close

Вход

Забыли?

вход по аккаунту

?

Correlation of clinical features and findings on cranial magnetic resonance imaging with urinary myelin basic protein-like material in patients with multiple sclerosis.

код для вставкиСкачать
Correlation of Clinical Features and Findings
on Cranial Magnetic Resonance Imagng
with Urinary Myelin Basic Protein-lke
Material in Patients with Multiple Sclerosis
John N. Whitaker,*TO Paula H. Williams,* Beverly A. Layton," Henry F. McFarland,ll Lael A. Stone,'I
M. E. Smith,ll R. David Kachelhofer," Edwin L. Bradley,$ Sheila Burgard,$ Guojun Zhao,'
and Donald W. Paty'
Immunoreactive material that appears to be a peptide encompassing all or a portion of residues 80 to 89 of myelin
basic protein is present in normal unconcentrated urine and is increased in certain patients with multiple sclerosis
(MS). Compared with normal controls, urines collected randomly from 158 MS patients or in a clinical research unit
from 8 patients with MS had higher mean values of urinary MBP-like material (MBPLM). The level of MBPLM in
urine showed no direct relationship to MBPLM in cerebrospinal fluid and did not correlate with clinical relapses of
disease. In the other neurological disease control group (26 patients), some patients with other inflammatory diseases,
but not stroke or early phase Guillain-Barre syndrome, also showed elevations. Among the subtypes of MS, those with
secondary chronic progressive disease had the highest values. Urinary MBPLM showed no definite correlation with
or effect of treatment with glucocorticoids and immunosuppressants except that a lower level of urinary MBPLM
showed a weak relationship with improvement following treatment with methylprednisoloneiprednisone. In a serial
study of 8 patients with unenhanced cranial magnetic resonance imaging and 20 patients with gadolinium-enhanced
cranial magnetic resonance imaging, urinary MBPLM did not show a direct correlation with new or enhancing lesions.
Urinary MBPLM does not parallel acute myelin damage hut appears to reflect an ongoing process, possibly linked to
attempted efforts at remyelination.
Whitaker JN, Williams PH, Layton BA, McFarland HF, Stone LA, Smith ME, Kachelhofer RD, Bradley EL,
Burgard S, Zhao G, Paty DW. Corrdation of clinical features and findings on cranial magnetic
resonance imaging with urinary myelin basic protein-like material in patients with
multiple sclerosis. Ann Neurol 1994;35:5'7-585
Given the variable course of multiple sclerosis (MS),
an objective and feasible means for monitoring activity
of this disease and predicting or paralleling its progression is highly desirable. The many subclinical events in
MS render clinical ascertainment of new lesions very
imprecise and insensitive f l , 21. Patterns of onset and
clinical features may suggest certain courses of natural
history but only in a general sense E3-51. Evoked potentials have proven unsatisfactory in monitoring disease activity but may show increasing abnormalities
with advancing disease [G, 7). Most attention has been
directed recently to the utilization of cranial magnetic
resonance imaging (MRI) to monitor disease activity
and course of MS [2, 8, 91. Cranial MRI without [2)
or with [3-11} gadolinium enhancement can provide
documentation of the dynamic and predominantly subclinical central nervous system (CNS) white matter lesions in MS. Although serial unenhanced cranial MRI
has been used to demonstrate a beneficial effect of
p-interferon in relapsing-remitting MS { 121, the role
of enhanced cranial MRI, although reflecting acute disease activity has not yet been fully tested in long-term
clinical trials. Pilot studies incorporating enhanced cranial MRI afford a possible prediction of beneficial effects to later be tried on larger numbers of patients
From the Departments of *Neurology, +Cell Biology, and tBiostatistics. and $Center for Neuroimmunoloev. UniversirV of Alabama ac
Birmingham, and §Neurology and Research Services of the Birmingham Veterans Medical Center, Birmingham, AL; INeuroimmunology Branch of the National Institutes of Health, Bethesda, MD;
and 'Division of Neurology. Vancouver General Hospital. Vancouver, British Columbia, C&ada.
Received Jul 19, 1991, and in revised form Oct 4. Accepred for
~ublicationOct .5., 1993.
. ...
I
C131.
As precise and potentially quantitative as cranial
MRI is or may become, other more feasible and less
expensive methods for accomplishing the same objectives are still needed. A number have been examined
i d d r e s s cor~espon~ence Dr Whicker, Department of Neurology, University of Alabama at Birmingham, UAB Starion, Birminghm, AL 35294-0007.
Copyright 0 1994 by the American Neurological Association
577
1141. The presence of myelin basic protein (MBPj-like
material (MBPLMj in cerebrospinal fluid (CSF) has
been known for over 15 years to be an index of disease
activity from CNS myelin damage { l S , 161. MBP, a
structural protein of the myelin sheath, has a molecular
weight of 18,500 and is comprised of 170 residues
C17J. There is evidence that fragments of MBP 118)
encompassing all or a portion of MBP peptide 80 to
87 [17] account for the CSF MBPLM. MBPLM in CSF
is an index of CNS myelin breakdown regardless of
cause [20, 211, but it may predict response of MS patients t o glucocorticoids f22). However, using CSF for
frequent sampling is not practical. Using a highly selective antiserum to an MBP fragment that recognizes a
cryptic epitope within the MBP molecule different
from that found in CSF, MBPLM in urine has been
detected [23]. The difficulties in developing this assay
and characterizing it have been previously reported
{24}. The clinical correlation of urinary MBPLM has
been much more difficult t o demonstrate than it has
for MBPLM in CSF. In the present investigation, results are reported for the optimization and validation
of the immunoassay for urinary MBPLM as well as its
correlation with clinical change and abnormalities on
cranial MRI. The findings demonstrate that urinary
MBPLM is always present though highly variable in
urine from adults, exists at higher levels in patients
with MS, especially those in a secondary chronic progressive phase, does not reflect acute disease activity
in MS as determined clinically, by CSF MBPLM levels
or changes on cranial MRI, and may serve as a predictor of clinical response to certain medications.
