Malignancy incidence in 5294 patients with juvenile arthritis

To determine cancer incidence in a large clinical juvenile-onset arthritis population.

We combined data from 6 existing North American juvenile-onset arthritis cohorts. Patients with juvenile-onset arthritis were linked to regional cancer registries to detect incident cancers after cohort entry, defined as first date seen in the paediatric rheumatology clinic. The expected number of malignancies was obtained by multiplying the person-years observed (defined from cohort entry to end of follow-up) by the geographically matched age, sex and calendar year-specific cancer rates. The standardised incidence ratios (SIR; ratio of cancers observed to expected) were generated, with 95% CIs.

The 6 juvenile arthritis registries provided a total of 5294 patients. The mean age at cohort entry was 8.9 (SD 5.0) years and 68% of participants were female. The mean duration of follow-up was 6.8 years with a total of 36 063 person-years spanning 1978-2012. During follow-up, 9 invasive cancers occurred, compared with 10.9 expected (SIR 0.82, 95% CI 0.38 to 1.5). 3 of these were haematological (Hodgkin's, non-Hodgkin's lymphoma and leukaemia). 6 of the patients with cancer were exposed to disease-modifying drugs; 5 of these had also been exposed to biological agents.

We did not clearly demonstrate an increase in overall malignancy risk in patients with juvenile-onset arthritis followed for an average of almost 7 years. 3 of the 9 observed cancers were haematological. 5 of the cancers arose in children exposed to biological agents. Longer follow-up of this population is warranted, with further study of drug effects.