GCA, TAK likely react differently to tocilizumab, despite disease similarities
Click Here to Manage Email Alerts
Although many rheumatologists view Takayasu’s arteritis as “basically the same disease” as giant cell arteritis, tocilizumab is more effective in the latter than the former, according to a presenter at the 2020 Congress of Clinical Rheumatology-East.
“I feel obligated to talk about this, because I think a lot of you are applying the transitive property of mathematics to clinical trials in vasculitis,” Philip Seo, MD, MHS, director of the Johns Hopkins Vasculitis Center, told attendees of the virtual conference. “What I mean by this, is if you have giant cell arteritis, tocilizumab seems to be a good thing for you. Now, I know many of you in your heads are probably thinking that giant cell arteritis and Takayasu’s arteritis are basically the same disease. They both have large vessel manifestations and they are both the major forms of large-vessel vasculitis; the only difference is the age of the patients.
“Because of that, and because you know that tocilizumab works for giant cell arteritis, most of you would be happy to assume that tocilizumab also works for Takayasu’s,” he added. “The problem is, I’m not exactly sure that is the case.”
Not Like the Other
To illustrate his point, Seo discussed the recent TAKT study, a phase 3, double-blind, placebo-controlled trial of tocilizumab (Actemra, Genentech) for TAK, which included patients with the disease who had relapsed in the prior 12 weeks and went into remission with oral glucocorticoids, with a primary endpoint of time to relapse. Published in 2018 in the Annals of the Rheumatic Diseases, the study failed to meet its primary endpoint, although the researchers wrote that the results suggested tocilizumab demonstrated favorable results over placebo in terms of time to relapse, with no new safety concerns.
However, according to Seo, this study had several issues, such as its small size — just 36 patients — and the fact that it included patients with TAK who had achieved remission with prednisone only, which, in terms of scope, he called “low-hanging fruit.”
“This was not a trial of refractory patients,” Seo said. “These were not patients who failed methotrexate and TNF inhibitors, or what have you. These were patients who you could treat just with prednisone. These were relatively mild patients, maybe the easiest patients you might see.”
Seo also pointed out that one of the study’s findings, that the hazard ratio for time to relapse was 0.41 (95.41% CI, 0.15-1.1), based on relapse in eight tocilizumab-treated and 11 placebo-treated patients, was not statistically significant, with a P value of .0596. This finding considered the full, 36-patient intent-to-treat population.
“That P value implies that this graph might have been created by chance, and it’s possible that if you just reran the same exact study, that these lines might have converged,” Seo said. “And in most clinical trials, that’s where the investigators would have stopped.”
“But remember this was a small trial, so these investigators actually knew all 36 patients who were enrolled, so what they did was something kind of clever,” he added. “They went back, and they drilled down. They tried to figure out what happened to each of these patients.”
Excluding the Outliers
Ultimately, the researchers excluded three of the participants for a per-protocol set, which included 16 treated with tocilizumab and 17 who received a placebo. Among the excluded patients, two experienced issues with prednisone treatments, while the third demonstrated an adverse event unrelated to the trial that require an interruption in therapy.
“It would probably be unfair to include these patients in the clinical analysis,” Seo said. “So, what they did was pull those patients out, asking us to give them a mulligan on those three patients, and said let’s see what happens when we rerun the numbers without those patients.”
The result, according to the TAKT study, was a hazard ratio of 0.34 (95.41% CI, 0.11-1) and a P value of .034.
“And imagine my surprise — if you pull out those three patients that the investigators didn’t want to include in the analysis, all of a sudden it’s statistically significant,” Seo said. “Now, I’m not saying what they did was deceptive at all, because they were very up-front in the analysis, describing exactly what they did and why they did it. I am just pointing out that if three patients, more or less, can swing the trial from being significant to not significant and back again, you have to wonder a little bit about how reliable these results are.”
Seo then compared the TAKT trial with the GiACTA trial, which found that both weekly and every-other-week doses of tocilizumab were efficacious for patients with new GCA. GiACTA also found that weekly tocilizumab may be more efficacious for patients with relapsing GCA and is associated with greater improvements in fatigue and quality of life, compared with placebo. In addition, the second part of the study found that weekly tocilizumab for 1 year may be associated with decreased rates of GCA relapse, up to 2 years after drug cessation.
According to Seo, these studies demonstrate that the magnitude of benefit is less for patients with TAK than for GCA.
“In an as-treated analysis, when you pull out those patients who maybe didn’t belong in the analysis, it is definitely true that tocilizumab seems to benefit patients with TAK,” Seo said. “But that benefit is small compared to the benefit that patients with GCA receive from tocilizumab, and you have to keep that in mind — GCA and TAK probably react differently to tocilizumab.”
“From my standpoint, I think that tocilizumab probably has a role in the treatment of TAK, but it is probably a second- or third-line therapy,” he added. “I still think that TNF inhibitors probably should be standard of care.”
Collapse
Seo P. Update on Vasculitis. Presented at: Congress of Clinical Rheumatology-East annual symposium; September 10-13, 2020 (virtual meeting).
Read more about
Click Here to Manage Email Alerts