Materials and Methods
H u m a n Urine Specimens
Specimens (2,832) of urine from 260 individuals were analyzed (Table l ) . The studies performed were approved by
the Institutional Review Boards. Normal controls and patients were grouped into 9 categories for purposes of identifying variables to be controlled for urine collection or interpretation of results and for determining the relationships of
urinary MBPLM with clinical course, CSF findings, treatment, and results of cranial MRI. Some of the results were
analyzed in different groupings accounting for the large number of specimens listed (see Table 1 ) . All samples were obtained by voiding or from a catheter. They were stored at
4°C for 1 to 24 hours before being frozen in aliquots at
- 70°C. The documented immunochemical stability of the
urinary MBPLM under these conditions has been noted previously [23].
Group I consisted of a total of 350 specimcns from 8 MS
patients and 8 controls, 4 of whom were spouses or siblings
of the MS patients, who were hospitalized in the General
Clinical Research Center (GCRC) of the University of Alabama Hospital. All patients received a 150-mg sodiumcontrolled diet, and urine was collected at 4-hour intervals
during the 72-hour period of hospitalization.
Group I1 consisted of 1 1 1 specimens from 10 normal controls (6 women, 4 men; ages 25-50 yr) from whom morning
(during the time of 0800- 1100 hr) urine was collected over
a 2-week period.
Group I11 consisted of 1,479 specimens from 158 patients
with MS. These patients were classified as relapsingremitting (RR),relapsing-progressive (RP), primary chronic
progressive (CPP), and secondary chronic progressive (CPS)
on the basis of clinical characteristics described [22]. Disability was expressed as a score based on the Kurtzke scale [ 2 5 ] .
Group IV consisted of 256 urine specimens from 31 dis-
Table 1. Groups of Subjects Studied for Measurement of Ut-inary Myelin Basic Protein-like iMaterir/l (MBPLM)
NumberiClinical Status
Group
Variable Examincd
No. of Subjects
I
Conditions for collection
8 controls
8 MS
10 controls
158 MS
26 OND; 5 renal
8 RR-MS
8 CPS-MS
20 RR-MS
205 ills
147 treated
58 untreated
13 CPS-MS
I1
111
IV
V
VI
VII
VIII
IX
Normal controls
Subtypes of MS
Disease controls
Cranial MRI relationship
Cranial MRI relationship
Cranial MRI relationship
Treatment effects MS
Cyclosporine effects MS
No. of Specimens
Urinary MBPLM
(ng of MBP 83-891
mg of Creatinine) (SD)
106.9” (56.7)
185.9 (120.4)
8’.2 ( 3 8 . 5 )
180.0 (91.4)
148.4 (70.4)
89.4 ( 4 5 1)
124.4 (465 )
121.7 (73.41
184.0 (105.8)
182.1 (94.3)
188.9 (131.4)
145.4 (48.0)
175
175
111
1,479
256
18
94
300
1,102
979
123
22 1
’Subject mean value.
MBP = myelin basic protein; MS = multiple sclerosis; OND
relapsing-remitting; CPS = secondary chronic progressive.
578 Annals of Neurology Vol 35 No 5
1
May 1994
other neurological diseases; MRI
=
magnetic resonance imaging; RR
=
ease controls (see Table I). The other neurological disease
(ONDj controls were Guillain-Barre syndrome (GBSj ( 158
specimens from 6 patients), myelopathy (26 specimens from
5 patients), stroke (18 specimens from 5 patients), other
CNS inflammatory diseases (15 specimens from 4 patients),
Devic’s syndrome (28 specimens from 1 patient), miscellaneous neurological diseases (6 specimens from 5 patients),
and renal disease ( 5 specimens from 5 patients).
Group V consisted of 18 specimens from 8 patients with
RR-MS who had serial cranial MRIs on a 0.3-T instrument
every 2 weeks for 8 weeks. Group VI consisted of 94 specimens from 8 patients with CPS-MS who had serial cranial
MRIs on a 0.3-Tinstrument and urine collections every 2
weeks for 24 weeks. In this group of 8 patients there was a
range of 0 to 12 new or enlarging lesions detected during
the study period. Group V11 consisted of 300 specimens
from 20 patients with RR-MS who had a monthly urine collection and cranial MRI,with and without gadolinium, on a
1.5-T instrument over an 18-month period. The serial MRIs
revealed a range of 0 to 12 new lesions and 0 to 18 enhancing
lesions { 111. All urine specimens were obtained within 24
hours of the performance of the cranial MRI.
Group VIlI consisted of 979 specimens from 147 patients
with MS of different clinical types and who were treated
with ACTH, glucocorticoids, or cyclophosphamide, and 123
specimens from 58 MS patients who had received no treatment with glucocorticoids or immunosuppressants for a minimum of 30 days. This group of treated patients included 62
MS patients who were worsening and were treated with a
regimen of intravenous methylprednisolone followed by oral
prednisone 1221. Early morning urine specimens were collected during the 5 days of treatment with intravenous methylprednisolone, and urinary MBPLM was analyzed for correlation with clinical response to treatment at day 5 and day
40 C22). Group IX consisted of 221 specimens from 13 patients with CPS-MS who received cyclosporine (6 patients)
or placebo (7 patients) at the NIH as part of the randomized,
double-blind trial of CPS-MS 1261.
