Antineutrophil cytoplasmic antibody (ANCA)-associated vasculitides are a heterogeneous group of diseases corresponding to necrotising inflammation of small vessels with a wide range of clinical presentations. At least two of the diseases are believed to exhibit a common ground of pathophysiological mechanisms. These are granulomatosis with polyangiitis (GPA, formerly known as Wegener's granulomatosis) and microscopic polyangiitis (MPA). ANCA directed against proteinase 3 (PR3) are preferentially associated with GPA, and anti-myeloperoxidase (MPO) ANCA are associated mainly with MPA and eosinophilic GPA (formerly known as Churg-Strauss syndrome). Anti-MPO and anti-PR3 antibodies can activate neutrophils in vitro. In vivo data are available for humans and mice on the pathogenicity of anti-MPO but it is more controversial for PR3-ANCA. A recent genome-wide association study of patients with ANCA-associated vasculitides confirmed the genetic contribution to the pathogenesis of these conditions, with significant association of PR3-ANCA and human leukocyte antigen-DP and the genes encoding α1-antitrypsin and PR3. MPO-ANCA were significantly associated with human leukocyte antigen-DQ. Thus, recent results from epidemiological studies, genome-wide association study and therapeutic trials have suggested that these entities are, in fact, distinct. We have summarised these results and discuss the idea that these two entities should be studied separately as the nature of the two auto-antigens suggests at a molecular level despite shared ANCA involvement.