 months. At month 4 (after three treatment periods), all patients had a double-blind clinical evaluation. Those who were found to be responders (according to the criteria defined below) remained on the same treatment for 3 more months, and those who were considered to be non-responders were switched to the alternative study drug for the 3 following months. Final evaluation was conducted at month 7. Adverse experiences

Immunotherapy in multifocal motor neuropathy were monitored throughout the study and were periodically reviewed by the Steering Committee.

Neurological assessments All patients were assessed by the same blinded observers. Clinical evaluation was conducted at baseline and every month for 7 months, as follows: (i) MRC score in 28 muscles (7 ⫻ 2 in the upper limbs, 5 ⫻ 2 in the lower limbs and 4 in the neck); (ii) self-evaluation scale, scored from 0 (normal) to 5 (impossible) for five motor activities of daily life, chosen for each patient together with the examiner at the baseline. At month 4, patients were considered as responders if they had at least 1 more MRC point in two affected muscles plus 1 point less in two activities of daily life compared with baseline. The same criteria were used to determine a successful response in patients of Group 2. Each assessment was made and recorded independently; the earlier records were not made available to the evaluating neurologist until the trial period was complete. Inter-observer agreement was tested before the study began and was found to be very good.

Electrophysiological studies Electrophysiological studies were conducted every month for 7 months in all patients using a standardized set of electrophysiological measurements. The measurement of motor nerve conduction velocities was made in six motor nerves (median, ulnar and peroneal bilaterally), including distal and proximal compound muscle action potential area/ amplitude, distal latency, forearm conduction velocity and F-wave latency. CB were defined as previously (Bouche et al., 1995). Sensory potentials were also assessed in six sensory nerves (median, radial and sural bilaterally). Needle electromyography of proximal (biceps, vastus medialis) and distal (first dorsal interosseous, tibialis anterior) muscles was carried out to evaluate the presence of fibrillation potentials and positive sharp waves at rest, and the motor unit configuration and recruitment.

Anti-GM1 antibody titres IgM anti-GM1 antibody titres were determined every month for 7 months in all patients. They were measured by enzymelinked immunosorbent assay (ELISA) according to technique B, as described previously (Bouche et al., 1995), with slight modifications. Briefly, the protocol was as follows: wells of Dynatech Immulon 2 plates were coated with 500 ng GM1 ganglioside (Sigma, reference G-7641) in 100 µl of 100% ethanol; the latter was added without GM1 to control wells to test for non-specific binding. The content of each well was evaporated to dryness at room temperature. All 96 wells were then blocked with 200 µl of 1% PBS–BSA, pH 7.4 (0.1 M phosphate-buffered saline; Sigma, reference P-4417; bovine serum albumin; Sigma, reference A-7906) and left overnight at 4°C. The sera were diluted in 1% PBS–BSA

147

and serially tested and incubated overnight at 4°C. The wells were then washed six times with PBS cooled to 4°C. After washing in PBS, peroxidase-conjugated rabbit antibodies to human µ chain (Dako reference P-0215) were added at dilution 1: 1000 in 1% PBS–BSA. The wells were left for 1 h at room temperature. Reaction products were visualized with O-phenylenediamine (Sigma, reference P-8412). After having been left 10 min in the dark, 50 µl of 3 M HCl was added to stop the reaction. The optical density (OD) of the reaction products was read immediately in an automatic plate reader (Dynatech MR 5000) and the titre was calculated by end-point analysis. Linear OD values were plotted against log test serum dilution as semi-log, and a best dilution curve was drawn. We considered as abnormal a titre greater than a value equivalent to the mean plus 2 SD observed in normal controls. The cut-off point corresponded to OD 0.230 and a mean titre of 800. However, as titres of 800 have been found in controls and other neurological diseases (Sadiq et al., 1990), we considered titres of 艌3200 clearly abnormal. Positive tests were always confirmed by showing the presence of IgM antiGM1 antibodies by immunodetection on silica gel chromatography after separation of gangliosides, according to the technique used in a previous study (Ben Younes-Chennoufi et al., 1992). The ELISA results presented in Table 5 were obtained with an improved ELISA method performed after the closure of the study database. For IgG anti-GM1 antibodies, the ELISA technique of Willison and colleagues (Willison et al., 1999) was used.

Laboratory studies Serum protein electrophoresis, serum immune electrophoresis, IgA, IgM and IgG subclasses, serum glucose, electrolytes, urea, creatinine, albumin, liver biochemistry profile, titres of antibodies directed against other gangliosides and glycolipids, quantitation of cryoglobulin in blood plasma and viral markers for HBV (hepatitis B virus), HCV (hepatitis C virus), HIV 1 and 2 (human immunodeficiency viruses) were performed at entry into the study and repeated once a month for 7 months. Renal function tests were performed after the third and sixth infusions of IVIg or placebo.

Statistical analysis Patients who had received IVIg as the first treatment (IVIg patients) and patients who had received placebo as the first treatment (placebo patients) were compared. The primary end-point was the difference in MRC score between baseline and month 4. The secondary end-points were the difference between baseline and month 4 in the self-evaluation scale, the electrophysiological parameters and anti-GM1 titres. These parameters were compared between IVIg and placebo patients. The changes in the MRC score and self-evaluation scale were calculated as medians with 95% confidence intervals

148

Jean-Marc Le´ ger et al.

Fig. 1 Trial profile. Patients in Group 2 had been treated with IVIg previously; those in Group 1 had not.

(CI) within each category of patients, and between IVIg patients and placebo patients. Changes were considered significant if the CI did not include zero. The clinical response rate was compared by Fisher’s exact test. For the electrophysiological study, changes in the amplitude of distal muscle compound action potentials (in mV), distal latencies (in ms) and F-waves (in ms) were compared between IVIg and placebo patients. Median differences and 95% CI for median differences were calculated for both sides in four motor nerves (median, ulnar, peroneal, tibial) to give a total of 24 estimates. CI were not adjusted for multiplicity. We additionally tried to evaluate the effect of the treatments on the severity of the CB. We considered as significant a reduction, for at least one CB, of both amplitude and area ratio of 15% in the distal segments and 25% in the proximal segments.

Results General results (Fig. 1) Nineteen patients (six women and 13 men), with a mean age of 54.6 years (range 40–65 years), were randomized for the trial. All patients had been shown by prior clinical and electrophysiological examinations to be suffering from a

multifocal motor neuropathy (mean duration of disease 9.04 years, range 1.3–31.7 years). The diagnosis of MMN was confirmed by typical findings on electrophysiological examination in all cases. All 19 patients had involvement of the upper limbs (bilateral involvement in eight patients) and nine of the 19 patients had involvement of the lower limbs (bilateral involvement in six). None had cranial nerve involvement or other neurological signs. Ten patients (five women and five men, aged 40–64 years) were included in Group 1 and nine patients (one woman and eight men aged 45–65 years) in Group 2. None of the patients in Group 1 had received prior treatment for MMN. All patients in Group 2 had received IVIg and four of them had previously been prescribed immunosuppressors (chlorambucil in Patient 12 and oral cyclophosphamide in Patients 15, 16 and 17). Nine of the 10 patients in Group 1 completed the study, one patient being lost to follow-up for personal reasons. All nine patients in Group 2 completed the study.

Overall response to the treatment In Group 1, four patients (Patients 2, 3, 7 and 8) received IVIg first: two showed improvement at both month 4 and

Immunotherapy in multifocal motor neuropathy Table 1 Baseline characteristics of placebo patients and IVIg patients

Table 2 Changes in MRC scores between baseline and month 4

Characteristic

MRC score

Placebo

IVIg

(n ⫽ 9)

(n ⫽ 10)

Sex (M/F) 6/3 Age (years)* 51.9 ⫾ 6.8 Disease duration (years)* 8.2 ⫾ 5.6 MRC score* 112.9 ⫾ 19.2 Self evaluation score* 18.7 ⫾ 2.9

7/3 57.1 ⫾ 6.6 9.8 ⫾ 8.7 118.4 ⫾ 11.2 19.1 ⫾ 2.0

P value 1 0.12 0.63 0.45 0.71

*Mean ⫾ SD.

