Key ACTEMRA/RoACTEMRA publication summaries
Indirect comparison of tocilizumab and other biologic agents in patients with rheumatoid arthritis and inadequate response to disease-modifying antirheumatic drugs
Bergman GJ, Hochberg MC, Boers M, Wintfeld N, Kielhorn A, Jansen JP.
Semin Arthritis Rheum 2010; 39:425–441.
In the absence of any published randomised controlled trials (RCTs) providing a direct head-to-head comparison of the efficacy of biologics used to treat RA, a mixed-treatment comparison (MTC) – an extension of meta-analysis that enables pair-wise comparisons across a range of interventions – is a powerful statistical alternative.
This first reported MTC in RA compared the patterns of ACR responses to tocilizumab (TCZ) with those of other biologics in patients who had had an inadequate response to conventional DMARDs.
Key study results are summarised below.
- Search criteria identified 18 published RCTs involving a total of 10,419 patients with RA.
- Three TCZ studies (LITHE, OPTION and TOWARD), two abatacept studies, two rituximab studies and 11 studies of anti-TNFs (adalimumab, etanercept, infliximab; grouped together as a single category as previous studies indicate similar efficacy) were included.
- Endpoints were ACR20/50/70 response rates at 22–30 weeks.
- A comparison of estimated responses for the non-overlapping ACR response rates showed TCZ to have a distinctive response pattern compared with those of other biologics in RA.
- TCZ was associated with a specifically high proportion of high-grade responders (>ACR70) compared with the other biologics.
- Pairwise comparisons indicated a higher estimated probability of response with TCZ vs. other biologics.
- The relative risk of response for TCZ vs. other biologics was >1 at all ACR response levels regardless of modelling assumptions.
- The ACR70 95% credibility intervals for TCZ indicated superiority vs. anti-TNFs and abatacept.
- Translated into absolute responses (using the average absolute placebo response as a baseline), TCZ was estimated to have 65% and 44% responses at ACR20 and ACR50, respectively, and these were comparable for anti-TNFs. However, the estimated ACR70 response rate was higher for TCZ (29%) than for the anti-TNFs (16%).
In conclusion, this MTC identified that in patients with an inadequate response to DMARD therapy, TCZ has a unique pattern of response compared with other biologics. The analyses suggest that the difference lies in a higher likelihood of ACR70 response with TCZ.