Primary Sjögren syndrome (pSS) is a chronic, slowly progressive, inflammatory, autoimmune disease that is characterised by lymphocytic infiltration of the exocrine glands (which reduces or eliminates glandular secretion) and marked B-lymphocytic cell hyper-reactivity, which initially is manifested by a variety of serum autoantibodies, including those against Ro (SSA) and La (SSB), and rheumatoid factor (RF) 1234.

Most patients with pSS present only with keratoconjunctivitis sicca and xerostomia, but approximately 40% develop extra-glandular musculoskeletal manifestations; lung, kidney, and skin involvement; vasculitis; neuropathy; and lymphoma 12. The most common are arthralgia and an intermittent non-erosive polyarthropathy affecting mainly the small joints, which means that the picture may mimic that of rheumatoid arthritis (RA), particularly as 50% to 80% of cases are RF-positive 45. However, the fact that tests for cyclic citrullinated peptide antibodies (anti-CCP) usually are negative may help to differentiate the two conditions. Anti-CCP antibodies, which were first described in 1998, were found to be highly specific in the diagnosis of RA (95%) and only slightly less sensitive than IgM RF (60% to 70%) 67, whereas the second generation of anti-CCP antibodies have a sensitivity of 80% and a specificity of 98%. Follow-up studies of patients with early RA have demonstrated that anti-CCP antibodies independently predict the development of erosions 89, but Goëb and colleagues 10 found anti-CCP autoantibodies in only 4% of 137 women and 16% of 11 men with pSS.

Gottenberg and colleagues 11 studied a cohort of 134 patients with pSS and found that 7.5% of the serum samples were positive for anti-CCP antibodies and 5.2% were positive for anti-keratin antibodies (AKAs) without any radiographic evidence of erosion after a long follow-up. They suggested that the anti-CCP-positive patients, who may be prone to developing RA, require cautious clinical and radiographic follow-up. The aims of this study were to verify the prevalence of anti-CCP antibodies in patients with pSS and to investigate any associations with their clinical and laboratory characteristics.

Fourteen (9.9%) of the patients with pSS had moderate to high levels of anti-CCP antibody, and 94 (66.7%) were RF-positive, 134 (95%) ANA-positive, 115 (81.6%) SSA-positive, and 55 (39%) SSB-positive (Table 1). The mean value for anti-CCP antibodies was 46 IU/mL (28 to 78), with a normal range (as defined by the manufacturer) of less than 5 IU/mL. None of the 14 patients fulfilled the ACR criteria for RA. The focus score at lip biopsy was greater than 1 in 94 patients (67.1%). Extra-glandular involvement was found in 81 patients (57.4%) (Table 2), and synovitis was found in 44 (31.2%) without any radiographic evidence of erosions (Table 1).

Musculoskeletal manifestations such as fatigue, myalgia, arthralgia, an intermittent non-erosive polyarthropathy affecting mainly the small joints, and usually mild synovitis are common in patients with pSS and may mimic RA, particularly in the presence of RF 2318, although the fact that tests for anti-CCP antibodies usually are negative may help to differentiate the two conditions 1920.

We found slightly high anti-CCP levels in 14 of our 141 patients with pSS (9.9%). Some studies have found that anti-CCP antibodies are an independent factor predicting the development of erosions in patients with RA 89, but none of our 44 pSS patients with synovitis (31.2%) showed any radiographic sign of erosions. After a long follow-up, Gottenberg and colleagues 11 found that 80 out of 134 pSS patients with no radiographic evidence of erosions (59%) were positive for IgM RF, 10 (7.5%) for anti-CCP, 7 (5.2%) for AKA, and 5 (3.7%) for both anti-CCP and AKA.