month 7 (Patients 3 and 7) and two did not respond at month 4 (Patients 2 and 8). The latter two patients were switched to placebo and did not respond at month 7. The five other patients in Group 1 (Patients 1, 4, 5, 9 and 10) received placebo as the first treatment: two responded at months 4 and 7 (Patients 1 and 9) and three did not respond at month 4 (Patients 4, 5 and 10). The latter three patients were switched to IVIg, and two of them (Patients 4 and 5) responded at month 7. In summary, four patients in Group 1 responded to IVIg, two responded to placebo and three did not respond to either treatment. In Group 2, five patients (Patients 13, 15, 16, 17 and 19) received IVIg as the first treatment: all 5 responded at both month 4 and month 7. The other four patients (Patients 11, 12, 14 and 18) received placebo as the first treatment: none responded at month 4, and they were all switched to IVIg. Three of these responded and one remained stable at month 7. In summary, eight patients in Group 2 improved on IVIg and one remained the same. No patient responded to placebo. When the 18 patients in the two groups were considered together, seven out of the nine patients who received IVIg as the first treatment (IVIg patients) were responders at month 4 versus two out of the nine patients who received placebo as the first treatment (placebo patients), a difference that was statistically significant (P ⫽ 0.03).

Comparison between IVIg and placebo patients Table 1 shows the characteristics at baseline of patients who received IVIg as the first treatment (IVIg patients) and patients who received placebo as the first treatment (placebo patients), irrespective of group. The mean age at the time of the study, the mean duration of the disease, the mean overall MRC score (maximum 140) and the mean self-evaluation score (maximum 25) did not differ significantly between IVIg patients and placebo patients. Table 2 shows the difference in MRC score in the two categories of patients between baseline and month 4: the median change was significant in IVIg patients, but did not differ significantly in placebo patients or between IVIg patients and placebo patients. Table 3 shows the difference in self-evaluation score in the two categories of patients between baseline and month 4. The median change differed significantly in IVIg patients

Baseline Median change at month 4 Confidence interval Significance

Placebo (n ⫽ 9) 112.9 ⫾ 19.2 3 –1, 10 NS

IVIg (n ⫽ 9) 118.4 ⫾ 11.2 3 2, 13 S

149

Difference

2 –2, 9 NS

NS ⫽ not significant; S ⫽ significant.

Table 3 Changes in self-evaluation scores between baseline and month 4 Self-evaluation score Baseline Median change at month 4 Confidence interval Significance

Placebo 18.7 ⫾ 2.9 0 –3, 3 NS

IVIg 19.1 ⫾ 2 –7 –14, –1 S

Difference -7 –12, –3 S

NS ⫽ not significant; S ⫽ significant.

and between IVIg patients and placebo patients, but there was no statistically significant difference in placebo patients.

Electrophysiological data Median differences between right- and left-side motor nerves were small and inconsistent (Table 4). High variability and small sample sizes resulted in only one out of 20 CI that did not include zero. For four nerves (right and left median, left ulnar, left peroneal nerves), not enough F-wave measurements were available to calculate 95% CI. In summary, the analyses did not show differences in any of these variables between IVIg patients and placebo patients. When comparing the amplitude/surface area ratio of distal and proximal motor evoked responses, four out of nine IVIg patients (Patients 7, 16, 17 and 19) had a reduction in the severity of CB (as defined above) at month 4, versus two out of nine placebo patients (Patients 1 and 5), which was not statistically significant. In addition, four other patients who received IVIg between month 4 and month 7 (Patients 4, 14, 11 and 15) had a reduction in the severity of CB at month 7.

Anti-GM1 antibodies (Table 5) Five out of 18 patients had significantly abnormal titres of IgM anti-GM1 antibodies at baseline (Patients 4, 5, 13, 16 and 18). The changes in titre between baseline, month 4 and month 7 were not significantly different in patients treated with IVIg and those who received placebo. For example, in Patients 1 and 9, who received only placebo, titres remained low. Patients 3, 7, 15, 17 and 19, whose overall response improved with IVIg, did not have significantly higher than

150

Jean-Marc Le´ ger et al.

Table 4 Electrophysiological parameters: median differences between baseline and month 4 Nerve

Latency (ms)

Median right Median left Ulnar right Ulnar left Peroneal right Peroneal left Tibial right Tibial left

Amplitude (mV)

F-wave latency (ms)

Median change

CI (range)

Median change

CI (range)

Median change

CI (range)

0 0 0.4 –0.1 –0.1 –0.3 –0.7 0.2

–1.8 to 1.0 –0.7 to 0.7 0.3 to 0.8 –1.1 to 0.5 –1.3 to 1.1 –1.1 to 0.5 –2.2 to 0.6 –1.5 to 1.7

0.5 0.3 1.0 0.0 –0.5 0.4 0.3 –0.8

–0.2 to 3.0 –1.4 to 3.4 –0.7 to 2.7 –1.1 to 2.6 –1.7 to 0.9 –1.1 to 2.0 –1.4 to 1.9 –3.5 to 1.3

–3.0 –3.0 –4.5 1.0 –3.0 –1.5 1.5 –2.5

NA NA –10.5 to –1.5* NA –9.0 to 28.1 NA –9.0 to 12.0 –5.0 to 6.0

CI ⫽ confidence interval; NA ⫽ not available. *Significant.

