Background: Several large population-based epidemiologic studies have shown that lymphoma risk is two to three times greater in patients with rheumatoid arthritis (RA) than in the general population. However, these studies may have underestimated the background risk for patients who will receive therapy with newer biologic agents. Compared to the general RA population, such patients are likely to have higher disease activity and to have had prior immunosuppressant treatment—both of which may further increase lymphoma risk.
Objectives: To assess the risk of Non-Hodgkin’s lymphoma (NHL) and Hodgkin’s disease (HD) in a sicker cohort of RA patients relative to the general population. Methods: We conducted a retrospective cohort study in the United Kingdom (U.K.) using General Practice Research Database (GPRD) data from 1988 to 2002. The GPRD contains anonymized electronic medical records from about 4% of the U.K. population. We identified a ‘sicker’ RA cohort defined as RA patients who received prescriptions for two or more Disease Modifying Anti-Rheumatic Drugs (DMARDs), excluding biologic agents. Follow-up began when the second DMARD was first prescribed and ended at the earliest of the following: lymphoma diagnosis, death, transfer out of practice, end of data, or prescription for a biologic. We estimated the relative risk of HD and NHL using standardized incidence ratios (SIRs), adjusting for age and gender, based on incidence rates derived from the general GPRD population of 5.5 million patients. We also evaluated lymphoma risk in the general RA population for whom follow- up began at the first diagnosis of RA.
Results: The ‘sicker’ RA cohort consisted of 8976 patients who were followed for an average of 3.4 years after cohort entry. Among the ‘sicker’ RA patients, the SIR compared to the general GPRD population was 7.2 (95%CI: 2.9, 14.9) for HD and 3.0 (95%CI: 2.0, 4.2) for NHL. The general RA cohort consisted of 40,547 patients who were followed for an average of 3.8 years after cohort entry. Among the general RA population, the SIR compared to the general GPRD population was 3.6 (95%CI: 2.1, 5.6) for HD and 2.2 (95%CI: 1.8, 2.6) for NHL, both of which are similar to estimates from population-based studies of RA.
Conclusions: In this large study of RA patients who received at least two DMARDs, the relative risks of NHL and especially HD were higher than those reported for the overall population. These results should be considered in evaluating lymphoma incidence rates from long-term studies of biologic therapy in RA.
