вход по аккаунту


Tumor necrosis factor receptor p75 correlates better than p55 with disease activity in juvenile rheumatoid arthritisComment on the article by Mangge et al.

код для вставкиСкачать
Vol. 39, No. 5 , May 1996, pp 883-886
0 1996, American College of Rheumatology
Tumor necrosis factor receptor p75 correlates better
than p55 with disease activity in juvenile rheumatoid
arthritis: comment on the article by Mangge et a1
To the Editor:
We have read with interest the article by Mangge and
coworkers (1) on the correlation of serum cytokines, interleukin-2 (IL-2), and tumor necrosis factor (TNF) p55 soluble
receptors with disease activity in juvenile rheumatoid arthritis (JRA), and we would like to make some comments.
In their study, the authors found a significant correlation between p55 soluble TNF receptors (sTNFR) and
disease activity in all JRA subtypes. Although TNFa was
reportedly significantly increased in the sera of JRA patients
compared with controls, no correlation was found between
TNFa serum levels and markers of inflammation. Other
recent reports (2,3) have shown that a low amount of TNFa
could be found in the sera and synovial fluids of JRA
patients, especially when compared with the levels of other
proinflammatory cytokines (i.e., IL-1 and IL-6). These findings have raised questions about the actual role played by
TNFa in the pathogenesis of JRA.
The elevation of serum sTNFR during disease activity has been previously reported in a study of adults with
rheumatoid arthritis (RA), and was explained as being secondary to a transient increase in TNFa synthesis, which can
be underestimated because of its short half-life. It is of note
that in RA, p75 sTNFR serum levels have been found to be
3-4-fold higher than those of p55 sTNFR (4).
So far, we have studied p55 and p75 sTNFR serum
levels in 36 JRA patients (12 systemic, 9 polyarticular, and
15 pauciarticular) during periods of high and low disease
activity (total of 74 sera) and in 20 healthy age-matched
controls. Clinical parameters for disease activity were
weekly fever score and/or swollen joint score (both ranging
from 0 to 3). C-reactive protein (CRP) levels, hemoglobin
(Hgb) concentration, and erythrocyte sedimentation rate
(ESR) were the laboratory parameters of disease activity.
High disease activity was assessed by fever and/or swollen
joint scores that ranged from 2 to 3, and ESR and/or CRP
elevation above the normal range. Low disease activity was
defined as fever and swollen joint scores that ranged from 0
to 1, and ESR and CRP levels within the normal range.
Healthy patients attending our clinic for a control visit were
used as controls after informed consent was given by the
parents. History of inflammatory or infectious disease in the
4 weeks before the examination was considered a criterion
for exclusion from the study.
Consistent with the data reported by Mangge et al,
we found that both p55 and p75 correlated positively with the
ESR (p55 r = 0.56, P < 0.001; p75 r = 0.64, P < 0.001) and
CRP levels (p55 r = 0.48, P < 0.001; p75 r = 0.52, P <
0.001). In contrast, the Hgb level correlated negatively with
both sTNFR (p55 r = -0.51, P < 0.001; p75 r = -0.67, P <
0.001). Furthermore, a significantly higher increase in the
p75 sTNFR serum levels, as compared with p55 sTNFR,
was found (Figure 1). Thus, in our experience, the p75
sTNFR seemed to be a more reliable disease activity parameter than the p55 sTNFR.
Ifi L
0High activity
13LOW activity
Figure 1. Comparison of serum p75 and p55 soluble tumor necrosis
factor receptor levels in systemic, polyarticular, and pauciarticular
juvenile rheumatoid arthritis patients, and in healthly controls.
Statistical analysis was performed using the Mann-Whitney U test.
* = P < 0.05, ** = P < 0,001, for high disease activity versus low
disease activity. + = P < 0.02, ++ = P < 0.001, for high disease
activity versus controls.
A possible explanation might be the recent in vitro
observation that when TNFa binds to its cellular receptors,
it causes the internalization of the p55 sTNFR at the same
time that the p75 sTNFR sheds from the surfaces (5).
Thus, consistent with the findings of Mangge et al,
both p55 and p75 serum sTNFR, as opposed to TNFa itself
(1-3), represent sensitive, albeit nonspecific, markers of
inflammation in JRA, and correlate with disease activity
parameters. Moreover, in our opinion, the study of sTNFR
may help us to better investigate the role of TNFa in the
pathogenesis of JRA, as has been shown for RA (6).
Marco Gattorno, MD
Paolo Picco, MD
Antonella Buoncompagni, MD
Franca Stalla
Vito Pistoia, MD
Institute for Children “G. Gaslini”
Genoa, Italy
1. Mangge H, Kenzian H, Gallistl S, Neuwirth G, Liebmann P,
Kaulfersch W, Beaufort F, Muntean W, Schauenstein K: Serum
cytokines in juvenile rheumatoid arthritis: correlation with conventional inflammation parameters and clinical subtypes. Arthritis Rheum 38:211-220, 1995
2. Lepore L, Pennesi M, Saletta S, Perticarari S, F’resani G, Prodan
M: Study of IL-2, IL-6, TNF-a, INFy and p in the serum and
synovial fluid of patients with juvenile chronic arthritis. Clin Exp
Rheumatol 12561-565, 1994
3. Madson KL, Moore TL, Lawrence JM, Osborn TG: Cytokine
levels in serum and synovial fluid of patients with juvenile
rheumatoid arthritis. J Rheumatol 21:2359-2363, 1994
4. Cope AP, Aderka D, Doherty M, Engelmann H, Gibbons D,
Jones AC, Brennan FM, Maini RN, Wallach D, Feldmann M:
Increased levels of soluble tumor necrosis factor receptors in the
sera and synovial fluid of patients with rheumatic diseases.