Table 5 Evolution of IgM anti-GM1 antibody titre during the IgIV trial Patient

Treatment

Month 1

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

Placebo/placebo IVIg/placebo IVIg/IVIg Placebo/IVIg Placebo/IVIg IVIg/– IVIg/IVIg IVIg/placebo Placebo/placebo Placebo/IVIg Placebo/IVIg Placebo/IVIg IVIg/IVIg Placebo/IVIg IVIg/IVIg IVIg/IVIg IVIg/IVIg Placebo/IVIg IVIg/IVIg

2

501 458 udt udt 1096 501 16 595 4786 5248 363 Lost to follow-up 660 549 151 263 158 109 1380 1380 831 udt 151 udt 95 499 120 226 udt udt 1380 1047 3200 275 251 229 8317 10 964 udt udt

3

4

5

6

7

363 udt 363 10 964 3464

436 udt 229 9772 10 000

1096 331 1819 50 118 7244

3162 1148 912 39 810 4168

2630 600 1905 36 307 6400

478 229 263 954 udt udt 63 095 udt 363 524 131 5248 udt

1148 831 251 2884 udt udt 66 069 udt 758 3981 316 5495 udt

691 575 275 3465 301 udt 69 183 udt 1148 3801 udt 3467 udt

954 208 319 1820 301 udt 151 356 udt 2398 5754 udt 5754 udt

1122 478 691 6367 229 udt 95 449 udt 660 1380 udt 16 595 udt

The titre was calculated by end-point analysis. For details, see Patients and methods. udt ⫽ undetectable.

normal IgM anti-GM1 antibody titres at baseline and titres were still low at months 4 and 7. Conversely, Patients 4 and 5, who did not respond to placebo as the first treatment but responded subsequently to IVIg, and Patients 13 and 16, who responded to IVIg as the first treatment, had significantly high titres at baseline (⬎3200). However, in these four cases the titres were not significantly correlated with clinical improvement. Patient 10, who had low titres at baseline, did not respond to placebo or to IVIg after switching, and his titres were significantly higher than normal at months 5 and 7. No IgG anti-GM1 antibodies were found in any patient.

Adverse events (Table 6) Some minor drug-related adverse events were noted in the two groups. Most of them occurred with IVIg treatment.

None of these events led to the withdrawal of therapy or to the unblinding of any patient’s file.

Discussion Patients with MMN CB do not respond to steroids or plasma exchanges, which may worsen the motor deficit in some cases (Donaghy et al., 1994). This constitutes a major distinction from other chronic immune-mediated polyneuropathies, such as CIDP (Hahn et al., 1996; van der Me´ che´ and van Doorn, 1997). On the other hand, several open studies (Kaji et al., 1992; Chaudhry et al., 1993; NobileOrazio et al., 1993; Comi et al., 1994; Le´ ger et al., 1994) have shown that IVIg infusions, given at 0.4 g/kg/day for 5 consecutive days, are followed by an early and often dramatic response, mainly consisting of an improvement in the motor

Immunotherapy in multifocal motor neuropathy Table 6 Adverse events in IVIg and placebo patients Adverse event