Human Cerebrospinal Fluid Specimens
Lumbar CSF specimens were available on 18 patients with
MS on whom urine specimens were collected within 24
hours of the lumbar puncture.
Detection of Myelilz Basic Protein-like Material in
Body Fluids
The level of MBPLM in CSF was determined by a double
antibody radioimmunoassay (RIA) in which radiolabeled human MBP peptide 69 to 89 served as the radioligand and
rabbit 7 3 (R79) antiserum was the first antibody 127). A
value of greater than 0.10 ngiml, more than 2 SD above the
mean {271,was considered to be elevated [22]. Some analysis
was also performed based on the value of 0.16 ngiml, 4 S D
above the mean [ 2 7 ] .
Urinary MBPLM was measured with a different RIA using
radiolabeled human MBP peptide 4 5 to 89 as the radioligand
and rabbit 110 (R110) antiserum as the first antibody [23].
Based on immunochemical studies (see Results below), MBP
peptide 8 3 to 89 was chosen as the inhibitor standard. Variation between and within assays as well as recovery were de-
termined for the assays with the standard of MBP peptide
83 to 89. The urinary creatinine level was measured with a
kit (Sigma, St Louis, MO) and the nanograms of urinary
MBPLM was expressed in relationship to milligrams of creatinine 1231.
Irnrnunochemical Analysis
The immunochemical nature of the urinary MBPLM was investigated using the principle of parallelism 1231. Human
MBP peptides 80 to 89, 81 to 89, 82 to 89, 83 to 89, and
84 to 89 were synthesized and their purity documented by
high-performance liquid chromatography and amino acid
analysis by Peninsula Laboratories (San Carlos, CA). Varying
concentrations of these peptides were analyzed for their comparative inhibitory effects on the reaction of K110 and radiolabeled MBP peptide 45 to 89 in the RIA for urinary
MBPLM 123,281. Dilutions of human urine were analyzed
for urinary MBPLM and compared for parallelism with the
displacement curves of five human MBP peptides. For studies of parallelism, results of RIA were expressed in a logitlog form t231.
StatisticaL Analyses
The average values of MBP, creatinine, and MBPicreatinine
were computed as subject means using all readings available
for himiher. Several methods were examined to express the
changes in urinary MBPLM over time f291. The mean and
standard deviation for urinary MBP, creatinine, and urinary
MBPLiM per creatinine war calculated for each subject. The
subject means among groups were compared using one-way
analysis of variance followed by Fisher’s protected least significant difference test. For correlation of urinary MBPLM
with changes after treatment with methylprednisoloneiprednisone, the xz method was used 122). Regression analysis
was used to test levels of urinary MBPLM versus CSF levels
of MBPLM as well as urinary MBPLM versus number of
lesions present in cranial MRI. A p value of <0.05 was considered statistically significant. Bonferroni adjustments were
made in an effort to avoid type I (false positive) errors C301.
Results
Quuntitative Optimization and Validation
of Radioimmmoassa y
To determine the MBP peptide that gave the must
accurate measure of urinary MBPLM and to serve as
the assay standard, the parallelism of urinary MBPLM
in serially diluted urines from 12 patients with MS, 4
patients with GBS, and 2 normals was compared with
the reaction of varying amounts of human MBP peptides 80 to 89, 81 to 89, 82 t o 89, 83 tu 89, and 84
to 89. Except for one RR-MS patient (Patient G, Fig
l), the best parallelism for urinary MBPLM was with
MBP peptide 83 to 89 for normals and for patients
with MS (see Fig 1). No characteristic group differences were evident when urinary MBPLM displacement curves in normals or GBS patients were compared with MS or RR-MS compared with CPS-MS. It
was concluded that the use of the standard of human
Whitaker et al: Urinary MBPLM in MS
579
encompassing the variations noted among the 4-hour
collections (data not shown).
With the lack of a clear diurnal pattern of urinary
MBPLM values, the collection of urine specimens on
a random basis was justified. In 10 normal controls
(see Table 1, Group 11; Table 2; Table 3), 111 urine
samples were collected randomly over a 14-day period.
There was a large variation among individuals with a
mean of 87.2 ? 38.5 (SD) ng/mg of creatinine, a value
similar to that for normals in the GCRC of 106.9 t
2
1
0
0
F
m
._
c
-1
0)
d
-2
-3
56.7.
-4
1
2
3
4
5
6
Ln (Amount - fm)
7
8
9
10
(volume - PI)
Fig I . Parallel displacementj effected bji seven urines (A-G)
from persons with multiple sclerosis and the human myelin basic
protein (MBPi peptides 80 to 89, 81 t o 89, 82 to 89. 83 t o
89, and 84 to 89. The best,fit parallelism, expressed after
logit-log tran.rJormation, for the urine MBP-like material is
uith human M B P peptide 83 t o 89.
hlBP peptide 83 to 89 was more accurate than that of
the previously used MBP peptide 80 to 89 [23]. With
the new assay standard, RIA displacement values of
samples measured all along the displacement curve of
the assay standard could be used more reliably, a problem not resolved when MBP peptide 80 to 89 was
used as standard [23}. Within assay variation was
t 11.3% ( 1 SD). Between assay variation in 23 separate assays performed over an 18-month period was
12.491 (1 SDj. Recovery of MBP peptide 83 to 89
added to dialyzed urine was 80 to 9456.The displacement curve of MBP peptide 83 to 89 was less steep
than that with MBP peptide 80 to 89 (see Fig 1) leading to higher values for urinary MBPLM than previously published [23}. The value of urinary MBPLM
was related to creatinine to correct for rend function
*
1231.
