© 2024 MJH Life Sciences™ and Patient Care Online. All rights reserved.
Results are in on the maintenance phase of the TAXIT trial that employed infliXImab to treat patients with Crohn Disease and ulcerative colitis. Details here.
Final Results of the Randomized Controlled TAXIT Trial: A presentation of a research abstract during the Plenary Session of Advances in IBD 2013
Presenter: Niels Vande Casteele, PharmD, PhD, University of California San Diego, and Katholieke University Leuven (Leuven, Belgium)
A recurrent theme at last year’s AIBD was the emerging strategy for the use of anti-TNF agents for IBD-that clinical results can be optimized by careful monitoring of response using scoring systems, objective endoscopic findings, C-reactive protein (CRP) levels, fetal calprotectin, development of antibodies to anti-TNF, imaging, and most importantly, measurement of trough drug levels to adjust anti-TNF dose (so-called dose-to-target strategies). Allowing trough level to go too low results in poor disease control, while allowing it to go too high increases the risk of immunosuppressive and other side effects (eg, antibody formation). Last year’s presenters didn’t show the results of new randomized trials supporting those strategies, so Dr Vande Casteele’s well-designed study of infliximab (IFX) fills that gap.
The optimization phase of the TAXIT (Trough-level Adapted infliXImab Treatment) trial has been previously reported, and showed that dose intensification of IFX in Crohn disease (CD) patients with trough level (TLl) <3 μg/mL resulted in better initial disease control, while dose reduction in CD and ulcerative colitis (UC) patients with TLl >7 μg/mL resulted in lower drug exposure and IFX cost while maintaining disease control.
In TAXIT’s maintenance phase reported today, these assumptions were studied over 1 year in 251 patients; 123 (82 with CD and 41 with UC) were randomized to the clinically based (CB) arm of the study, and 128 were randomized to the trough-level–based (LB) arm (91 with CD and 37 with UC). The primary endpoints of the study were the rates of clinical remission (Harvey-Bradshaw/partial Mayo score) and biological remission (C-reactive protein ≤5 mg/L) 1 year after randomization. Secondary endpoints were the number of patients with TLl between 3 and 7 μg/mL and antibody to IFX. Baseline TLl and CRP were not significantly different in the CB and LB arms.
A total of 226 of 251 patients completed the study. After 1 year, 56% in the CB versus 78% in the LB arm had TLl between 3 and 7 μg/mL (P < .001). The trough-level–based algorithm clearly improves drug level maintenance compared with clinically driven dosing. Undetectable trough level (too low to measure) was 3.7 times as likely in the patients randomized to clinically driven dosing. Development of antibody was 3.3 times as likely in the clinically driven arm. Unfortunately, despite better numbers, patients randomized to dose-to-target optimization did not show superior durable remission rates.
But the study is exciting when you consider the extreme cost of IFX. The dose-to-target algorithm allowed the drug to be used more efficiently, preventing high trough levels (I’d have liked to see a comparison of the cost of the drug in the two arms). The fact that dose-to-target patients had less antibody formation suggests they were being exposed to less of this very, very expensive drug. Longer-term follow-up may establish lower rates of malignancy and infection, which would be the welcome news.
For now, the maintenance phase of the TAXIT trial shows that the drug can be used more efficiently without sacrificing remission durability. Cost-effectiveness may well drive future accessibility to this highly effective medication.
Related Content: