Background/Purpose: Swedish and Danish national biologics registers (*) have reported a possible increase in melanoma risk with TNF inhibitors.  Since melanomas are uncommon, the association is difficult to evaluate in other individual registers. We therefore planned a EULAR collaborative project.

Methods: Patients with RA from 11 European biologics registers in 9 countries were included. Patients were followed prospectively from start of a new biologic treatment until the occurrence of first invasive histology-confirmed cutaneous melanoma, using an ever-exposed approach. For the TNFi cohort, prior exposure to biologic drugs was not permitted. Prior exposure to TNFi was allowed for other biologic drugs. For each register, incidence rates and standardized incidence ratios (SIR) of melanoma were calculated by using age- sex- and calendar year-specific rates from the general population of the corresponding country as reference. Poisson regression models were used to summarize the register-specific SIRs to overall SIR estimates. Rates of melanoma in biologic exposed patients were compared to those in biologic-naïve patients enrolled in participating registers by calculating incidence rate ratios (IRR). Overall SIRs and IRRs were calculated, taking the size of the registers into account.

Results: Overall, 114,291 patients were available for analysis: mean age 58 years; 74% female. 287 developed a first invasive melanoma. Background population rates varied due to differences in the incidence by country, calendar years and differences in the age and sex distribution of the corresponding RA cohorts (Table). The SIRs for biologic naϊve patients were similar across the registers whereas there was variation in SIRs between TNFi cohorts (table). The overall IRRs did not show a significantly increased melanoma risk for any of the biologic therapies compared to biologic-naïve patients. Similar results were found with other drug exposure models (data not shown).

Conclusion: This large European collaborative project of 11 registers from 9 countries did not confirm an increased risk of melanoma following exposure to TNFi, although an association cannot be completely ruled out with these data.

Table. Number of melanomas, standardized incidence ratios and incidence rate ratios in different treatment groups of RA

Table: Person-years, incidence rate in the general population(s) (Pop. rate), observed (obs.) and expected (exp.) melanoma cases, SIRs, and IRRs for biologic-naive, anti-TNF (TNF), rituximab (RTX), tocilizumab (TOC), and abatacept (ABA) exposed RA patients; n.a. not available (no biologic naïve cohort)         

(*)Raaschou P et al., BMJ 2013;346:f1939; Dreyer L et al. Ann Rheum Dis 2013;72:79-82