Background/Purpose: Mitochondria are key regulators of metabolism, inflammation and cell death, with mitochondrial dysfunction being associated with several diseases. Though mainly found intracellularly, we recently made the observation that mitochondria (Mt) were extruded upon formation of neutrophil extracellular traps (NETs), a neutrophil cell death process commonly observed in autoimmune rheumatic diseases, including rheumatoid arthritis (RA). Other than the inflammatory potential of mitochondria, they are highly immunogenic, evoking an immune response with development of anti-mitochondrial antibodies (AMA). Anti-mitochondrial antibodies are also found in RA patients, with reactivity towards the mitochondrial-specific phospholipid cardiolipin having major clinical significance given the capacity to predispose to thrombosis and miscarriage. However, the potential pathogenic role of other AMAs in RA pathogenesis has not been carefully addressed. The aim of the current study was to investigate prevalence and clinical utility of AMAs in RA patients.

Methods: Mitochondria (Mt) were isolated from the human hepatic cell line HepG2 and analyzed for binding of AMAs using a state-of-the art flow cytometry-based method developed in-house. Two cohorts of healthy controls (HC) and RA patients were used: one cross-sectional (HC, n=30; RA, n=101) and one longitudinal inception cohort (HC, n=100; RA n=247). Purified whole Mt lysate was used to detect proteins targeted by RA AMAs via Western Blot.

Results: Using Western blot analysis, several mitochondrial proteins were identified to which RA patients, but not healthy individuals had reactivity. Of note, most RA patients (67%) recognized a 70 kDa mitochondrial protein (p< 0.05). Also by flow cytometry, assessing binding of RA IgG to highly purified mitochondria, AMA IgG levels were markedly elevated in the two RA cohorts as compared to HCs (p< 0.0001 and p=0.02). In the cross-sectional RA cohort, levels of AMAs were associated with a severe erosive disease (p=0.01). We next asked whether AMA positivity could predict development of erosive disease using the longitudinal inception cohort with a median follow-up of 8 years. Consistent with prior studies, ACPA positivity predicted future erosive disease (OR=6.7, p< 0.0001). Of note, AMA positivity could also predict erosive disease independent of ACPA positivity (OR=4.6, p=0.006). Being positive for either of the autoantibodies further increased the odds ratio (10.4, p=0.002), suggesting additive value of the two markers. Further, in contrast to ACPA (OR=1.5, p=0.43), AMA could also predict joint space narrowing (OR=3.3, p=0.02).

Conclusion: Our results demonstrate the presence of an important, yet to be identified, novel mitochondrial autoantibody prevalent in RA patients. Further, these AMAs, by themselves, as well as in combination with established biomarkers, e.g. ACPA, provide significant clinical value in identifying patients at risk of developing severe erosive disease, allowing for preventative therapeutic strategies.

« Back to 2019 ACR/ARP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/anti-mitochondrial-antibodies-predict-severe-erosive-disease-in-rheumatoid-arthritis/