Possible Physiological Variations of Urinury Myelin
Busic Protein-Like Material
Because of the fluctuation of values of the urinary
MBPLM, there was concern about its diurnal variations
as well as dietary effects. Eight patients with MS ( 3
with RR-MS, 2 with RP-MS, 1 with CPP-MS, and 2
with CPS-MS) and 8 controls (see Table 1, Group I;
Table 2) were admitted to the GCRC during which
time urine was collected in 4-hour increments. There
was no evidence of a relationship of urinary MBPLM
to time of day over a 72-hour period in MS or controls
(data not shown). The MS patient group had a higher
value of 185.9 of urinary MBPLMicreatinine (ngimg)
than the control group value of 106.9 (see Tables 1
and 2). The comparison studies of 24-hour and 72hour values of urinary MBPLM revealed no additional
trend or change but provided a more averaged value
580 Annals of Neurology Vol 35 No 5
May 1994
Urinary MBPLM in Dzfferent CLinicul Tj@es qf
Multiple Sclerosis und Other Neurological Diseaes
As with normals, the urinary MBPLM values in MS
subtypes (see Table 1, Group 111) varied broadly (see
Table 2). By using mean values, the levels of urinary
MBPLM were higher in CPS-MS (208.8 ngimg) than
in RR-MS (150.1 ng/mg), RP-MS (184.3 ngimg), and
CPP-MS (185.1 ngimg) (see Table 3j. When compared
with the control group, all subtypes of MS had significantly higher values (see Table 3).This difference just
attained significance for RR-MS ( p = 0.0456) and was
most evident for CPS-MS ( p = 0.0003). Due to the
overlap of the ranges of values, there were no statistically significant differences among the MS subtypes.
For the small group of MS patients studied in the
GCKC, a clear difference from normals could only be
demonstrated for the CPS-MS cases (see Table 2).
The 31 patients with O N D or renal disease (see
Table 1, Group IV; Table 2) also showed a broad range
of values. When analyzed as a group, the disease
controls had a significantly higher value of urinary
MBPLM ( p = 0.0128) (see Table 3). When compared
with the normals studied in the GCRC (Group I )
whose value was 106.9 ngimg or the normals providing
random morning urines (Group 11) whose value was
87.2 ng/mg, the urinary MBPLM values for patients
with stroke (104.9 ngimg), inflammatory CNS diseases
( 140.4 ngimgj, miscellaneous neurological disorders
(142.6 ngimg), and renal disease (1 16.0 ngimg) were
not significantly different. Patients with myelopathy
(200.0 ngimg), GBS (159.3 ngimg), and the 1 patient
with Devic’s syndrome (265.4 ng/mg) had higher values. Although the categorization of these disorders in
Group IV as OND might be questioned, they are distinctive clinically or different from the MS subtypes
already mentioned.
Serial studies of the GBS patient group disclosed a
delayed increase of MBPLM. During the first few
weeks of hospitalization when the patients were bed
bound and most were on a respirator, there was a range
of values, but the mean values were not different from
normals recorded. Later in their disease course, GBS
patients had higher mean urinary MBPLM values (Fig
2), near those of CPS-MS. Analysis of the results
Table 2. Urinary Specimem Anahzed for Myelin Bmic Protein-like Alaterial
No. of Patients
Clinical Classification
Controls
GCRC
Random
MS patients
GCRC urine collection
Relapsing-remitting
Relapsing-progressive
Chronic progressive, primary
Chronic progressive, secondary
Random urine collection
Relapsing-remi tting
Relapsing-progressive
Chronic progressive, primary
Chronic progressive, secondary
MS patients and treatment
ACTH
CyclophosphamideiACTH
Dexamethasone
Methylprednisolone
Prednisone
No treatment for 1 month
Other diseases
Guillain-Barre syndrome
Myelopath y
Stroke
Inflammatory CNS
Devic's syndrome
Miscellaneous
Renal disease
All non-MS
OND
Renal disease
MBP = myelin basic protein; MS
= other neurological disease.
=
ng of MBP Peptide 83-89i
mg of Creatinine ( SD)
*
No. of Specimens
8
10
175
111
106.9 (56.7)
87.2 (38.5)
8
175
3
66
44
185.9 ( 1 20.4)
100.8 (73.3)
165.2 (82.2)
139.2 357.7 (15.7)
180.0 (91.4)
150.1 (65.7)
184.1 (83.2)
185.1(1189)
208.8 (99.9)
2
1
2
158
39
63
30
26
21
44
1479
379
763
215
122
3
9
152.6 (57.0)
228.5 (87.3)
200.2 (49.')
1?6.4 (87.4)
220.0 (176.2)
188.9(131.4)
39
161
15
75 1
13
123
2
124
9
58
6
5
5
4
158
26
18
15
28
1
5
5
6
5
31
26
5
256
25 1
5
159.3
200.0
104.9
140.4
265.4
142.6
116.0
148.4
154.6
116.0
(95.0)
(46.6)
(6'.9)
(59.5)
-
(48.6)
(58.0)
(70.4)
(71.8)
(58.0)
multiple sclerosis; GCRC = General Clinical Research Center; CNS = central nervous system; OND
Table 3. Comparison of Lecels of Urznaq iMBPLM in Normal Controls. MS Subtypes, and Dueare Contro6
No. of
Subjects
Group
No. of
Specimens
10
Normals
MS
111
1479
158
MBP
(ngiml) p
Creaciiiine
(mgiml)
132.3"
(66.0)b
(0.341
169.6
1.50
0.1386'
1.07
(0.48)
0.0034
0.3347
1.19
(0.53)
1.08
(0.47)
0.0558
0.2927
0.99
(0.4
2)
0.0028
0.0976
0.94
(0.43)
0.78
(0.34)
0.0015
(74.9)
MS subtypes
Relapsing-remitting
379
39
Relapsing-progressive
61
763
Chronic progressive, primary
30
215
Chronic progressive, secondary
26
122
Disease controls
256
31
155.3
(66.3)
179.1
(77.1)
160.9
(72.3)
178.4
(84.0)
120.9
(89.4)
p
0.0663
0.7146
MBPi
Creatinine
(ngimg)
87.2
(38.5)
180.0
(91.4)
150.1
p
0.0012
0.0456
(65.7)
0.0083
184.3
(83.2)
185.1
(118.9)
208.8
0.0015
0.0028
0.0003
(99.9)
0.0001
148.4
(70.4)
0.0128
'Subject means.
