We evaluated the diagnostic value of anti-cyclic citrullinated peptide 2 (anti-CCP2) antibodies and other potential diagnostic biomarkers (IgM rheumatoid factor, anti-agalactosyl IgG antibodies, matrix metalloproteinase 3, C-reactive protein) for predicting early development of rheumatoid arthritis (RA). Patients were defined as having recent-onset undifferentiated arthritis (UA) if they had developed arthritis in two or more joints within the previous 2 years and could not be classified with a well-defined arthropathy. Baseline levels of biomarkers were measured in blood samples collected at the entry of the study and the patients were followed for 1 year to monitor development of RA. Diagnoses of RA and non-RA arthropathies were made according to individual standard diagnostic criteria. A total of 146 patients were enrolled in the study. In the follow-up year, 18 patients developed RA, 54 developed non-RA arthropathies, and 60 remained in the UA category. The sensitivity and specificity of the presence of anti-CCP2 antibodies for the diagnosis of RA were 83.3 and 93.0%, respectively. The positive predictive value (PPV), negative predictive value (NPV), and diagnostic accuracy of anti-CCP2 antibodies for RA (65.2, 97.2, and 91.7%, respectively) were higher than for any other biomarker. Combination of anti-CCP2 with any other biomarker only slightly improved each diagnostic value compared to the presence of anti-CCP2 alone. Among the anti-CCP2-positive patients, the average titer was significantly higher in those with RA than in non-RA or UA patients (163.7 +/- 138.4 vs 55.2 +/- 72.0 U/ml, p = 0.017). Anti-CCP2 antibodies are superior to any other single biomarker for predicting early development of RA in patients with recent-onset UA and the diagnostic value of anti-CCP2 alone is similar to that for biomarker combinations. Moreover, the anti-CCP2 antibody titer is useful to discriminate between patients at high risk for early developing RA from those at risk of developing non-RA arthropathies.