IVIg

Placebo

Headache Flushing Shivering Fever Cold feet Visual blur Eczema

3 1 2 1 0 2 1

0 0 0 0 1 0 0

deficit in the affected nerves and sometimes of a reduction in the CB in electrophysiological studies. On the other hand, IVIg treatment does not affect IgM anti-GM1 antibody titres. Few controlled studies have been conducted with IVIg in MMN. Azulay and colleagues gave a course of IVIg infusions or placebo, followed by the other treatment after cross-over, in 12 patients with motor neuropathy and high IgM antiGM1 titres, of whom only five had CB. They found that the only patients with CB responded to IVIg (Azulay et al., 1994). Van den Berg and colleagues published a doubleblind, placebo-controlled study in six patients with MMN, all responders to IVIg in open treatment: four patients received two infusions of either IVIg (0.4 g/kg/day) or placebo, and the two remaining patients received one infusion. A significant clinical improvement, assessed by MRC score in 16 affected muscles, was observed in five out of six patients with IVIg, and no clinical improvement was found with placebo. In addition, electrophysiological studies showed a significant reduction of only one CB in the median nerve in one patient. No significant change in IgM anti-GM1 titres was observed (Van den Berg et al., 1995). In our prospective, controlled and blinded assessment of IVIg versus placebo in 18 patients, we demonstrated therapeutic benefit in four out of nine patients in Group 1 (who had never been treated with IVIg) and eight patients in Group 2 (responders to IVIg in a prior open study). The spontaneous improvement in two patients of Group 1 may be explained by the beneficial effect of rest observed by the patients during the 6 months of the trial, which may occur in immune-mediated diseases. However, there was no statistically significant difference in the primary end-point, which was the change in the MRC score between baseline and month 4. On the other hand, there was a significant improvement in the self-evaluation score in IVIg patients when compared with placebo patients. The self-evaluation score was in better agreement with the clinical improvement than the MRC score. There was also a significant improvement at month 4 in both MRC score and self-evaluation score in IVIg patients, which was not found in placebo patients. Interestingly, the three patients who did not respond to IVIg and the two patients who responded to placebo were all in Group 1, the group which was naı¨ve to IVIg treatment. This suggests that prior treatment with IVIg allows patients to respond more readily to its beneficial effect. The fact that five (two in Group 1 and three in Group 2) of the seven non-

151

responders to placebo in the first phase of the trial improved when switched to IVIg tends to confirm the results of the first phase, despite the fact that there were no comparisons with control subjects. The differences in MRC score were not closely correlated with responder status, but reflected the clinical improvement of the patients. This suggests that the self-evaluation scale was more effective in measuring the response to treatment. Whereas the MRC score reflects the examining physician’s assessment of motor deficit, the self-evaluation scale takes into account the patient’s perception of his or her difficulties in performing everyday activities. This scale seems a useful tool to evaluate the quality of life, such as writing, buttoning, walking and locking a door. The self-evaluation scale may indeed be a more appropriate tool for use in future trials in this disease. Electrophysiological studies failed to show a difference between IVIg patients and placebo patients in any of the variables tested. This is not surprising, as there does not appear to be a close correlation between clinical improvement and electrophysiological results in chronic dysimmune polyneuropathies (Maisonobe et al., 1996; van der Me´ che´ and van Doorn, 1997). Another observation in previous studies (Comi et al., 1994; Bouche et al., 1995) is the poor improvement in CB in clinically improving patients with MMN. This may be explained by the fact that electrophysiological studies explore mainly distal motor nerve segments and do not allow a significant reduction in very proximal CB to be observed. However, in our study we found a reduction in CB in four IVIg patients between baseline and month 4, and in CB in four other patients between months 4 and 7, although the level of improvement was not statistically significant. Another interesting point was the lack of any significant change in IgM anti-GM1 antibody titre from baseline to month 7 in our clinically improving patients. The presence of these antibodies at significantly high titres seems to be well correlated with MMN (Willison, 1994), but their significance remains unclear. However, in our study, all patients but one with significantly high titres responded to IVIg (the last one remaining unchanged). However, given the wide fluctuations of the measured titres, there is a need for more precise measurements in larger cohorts. Lastly, our study showed that IVIg remains efficacious when given over a 6 month period. After this period, few patients seem to have long-lasting remission of the disease, and the majority need repeated IVIg infusions to maintain their motor improvement. Treatment with cyclophosphamide has been proposed as a way of delaying IVIg infusions and stabilizing the disease (Pestronk et al., 1988; Feldman et al., 1991), but the results of retrospective studies conducted by several groups in which this drug was given orally or by monthly intravenous infusions seem to be disappointing (Azulay et al., 1997; Meucci et al., 1997; Van den Berg et al., 1998). Prospective studies are now needed to find out whether immunosuppressive treatments such as cyclophos-

152

Jean-Marc Le´ ger et al.

phamide, azathioprine and β1a interferon (Martina et al., 1999) are able to lead to long-lasting remission of the disease.