'One SD.
'All p values derived from comparison with normals
MBPLM
=
myelin basic protein-like material; MS
=
multiple sclerosis; MBP
=
myelin basic protein.
Whitaker et al: Urinary MBPLM in MS
581
500
450
400
h
.-c
350
.-c
0
2
0
.
F
tn
C
300
00
m
I
250
03
04
05
06
Fig 3. The reiutionship of cerebro.rpinu1Rr4id (CSF) myelin basic protein-like material (MBPL.Mi was compared with the levels of urinavy MBPLM on the same patient during the same
200
24-hour period. CSF and urine samples were collected on 18
multiple .rclerosis patients. both relapsing-remitting and seconda.Py chronic-pwgressiw types. Urine and CSF values showed a
correhtion of - 0.2428. This did not reach statistical sign$cunce, but a trend Ip 0.101of negative correlation uiaJ
suggested.
150
.-C
5
02
CSF MBP (nglml)
v
3o.
01
100
-
50
0
1
2
3
4
5
6
7
8
Week of illness
Fig 2 . [Jrinavy myelin basic protein-like muterial in u total of
6 patients with Gzdain-Bavri syndrome (CBS, during their
disease courses. During the 8 weeki o f ~ t u d yI28 urinay specimens were az;ailublefor anabssis. These were spread among the
depicted means (SD)for weeL 1 ( 5 patients, 19 specimens), week
2 (5 putients. 20 specimens). week 3 (5 patients, 21 .specimens),
week 4 (5 patients, 21 specimens). week 5 14 patients. 14 specimens). week 6 (3 patients. 13 specimens). week 7 (2patients,
12 specimem), and week 8 (2patients, 8 specimens). Thirty
urine specimem on GAS patients (Table 21 were obtained
beyond week 8 of the diseuse und were not included in this
unalysiJ.
showed that the subject means of urinary MBPLM for
the GBS patients at 6 to 8 weeks of hospitalization
were significantly higher ( p = 0.039) than controls,
whereas the values for GBS patients in weeks 1
through 5 were not.
Relationship of Urinay and Cerebrospinal
Fluid MBPLM
The relationship of MBPLM in urine and CSF was
investigated by comparing values on each in 18 patients
with either RR or CPS forms of MS in whom CSF and
urine were available from the same 24-hour period.
Ten of the 18 had clearly elevated values of N . 1 6 ngi
ml of CSF MBPLM (271. Whether urinary MBPLM
was expressed either as nanograms of MBP peptide 83
to 89 per milliliter of urine (Fig 3) or as nanograms of
582 Annals of Neurology Vol 35 No 5 May 1994
MBP peptide 83 to 89 per milligram of creatinine (data
not shown), there was no relationship of CSF and urinary levels of MBPLM (see Fig 3). In fact, there was an
insignificant trend ( p - 0.10) for a negative correlation.
U r i n a v MBPLM Levels and FindingJ on Cranial
Magnetic Resonance Imaging
The lack of a relationship between the level of urinary
MBPLM and either clinically ascertained disease activity or the CSF level of MBPLM, which does correlate
well with acute CNS myelin damage (16, 211, led to
an initiation of attempts to correlate the level of urinary
MBPLM with cranial MRI. In the first preliminary
study (Group V), 8 RR-MS patients were subjected to
unenhanced cranial MRI on a 0.3-T instrument every 2
weeks over 2 months. In this small group the individual
urinary MBPLM levels on a total of 18 urine specimens
revealed no apparent relationship of MBPLM level and
the appearance of new areas of increased signal on T2weighted images. The mean for all specimens was 89.4
& 45.1 ng/mg (1 SD). The second study (Group VI)
was on 8 MS patients with a CPS course who had an
unenhanced cranial MRI on a 0.3-T instrument every
2 weeks over 6 months. Eleven to 14 urine specimens
were collected on 7 patients and 8 urine specimens
were collected on 1 patient at the time of cranial MRI.
The subject mean for the 8 patients was 124.4 & 46.5
ng/mg ( 1 SD). When separated into two groups based
on cranial MRI appearance of new lesions, the group
(n = 4 ) with 5 to 12 new lesions had a subject mean
urinary MBPLM value of 88.1 & 12.0 ng/mg (SD).
The group (n = 4 ) with 0 to 4 new lesions over the
6-month period had a subject mean of 160.7 & 37.2
ngimg (SD). Because of the small number of cases
available for analysis, no difference ( p = 0.2313) between the values of these two groups could be demonstrated.
