Background Anti-centromere pattern (ACA) is infrequently seen among antinuclear antibodies (ANA) detected by indirect immunofluorescence (IFI). ACA is associated with systemic autoimmune diseases (SAD), especially systemic sclerosis (SSc). Some studies had recently related ACA with cancer, mostly with breast and lung cancer [1]. However, most published series of patients with ACA lack information about cancer occurrence. A prevalence of cancer of 11.1% was the only reported data, on a series of 45 unselected patients with ACA [2]

Objectives Our aim was to study the prevalence of cancer in the largest series of patients with ACA, with a long follow-up. Our second objective was to make a cohort to calculate the incidence of cancer and to try to identify risk markers of cancer in these patients

Methods We included consecutive patients with at least 2 positive determinations of ANA with ACA by IFI on Hep2 cells between January 1st of 2011 and June 30th of 2015 in 6 Galician hospitals. The authors reviewed each patient's chart to determine the presence of cancer and its type among numerous variables. Then, patients with cancer at the moment of the first positive determination of ANA with ACA were excluded. We checked the presence of any tumour in patients of this cohort at the end of follow-up to calculate the incidence of cancer. Finally, we compared patients with and without cancer by multivariate analysis, with the SPSS 20.0. The ethics committees of each hospital approved the study and patients gave their consent

Results 369 patients with ACA were studied, of which 333 were women (90.2%), with a mean age of 64.7 years (range: 22–92). The mean follow-up from the first positive ACA determination was 67.6 months. 283 patients (76.7%) had a SAD: 46.3% SSc (79% of whom were lc-SSc), 8.13% primary biliary cholangitis, 7.05% Sjögren's syndrome, 4.6% autoimmune hepatitis and 11 other SAD (polyarthritis, SLE, Raynaud phenomenon, sarcoidosis, mixed connective tissue disease, etc). 45 of these patients (15.9%) had any overlap syndrome. 39 patients had cancer at some time (10.6%), 3 of them with 2 types of cancer. The most frequent were: breast in 9 (23.1%), lung in 5 (11%), NHL in 5 (11%) and colorectal in 4 (10.3%). Cancer preceded the diagnosis of ACA in 19 patients (48.7%), with a mean time to diagnosis of 9.1 years (range 1–18). In the cohort of 350 patients with ACA the incidence was 1.14 per 100 patients per year (n 20). The mean time to diagnosis of cancer was 7.3years (range 1–27). The oldest age was the only risk marker of cancer identified (70.8±13.29 years vs. 63.9±14.32, p 0.005). There were no differences in the other variables analysed including sex, tobacco, diagnosis of SAD, capillaroscopy pattern, ANA titrations and mortality

Conclusions Cancer was frequent in patients with ACA, with a prevalence of 10.6% and incidence of 1.14 per 100 patients per year. The only risk marker of cancer identified in this population was the oldest age

References

1. Briasoulis E, Kamposioras K, Tzovaras V et al. CENP-B specific anti-centromere autoantibodies heralding small-cell lung cancer. A case study and review of the literature. Lung cancer 2008;60:302–306.

2. Zuber M, Gotzen R, Filler I. Clinical correlation of anticentromere antibodies. Clinical rheumatology 1994;13(3):427–432.

References

Disclosure of Interest None declared