LAUSR

We read with interest the article by Dahlqvist et al. about individuals with anti-mitochondria antibodies (AMAs) and nonestablished primary biliary cholangitis (PBC). The fortuitous finding of a positive AMA test in patients with no clinical and biochemical evidence of PBC has become more frequent due to the widespread testing for antinuclear antibodies using the HEp-2 indirect immunofluorescence (ANA-HEp-2) test. The suspicion of AMAs is elicited by the observation of the ANAHEp-2 cytoplasmic reticular pattern, which is highly suggestive of AMAs. These cases should be then confirmed in specific AMA assays (e.g., indirect immunofluorescence on rodent tissue and solid phase immunoassay). Autoantibodies precede clinical/laboratory evidence of numerous autoimmune diseases, and AMAs may precede the clinical onset of PBC by decades (reviewed by Dellavance and Coelho Andrade). The fact that some AMA-positive individuals eventually develop PBC, as demonstrated by Dahlqvist et al., indicates that the preclinical occurrence of AMA may represent part of an evolving and dynamic autoimmune disorder. If this is the case, it might be that the humoral autoimmune response would become stronger with the transition to the clinical stage of the disease. The data by Dahlqvist et al (1) lend credit to this hypothesis by showing lower AMA titer and lower frequency of PBC-specific antibodies (gp210 and sp100) in individuals with nonestablished PBC compared with patients with established PBC. These findings are in agreement with previous observations from our group in which we demonstrated a more robust autoantibody profile in established PBC than in nonestablished PBC. In our study, AMA-positive individuals with nonestablished PBC (n 5 151) presented lower indirect immunofluorescence AMA titer, lower levels of anti-pyruvate dehydrogenase complex E2 subunit (PDC-E2) antibody on enzyme-linked immunosorbent assay, lower avidity of anti–PDC-E2 antibodies, lower frequency of anti-sp100, anti-gp210, and anti–CENP-B antibodies, and a more restricted antibody isotype use for AMA than patients with established PBC (n 5 61). In a multivariate analysis, three variables were independently and strongly associated with the possibility of any given AMA-positive sample belonging to a patient with established PBC: high titer AMA, high avidity anti–PDC-E2 antibodies, and the occurrence of three or more PBC-specific antibodies. These parameters may be helpful in spotting apparently healthy AMA-positive individuals who are at risk of developing overt PBC over time.