Peripheral neuropathy in monoclonal gammopathy of undetermined significance: prevalence and immunopathogenetic studies Nobile-Orazio E, Barbieri S, Baldini L, Marmiroli P, Carpo M, Premoselli S, Manfredini E, Scarlato G. Peripheral neuropathy in monoclonal gammopathy of undetermined significance: prevalence and immunopathogenetic studies. Acta Neurol Scand 1992: 85: 383-390. In an unselected series of patients with monoclonal gammopathy of undetermined significance (MGUS) we found neuropathy in 2 of 34 patients with IgG ( 6 % ) , 2 of 14 with IgA (14%), and 8 of 26 with IgM MGUS (3 1 %). The neuropathy was subclinical in 6 patients (1 IgG, 1 IgA, and 4 IgM). Patients with IgG or IgA MGUS had a prominent motor impairment with electrophysiologic and morphologic findings suggestive of predominant axonal degeneration. No deposit of the M-protein in sural nerve and no reactivity of the M-protein with nerve was detected in these patients. Patients with IgM MGUS had a prominent sensory impairment with evidence of predominant demyelination. In 6 of these patients the M-protein reacted with the myelin-associated glycoprotein (MAG). The higher prevalence of neuropathy in patients with IgM MGUS may be related to the frequent reactivity of IgM M-proteins with MAG.
Neuropathy has been frequently reported in patients with IgM monoclonal gammopathy, including Waldenstrom’s macroglobulinemia and IgM monoclonal gammopathy of undetermined significance (MGUS) (1-4). In these patients the neuropathy has frequently homogeneous features and is often associated with IgM M-proteins that react with the myelin-associated glycoprotein (MAG) and related glycoconjugates (1-10) or, less frequently, with other antigens in nerve (2, 10-16). Neuropathy has been also reported in patients with IgG MGUS (17-28) but the significance of this association is less clear as paraproteins, mostly IgG, are frequent in adults with a prevalence of up to 4% after 80 (29). Furthermore different forms of neuropathy have been described in these patients (20-28), while reactivity of IgG Mproteins with nerve has been rarely found (25, 28). Much less is known on neuropathy associated with IgA MGUS as only few patients have been reported in detailed (26, 30). We studied an unselected series of 74 patients with IgG, IgA or IgM MGUS in order to determine: (1) the prevalence of neuropathy in the different forms of MGUS; (2) the clinical, electrophysiologic and morphologic aspects of the neuropathy in relation to different Ig isotypes; (3) the antigen specificities of M-proteins and their relation with the presence of neuropathy.
E. Nobile-Orazio ’, S.Barbieri ’, L. Baldini’, P. Marmiroli ’, M. Carpo ’,
’,
S. Premoselli E. Manfredini G. Scarlato’
’,
’
Institutes of Clinical Neurology, Din0 Ferrari Centre, Medical Sciences, Department of Hematology, University of Milan, Ospedale Maggiore Policlinico, Milan, Italy
Key words: neuropathy: monoclonal gammopathy; autoantibodies; immunoglobulins: myelin-associated glycoprotein
E. Nobile-Orazio, Institute of Clinical Neurology, University of Milan, Via F. Sforza 35, 20122, Milan, Italy Accepted for publication October 12, 1991
Material and methods Patients
We studied 34 consecutive outpatients (14 men, 20 women), aged 26 to 77 years (mean, 57.4) with IgG MGUS, 14 (8 men, 6 women), aged 50 to 77 (mean, 60.1) with IgA MGUS and 26 (15 men, 11 women) aged 40 to 78 (mean, 61.7) with IgM MGUS, followed at the Department of Hematology of our University. All the patients had been referred to that department for the investigation of a serum Mprotein found in the evaluation of unrelated diseases except for a patient with IgM MGUS in whom the M-protein was originally discovered in another hospital during the evaluation of symptoms of neuropathy and