Rheumatoid factor (RF) is found commonly in patients with systemic lupus erythematosus (SLE), and has been associated with a more benign disease course.^1,2 Anti-citrullinated peptide antibodies (ACPA) are more specific for rheumatoid arthritis (RA).^3–5 Several assays for ACPA detection have been developed: among others, an enzyme linked immunosorbent assay (ELISA) for anti-cyclic citrullinated peptide (anti-CCP) antibodies³ and a line immunoassay (LIA) for antibodies to peptide A (pepA) and peptide B (pepB), two synthetic citrullinated peptides.⁴ Few reports exist about the presence of ACPA in SLE. Although patients with SLE are often part of the control group when determining the specificity of ACPA for RA, SLE alone is seldom studied. Mediwake et al found that 3/66 patients with SLE were positive for anti-CCP1 antibodies; two of them had erosive arthritis.⁶ We investigated the presence of RF and three different ACPA (anti-CCP, anti-pepA, and anti-pepB antibodies) in SLE.

Two hundred and thirty five patients with SLE, meeting American College of Rheumatology (ACR) revised criteria for classification of SLE,^7,8 were prospectively included in four European centres. The study investigated associations between symptoms and specific antinuclear reactivities and has been reported elsewhere.⁹ Serum was available for further analysis in 201 patients. The male to female ratio was 25:176. The mean age was 40 years. The study was approved by the local ethics committees. Informed consent was obtained from all patients.

Fine antinuclear reactivities were determined with INNO-LIA-ANA Update (Innogenetics, Ghent, Belgium) and by indirect immunofluorescence on Crithidia luciliae. RF was detected using the latex fixation method (Becton Dickinson, Sparks, Maryland, USA). Titres ⩾160 were considered positive, which corresponds to a specificity for RA of 95.9% in an independent control cohort, consisting of 146 patients with rheumatic complaints but no RA (data not shown). Anti-CCP2 antibodies were detected by ELISA (Immunoscan RA, mark 2, Eurodiagnostica, Arnhem, Netherlands). A cut off value of 42 U/ml was used. Anti-pepA and anti-pepB antibodies were detected by a research LIA (Innogenetics).⁴ During each run, a strip was developed using a control serum, providing a cut off intensity for each antigen line. In the control population mentioned earlier, all three ACPA had a specificity of at least 98.5%.⁵ The RA associated HLA-DR shared epitope (SE) was determined with INNO-LiPA (Innogenetics).

χ² Tests were used to determine associations. Antibody frequencies were compared using the McNemar test.

Anti-CCP2 antibodies were found in 11/201 (5.5%) patients, anti-pepA antibodies 3 (1.5%) patients, and anti-pepB antibodies in 5 (2.5%) patients. Table 1 shows the characteristics of patients positive for ACPA. Anti-CCP2 antibodies were significantly more frequent then anti-pepA antibodies (p = 0.008), but not anti-pepB antibodies (p = 0.109). It is important to notice that in an independent control cohort all three ACPA obtained comparable specificities of at least 98.5%.⁵ Apparently, the different substrates behave differently in SLE. RF was found in 26 (12.9%) patients, which was significantly more frequent than anti-pepA (p<0.001), anti-pepB (p<0.001), and anti-CCP2 antibodies (p = 0.006). Although the diagnosis in the ACPA positive patients was SLE, and all fulfilled classification criteria for SLE,^7,8 ACR criteria for RA¹⁰ were also fulfilled in 6/10 evaluable patients, with 3/10 carrying an SE allele; radiographic erosions were present in 3/7 evaluable patients.

Our data suggest that the presence of ACPA does not exclude a diagnosis of SLE. It remains to be evaluated whether ACPA in SLE predispose for a chronic RA-like arthritis in this case.