Effectiveness of Combination Therapy vs. Monotherapy with Etanercept or Infliximab in Real-World Clinical Practice
Efficacy studies of TNF inhibitors in rheumatoid arthritis (RA), such as TEMPO, have shown that initial therapy with a TNF inhibitor + methotrexate combination is superior to either drug used as monotherapy. However, in clinical practice situations may arise preventing the specific selection, timing, and sequence of the agents in the manner tested in randomized clinical trials, such as with prior DMARD failure or combination therapy with a non-methotrexate DMARD. Here, Hyrich et al (Arthritis Rheum 2006; 54(6): 2006) report the results of an effectiveness study (i.e. observational study of real world patients) examining monotherapy with etanercept or infliximab against therapy with either of these agents in combination with methotrexate or non-methotrexate conventional DMARDs.
Methods: Subjects were all those with RA enrolled since October 2001 in the British Society for Rheumatology Biologics Register (BSRBR), a national United Kingdom based registry of patients age 16 years and older receiving biologic DMARDs for any rheumatic disease. Subjects initiating therapy with either etanercept or infliximab were analyzed separately and divided into three comparison groups:
- TNF inhibitor monotherapy
- TNF inhibitor + methotrexate
- TNF inhibitor + non-methotrexate DMARD
The primary endpoint was the clinical EULAR response at 6 months.
Results: 2,711 patients were included (1,258 prescribed etanercept and 1,453 infliximab). Of the etanercept treated patients, 61% received monotherapy, 20% combination therapy with methotrexate, and 19% combination therapy with a non-methotrexate DMARD. Of the infliximab treated patients, 9% received monotherapy, 83% combination therapy with methotrexate, and 8% combination therapy with a non-methotrexate DMARD. The most common non-methotrexate DMARDs used, in order of frequency, were leflunomide, azathioprine, and sulfasalazine. Combination therapies with hydroxychloroquine, cyclosporine, penicillamine, gold, or minocycline were used in less than 1% of subjects.
Baseline Characteristics. Subjects were predominantly female (78%) with a mean age of 57 years and a mean RA disease duration of 15 years at entry. Baseline disease activity and reported disability at entry were high, with a mean DAS28 of 6.7 and HAQ of 2.2. The frequency of prednisone and NSAID use was 51 and 64%, respectively. There were no baseline differences in demographic or RA disease characteristics between subjects receiving infliximab or etanercept. However, there were differences in subjects receiving combination DMARD therapy compared to those receiving monotherapy with a TNF inhibitor. For etanercept users, subjects receiving combination therapy were younger (by 3-4 years), with shorter disease duration (by 1-3 years), higher frequency of NSAID use and lower frequency of corticosteroid use. Disease activity and disability measures were comparable. Similar relationships were found for subjects receiving combination therapy with infliximab compared to those receiving infliximab monotherapy.
Discontinuation Rates. At 6 months, approximately 20% of subjects had discontinued TNF inhibitor therapy, with no differences in discontinuation rates found between combination therapies vs. monotherapy.
Effectiveness. Subjects receiving combination therapy with etanercept + methotrexate were almost 2 times more likely to achieve a favorable clinical response at 6 months compared to those receiving monotherapy, even after adjustment for demographic and RA disease characteristics (OR 1.98 (95% CI 1.45 2.71)) and more than 2 times more likely to meet EULAR remission criteria (OR 2.24 (95%CI 1.24 4.06)). Compared to subjects receiving therapy with etanercept + a non-methotrexate DMARD, those receiving etanercept + methotrexate were also significantly more likely to achieve a favorable clinical response, after adjustment (OR 1.66 (95%CI 1.14 2.42)). Combination therapies with etanercept + non-methotrexate DMARDs were not statistically different from etanercept monotherapy in the odds of achieving a favorable clinical outcome (OR 1.20 (95%CI 0.89 1.61)).
For infliximab, there were no statistical differences in response or remission rates between those receiving monotherapy and those receiving combination therapies with either methotrexate or non-methotrexate DMARDs.
Conclusions: Combination therapy with etanercept + methotrexate is preferable over monotherapy or combination therapy with non-methotrexate DMARDs. Although not recommended, infliximab monotherapy appears to be as effective as combination therapy.
Editorial Comment: Effectiveness studies (of real-world patients), such as this one, add greatly to the information gained in efficacy studies (i.e. randomized clinical trials). In contrast to efficacy studies, in which the constraints of patient selection and follow-up may limit the generalizability of results, effectiveness studies give a better idea of how a drug performs (often better) in clinical practice. The downside of these studies is the lack of randomization, with confounding by indication, often difficult to measure, playing a large role.
The primary practice point gained form this large and well analyzed effectiveness study is that unless contraindicated, etanercept should be used in combination with methotrexate. Though not fully supported by this analysis, combination therapy with a non-methotrexate DMARD is probably preferable to monotherapy. Assuringly, and supporting data from efficacy studies, toxicity and discontinuation rates do not appear to be increased in patients receiving combination therapy.
The results presented here for infliximab are less definitive. For one, since methotrexate is often used in low dose (e.g. 7.5 mg per week) in combination with infliximab (for the prevention of neutralizing anti-infliximab antibodies), one might not expect much difference in efficacy between this form of combination therapy and monotherapy. Separating out the subjects by methotrexate dose would be helpful in clarifying this issue. In addition, follow-up past six months may show a decrease in efficacy for the same dose within the infliximab monotherapy group due to the development of the aforementioned neutralizing antibodies.