A detailed description of the first 24-week study phase, initiated in June 2009 in five countries in Latin America (i.e. Argentina, Chile, Colombia, Mexico, and Panama), has been published elsewhere [12]. In brief, adult RA patients who had active disease (≥8 tender/≥6 swollen joints and erythrocyte sedimentation rate [ESR] ≥28 mm/h) despite treatment with methotrexate 7.5 to 25 mg/week for at least 3 months were randomized in a 2:1 ratio to receive etanercept (50 mg/week) plus methotrexate or the conventional DMARD hydroxychloroquine or sulfasalazine (at the investigator’s discretion) plus methotrexate for 24 weeks. Methotrexate doses could be titrated between baseline and week 4, from a minimum dosage of 7.5 mg/week to a maximum dosage of 25 mg/week, but stable doses were required thereafter. In patients with treatment-related adverse events, methotrexate doses could be reduced, with doses re-titrated to previous levels when possible. Hydroxychloroquine and sulfasalazine were titrated to optimal doses by week 4 based on investigator judgment to reflect local standard of care and product labeling. Permitted concomitant medications included stable doses of oral corticosteroids (≤10 mg/d prednisone or equivalent), intra-articular corticosteroid injections (administered >28 days prior to study visits), and a single non-steroidal anti-inflammatory drug (as-needed, with dose adjustments and substitution allowed).

Patients who completed the 24-week randomized phase could enter an open-label, 104-week extension phase (for a total of 128 weeks of treatment). At the start of the optional study extension, investigators selected a treatment regimen (i.e. etanercept, methotrexate, hydroxychloroquine, or sulfasalazine in any combination at the desired dosage) based on patients’ previous response to randomized treatment, their preference, and local product labeling. Biologic agents other than etanercept or conventional DMARDs other than hydroxychloroquine, sulfasalazine, or methotrexate could not be used. Doses of methotrexate, hydroxychloroquine, or sulfasalazine selected at week 24 could be titrated at the discretion of the investigator, but no new treatment could be initiated after treatment selection for patients entering the extension. Although the selected treatment was to remain as stable as possible throughout the subsequent 104 weeks of follow-up, the investigator could make therapeutic changes based on adverse events (i.e. titration or discontinuation of a selected treatment or substitution with an alternate study DMARD) or disease activity (titration or discontinuation of a selected treatment).

This study was conducted in compliance with the Declaration of Helsinki principles, the International Conference on Harmonisation Good Clinical Practice guidelines, and local regulatory requirements. All patients signed informed consent forms, which were reviewed and approved by independent ethics committees or institutional review boards.

Changes in the initial randomized treatment regimens were recorded at the beginning of the optional extension. Treatment-emergent adverse events and serious adverse events were assessed at each study visit and classified by system organ class and preferred terms of the Medical Dictionary for Regulatory Activities. Clinical efficacy end points in the extension included the proportions of patients achieving the American College of Rheumatology (ACR) 20 (ACR20), 50 (ACR50), and 70 (ACR70) responses (20%, 50%, and 70% improvement in ACR response criteria); low disease activity (LDA), defined as a Disease Activity Score in 28 joints (DAS28) based on ESR of <3.2; and remission, defined as DAS28 <2.6. Mean changes from baseline to weeks 76 and 128 were assessed for DAS28; tender and swollen joint counts; physician global assessment; and duration of morning stiffness.

PROs were evaluated using the subject global assessment; visual analog scales for general health, pain, and fatigue; the total Health Assessment Questionnaire Disability Index (HAQ-DI); the physical and mental component summaries (PCS, MCS) and vitality domain of the 36-item Short-Form (SF-36) Health Survey; and the Work Productivity and Activity Impairment:RA (WPAI:RA) questionnaire. The proportions of patients achieving a normal score on the HAQ-DI (i.e. ≤0.5) and a minimally clinically important difference (MCID; i.e. ≥5) in improvement from baseline in the SF-36 scores for the PCS, MCS, and vitality domain were assessed. Physician and patient satisfaction with RA control were evaluated throughout the 104-week extension using questionnaires. The Caregiver Burden and Resource Utilization (CBRU) questionnaire evaluated current employment and utilization of healthcare resources (i.e. rheumatologist and emergency department visits and caregiver assistance).