While these results failed to show an immediate
change in urinary MBPLM with changes on serial cranial MRI, they suggested that levels of urinary
MBPLM might reflect some other process beyond
acute CNS myelin damage. I n 20 mild (mean GDSS
of 2.66) RR-MS patients (Group VII), the results of
gadolinium-enhanced cranial MRIs were correlated
with levels of urinary MBPLM. The mean of 12 1.7 ng/
mg of urinary MBPLM was not significantly different
from normal or disease control values or from the values of RR-MS patients in the other groups. Furthermore, there was no relationship apparent between the
mean level of urinary MBPLM with the number of
enhancing lesions ( p = 0.070) (Fig 4), new lesions ( p
= 0.088), clinical exacerbation ( p = 0.298), and EDSS
( p = 0.137).
Short-term Changes in Urinuy MUPLM
Related to Treutmenf
The 1,102 urine specimens from 205 patients were
analyzed as to a relationship of MBPLM with treatment. When patients were grouped, irrespective of M S
subtype, by treatment with intramuscular ACTH, intravenous methylprednisolone, oral prednisone, oral
dexamethasone, and intravenous cyclophospharnidei
ACTH and no treatment for 30 days, no significant
differences among groups were noted. Nor was the
mean urinary MBPLM value of the 979 specimens
from treated patients significantly different from all MS
patients studied in the GCRC (Group I ) (185.9 ngi
Fig 4. Correlation of the level ojurinao myelin basic proteinlike material with the number ~‘adoliniurn-enhancingle~ion.r
on serial crarzial magnetic reioname imaging. There is no statistical(y .signijicant correlation.
150
t
I
0
1
2
3
4
5
6
7
Number of enhancing lesions
0
9+
mg) or with regard to MS subtype (Group 111) (130.0
ngi mg ).
The previously reported group [22) of MS patients
who received a standard regimen of methylprednisoloneiprednisone was restudied for the correlation or
predictive rolc of urinary MBPLM and treatment outcome. The results were analyzed by studying the relationship to improvement by at least one Kurtzke level
between days 1 to 5 , 5 to 40, and 1 to 40 1221 of daily
urinary MBPLM on days 1 to 5 or the average value
of urinary MBPLM on days 1 to 5. This type analysis
used as a reference point the median of the 5-day average of urinary MBPLM, which was 161 ngimg of creatinine, for all 62 patients so treated. Of those patients
with an admission (day 1) urinary MBPLM value of
2161 ngimg, only 2 6 q showed an improvement of at
least one Kurtzke score point at day 5. However, in
those patients with a day 1 value of < 161 ngimg, 55%
improved to the same extent by day 5. In those patients showing no improvement by day 5 , 67% had a
urinary MBPLM 2161 ngimg, while in those patients
showing improvement at day 5 only 37% had a value
2161 ng/mg ( p = 0.041). Thus, the lower urinary
MBPLM on day 1 correlated with improvement at day
5. Day 1 values did not correlate with improvement
between days 1 and 40 nor days 5 and 40. Urinary
MBPLM on days 2, 3, 4, or 5 showed no relationship
to improvement at day 5 or day 40. The 5-day average
of urinary MBPLM did not correlate with improvement between days 1 and 5 or days 1 and 40, but those
with 5-day averages of <161 ngimg were more likely
( p = 0.015) to show improvement between days 5
and 40. The reported significant correlations of first
day and 5-day average values with response to treatment should be viewed cautiously in that neither met
the p value of 0.003 required for the Bonferroni adjustments. When examining the serial changes, the
slope, either positive or negative, showed no consistent
change with outcome to treatment (data not shown).
Changes in Urinuv MBPLM During Cyclosporine A
Therapeutic Trzal
From 13 patients with CPS-MS (see Table 1, Group
IX), 221 urine specimens were collected monthly for
25 months during the cyclosporine trial [26]. Six patients received placebo and ? received cyclosporine.
The mean values of urinary MBPLM were 141.2 ?
40.7 ngimg (SD) for the placebo controls and 149.1 ’-+
50.1 ngimg (SD) for the cyclosporine-treated group.
These measurements were not significantly different.
The slopes of longitudinal changes calculated were
1.7038 2 3.0139 (SD) for the cyclosporine-treated
group and - 1.0600 k 1.0990 (SD) for the placebo
controls. Neither the increasing values in the cyclosporine-treated group ( p = 0.0549) nor decreasing
slope in the placebo controls ( p = 0.0578) was sigWhitaker er al: LJrinary MBPLM in MS
583
nificantly different from zero or each other but approached significance from each other with the treated
group rising and the placebo group falling. There was
no relationship of values of urinary MBPLM and clinical measures of outcome.
Discussion
The first goal of this study was to enhance the validity
and accuracy of the RIA used for the detection of urinary MBPLM. This was accomplished through defining
the peptide of MBP which more suitably reflected the
MBPL.M. This proved to be MBP peptide 83 to 89,
and the utilization of this peptide as a standard led to
more accurate meaurements with greater recovery
and a numerical elevation of normal values previously
reported r231. Expression of the MBPLM in relation
to creatinine furnished a value that was less variable
and corrected for renal function. Even with these steps,
urinary MBPLM is highly variable although it is always
present in urine of adults. Based on studies of patients
in a controlled clinical research unit, the MBPLM does
not vary on a diurnal pattern throughout a 72-hour
period. This finding permitted greater flexibility for
choosing the time of urine collection for measurement
of MBPLM. It was shown that patients with MS, when
studied as a group, had values that were significantly
higher for MBPLW than were normal controls. Furthermore, the extent of the difference between MS
patients and related controls appeared even greater in
a small number of patients studied in a clinical research
unit so that the differences noted of urinary MBPLM
o n random urines may be less striking in their difference of MS and controls than might be noted had urine
been collected from subjects under identical conditions. ln persons in the research unit, 4 patients with
chronic progressive MS, especially of a secondary variety, had higher values than those who were in an early
RR phase of disease. This trend continued for all CPSM S patients, but the differences were less significant
because of the broad range of values. There was no
correlation with acute disease activity. This lack of correlation was demonstrated by showing that urinary
MBPLM did not correlate with a measure of acute
activity by CSF MBPLM, by cranial MR1, or by clinical
assessment.
