Article Text


Is soluble CD44 variant isoform 5 useful as a predicting factor and a parameter for long term observation in rheumatoid arthritis?
  1. M Skoumal1,
  2. G Kolarz1,
  3. G Haberhauer2,
  4. J Feyertag2,
  5. A Wottawa1
  1. 1Institute for Rheumatology in cooperation with the Donau Universität Krems, Marchetstrasse 78, 2500 Baden, Austria
  2. 25th Department of Internal Medicine with Rheumatology, Wilhelminenspital der Stadt Wien, Montleartstrasse 37, 1160 Vienna, Austria
  1. Correspondence to:
    Dr M Skoumal, Rheumasonderkrankenanstalt der SVA gewerbliche Wirtschaft, Adolfine Malchergasse 1, 2500 Baden, Austria;

Statistics from

In addition to its widespread expression, CD44 is a homing receptor for human T cells and an adhesion molecule in cell-cell and cell-matrix interaction.1,2 CD44 is a receptor for hyaluronan and other extracellular matrix and cell surface proteins. CD44 has been characterised in detail on lymphocytes, macrophages, fibroblasts, epithelial cells, and keratinocytes. It is part of the regulation of immune mediated inflammation. The binding of CD44 leads to lymphocyte homing, lymphocyte binding to high endothelial venules, myelopoiesis, lymphocyte activation, natural killer target lysis, B cell lymphopoiesis, cell adhesion, inflammation, and tumour dissemination.3 Binding of CD44 monoclonal antibodies or hyaluronan induces intracellular signals that enhance T lymphocyte mitogenesis and releases cytokines (interleukin 1, tumour necrosis factor α) from macrophages. Furthermore, it has been suggested that extravasation of T lymphocytes is mediated by CD44.

The soluble CD44 isoform variant 5 (sCD44v5) was found to be increased in the serum of patients with active rheumatoid arthritis (RA).4,5 So far no longitudinal studies have been carried out.

We examined whether sCD44v5 correlates with inflammation or joint destruction, or both, in the long term treatment of patients with RA.


Serum levels of sCD44v5 were measured in 81 patients with RA according to the criteria of the American College of Rheumatology, who were admitted at least three times during an observation period of 10 years (table 1). A commercially available enzyme linked immunosorbent assay (ELISA), developed by Bender-Med Systems, Vienna, was used to measure sCD44v5. The influence of rheumatoid factor (RF) on the measurements of sCD44v5 was ruled out by dilution experiments. The results of sCD44v5 at the first visit and the area under the curve as measure for the long term course were compared with the disease activity score, 28 joint count, 30 swollen joint count, Ritchie index, proximal interphalangeal score, Stoke index, Larsen score, Steinbrocker stage, C reactive protein, erythrocyte sedimentation rate, RF, and treatment with disease modifying antirheumatic drugs (DMARDs) as well.

Table 1

Clinical data of 81 patients (57 female, 24 male) at the first visit

When appropriate, results were analysed by repeated measures analysis of variance, general linear model, and Spearman’s correlation coefficient.

We detected raised serum levels of sCD44v5 in 50/356 measurements, with a range from 58.4 to 312.9 ng/ml. The mean sCD44v5 was 42.3 ng/ml. We could not find a significant correlation between sCD44v5 and the clinical and laboratory data of visit 1, except for a correlation with RF (table 2). Increased Steinbrocker stages and Larsen scores did not show a correlation with sCD44v5. To demonstrate sCD44v5 as a predicting factor we therefore correlated the sCD44v5 values of the first visit with Δ Larsen, but the result was not significant.

Table 2

Spearman correlation coefficient of sCD44v5 at visit 1 and AUC sCD44v5 compared with clinical and laboratory parameters


SCD44v5 is described as a measure of inflammation and as an indicator for treatment response in RA.6

We aimed at examining whether sCD44v5 might be useful as a predicting factor and measure of treatment response to different DMARDs. Although sCD44v5 is known to be raised in patients with rheumatoid flare, such a correlation could not be shown in our study of patients with weakly to moderately inflamed disease. Soluble CD44 seems to be raised only in severe inflammation.7 We found no evidence that sCD44v5 might be a predicting factor. In particular, no significant difference was found between sCD44v5 and Steinbrocker stages or the Larsen score.

Earlier studies have shown that sCD44v5 has an important role as an indicator of inflammation in RA, but in our long term study sCD44v5 had no clinical relevance.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.