Background Multiple studies have observed unfavourable changes in lipid profile associated with tocilizumab (TCZ, anti-IL6 receptor antagonists) and some other rheumatoid arthritis (RA) therapies. The real-world cardiovascular disease (CVD) risk associated with the first anti IL-6R medication for RA, TCZ, remains uncertain.
Objectives The objective of this study was to assess the CVD risk associated with TCZ compared to individual tumour necrosis inhibitor (TNFi) therapies, as well as to other biologics used for RA (e.g. rituximab, abatacept).
Methods Using 2006–2015 Medicare and MarketScan claims data, we conducted a retrospective cohort study among RA patients who initiated biologic disease-modifying antirhematic drugs (bDMARDS) after January 1, 2010 and had at least 365 days medical and pharmacy coverage before initiation. The primary outcome was a composite of myocardial infarction (MI), stroke, and fatal CVD assessed using a validated method. Subgroups analyses were done for RA patients experienced to other bDMARDs before initiation and by stratifying patients with respect to key CVD risk factors to identify both higher and lower CVD risk patients. We provided descriptive statistical for each exposure group. Incidence rates and 95% confidence intervals were calculated using Poisson regression. COX regression was used to generate unadjusted and adjusted hazard ratio.
Results We identified 3 54 486 (Medicare 206,275+MarketScan 148,211) RA patients and 4 63 446 (Medicare 271,832+MarketScan 191,614) initiations of bDMARDS. After applying inclusion and exclusion criteria, the final cohort contained 88 463 (Medicare 46,648+MarketScan 41,815) RA patients and 1 17 493 (Medicare 61,715+MarketScan 55,778) episodes. The mean (SD) age was 64.7 (12.1) in Medicare and 52.2 (12.3) in MarketScan. The majority of patients were female (83.9% in Medicare and 80.5% in MarketScan), and 68.6% were non-Hispanic White in Medicare. TCZ users were similar to abatacept and rituximab users except that TCZ users were less likely to be naïve to bDMARDS. Compared to TNFi users, TCZ users were more likely to be white, with history of CVD (other than MI or stroke), heart failure, atrial fibrillation, hospitalisation and had more physician visits in baseline. TCZ users were less likely to be diabetic, use methotrexate in the baseline, and to be naïve to bDMARDS.
The crude incidence rate (IR) per 1000 patient-years for composite CVD among Medicare patients ranged from 13.3 (95% CI: 11.1 to 16.0) for etanercept to 19.4 (95% CI: 16.3 to 20.9) for rituximab users. The crude incidence rate for pooled TNFi users was 16.4 (15.2–17.7). Compared to TCZ, the adjusted hazard ratios were 1.03 (0. 82–1.29) for abatacept, 1.25 (0.96–1.61) for rituximab, 1.13 (0.84–1.52) for etanercept, 1.33 (0.99–1.80) for adalimumab, and 1.57 (1.21–2.05) for infliximab (figure 1). There were no significant differences in CVD risk between tocilizumab and any other biologic using MarketScan data. Results were robust in numerous subgroup analyses.
Conclusions Consistent with findings of a recently completed safety trial in RA, tocilizumab was associated with a comparable CVD risk compared to etanercept, as well as a number of other RA biologics, in two large data sources.
Disclosure of Interest F. Xie: None declared, H. Yun Grant/research support from: BMS, E. Levitan Grant/research support from: Amgen, Consultant for: Amgen, Novartis, P. Muntner: None declared, J. Curtis Grant/research support from: AbbVie, Amgen, BMS, Corrona, Janssen, Lilly, Myriad, Pfizer, Roche/Genentech, UCB, Consultant for: AbbVie, Amgen, BMS, Corrona, Janssen, Lilly, Myriad, Pfizer, Roche/Genentech, UCB
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.