Polymyositis/dermatomyositis and malignancy risk: a metaanalysis study.

Abstract

To investigate the association between polymyositis (PM)/dermatomyositis (DM) and risks of malignancy. We searched Pubmed for articles dated before August 16, 2013. Studies were included if they met the following criteria: (1) a cohort or observational study; (2) PM or DM as one of the exposures of interest; (3) cancer as an outcome of interest; and (4) the rate ratio (RR) or standardized incidence ratio (SIR) were available with their 95% CI. We used random-effects or fixed-effects models to calculate the pooled RR according to the heterogeneity test. Twenty publications were included. Compared with the general population, the pooled RR for patients with PM, DM, and PM/DM were 1.62 (95% CI 1.19-2.04), 5.50 (4.31-6.70), and 4.07 (3.02-5.12), respectively. The increased risks were more significant in patients within the first year of myositis diagnosis, male patients, and population-based studies (for DM). A significant association was also found between PM or DM and most site-specific malignancies. However, both PM and DM were not associated with stomach and prostate cancers. Significant heterogeneity was found between studies on association between PM/DM and overall malignancy, but not between PM/DM and the majority of site-specific malignancies, suggesting that that inherent malignancy difference may be a major source of heterogeneity. The present metaanalysis indicates that PM and DM are significantly associated with increased risks of overall malignancy and most site-specific malignancies. The number of studies on association between PM or DM and some malignancies is too small to draw a firm conclusion. Accordingly, more research is needed for these malignancies.