POS0844 LONG-TERM AND SHORT-TERM RISK OF MALIGNANCY IN IDIOPATHIC INFLAMMATORY MYOPATHY

Abstract

BackgroundIdiopathic inflammatory myopathy (IIM) is associated with a high prevalence of malignancy, in which anti-transcriptional intermediary factor 1-γ(TIF1-γ) antibody is at a high risk of malignancy compared to the other myositis-specific antibodies, including anti-aminoacyl-tRNA synthetase (ARS), melanoma differentiation-associated gene 5 (MDA5), and Mi-2 antibody [1]. Although some subgroups of IIM have a high risk of malignancy up to 10 years after diagnosis [2], long-term relationships between malignancy and each antibody are not elucidated so far.ObjectivesTo clarify the risk of malignancy in IIM in each antibody type with standard incidence ratios (SIR) of malignancy and the long-term occurrence of malignancy.MethodsFrom our IIM cohort at the University of Tokyo hospital, we enrolled patients who fulfilled with Bohan and Peter criteria for dermatomyositis/polymyositis or 2017 European League Against Rheumatism/American College of Rheumatology classification criteria for IIMs and with positivity for anti-ARS, MDA5, Mi-2, or TIF1-γ antibody. We retrospectively collected the data and calculated SIRs by using the database of Cancer Statistics, Cancer Information Service, National Cancer Center, Japan. Fisher exact test and Wilcoxon test were used to compare the clinical features in each group. Cumulative incidence of cancer occurrence was estimated by Kaplan-Meier methods, and Log-rank test was used to compare the cancer-free survival.ResultsThe study included 111 patients, including 16 patients being diagnosed with a malignancy within a three-year before and after the occurrence of IIM. Patients with malignancy were significantly older age at the onset of IIM (median age (interquartile range) 67.5 (62.3-70.8) vs. 52.0 (41.5-61.5). p<0.001) and more frequently with anti-TIF1-γ antibody (25.0% vs. 4.2%, p=0.015). After adjustment for age, the proportion of anti-TIF1-γ antibody remains significantly higher in the malignancy group (p=0.01). SIRs (95% confidence interval) in each type of antibody were anti-ARS of 2.21 (1.03-4.55), anti-MDA5 of 1.89 (0.49-6.02), anti-Mi-2 of 2.75 (0.14-17.84), and anti-TIF1-γ of 14.73 (4.72-40.50) (Table 1). Types of malignancy included four gastric cancers, two bladder cancers, breast cancers, cervical cancers, ovarian cancers, and lung cancer, one pancreatic cancer, rectal cancer, and renal cancer. Kaplan-Meier estimated cancer-free survival was significantly lower in patients with anti-TIF1-γ antibody than patients with anti-ARS antibody (p<0.01) (Figure 1). The frequencies of malignancy of IIM patients with anti-ARS, MDA5, Mi-2, or TIF1-γ antibody through the follow-up period were 15.6%, 20.0%, 11.1%, 50.0%, respectively.Figure 1.Comparison of malignancy rate of idiopathic inflammatory myopathy with anti-ARS, and TIF1-γ antibodies in 10 years.Table 1.Standard incident ratio (SIR) of malignancy in idiopathic inflammatory myopathy according to myositis-specific antibodies.Total (Person-years)ObservedExpectedSIR (95% CI)Anti-ARS antibody51083.622.21 (1.03-4.55)Anti-MDA5 antibody21031.591.89 (0.49-6.02)Anti-Mi-2 antibody5610.362.75 (0.14-17.84)Anti-TIF1-γ antibody3940.2714.73 (4.72-40.50)Total815165.842.74 (1.62-4.56)ConclusionIIM patients with anti-TIF1-γ antibody and with anti-ARS antibody had a higher prevalence of malignancy than the general population. In addition, anti-TIF1-γ antibody are more associated with malignancy rather than anti-ARS antibody in 10 years as well. Malignancy associated with anti-TIF1-γ positive IIM occurred exclusively within the one year before and after the IIM diagnosis, and cancer should be particularly investigated at the IIM diagnosis.