that was later referred to our Hematology Dept. All patients were first examined by us in the occasion and as a part of this study while patients originally referred to our Institute for the evaluation of neuropathy and subsequently found to have MGUS were not included. The diagnosis of MGUS was made when all subsequent criteria were fulfilled at the time of neurological evaluation: serum Mprotein levels of less than 3 g/dl; no or small amounts of Bence Jones proteinuria; less than 10% ofplasma cells in the bone marrow; no evidence of osteolytic lesions, lymphocytosis, anemia, hypercalcemia, organomegaly, lymphadenopathy and renal insuffi383
Nobile-Orazio et al. ciency (3 1). All the patients were followed for at least 2 years after completing the study. In none of them evidence of malignancy became apparent. Neurologic evaluation and diagnostic criteria
Clinical evaluation of the neuropathy and electrodiagnostic studies, including needle EMG and nerve conduction studies, were performed in all patients as previously described (2). The diagnosis of peripheral neuropathy was made in patients with symptoms of neuropathy in whom at least two bilaterally symmetric clinical signs and two electrophysiologic abnormalities were found. Patients without definite symptoms or signs of neuropathy were deemed to have a subclinical neuropathy when two or more electrophysiologic abnormalities were found. Electrophysiologic studies were classified (32) as suggestive of predominant demyelination (D) or axonal degeneration (A) or were considered mixed (A/D). Morphologic studies on sural nerve biopsies were performed in 5 patients with neuropathy (3 IgM, 1 IgG and 1 IgA MGUS) and included histologic and histometric analysis on nerve sections (2) and teased myelinated fibers analysis (33). The presence of concomitant other possible causes of neuropathy was excluded in all the patients by clinical and laboratory investigations (2). Immunologic studies
Immunoblot - Peripheral nervous system (PNS) myelin and non-myelin fractions and central nervous system (CNS) myelin (34) were prepared from bovine cauda equina and human autopsy brain, respectively, obtained within 6 hours after death. Whole cauda equina was also lyophilized and homogenized in 2 % SDS. Proteins (100 ug) were fractionated by SDS polyacrylamide gradient (4-20 %) gel electrophoresis (PAGE) (35) and patient’s Ig reactivity with nerve tested by immunoblot (36) at serum dilutions of 1:200 using peroxidase conjugated goat immunoglobulins to human IgG, IgA or IgM and, when positive, kappa and lambda light chains (DAKO, Denmark) (2). Anti-MAG IgM antibody titers were determined by immunoblot after SDS-PAGE of CNS myelin at increasing twofold serum dilutions starting from 1:25 (6). Immunostaining after HPTLC - Bovine PNS glycolipids were prepared from cauda equina (37), separated by HPTLC on aluminium backed silica gel 60 HPTLC plates (Merck, Germany) and patients’ Ig reactivity was tested by immunostaining (2) at a serum dilution of 1:100, using the above mentioned antibodies. 384
ZmmunoJluorescence studies - Binding of patients’ Ig to nerve was studied by indirect immunofluorescence (2, 38) on 6 pm cryostat cut sections of normal human sural nerve. Sections were incubated with patients’ sera diluted 1:25 and counterstained with fluorescein conjugated rabbit immunoglobulins to human IgG, IgA or IgM antibodies and, when positive, kappa and lambda light chains (DAKO, Denmark). Ig deposits on patients’ nerve were detected by direct immunofluorescence (2,38). Immunoabsorption studies - Sera from patients with anti-MAG IgM were absorbed with PNS myelin (2). Absorbed sera were tested for IgM reactivity with MAG by immunoblot and for the M-component by agarose gel electrophoresis (AGE). Sera from patients with neuropathy and M-protein not anti-MAG were absorbed with whole cauda homogenate and with PNS myelin and subsequently tested for the M-protein by AGE (6). Results Prevalence of neuropathy
We found clinical and electrophysiologic signs of neuropathy in 1 patient with IgG (2.9%), 1 with IgA (7.1) and 4 with IgM MGUS (15.4%). Six additional patients, 1 with IgG, 1 with IgA and 4 with IgM MGUS had electrophysiologic signs of neuropathy in the absence of clinical manifestations. The overall prevalence of neuropathy was significantly higher in patients with IgM MGUS (30.8%) as compared to patients with IgG MGUS (5.9%) = 6.57; P < 0.025) and to the combined groups of patients with IgG and IgA M G U S (x2=6.24; P < 0.025), while no significant dflerence between patients with IgM and IgA MGUS (14.2%) = 1.3 1) and between patients with IgG and IgA MGUS (x2= 0.91) was observed. Patients with IgG MGUS and neuropathy (Table 1) did not differ from those without neuropathy in relation to sex, duration of MGUS, mean serum IgG levels or light chain type, but their mean age was higher than in patients without neuropathy. No difference in the same parameters was observed between IgA patients with and without neuropathy while in patients with IgM MGUS the neuropathy was more frequent in men (46.6%) than in women (9%) (Table 1). Patients with clinical neuropathy and IgA or IgM MGUS, had symptoms of neuropathy for, respectively, 10 years and a mean of 2.8 years (range 0.57), before gammopathy was discovered but in only one patient, symptoms of neuropathy led to the diagnosis of MGUS. The clinically affected patient with IgG MGUS developed symptoms of neuropathy one year after MGUS was diagnosed.
(xz
(x2
Neuropathy in MGUS Table 1. Hematological findings in patients with MGUS with and without neuropathy IgG MGUS
No. of patients Mean age (range)
IgA MGUS
IgM MGUS
PN
no PN
PN
no PN
PN
no PN
2
32
2
12
8
18
74.01 (71-77)
56.4 (26-77)
60.5 (60-6 1)
60.1 (50-75)
61.1 (45-78)
61.0 (40-77)
u**
8/10
111
13/19
l/l
715
Mean time from MGUS diagnosis, months (range)
24.5 (27-36)
34.7 (1-1 20)
49 (28-60)
28.3 (1-1 08)
19.9 (1-46)
21.3 (1-96)
Mean lg levels at diagnosis (g/dl) (range)
1.9 (1.7-2.1)
1.9 (1.5-2.8)
1.6 (1.2-2.0)
0.9 (0.2-2.1)
1.2 (0.8-1.8)
1.4 (0.4-2.9)
111
2318 (1 k&?J
l/l
616
71 1
12/6
Men/women
Light chain of M-proteins ( k h )
PN: neuropathy; no PN: no neuropathy.
* : Student's t test: p=0.0504. **: 1~4.20;p
Neuropathy has been frequently reported in patients with IgM monoclonal gammopathy, including Waldenstrom’s macroglobulinemia and IgM monoclonal gammopathy of undetermined significance (MGUS) (1-4). In these patients the neuropathy has frequently homogeneous features and is often associated with IgM M-proteins that react with the myelin-associated glycoprotein (MAG) and related glycoconjugates (1-10) or, less frequently, with other antigens in nerve (2, 10-16). Neuropathy has been also reported in patients with IgG MGUS (17-28) but the significance of this association is less clear as paraproteins, mostly IgG, are frequent in adults with a prevalence of up to 4% after 80 (29). Furthermore different forms of neuropathy have been described in these patients (20-28), while reactivity of IgG Mproteins with nerve has been rarely found (25, 28). Much less is known on neuropathy associated with IgA MGUS as only few patients have been reported in detailed (26, 30). We studied an unselected series of 74 patients with IgG, IgA or IgM MGUS in order to determine: (1) the prevalence of neuropathy in the different forms of MGUS; (2) the clinical, electrophysiologic and morphologic aspects of the neuropathy in relation to different Ig isotypes; (3) the antigen specificities of M-proteins and their relation with the presence of neuropathy.
