Is Non-TNF Targeted Tx Better Than a Second Anti-TNF Drug?

SAN FRANCISCO, CA—A non-TNF-targeted biologic showed better efficacy than a second anti-TNF agent for patients with rheumatoid arthritis and insufficient response to TNF-inhibitors, according to research presented at the 2015 ACR/ARHP Annual Meeting.

Up to one-third of patients demonstrate insufficient response to their first TNF inhibitor, defined as “TNF-IR.” However, “the literature provides little guidance for choosing a new treatment,” Jacques-Eric Gottenberg, MD, PhD, of Hautepierre Hospital in Strasbourg, France, told meeting attendees. “Recently, the EULAR recommendations acknowledged the absence of evidence to guide the choice of a second biologic in TNF-IR patients and the need for randomized trials.”

Observational studies and two randomized, placebo-controlled trials have suggested that half of TNF-IR patients might respond to a second anti-TNF drug. Another possible strategy: switch to a non-TNF-targeted therapy.

To determine which strategy demonstrated more efficacy, Dr. Gottenberg and colleagues conducted ROC (Rotation of anti-TNF Or Change of class of biologic), a 48-week, multi-center, investigator-initiated, open, parallel-group, randomized controlled trial.

A total of 300 patients with inadequate response to a first anti-TNF were randomized to receive a non-TNF–targeted biologic (abatacept, rituximab, or tocilizumab; n=150) or a second anti-TNF agent (adalimumab, certolizumab, etanercept, or infliximab; n=150). The treating clinician selected the type of biologic within each study group (eg, non-TNF-targeted biologic or a second anti-TNF agent). The primary outcome was EULAR good or moderate response at 6 months.

The majority of patients in each arm was female, with disease duration of at least 10 years.

At Week 12, 64.2% of patients in the non-TNF–targeted biologic arm and 47.8% of patients in the second anti-TNF agent arm achieved good or moderate EULAR response (P=0.005); at Week 24, these results were 69.7% vs. 52.1%, respectively (P=0.003); and, at Week 52, 59.5% vs. 42.5% (P=0.006).

At Week 52, low disease activity (DAS28-ESR <3.2) was 40.8% in the non-TNF–targeted biologic arm and 23.3% in the second anti-TNF agent arm (P=0.003); remission (DAS28 <2.6) was 26.9% and 13.5%, respectively (P=0.008).

At Week 24, EULAR good or moderate response among the non-TNF-targeted biologics was 66.8% for abatacept, 61.1% for rituximab, and 79.9% for tocilizumab.

During the trial, 27 patients in the non-TNF-targeted biologic arm and 48 in the second anti-TNF agent arm initiated a new biologic that differed from the allocated biologic, the majority for lack of efficacy. At discontinuation of the allocated biologic, the DAS28 was 4.54 in the non-TNF-targeted biologic arm and 5.25 in the second anti-TNF arm.

The investigators observed “no significant difference in serious adverse events or serious infections” between the two arms, Dr. Gottenberg said. A total of 18 serious adverse events (12.3%) occurred in 16 patients in the non-TNF-targeted biologic arm and 13 (8.9%) in the second anti-TNF arm (P=0.10). There were 7 serious infections (4.8%) in 7 patients in the non-TNF-targeted biologic arm and 10 serious infections (6.8%) in 8 patients in the second-anti-TNF arm.

He noted that these results represent the first confirmation in a setting that nears “common practice” that “approximately 50% of TNF-IR patients might respond to a second anti-TNF agent”; however, “the therapeutic response was more frequent with non-TNF-targeted biologics.” In addition, “this randomized controlled pragmatic ROC trial demonstrated better efficacy of a non-TNF-targeted biologic than a second anti-TNF agent for TNF-IR patients.”

“This superiority was consistent throughout the study period and across numerous outcome criteria,” Dr. Gottenberg concluded.