Risk of malignancy including non-melanoma skin cancers with anti-tumor necrosis factor therapy in patients with rheumatoid arthritis: meta-analysis of registries and systematic review of long-term extension studies.

Objectives: To assess the risk of malignancy in patients with rheumatoid arthritis (RA) receiving tumour necrosis factor (TNF) antagonists through a meta-analysis of data from registry studies and systematic review of long-term extension (LTE) studies.

Methods: We systematically reviewed the literature up to January 2010 in the Embase and Medline databases, as well as abstracts from the 2008 and 2009 annual meetings of the EULAR and the ACR. The Mantel-Haenszel method was used to provide a common odds ratio (OR). Statistical heterogeneity was assessed by the chi-square Q test (χ²). Standardised incidence ratio (SIR) was extracted for post-marketing studies and registries.

Results: The literature search identified 634 articles and 110 abstracts, of which 12 and 5, respectively, were selected for analysis. We could perform a meta-analysis of data from 4 and 3 registries for risk of total malignancy and non-melanoma skin cancers (NMSC), respectively. The pooled OR for total malignancy and for NMSC was 0.81 [95% confidence interval (CI) 0.71-0.94] and 0.79 [0.62-1.02] in TNF antagonist group versus DMARD group, respectively. There was no significant heterogeneity. Among 4 LTE studies and 4 registries, no significant increase in the incidence of total malignancy was noted versus the general population. The only signal may be an increased risk of non-melanoma skin cancers.

Conclusions: Our meta-analysis of data from registries and systematic review of LTE studies did not reveal an increased risk of total malignancy in RA patients receiving anti-TNF therapy.