E. Nobile-Orazio ’, S.Barbieri ’, L. Baldini’, P. Marmiroli ’, M. Carpo ’,
’,
S. Premoselli E. Manfredini G. Scarlato’
’,
’
Institutes of Clinical Neurology, Din0 Ferrari Centre, Medical Sciences, Department of Hematology, University of Milan, Ospedale Maggiore Policlinico, Milan, Italy
Key words: neuropathy: monoclonal gammopathy; autoantibodies; immunoglobulins: myelin-associated glycoprotein
E. Nobile-Orazio, Institute of Clinical Neurology, University of Milan, Via F. Sforza 35, 20122, Milan, Italy Accepted for publication October 12, 1991
Material and methods Patients
We studied 34 consecutive outpatients (14 men, 20 women), aged 26 to 77 years (mean, 57.4) with IgG MGUS, 14 (8 men, 6 women), aged 50 to 77 (mean, 60.1) with IgA MGUS and 26 (15 men, 11 women) aged 40 to 78 (mean, 61.7) with IgM MGUS, followed at the Department of Hematology of our University. All the patients had been referred to that department for the investigation of a serum Mprotein found in the evaluation of unrelated diseases except for a patient with IgM MGUS in whom the M-protein was originally discovered in another hospital during the evaluation of symptoms of neuropathy and that was later referred to our Hematology Dept. All patients were first examined by us in the occasion and as a part of this study while patients originally referred to our Institute for the evaluation of neuropathy and subsequently found to have MGUS were not included. The diagnosis of MGUS was made when all subsequent criteria were fulfilled at the time of neurological evaluation: serum Mprotein levels of less than 3 g/dl; no or small amounts of Bence Jones proteinuria; less than 10% ofplasma cells in the bone marrow; no evidence of osteolytic lesions, lymphocytosis, anemia, hypercalcemia, organomegaly, lymphadenopathy and renal insuffi383
Nobile-Orazio et al. ciency (3 1). All the patients were followed for at least 2 years after completing the study. In none of them evidence of malignancy became apparent. Neurologic evaluation and diagnostic criteria
Clinical evaluation of the neuropathy and electrodiagnostic studies, including needle EMG and nerve conduction studies, were performed in all patients as previously described (2). The diagnosis of peripheral neuropathy was made in patients with symptoms of neuropathy in whom at least two bilaterally symmetric clinical signs and two electrophysiologic abnormalities were found. Patients without definite symptoms or signs of neuropathy were deemed to have a subclinical neuropathy when two or more electrophysiologic abnormalities were found. Electrophysiologic studies were classified (32) as suggestive of predominant demyelination (D) or axonal degeneration (A) or were considered mixed (A/D). Morphologic studies on sural nerve biopsies were performed in 5 patients with neuropathy (3 IgM, 1 IgG and 1 IgA MGUS) and included histologic and histometric analysis on nerve sections (2) and teased myelinated fibers analysis (33). The presence of concomitant other possible causes of neuropathy was excluded in all the patients by clinical and laboratory investigations (2). Immunologic studies
Immunoblot - Peripheral nervous system (PNS) myelin and non-myelin fractions and central nervous system (CNS) myelin (34) were prepared from bovine cauda equina and human autopsy brain, respectively, obtained within 6 hours after death. Whole cauda equina was also lyophilized and homogenized in 2 % SDS. Proteins (100 ug) were fractionated by SDS polyacrylamide gradient (4-20 %) gel electrophoresis (PAGE) (35) and patient’s Ig reactivity with nerve tested by immunoblot (36) at serum dilutions of 1:200 using peroxidase conjugated goat immunoglobulins to human IgG, IgA or IgM and, when positive, kappa and lambda light chains (DAKO, Denmark) (2). Anti-MAG IgM antibody titers were determined by immunoblot after SDS-PAGE of CNS myelin at increasing twofold serum dilutions starting from 1:25 (6). Immunostaining after HPTLC - Bovine PNS glycolipids were prepared from cauda equina (37), separated by HPTLC on aluminium backed silica gel 60 HPTLC plates (Merck, Germany) and patients’ Ig reactivity was tested by immunostaining (2) at a serum dilution of 1:100, using the above mentioned antibodies. 384
