The value of anti-neutrophil cytoplasmic antibody (ANCA) detection for monitoring disease activity in ANCA-associated systemic vasculitis (AASV) remains controversial. The aim of our work was to rate the performance of a new automated fluorescence PR3 and MPO-ANCA immunoassay (EliA) for monitoring disease activity in AASV. We evaluated 100 serum samples from 71 AASV patients (with Wegener's granulomatosis, microscopic polyangiitis, and Churg-Strauss syndrome) as well as sera from 58 pathological and 35 normal controls. In addition to PR3 and MPO-ANCA EliA, we performed indirect immunofluorescence and "homemade" PR3 and MPO-ANCA ELISA tests. In AASV patients, ANCA levels were correlated with disease activity, according to the Birmingham Vasculitis Activity Score (BVAS). We derived cutoff limits from receiver operating characteristic (ROC) curve analysis comparing AASV with pathological controls. Our results showed that EliA and ELISA had comparable sensitivity (76%) and specificity (95%). The analysis of active versus inactive status and correlation with ANCA levels showed a clear difference between BVAS Group I (score < or = 4) and BVAS Group II (scores > 4) (AUC = 0.86 vs. 0.72; relative risk [RR] = 2.4; P < 0.0001) for PR3-ANCA, but not for MPO-ANCA (AUC = 0.94 vs. 0.87; RR = 1.48; P = 0.46). Serial serum samples from 16 patients were examined in detail. For the majority of patients, for both PR3 and MPO-ANCA, change in titer was strongly associated with change in BVAS score. Our data showed a good correlation between ANCA titer (especially for PR3) and AASV disease activity. We recommend that ANCA titer be used to monitor AASV disease activity with the caveat that a few exceptions, in particular with MPO-ANCA, are possible.