FRI0055 Safety of etanercept in elderly patients with rheumatoid arthritis: a pooled analysis

Abstract

Background: Both ageing and rheumatoid arthritis (RA) are associated with a higher risk of comorbidities, some of which may be increased by current therapies. However, failure to control RA can lead to a particularly rapid loss of function and independence in the elderly population. Objectives: To assess the safety of etanercept (ETN) in older (aged >65 years) vs younger (aged ≤65 years) patients with RA. Methods: Patient‐level data were pooled from the double‐blind, placebo (PBO)‐ controlled phases of all completed, randomised, Pfizer‐ or Amgen‐sponsored ETN studies in patients with RA. The occurrence of adverse events (AEs,%) in PBO‐ and ETN‐treated patients was analysed by age (≤65 years vs >65 years), using the exact Poisson method to calculate 95% confidence intervals (CI). Fisher's Exact tests and Breslow Day tests for interaction were used to assess statistical differences. Results: Data on 6418 patients were collected from 19 studies (≤65 years: PBO, n=1910; ETN, n=3497;>65 years: PBO, n=364; ETN, n=647). In PBO‐treated patients, the occurrence of congestive heart failure (CHF), interstitial lung disease (ILD), and angioedema was significantly higher in older than younger patients (CHF and ILD: figure 1; angioedema: 0.82%, 95% CI 0.17‐2.41% and 0%, 95% CI 0‐0.19, respectively, p<0.0001). In patients treated with ETN, the occurrence of CHF, serious infections (SI), and non‐melanoma skin cancers (NMSC) was significantly higher in older than younger patients (all p<0.0001, figure 1), as was the occurrence of anaphylactic reactions (p=0.02), central nervous system and peripheral demyelinating events (both p=0.001), and seizures (p=0.02). In contrast, the occurrence of non‐serious infections and injection site reactions (ISRs) was significantly lower in older than younger ETN‐treated patients (infections: 43.12%, 95% CI 38.21‐48.49% and 48.27%, 95% CI 45.99‐50.63, respectively, p=0.016; ISRs: 17.31%, 95% CI 14.25‐20.83% and 22.36%, 95% CI 20.82‐23.99, respectively, p=0.004). The difference between PBO‐ and ETN‐treated patients was statistically significant in the 2 age groups combined (Breslow Day interaction) for angioedema (p=0.003), ISRs (p=0.042), and NMSC (p=0.042). Older, but not younger, patients had greater (but not statistically significant) occurrence of CHF, SI, and NMSC when taking ETN vs PBO. There was an interesting, non‐significant, reduction of ILD occurrence in older, but not younger ETN‐treated vs PBO‐treated patients (figure 1). There were no significant agerelated increases in the occurrence of other AEs of interest such as uveitis, opportunistic infections (invasive fungal, protozoal, bacterial, and atypical mycobacterial infections and Legionella), tuberculosis, lymphoma, melanoma, or leukaemia. The smaller number of patients in the older cohort was a study limitation, and further stratification of the data in the higher age brackets may provide key information on the age at which the risk of particular AEs increases. SI included pneumonia, cellulitis, septic arthritis, sepsis, and parasitic infection. Conclusions: The occurrence of CHF, SI, and NMSC was higher in older vs younger patients taking ETN. While the analysis suggests an overall acceptable safety profile of ETN, the relative risks and benefits need to be considered when treating older patients.