Home

Current Issue

Archives

Guidelines for Authors

Classified Ads

Links

Search PubMed

Subscriptions

Subscriber Registration

Guidelines for Website Users

JRheum Update Service

Contact Info

Antibodies to Cyclic Citrullinated Peptides in Psoriatic Arthritis: Do Classification Criteria Affect Study Results?

To the Editor:

We read with interest the article concerning the presence of antibodies to cyclic citrullinated peptides (CCP)¹ and as we are performing a similar study of a population from virtually the same geographical area, we would like to describe our interim data and make some comments.

Our aim is to establish the usefulness of anti-CCP antibodies in discriminating psoriatic arthritis (PsA) and rheumatoid arthritis (RA). Our inclusion criteria were the same as those used by Bogliolo, et al¹: consecutive outpatients with a diagnosis of PsA according to Moll and Wright². As this definition leaves it up to the researcher to decide whether or not to include cases positive for rheumatoid factor (RF), its specificity depends on how patients with RA with concomitant psoriasis are ruled out. In our study, patients with RA with psoriasis and RF-positive patients with true PsA were distinguished on the basis of an expert's opinion. More precisely, the patients with features characteristic of RA (rheumatoid nodules, vasculitis, typical RA radiographic changes) were excluded from the study, whereas RF-positive patients with features characteristic of PsA (dactylitis, distal interphalangeal involvement, axial involvement, and typical PsA radiographic changes) were included. Anti-CCP antibodies were measured in sera using a commercially available second generation ELISA kit (INOVA kit, Quantalite Ltd. San Diego, CA, USA), for which the upper normal limit suggested by the manufacturer is 20 UI/ml.

We have so far enrolled 129 patients with PsA whose main demographic and clinical characteristics are shown in Table 1. At the threshold value of 20 UI/ml, only 4 (3.5%) patients were anti-CCP positive, with serum levels of 32, 43, 126, and 159 UI/ml. Six patients had anti-CCP values of 10-20 UI/ml, and the remaining 119 values were < 10 UI/ml. All anti-CCP positive patients had polyarticular disease, and no correlation was found between the presence and titer of these antibodies and the markers of disease severity (number of irreversibly damaged joints, erosive disease, disability as measured by the Health Assessment Questionnaire, and therapy with disease modifying antirheumatic agents). Interestingly, the 35 patients (27.1%) in our series who were positive for RF (> 20 UI/ml) included the 2 patients with the highest anti-CCP values.

Table 1. Main demographic and clinical characteristics of 129 study patients with PsA. Values are number of patients (%), unless otherwise indicated.

In the study by Bogliolo, et al¹, 16 out of 102 (15.7%) patients with PsA were positive for anti-CCP antibodies, a significantly higher percentage than we found (p ≤ 0.001 by Fisher's exact test). All positive values were > 40 UI/ml, indicating that the large number of positive anti-CCP cases was not due to an exceedingly low normal limit (5 UI/ml). Finally, the median positive anti-CCP value was higher in Bogliolo's study population than ours: > 100 UI/ml versus 84 UI/ml, respectively.

Comparison of the 2 studies reveals significant differences in the number of patients with PsA with a positive anti-CCP test (15.7% vs 3.5%), which is unlikely to have been due to mere chance because the patients were comparable in terms of ethnicity, demographics, the main clinical features (Table 1), and the level of the attended center (both tertiary). There are 2 reasonable explanations: a difference in the sensitivity of the tests and/or the inclusion of patients with RA and concomitant psoriasis. The first possibility cannot be verified (the same sera should be evaluated by both tests) but seems to be quite remote as both studies used second-generation enzyme-linked assays; a misclassification of some patients appears to be much more likely. It has been shown that Moll and Wright's definition of PsA is sensitive and specific when RF-positive cases are excluded³, but it clearly loses part of its specificity if they are included. Bogliolo, et al¹ stated that only 6 out of the 16 anti-CCP positive patients had unmistakable evidence of PsA, while the remaining patients could have had RA coexisting with psoriasis, thus confirming the relative lack of specificity of diagnosis. We used the same diagnostic definition in our study, but also classified the patients as having PsA only if RA could be reasonably ruled out on sound clinical and radiographic grounds. In this regard, as about 27% of the patients were RF positive, our expert considered that this positivity had poor discriminant value. Comparison of the 2 studies seems to indicate that this is a classical case of a difference in inclusion criteria leading to a significant difference in results, which is very important when the sensitivity and specificity of a new test in a specific rheumatic disorder are still unknown.

The main conclusion we can draw is that great attention needs to be paid to diagnostic/classification criteria when performing this kind of study. In the case of PsA, the relative confusion generated by the existence of at least 7 sets of classification criteria³ should be reduced by the recent proposal of new criteria⁴. Another possible conclusion is that it is indeed virtually impossible to discriminate PsA and RA in a few cases. The coexistence of PsA and RA may have played a role but only in a few cases because, given the approximately 1% prevalence of RA, mere chance would have meant that only one of Bogliolo's 102 patients with PsA should have had coexisting RA. Finally, whether anti-CCP antibodies will be as helpful in predicting disease severity in PsA as in RA⁵ will depend on their true frequency in PsA.

ANTONIO MARCHESONI, MD; ALFREDOMARIA LURATI, MD; FRANCESCA DESIATI, MD; VALERIA ROSSI, MD; NORMA BATTAFARANO, MD, Rheumatology Department, G. Pini Orthopedic Institute, University of Milan, Via Pini 9, 20122 Milan, Italy. Address reprint requests to Dr. Marchesoni.

REFERENCES

Search PubMed for: