ALPHA(1)-ANTITRYPSIN DEFICIENCY AND ANTIPROTEINASE 3 ANTIBODIES IN ANTINEUTROPHIL CYTOPLASMIC ANTIBODY (ANCA)-ASSOCIATED SYSTEMIC VASCULITIS

alpha(1)-antitrypsin (alpha(1)-AT) is a naturally occurring inhibitor of proteinase 3 (PR3) and elastase, two of the target antigens of anti -neutrophil cytoplasmic antibodies (ANCA). An increased incidence of a lpha(1)-AT phenotypes associated with dysfunctional alpha(1)-AT or low serum levels has been reported in patients with anti-PR3 antibodies. We have studied the relationship between ANCA, and phenotypes and seru m levels of alpha(1)-AT. Phenotypes usually associated with a moderate or severe reduction in alpha(1)-AT serum levels or in dysfunctional a ctivity were found more often in individuals with anti-PR3 antibodies than in the general population: four of the 31 patients (13%) with ant i-PR3 antibodies had phenotypes MZ (n = 2), S (n = 1) or Z (n = 1) (P < 0.05). However, the corresponding alpha(1)-AT serum levels were norm al (n = 3) or elevated (n = 1). None of the 31 sera with anti-PR3 anti bodies had low levels of alpha(1)-AT. No abnormal alpha(1)-AT phenotyp e was demonstrated in seven patients with anti-elastase antibodies, de spite a low level of alpha(1)-AT in one serum. Anti-myeloperoxidase an tibodies are common in patients with ANCA, but no abnormal phenotype o r low serum alpha(1)-AT level was demonstrated in any of 29 sera conta ining these antibodies. Finally anti-glomerular basement membrane (GEM ) antibodies occur occasionally in patients with ANCA-associated disea ses, but again none of 10 sera had an abnormal alpha(1)-AT phenotype o r low serum level. ANCA were not demonstrated by indirect immunofluore scence in any serum from 73 patients with abnormal alpha(1)-AT phenoty pes. These results confirm that patients with anti-PR3 antibodies ofte n have alpha(1)-AT phenotypes that are usually associated with low ser um levels of alpha(1)-AT or with dysfunctional protein. Nevertheless, the incidence of anti-PR3 antibodies in patients with abnormal alpha(1 )-AT phenotypes is very low. This probably reflects the rarity of Wege ner's granulomatosis, the major disease associated with anti-PR3 antib odies, and the relative frequency of abnormal alpha(1)-AT phenotypes. The mechanism for the development of anti-PR3 antibodies in patients w ith abnormal alpha(1)-AT phenotypes is not clear, but may relate to th e increased propensity of unbound and uninhibited PR3 to stimulate aut oantibody production.