Where do we go after GiACTA?

The GCA Actemra (GiACTA) trial fundamentally changed how many clinicians manage GCA. This phase 3 randomized controlled trial compared a 26-week prednisone taper and s.c. tocilizumab at weekly or every other week intervals to either a 26-week or a 52-week prednisone taper, both paired with placebo injections [1]. Patients randomized to either regimen of tocilizumab were significantly more likely to be in sustained remission at week 52 as compared with either placebo arm, received significantly less prednisone and reported meaningful improvements in health-related quality of life [1, 2]. Clinical practice guidelines from the European Alliance of Associations for Rheumatology (EULAR) and ACR have both recommended its use, though with some differences. While ACR recommends tocilizumab upfront for all eligible patients, EULAR recommends it as initial therapy for patients at a high risk of glucocorticoid-related adverse events and for patients with relapsing disease [3, 4]. The degree to which upfront treatment with tocilizumab improves remission beyond the first year of therapy with respect to disease status (new onset or relapsing) or treatment strategy (weekly or every other week injections) is the subject of the open-label extension study from GiACTA that is reported in a recent edition of Rheumatology [5].

Open-label extension studies can be informative, but they are often underpowered to identify novel safety signals and suffer from the very biases their parent studies were designed to avoid [6, 7]. These issues can be seen in the design of part 2 of the GiACTA study, which was a 104-week open-label extension study of patients who completed part 1 and elected to continue with part 2. Patients could receive weekly tocilizumab, prednisone, MTX, no further treatment, or any combination of these therapies at the investigator’s discretion. As is typical for open-label extension studies, no blinding or randomization occurred, but patients were followed at regular intervals. Relapses were defined as recurrent GCA symptoms or signs or an elevated ESR that was attributable to GCA, as determined by the investigator. The analysis of part 2 outcomes were exploratory and blinded adjudicators were not used, but investigators remained blinded to the original treatment assignment.

The overall results of part 2 were published earlier this year in Lancet Rheumatology [8]. From the 250 participants in part 1, 215 (86%) patients entered part 2 and were eligible for analysis, which was conducted based upon the initial randomization from part 1, regardless of treatments received in part 2. At the end of 3 years, patients randomized to either tocilizumab regimen had a lower median cumulative glucocorticoid dose as compared with those who had been randomized to placebo and prednisone tapers. There were no new safety signals and no patients experienced vision loss. The absence of irreversible sight loss is encouraging at face value, but neither the treatment groups nor the placebo groups from GiACTA experienced vision loss at rates approximating clinical practice, which are as high as 10%. Selection bias related to patient eligibility likely brought about this result [9].

The linked report from part 2 of GiACTA in this edition of Rheumatology extends these analyses, evaluating whether disease status (new onset or relapsing) or treatment strategy (weekly or every other week injections) affect rates of remission over 3 years. Subgroup analyses from part 1 of GiACTA suggested that newly diagnosed patients did not benefit from weekly tocilizumab dosing as opposed to every other week dosing. In part 2 of GiACTA, patients who received weekly dosing had a lower rate of flare over the 3 year period, regardless of disease presentation. Those assigned to tocilizumab bi-weekly (27% new onset GCA, 35% relapsing GCA) or placebo at baseline (28% new onset, 31% relapsing) had lower rates of remission as compared with those initially assigned to weekly tocilizumab (49% new onset, 47% relapsing). Similarly, the longest times to first flare were observed in the tocilizumab weekly group for both new-onset disease (577 days vs 479 days for bi-weekly tocilizumab vs 179 days for placebo) and for relapsing disease (575 days vs 428 days for bi-weekly tocilizumab vs 224 days for placebo). These results support weekly prescribing of tocilizumab as initial therapy for both new onset and relapsing disease, a finding that has important implications for both practicing clinicians and future clinical practice guidelines.

Ultimately, however, part 2 of GiACTA has not answered critical questions about tocilizumab prescribing. First, the general lack of vision loss in both treatment and placebo arms persisted, preventing any conclusions from being drawn with regard to arguably the most important consequence of GCA. Second, surprisingly little has been learned about the role of tocilizumab in managing large vessel GCA, which is associated with more flares and higher glucocorticoid requirements [10]. Finally, no conclusions about the duration of therapy can be made. Patients in part 2 were analysed according to their initial treatment assignment from part 1, but neither the present report nor the previously published report in Lancet Rheumatology provides data about outcomes from subsequent therapies. What were relapse rates among patients who received tocilizumab for all three years, and how do these compare with those who received it for only 1 year or for rescue therapy? The lack of blinding and randomization in part 2 place inherent limitations upon such data, but answering these questions will be important for patients, physicians and public policy makers. In one current example, the National Health Service in the UK intends to discontinue a COVID-19-related extension of tocilizumab prescribing beyond the first year of therapy. In the absence of clarifying long-term data from GiACTA, it is difficult to determine the appropriateness of this policy.

A number of additional unmet needs in GCA should also be considered. First, it is notable that over half of patients treated in GiACTA with tocilizumab still flared over 3 years. Prior efforts to identify these patients at diagnosis have been largely unsuccessful, but novel techniques in cellular and molecular phenotyping may be promising. Second, pragmatic head to head trials should be conducted to evaluate lower-still doses of prednisone, and comparative effectiveness trials against tocilizumab should be conducted for existing agents [11] as well as novel therapeutics. Finally, predictors of response to specific immunosuppressive agents should be identified to ensure that patients receive a treatment strategy that is tailored to their particular GCA phenotype. These may be ambitious goals, but the GiACTA trial demonstrated that practice-changing large-scale randomized controlled trials are feasible in GCA. In addition to the tangible benefits for patients, this may be the enduring legacy of GiACTA.

Acknowledgements

M.P. is supported by a Rheumatology Research Foundation Scientist Development Grant.

Funding: No specific funding was received from any bodies in the public, commercial or not-for-profit sectors to carry out the work described in this article.

Disclosure statement: M.P. receives research funding related to clinical trials by Abbvie (SELECT-GCA) and AstraZeneca (MANDARA). S.S. receives research funding related to clinical trials by AstraZeneca (MANDARA). R.C. receives research funding related to clinical trials by Abbvie (UPHOLD and CONTExT-RA), and speaking fees from Abbvie, Sanofi, Roche and Janssen.

Data availability statement

Data will be made available upon reasonable request.

Patient and public involvement

There was no patient or public involvement with this manuscript.

References