There was no correlation with treatment and overall
subject mean values or with serial changes during
cyclosporine treatment of chronic progressive MS patients. The level of urinary MBPLM correlated marginally with the response to a course of intravenous methylprednisolone followed by oral prednisone EX!}.
Those patients with lower urinary MBPLM levels on
the first day of treatment or from a 5-day average
showed a greater likelihood of improvement between
days 1 and 5 or between days 5 and 40, respectively,
584 Annals of Neurology Vol 35
No 5
May 1994
during the treatment period. This may be reflecting
the fact that those patients with higher values of urinary
MBPLM are from a chronic progressive cohort who
would be expected to be less responsive to this treatment with glucocorticoids.
The goals of investigation on urinary MBPLM have
been primarily to correlate it with disease activity and
to define its exact chemical nature 124). The clinical
correlation has proved to be far more complex than
the direct relationship between disease activity, whatever the cause, and the appearance of elevated levels
of CSF MBPLM (14, 16). The solution of technical
requirements for the urine assay, the need to define
the specific epitopes involved 1231, awareness of the
crucial dependence of conformation on the immunochemical detection of MBPLM [28), and the correction
of the sequence in the peptide region of MBP residues
80 to 89 131) were requisite steps to make this investigation possible. Problems in defining the demands on
this assay have led to considerable problems in all laboratories addressing this issue [24}.
The exact chemical features of the MBPLM has resisted solution to date, mainly because of the difficulty
in obtaining enough purified material and because of
an apparently blocked amino terminus (J. N. Whitaker
et al, unpublished observations). Such an NH,
blockage could occur through binding or chemical
change in that the peptide of MBP irnmunochemically
detected should have no blocked NH, group. Of interest is that the region of MBP, i.e., peptide 80 to 89,
cross-reactive with MBPLM in CSF and urine [ 2 3 ] , is
from an area for a dominant epitope for cellular E32,
331 and humoral [34] immunity in MS patients.
Whether the urinary MBPLM is truly a peptide of
MBP or another cross-reacting or binding substance
cannot be determined unequivocally. Assuming that
the CSF and urinary MBPLM is a MBP peptide(s), its
characteristics and the mechanism for its entry into the
two body fluids are different. In studies of brain tissue
from MS patients, there is evidence that early in the
disease course oligodendrocytes can proliferate 135371 and remyelinate 138). Later, this proliferative response of oligodendrocytes subsides {37}. Since it is in
that latter phase where urinary MBPLM seems to be
the highest, it is possible that the urinary MBPLM
arises, not from direct injury to oligodendrocytes or
CNS myelin, but rather reflects the degradation of
MBP that has been synthesized by oligodendrocytes
but blocked in some manner from being incorporated
into the myelin sheath.
This investigation was supported by NIH grant NS 23240, by NIH
grant RK00032 to the General Clinical Research Center of UAB.
by the Research Program of the Veterans Administration, and by
the Thompson Fund for Multiple Sclerosis Research.
John J. Curtis, MD, and Katharine A. Kirk, PhD, furnished helpful
advice in regard to measurements of-urinary creatinine and expression of results and analysis of urine-CSF relationships, respectively.
Ms Joanne Cage and Ms Linda Brent provided excellent secretarial
assistance in the preparation of this manuscript.
References
1. Namerow NS, Thompson LR. Plaques, symptoms, and the remitting course of multiple sclerosis. Neurology 1969;19765-
774
2. Paty DW. Magnetic resonance imaging in the assessment of
disease activiry in multiple sclerosis. Can J Neurol Sci 1988;lS:
266-2 72
3. Weinshenker BG, Bass B, Rice GPA, et al. The natural history
of multiple sclerosis: a geographically based study. 2. Predictive
value of the early clinical course. Brain 1989;112:3419-1428
4. Weinshenker BG, Ricc GPA, Noseworthy JH, e t al. The natural history of multiple sclerosis: a geographically based study. j.
Multivariate analysis of predictive factors and models of outcome. Brain 1991;114:1045- 1056
5. Runmarker B, Andcrsen 0. Prognostic factors in a multiple
sclerosis incidence cohort with twenty-hvc years of follow-up.
Brain 1993;116:117-134
6. Aminoff MJ, Davis SL, Panitch NS. Serial evoked potential
studies in patients with definite multiple sclerosis: clinical relevance. Arch Neurol 1984;41:1197-1202
7. Iragui VJ, Wiederholt WC, Komine JS. Serial recordings of
inultimodality evoked potentials in multiple sclerosis: a four year
follow-up study. Can J Neurol Sci 1986;13:320-326
8. Paty DW, McFarlin DE, McDonald WI. Magnetic resonance
imaging and laboratory aids in the diagnosis of multiple sclerosis.