Arthritis Rheum 35:116&1169, 1992
5. Higuchi M, Aggarwaal BB: TNF induces internalization of p60
receptor and shedding of the p80 receptor. J Immunol 152:355&
3558, 1994
6. Cope P, Maini RN: Soluble tumor necrosis factor in arthritis. J
Rheumatol22:382-384, 1995
To the Editor:
We are pleased to learn that Gattorno and coworkers
essentially confirm the results of our study on serum cytokines in different subsets of juvenile rheumatoid arthritis
(JRA). We also agree with their conclusion that both p55 and
p75 sTNFR represent sensitive, but nonspecific, markers of
inflammation in JRA, and that the study of sTNFR may help
us to better understand the role of TNFa in the pathogenesis
of juvenile rheumatoid disease.
However, their results as presented in Figure 1 show
a discordance with our observations concerning p55 sTNFR
levels in pauciarticular JRA. These levels were found by
Gattorno et a1 to be within the normal range, in comparison
with healthy controls. In our study, patients with clinically
active pauciarticular arthritis of both type I and I1 showed
significantly elevated serum levels of p55 sTNFR, but normal conventional markers of inflammation in type I1 pauciarticular JRA. Furthermore, we observed significant differences in p55 sTNFR levels and other cytokines between type
I and type I1 JRA. One of the messages of our report was
that these observations support a different pathogenesis of
these two types of JRA, and may be of clinical usefulness in
making a differential diagnosis. We therefore suggest separate analysis of p75 and p55 sTNFR levels in type I and type
I1 pauciarticular JRA, since there is growing evidence that
these subsets are different disease entities according to their
clinical presentation as well as their laboratory features.
Harald Mangge, MD
University of Graz School of Medicine
Graz, Austria
Serum prwmatrix metalloproteinase 3 in rheumatoid
arthritis: a reflection of local or systemic
To the Editor:
We read with interest the article by Manicourt et a1
(1) in which they suggest that, within their rheumatoid
arthritis (RA) study population, a significant proportion of
matrix metalloproteinase 3 (MMP-3 ; stromelysin-1) may
originate from other sources besides the inflamed joints. To
the contrary, we believe that the majority of MMP-3 measured in the serum of patients with arthritis is actually
produced locally in the joints, where inflammation is most
When examining paired serum and synovial fluid
(SF) samples from 27 RA patients (10 men, 17 women, mean
age 61.7 years, mean k SD disease duration 57.5 k 13.5
Table 1. Levels of pro-matrix metalloproteinase 3 (proMMP-3)
and p75 and p55 tumor necrosis factor receptors (TNFR) in paired
serum and synovial fluid (SF) samples from 27 patients with rheumatoid arthritis
p75 TNFR
p55 TNFR
months, 17 patients taking second-line agents), we found the
mean MMP-3 concentration in SF to be -290-fold higher
than that in serum. This finding is in accordance with other
previous reports (2,3) and supports the concept of MMP-3
being produced locally within inflamed joints.
In 15 patients with inflammatory arthritides and
single joint flares (11 with RA, 3 with seronegative spondylarthropathy, 1 with acute gout), serum levels of MMP-3 were
reduced significantly (P = 0.017) 14 days after intraarticular
steroid injection. Our finding that a single steroid injection
into just 1joint reduces serum levels of proMMP-3 indicates
that it originates predominantly from the inflamed joint.
In addition, we measured the level of tumor necrosis
factor (TNF) receptors, p55 TNFR and p75 TNFR, in the
paired RA samples, since the TNFa cytokine system is
central to the inflammatory process in RA and is a potent
inducer of metalloproteinase synthesis by synovial fibroblasts. The median SF concentration of both TNFR types
was -3-5-fold higher than that in serum (4), supporting the
notion of localized production. SF and serum levels were
significantly correlated, in the order of MMP-3 > p75TNFR
> p55TNFR (Table 1). Thus, the higher SF-to-serum ratio
and closer correlation provides further evidence that
changes in serum MMP-3 reflect the extent of local joint
inflammation more than do changes in either of the soluble
TNF receptors.
Most authorities would agree that, in clinical practice, C-reactive protein (CRP) is a good marker of systemic
inflammation and is sensitive to change. Its positive correlation with MMP-3 levels led Manicourt et al to suggest that
both are markers of a systemic response to cytokines present
in blood or other body fluids. However, our measurements
in patients with other diseases suggest that proMMP-3 is not
another nonspecific marker of the acute-phase response to
inflammation. In diseases in which inflammation exists and
joints are not involved, for example in a group of critically ill
patients (n = 20) with multiple organ failure due to septic
shock, severe trauma, or adult respiratory distress syndrome, we found no correlation between levels of serum
proMMP-3 and CRP. These patients had greatly increased
serum CRP levels (68-326 Erglml), while serum proMMP-3
levels were not significantly greater than in a normal control
population (5).
Furthermore, we have observed no correlation between serum CRP and proMMP-3 levels in acute graftversus-host disease (GVHD), which developed in patients
following bone marrow transplant during treatment for various hematologic malignancies. Interleukin-1 (IL-1) and
TNFa are thought to be central in this disease and, as in RA,
Без категории
Размер файла
258 Кб
better, articles, p55, p75, disease, mangga, necrosis, factors, arthritiscomment, correlates, activity, juvenile, receptov, tumors, rheumatoid
Пожаловаться на содержимое документа