Acknowledgements The authors would like to thank Drs M. Bagot d’Arc and K. Koppers for coordination between Assistance PubliqueHoˆ pitaux de Paris and Immuno AG, now Baxter AG (Vienna, Austria), Dr G. Eder (Baxter AG) for documenting samples, M. Kunschak, W. Engl, F. Maritsch, M. Donabauer, H. Leibl and G. Mayer (Baxter AG) for preparation of the study report, Dr F. Keime-Guibert (Fe´ de´ ration de Neurologie Mazarin, Hoˆ pital de la Salpeˆ trie`re) for collecting the patient data, Drs M. Eibl, H. Leibl and H. Ehrlich (Baxter AG) for study management and safety assessments, and M. L. Harpin and Y. Marie (INSERM U 495) for the immunochemical studies. This study was supported by Assistance PubliqueHoˆ pitaux de Paris (CEDIT), which concluded convention ST2 1898 with Immuno AG, now Baxter AG (Vienna) for logistic and financial support, and by a BIOMED contract (INCAT) of the European Community, Brussels (BMH4CT96–0324).

References Azulay JP, Blin O, Pouget J, Boucraut J, Bille´ -Turc F, Carles G, et al. Intravenous immunoglobulin treatment in patients with motor neuron syndromes associated with anti-GM1 antibodies: a doubleblind, placebo-controlled study. Neurology 1994; 44: 429–32. Azulay JP, Rihet P, Pouget J, Cador F, Blin O, Boucraut J, et al. Long term follow up of multifocal motor neuropathy with conduction block under treatment. J Neurol Neurosurg Psychiatry 1997; 62: 391–4. Ben Younes-Chennoufi A, Le´ ger JM, Hauw JJ, Preud’homme JL, Bouche P, Aucouturier P, et al. Ganglioside GD1b is the target antigen for a biclonal IgM in a case of sensory–motor axonal polyneuropathy: involvement of N-acetylneuraminic acid in the epitope. Ann Neurol 1992; 32: 18–23. Bouche P, Moulonguet A, Ben Younes-Chennoufi A, Adams D, Baumann N, Meininger V, et al. Multifocal motor neuropathy with conduction block: a study of 24 patients. J Neurol Neurosurg Psychiatry 1995; 59: 38–44. Chaudhry V, Corse AM, Cornblath DR, Kuncl RW, Drachman DB, Freimer ML et al. Multifocal motor neuropathy: response to human immune globulin. Ann Neurol 1993; 33: 237–42. Comi G, Amadio S, Galardi G, Fazio R, Nemni R. Clinical and neurophysiological assessment of immunoglobulin therapy in five patients with multifocal motor neuropathy. J Neurol Neurosurg Psychiatry 1994; 57 Suppl: 35–7. Donaghy M, Mills KR, Boniface SJ, Simmons J, Wright I, Gregson N, et al. Pure motor demyelinating neuropathy: deterioration after steroid treatment and improvement with intravenous immunoglobulin. J Neurol Neurosurg Psychiatry 1994; 57: 778–83.