Ann Neurol 1991;2‘9:3-5
9. Thompson AJ, Miller D, You1 B, ct al. Serial gadoliniunienhanced MRI in relapsingiremitring multiple sclerosis of varying disease duration. Neurology 1992;42:60-63
10. Harris JO, Frank JA, Patronas N, et al. Serial gadoliniumenhanced magnetic resonance imaging scans in patients with
rarly, relapsing-remitting multiple sclerosis: implications for
clinical trials and natural history. Arch Neurol 1391;29:548-
555
11. Smith ME, Katz DA, Harris JO, e t al. Systemic histiocytosis
presenting as multiple sclerosis. Ann Neurol 1993;33:549-554
12. Paty DW, Li DKB, Duquette P, e t al. Interferon beta-lb is
effective in rclapsing-remitting multiple sclerosis. 2 . MRI analysis results of a multicenter, randomized, double-blind, placcbocontrolled trial. Neurology I993;43:662 -667
13. Smith ME, Stone LA, Frank JA, et al. MRI used to monitor
efficacy of early treatment of R/R MS. Neurology 1993;43:
A183 (Abstract)
14. Whitaker JN, Snyder DS. Myelin components in the cerebrospinal fluid in diseases affecting central nervous system myelin.
Clin lmmunol Allergy 1982;2:469-482
15. Cohen SK, Herndon RM, McKhann GM. Myelin basic protein
in Cerebrospinal fluid as an indicator of active demyelination.
N Engl J Med 1976;295:1454-1457
16. Whitaker J N . hfyelin encephalin)genic protein fragments in cerebrospinal fluid ot persons with multiple sclerosis. Keurology
1977;27:311-920
17. Kamholz J , Wrabetz L. Molecular genetics of myelin basic protein. In: Martenson RE, ed. Myelin: biology and chemistry. Boca
Katon, FL: CKC Press, 1932:597-607
18. Bashir KM, Whitaker JN. Molecular features of immunoreactive
myelin basic protein in cerebrospinal fluid of persons with multiple sclerosib. Ann Neurol 1980;7:30-57
19. Whitakcr JN, Gupta M, Smirh OF. Epitopes of immunoreactive
myelin basic protein in human cerebrospinal fluid. Ann Neurol
1Y86;20:3Z-i 36
20. Cohen SR, Brooks BR, Herndon RM, McKhann GM. A d i g nostic index of active demyeliimion: myelin basic protein in
cerebrospinal fluid. Ann Neurol 1980;8:25-11
21. Whitaker J N , Lisak RP, Bashir KM, et al. Inimunoreactive myelin basic protein in the cerebrospinal fluid in neurological disorders. Ann Neurol 1980;?:58-64
22. Whitaker JN, Layton BA, Herman PK, et al. Correlacion of
myelin basic protein-like material in cerebrospinal fluid of multiple sclerosis patients with their response to glucocorricoid treatment. Ann Neurol 1993;33:10-17
23. Whitaker JN. The presence of iinmunoreactive myelin basic
protein peptide in urine of persons with multiple sclerosis. Ann
Neurol 1987;22:648-655
24. Whitaker JN. Myelin basic protein peptides in urine. Ann Neurol 1992;11:345-348
25. Kurtzke JF. O n the evaluation of disability in multiple sclerosis.
Neurology 1961;11:686-694
26. The Multiple Sclerosis Study Group. Efficacy and toxicity of
cyclosporine in chronic progressive multiple sclerosis: a ranciomized, double-blinded, placebo-controlled clinical trial. Ann Neurol 1 Y90;2 7: 591-605
27. Whitaker JN, Herman PK. IHuman myelin basic protein peptide
69-89: imniunochemical features and use in immunoassays of
cerebrospinal fluid. J Neuroiinmunol 1988;19:47-57
28. Whitaker J N , Mvscarello MA, Herman PK, et al. Conformational correlates of the epitopes of human myelin basic protein
peptide 80-89.J Neurochem 1990;55:568-576
29. Snedecor GW, Cochran WG. Statistical methods. 7th cd. Ames,
Iowa: Iowa State University Press, 1980:2 15-222
30. Klcinbauni DG, Kupper LL, Mullcr KE. Applied regression
and other multivariable methods. 2nd ed. Boston: PWS-Kent,
1988:32
31. Gibson BW, Gilliom RD, Whitaker JN, Uieinann K. Amino
acid sequence of human myelin basic protein peptide 45-83 as
dcterinined by mass spectrometry. J Biol Chem 1984;259:
5028-5031
32. Om K, Matsui M, Milford EL, e t al. T-cell recognition of an
iminunodominant myelin basic protein epitope in multiple sclerosis. Nature 1990;146:183-187
33. Martin R, Jaraquemada D, Flerlage M, et al. Fine specificity and
HLA restriction of myelin basic protein-specific cytotoxic 7’ cell
lines from multiple sclerosis patients and healthy individuals.
J Immunol 1990;145:540-548
34. Martino G, Olsson T, Fredrikson S, e t al. Cells producing antibodies specific for myelin basic protein region 70-89 are predominant in cerebrospinal fluid from patients with multiple sclerosis. Eur J Jmmunol 1991;21:2971-2976
35. Raine CS, Scheinberg L, Waltz JM. Multiple sclerosis: oligollendrocytr survival and proliferation in an active established lesion.
Lab Invest I98 1;45:5 3 4-546
36. Prineas JW, Kwon EE, Goldenberg PZ, et al. Multiple sclerosis:
oligoclencirocyte proliferation and differentiation in fresh Iesions. Lab Invest 1989;61:489-503
37. Prineas JW’, Barnard RO, Kwon EE, et al. Mdtiple sclerosis:
rrniyelination of nascent lesions. Ann Neurol 199
38. Prineas JW, Connell HS. Remyelination in multiple sclerosis.
Ann Neurol 1979;5:22-31
Whicaker e t al: Urinary MBPLM in M S
585
Документ
Категория
Без категории
Просмотров
3
Размер файла
948 Кб
Теги
features, patients, material, findings, like, correlation, urinary, cranial, clinical, basic, magnetic, protein, imagine, myelin, sclerosis, multiple, resonance
1/--страниц
Пожаловаться на содержимое документа