Incidence of malignant disease in biopsy-proven inflammatory myopathy. A population-based cohort study.

BACKGROUND: The validity and magnitude of an association between myositis and malignant disease continue to be debated. Such issues as the legitimacy of a myositis diagnosis and distinction among myositis subgroups in previous population-based studies remain unresolved. OBJECTIVE: To determine the risk for malignant disease in patients with biopsy-proven inflammatory myopathies. DESIGN: Population-based, retrospective cohort study. SETTING: Victoria, Australia. PATIENTS: 537 patients in whom a biopsy-positive idiopathic inflammatory myopathy was first diagnosed from 1981 through 1995. MEASUREMENTS: Standardized incidence ratios were calculated to compare the incidence of malignant disease in patients with inflammatory myopathy and the general population. RESULTS: A total of 116 cases of malignant disease were found in 104 patients. Seventy-four cases were identified concurrently with (within 7 days) or after diagnosis of myositis. The highest risk for malignant disease was associated with dermatomyositis (standardized incidence ratio, 6.2 [95% CI, 3.9 to 10.0]). The risk was also increased in polymyositis (standardized incidence ratio, 2.0 [CI, 1.4 to 2.7]), although the relative risk for malignant disease in dermatomyositis compared with polymyositis was 2.4 (CI, 1.3 to 4.2). An increased risk for malignant disease was also found in inclusion-body myositis (standardized incidence ratio, 2.4 [CI, 1.2 to 4.9]). The excess risk for malignant disease diminished with time (standardized incidence ratio, 4.4 [CI, 2.7 to 7.1] in the first year; 3.4 [CI, 2.3 to 5.1] between 1 and 3 years; 2.2 [CI, 1.3 to 3.9] between 3 and 5 years; and 1.6 [CI, 1.0 to 2.6] beyond 5 years [ P for trend, 0.002]). CONCLUSION: The risk for malignant disease is increased in biopsy-proven dermatomyositis and polymyositis and also appears to be increased in inclusion-body myositis.