Combination Therapy with Etanercept and Anakinra for Rheumatoid Arthritis
Sustained inflammation observed in RA is largely mediated by the actions of two macrophage-derived cytokines, TNF-a and IL-1. In clinical trials, antagonism of either TNF-a or IL-1 has proven effective in reducing the signs and symptoms, and in slowing radiographic progression of RA. While combination therapy targeting these two important cytokines would seem attractive given the possibility of a synergistic effect on blocking inflammation, the possibility of increased risk of infection has raised concerns about the safety of this combination. In fact, a small, open-label study (see ACR 2001 highlights) of etanercept + anakinra (without concomitant DMARD) identified a 7% rate of serious infection at week 24 of treatment. Here in a larger placebo-controlled study, Genovese et al (Arthritis & Rheum 50(5); 1412, 2004) examine the safety and efficacy of the combination of TNF-aand IL-1 inhibitor therapy.
Methods: In this multicenter, double blind, placebo-controlled trial, RA patients with an inadequate response to methotrexate monotherapy (at a dose of 10-25 mg per week for at least 16 weeks) manifested by at least 6 swollen and 9 tender joints and any two of the following: morning stiffness > 45 minutes, ESR > 28 mm/hr, or CRP > 1.5 mg/dL were randomized to one of three groups:
- etanercept 25 mg twice a week + anakinra placebo QD + background methotrexate
- etanercept 25 mg QW + anakinra 100 mg QD + background methotrexate
- etanercept 25 mg BIW + anakinra 100 mg QD + background methotrexate
Subjects with significant comorbid illnesses or history of serious infections were excluded.
Subjects were evaluated for safety and efficacy every two weeks for the first month and monthly thereafter for the 24 weeks of treatment. The primary efficacy endpoint was the proportion of patients achieving an ACR50 response at 24 weeks. Secondary efficacy endpoints included ACR70 and ACR20 responses at 12 weeks, sustained ACR20 response (for four months), and EULAR good or moderate response at week 24.
Results: 242 subjects were randomized (80 in group 1, 81 each in groups 2 and 3). The three groups were balanced in terms of baseline characteristics and were typical for this type of trial (mean age approximately 50 years, 85% white, 80% female, 70% Seropositive for RF). 75 (93%) of patients in group 1 completed treatment to 24 weeks, there were no discontinuations for adverse events. 63 (78%) of patients in group 2 completed treatment to 24 weeks, 7 discontinued due to adverse events. 66 (80%) of patients in group 3 completed treatment to 24 weeks, 6 discontinued due to adverse events.
ACR20, 50, and 70 responses at 24 weeks in the intention to treat analysis are summarized below. There was no significant difference between groups 1 and 3 in the proportion of subjects achieving an ACR50 response at 24 weeks (p=0.914). These relationships were maintained in completers-only and last- observation-carried-forward analyses.
Percent of patients achieving ACRn
response at 24 weeks
| Group 1 | Group 2 | Group 3 | |
| ACR 20 | 68% | 51% | 62% |
| ACR 50 | 41% | 39% | 31% |
| ACR 70 | 21% | 24% | 14% |
90-95% of subjects in all 3 groups reported adverse events, most of which were injection site reactions (40% for group 1 vs. approximately 70% for groups 2 and 3). No serious infections (requiring antibiotics or hospitalization) were observed in group 1. Three serious infections were observed in group 2 (cellulitis, pneumonia, and pneumonitis). Six serious infections were observed in group 3 (herpes zoster, pyelonephritis, and two each of pneumonia and cellulitis). One malignant lymphoma was reported in group 3. Two cases of neutropenia, not leading to infection, were noted in group 3. No cases of tuberculosis or other opportunistic infections were observed.
Conclusions: Combination therapy with etanercept and anakinra for RA refractory to methotrexate monotherapy provides no added benefit to therapy over etanercept alone. While the addition of anakinra to etanercept did not enhance efficacy, there was a trend towards a higher rate of serious infections in patients treated with the combination.
Editorial Comments: This is the first double-blind, placebo-controlled trial appropriately powered to evaluate the efficacy of this combination regimen. Due to the relatively small numbers of infections, this study is not appropriately powered to evaluate the safety of the combination versus etanercept alone. However, the clear lack of added efficacy conveyed by the addition of anakinra to etanercept is reason enough to avoid this regimen in clinical practice.
It is not clear why an additive or synergistic effect was not observed with the etanercept + anakinra combination. Anakinra’s clinical efficacy in RA is relatively modest in comparison to etanercept, perhaps due to its relatively short half-life. A more potent and durable IL-1 inhibitor, such as the IL-1 trap molecule, may have greater efficacy in combination with etanercept. However, the high rate of serious infection, now shown in two etanercept + anakinra combination studies, mitigates against the use of any combination of TNF inhibitor + IL-1 inhibitor for now.