Feldman EL, Bromberg MB, Albers JW, Pestronk A. Immunosuppressive treatment in multifocal motor neuropathy. Ann Neurol 1991; 30: 397–401. Hahn AF, Bolton CF, Pillay N, Chalk C, Benstead T, Bril V, et al. Plasma-exchange therapy in chronic inflammatory demyelinating polyneuropathy. A double-blind, sham-controlled, cross-over study. Brain 1996; 119: 1055–66. Kaji R, Shibasaki H, Kimura J. Multifocal demyelinating motor neuropathy: cranial nerve involvement and immunoglobulin therapy. Neurology 1992; 42: 506–9. Kaji R, Oka N, Tsuji T, Mezaki T, Nishio T, Akiguchi I, et al. Pathological findings at the site of conduction block in multifocal motor neuropathy. Ann Neurol 1993; 33: 152–8. Lange DJ, Trojaborg W, Latov N, Hays AP, Younger DS, Uncini A, et al. Multifocal motor neuropathy with conduction block: is it a distinct clinical entity? Neurology 1992; 42: 497–505. Le Forestier N, Chassande B, Moulonguet A, Maisonobe T, Schaeffer S, Birouk N, et al. Neuropathies motrices multifocales avec blocs de conduction: 39 cas. Rev Neurol (Paris) 1997; 153: 579–86. Le´ ger JM. Multifocal motor neuropathy and chronic inflammatory demyelinating polyradiculoneuropathy. [Review]. Curr Opin Neurol 1995; 8: 359–63. Le´ ger JM. Treatment of chronic dysimmune polyneuropathies. Transfus Sci 1999; 20: 63–7. Le´ ger JM, Ben Younes-Chennoufi A, Chassande B, Davila G, Bouche P, Baumann N, et al. Human immunoglobulin treatment of multifocal motor neuropathy and polyneuropathy associated with monoclonal gammopathy. J Neurol Neurosurg Psychiatry 1994; 57 Suppl: 46–9. Lewis RA, Sumner AJ, Brown MJ, Asbury AK. Multifocal demyelinating neuropathy with persistent conduction block. Neurology 1982; 32: 958–64. Maisonobe T, Chassande B, Ve´ rin M, Jouni M, Le´ ger JM, Bouche P. Chronic dysimmune demyelinating polyneuropathy: a clinical and electrophysiological study of 93 patients. J Neurol Neurosurg Psychiatry 1996; 61: 36–42. Martina ISJ, van Doorn PA, Schmitz PI, Meulstee J, van der Me´ che´ FG. Chronic motor neuropathies: response to interferon β1a after failure of conventional therapies. J Neurol Neurosurg Psychiatry 1999; 66: 197–201. Meucci N, Cappellari A, Barbieri S, Scarlato G, Nobile-Orazio E. Long term effect of intravenous immunoglobulins and oral cyclophosphamide in multifocal motor neuropathy. J Neurol Neurosurg Psychiatry 1997; 63: 765–9. Nobile-Orazio E. Multifocal motor neuropathy [editorial]. [Review]. J Neurol Neurosurg Psychiatry 1996; 60: 599–603. Nobile-Orazio E, Meucci N, Barbieri S, Carpo M, Scarlato G. Highdose intravenous immunoglobulin therapy in multifocal motor neuropathy. Neurology 1993; 43: 537–44. Parry GJ, Clarke S. Multifocal acquired demyelinating neuropathy masquerading as motor neuron disease. Muscle Nerve 1988;11: 103–7.

Immunotherapy in multifocal motor neuropathy

153

Pestronk A. Multifocal motor neuropathy: diagnosis and treatment. [Review]. Neurology 1998; 51 (6 Suppl 5): S22–4.

high dose intravenous immunoglobulins: a double blind, placebo controlled study. J Neurol Neurosurg Psychiatry 1995; 59: 248–52.

Pestronk A, Cornblath DR, Ilyas AA, Baba H, Quarles RH, Griffin JW, et al. A treatable multifocal motor neuropathy with antibodies to GM1 ganglioside. Ann Neurol 1988; 24: 73–8.

Van den Berg LH, Franssen H, Wokke JH. The long-term effect of intravenous immunoglobulin treatment in multifocal motor neuropathy. Brain 1998; 121: 421–8.

Roth G, Rohr J, Magistris M, Ochsner F. Motor neuropathy with proximal multifocal persistent conduction block, fasciculations and myokymia. Evolution to tetraplegia. Eur Neurol 1986; 25: 416–23. Sadiq SA, Thomas FP, Kilidireas K, Protopsaltis S, Hays AP, Lee KW, et al. The spectrum of neurologic disease associated with antiGM1 antibodies. Neurology 1990; 40: 1067–72. Van den Bergh P, Logigian EL, Kelly JJ Jr. Motor neuropathy with multifocal conduction blocks. Muscle Nerve 1989; 12: 26–31. Van den Berg LH, Kerkhoff H, Oey PL, Franssen H, Mollee I, Vermeulen M, et al. Treatment of multifocal motor neuropathy with

van der Me´ che´ FG, van Doorn PA. The current place of high-dose immunoglobulins in the treatment of neuromuscular disorders. [Review]. Muscle Nerve 1997; 20: 136–47. Willison HJ. Antiglycolipid antibodies in peripheral neuropathy: fact or fiction? [editorial]. [Review]. J Neurol Neurosurg Psychiatry 1994; 57: 1303–7. Willison HJ, Veitch J, Swan AV, Baumann N, Comi G, Gregson NA, et al. Inter-laboratory validation of an ELISA for the determination of serum anti-ganglioside antibodies. Eur J Neurol 1999; 6: 71–7. Received May 22, 2000. Revised July 30, 2000. Accepted August 22, 2000