We also found no significant differences between our anti-CCP-positive and -negative patients in terms of demographic factors or extra-glandular manifestations. However, the presence of synovitis in pSS patients did seem to be closely associated with the presence of anti-CCP antibodies as the odds of having synovitis were 1.475 times greater in our anti-CCP-positive patients, regardless of the other covariables in the model. There were no other significant associations with clinical or laboratory parameters between the anti-CCP-positive and -negative patients.

In addition, anti-SSB positivity was weakly and inversely associated with the prevalence of synovitis. Antibodies to the Ro/SSA and La/SSB ribonucleoprotein particles usually are found in the serum of pSS patients and are associated with a longer disease duration, more frequent non-exocrine manifestations, and more intense lymphocytic infiltration of the minor salivary glands 21. The prevalence of anti-SSB is higher in pSS, but in our patient cohort it was also inversely related to the presence of synovitis.

The possibility that our anti-CCP patients represent a subgroup of patients with RA and secondary SS was ruled out by excluding patients fulfilling ACR criteria at the beginning of the study and re-evaluating all of the anti-CCP-positive patients. In our study, we decided to be very stringent in the application of ACR criteria, and we classify as RA those patients in whom clinical symptoms (symmetrical arthritis, arthritis of the hands, more than three groups of joints involved, and morning stiffness) were simultaneously present at the moment of evaluation or for at least 6 weeks in the past. In fact, even if we had considered the possibility of a diagnosis of RA for symptoms that were successive and not simultaneously present, none of the 14 patients would have been classifiable as having RA (data not shown).

Indeed, the possibility that patients with anti-CCP antibodies may develop RA cannot be ruled out by the present study, and it is possible that in the future anti-CCP will be interpreted as an additional marker of definite RA. It should be noted, however, that anti-CCP antibodies are strictly associated with the development of erosions in RA and that none of our patients developed erosions after a mean follow-up of 2.4 years.

As a matter of fact, anti-CCP antibodies may also be observed in a minority of patients with other systemic autoimmune diseases. Takasaki and colleagues 22 detected anti-CCP antibodies in patients with mixed connective tissue disease (MCTD) (9%), systemic lupus erythematosus (14%), systemic sclerosis (13%), polymyositis/dermatomyositis (14%), and SS (18%). Some of these patients presented overlapping RA and connective tissue diseases (particularly MCTD) and so, as in the case of other autoimmune diseases, anti-CCP positivity in pSS may be a marker of synovial tissue inflammation. The results of our study suggest that the production of anti-CCP antibodies may be less related to the pathogenesis of RA. It is possible that marked B-lymphocyte hyper-reactivity (a characteristic of pSS) may explain the presence of anti-CCP antibodies, as usually is observed in the case of RF and anti-SSA and anti-SSB antibodies 23.

Only a minority of patients with pSS are anti-CCP-positive, but the prevalence of synovitis seems to be closely associated with the presence of anti-CCP antibodies as multivariate analysis confirmed the association between anti-CCP positivity and an increased prevalence of synovitis. However, the possibility that patients with anti-CCP antibodies may develop RA cannot be ruled out, particularly in patients with the concomitant presence of IgM RF.

ACR = American College of Rheumatology; AKA = anti-keratin antibody; ANA = antinuclear antibody; anti-CCP = cyclic citrullinated peptide antibody; CI = confidence interval; ELISA = enzyme-linked immunosorbent assay; ENA = anti-extractable nuclear antigen; ESR = erythrocyte sedimentation rate; MCTD = mixed connective tissue disease; OR = odds ratio; pSS = primary Sjögren syndrome; RA = rheumatoid arthritis; RF = rheumatoid factor; SS = Sjögren syndrome.

The authors declare that they have no competing interests.

FA and PS-P contributed to the conception of the study and the acquisition, analysis, and interpretation of data and participated in drafting the manuscript. NL participated in the analysis and interpretation of data. EB and FB-P contributed to the acquisition of data. RC contributed to the interpretation of data and to the critical review of the manuscript. CM provided the final approval of the version of the manuscript to be published. All authors read and approved the final manuscript.