RESUMEN
OBJECTIVE: We recently recorded a high prevalence of inclusion body myositis (IBM) in patients with Sjögren's syndrome (SS). Whether myositis patients with SS differ from myositis patients without SS in terms of the characteristics of the myositis is currently unknown. Anti-cytosolic 5'-nucleotidase 1 A (cN1A) has recently been proposed as a biomarker for IBM but is also frequent in SS. Whether anti-cN1A is independently associated with IBM is still an open question. We aimed to assess the significance of SS and anti-cN1A in myositis patients. METHODS: Cumulative data on all myositis patients (EULAR/ACR 2017 criteria) screened for SS (ACR/EULAR 2016 criteria) in a single centre were analysed. Ninety-nine patients were included, covering the whole spectrum of EULAR/ACR 2017 myositis subgroups and with a median follow-up of 6 years (range 1.0-37.5). The 34 myositis patients with SS (myositis/SS+) were compared with the 65 myositis patients without SS (myositis/SS-). RESULTS: . IBM was present in 24% of the myositis/SS+ patients vs 6% of the myositis/SS- group (P = 0.020). None of the IBM patients responded to treatment, whether they had SS or not. Anti-cN1A was more frequent in myositis/SS+ patients (38% vs 6%, P = 0.0005), independently of the higher prevalence of IBM in this group (multivariate P value: 0.02). Anti-cN1A antibody specificity for IBM was 0.96 (95% CI: 0.87, 0.99) in the myositis/SS- group but dropped to 0.70 (95% CI: 0.48, 0.85) in the myositis/SS+ group. INTERPRETATION: In myositis patients, SS is associated with IBM and with anti-cN1A antibodies, independently of the IBM diagnosis. As a consequence, anti-cN1A has limited specificity for IBM in myositis patients with SS.