ZmmunoJluorescence studies - Binding of patients’ Ig to nerve was studied by indirect immunofluorescence (2, 38) on 6 pm cryostat cut sections of normal human sural nerve. Sections were incubated with patients’ sera diluted 1:25 and counterstained with fluorescein conjugated rabbit immunoglobulins to human IgG, IgA or IgM antibodies and, when positive, kappa and lambda light chains (DAKO, Denmark). Ig deposits on patients’ nerve were detected by direct immunofluorescence (2,38). Immunoabsorption studies - Sera from patients with anti-MAG IgM were absorbed with PNS myelin (2). Absorbed sera were tested for IgM reactivity with MAG by immunoblot and for the M-component by agarose gel electrophoresis (AGE). Sera from patients with neuropathy and M-protein not anti-MAG were absorbed with whole cauda homogenate and with PNS myelin and subsequently tested for the M-protein by AGE (6). Results Prevalence of neuropathy
We found clinical and electrophysiologic signs of neuropathy in 1 patient with IgG (2.9%), 1 with IgA (7.1) and 4 with IgM MGUS (15.4%). Six additional patients, 1 with IgG, 1 with IgA and 4 with IgM MGUS had electrophysiologic signs of neuropathy in the absence of clinical manifestations. The overall prevalence of neuropathy was significantly higher in patients with IgM MGUS (30.8%) as compared to patients with IgG MGUS (5.9%) = 6.57; P < 0.025) and to the combined groups of patients with IgG and IgA M G U S (x2=6.24; P < 0.025), while no significant dflerence between patients with IgM and IgA MGUS (14.2%) = 1.3 1) and between patients with IgG and IgA MGUS (x2= 0.91) was observed. Patients with IgG MGUS and neuropathy (Table 1) did not differ from those without neuropathy in relation to sex, duration of MGUS, mean serum IgG levels or light chain type, but their mean age was higher than in patients without neuropathy. No difference in the same parameters was observed between IgA patients with and without neuropathy while in patients with IgM MGUS the neuropathy was more frequent in men (46.6%) than in women (9%) (Table 1). Patients with clinical neuropathy and IgA or IgM MGUS, had symptoms of neuropathy for, respectively, 10 years and a mean of 2.8 years (range 0.57), before gammopathy was discovered but in only one patient, symptoms of neuropathy led to the diagnosis of MGUS. The clinically affected patient with IgG MGUS developed symptoms of neuropathy one year after MGUS was diagnosed.
(xz
(x2
Neuropathy in MGUS Table 1. Hematological findings in patients with MGUS with and without neuropathy IgG MGUS
No. of patients Mean age (range)
IgA MGUS
IgM MGUS
PN
no PN
PN
no PN
PN
no PN
2
32
2
12
8
18
74.01 (71-77)
56.4 (26-77)
60.5 (60-6 1)
60.1 (50-75)
61.1 (45-78)
61.0 (40-77)
u**
8/10
111
13/19
l/l
715
Mean time from MGUS diagnosis, months (range)
24.5 (27-36)
34.7 (1-1 20)
49 (28-60)
28.3 (1-1 08)
19.9 (1-46)
21.3 (1-96)
Mean lg levels at diagnosis (g/dl) (range)
1.9 (1.7-2.1)
1.9 (1.5-2.8)
1.6 (1.2-2.0)
0.9 (0.2-2.1)
1.2 (0.8-1.8)
1.4 (0.4-2.9)
111
2318 (1 k&?J
l/l
616
71 1
12/6
Men/women
Light chain of M-proteins ( k h )
PN: neuropathy; no PN: no neuropathy.
* : Student's t test: p=0.0504. **: 1~4.20;p