Results of all analyses from the extension were summarized using descriptive statistics; no statistical comparisons of findings from the 104-week extension were performed. Findings from the extension are summarized using the treatment groups to which patients were randomized in the initial randomized phase (i.e. etanercept plus methotrexate and DMARD [hydroxychloroquine or sulfasalazine] plus methotrexate), which may have been modified at the start of the extension. All patients who took at least one dose of study drug in the extension phase are included in the extension analyses.

The number of treatment-emergent adverse events per patient-year (PY) was assessed in patients exposed to etanercept in the study, from the start of the initial randomized phase for those in the etanercept-plus-methotrexate group and from the start of the extension for those in the conventional DMARD-plus-methotrexate group. The last observation carried forward approach was used for analyses of clinical efficacy, PROs, and work productivity. Observed cases were included in analyses of the physician and patient satisfaction and CBRU questionnaires.

Of 424 randomized patients evaluated in the first phase of the study, 398 (94%) completed that phase and 386 (91%) entered the optional extension. Of the latter 386 patients, 260 patients had been randomized to receive etanercept plus methotrexate and 126 patients to receive the conventional DMARD plus methotrexate in the first phase of the study. Twenty-four patients (6%) were discontinued during the first 52 weeks of the extension (see Fig. 1, Supplementary Content 1). Of the 361 patients (94%) who continued after week 76 of the extension (1 additional patient completed the first 52-week extension period but discontinued before entering the second 52-week extension period), 331 patients (92%) completed the study: 218 patients (90%) from the randomized etanercept-plus-methotrexate group, and 113 (94%) from the conventional DMARD-plus-methotrexate group in the first phase of the study.

Demographic and baseline disease characteristics of patients who entered the 104-week extension are shown in Table 1. Overall, the mean (standard deviation [SD]) age of patients was 48.4 (11.6) and the majority (89%) were female; 45% of patients were white, 23% Mestizo, and 15% African-Latin American. The mean (SD) disease duration was 8.2 (7.2) years. Before entering the study, all patients had been treated with methotrexate, and most had been treated with nonsteroidal anti-inflammatory drugs (85%) and corticosteroids (74%).

As previously described, investigators were permitted to modify patients’ treatment regimens at the start of the 104-week extension. Few changes were made to the treatment regimens of patients who received etanercept plus methotrexate in the first study phase: 259/260 patients in this group continued to receive etanercept in the extension (see Table 1, Supplementary Content 2). The additional patient in this treatment group had been assigned to etanercept-plus-methotrexate treatment for the randomized phase but was given sulfasalazine in error; this patient’s treatment was subsequently changed from sulfasalazine to etanercept at the beginning of the extension. All patients in this group continued to receive methotrexate. Hydroxychloroquine was added to the treatment regimen in eight patients (3%) and sulfasalazine was added in three patients (1%) in the extension.

In contrast, the treatment regimen was changed at the beginning of the extension in most patients who had been randomized to receive conventional DMARD plus methotrexate in the first phase of the study, with etanercept added as a treatment in 105/126 patients (83%). Hydroxychloroquine was discontinued in 35 patients (55%) and sulfasalazine was discontinued in 51 patients (82%). Methotrexate was continued in all 126 patients in the group.

Fig. (1). Proportion of patients achieving (A) ACR20 response, (B) ACR50 response, (C) ACR70 response, (D) DAS28 LDA, (E) DAS28 remission, and (F) normal HAQ (≤0.5). Treatment modification was permitted at the start of the extension (i.e., after week 24); during the extension, 259/260 patients in the ETN + MTX group continued to receive etanercept and 105/126 patients in the DMARD + MTX group received etanercept. ACR, American College of Rheumatology; DAS28, Disease Activity Score in 28 joints; DMARD, disease-modifying antirheumatic drug; ETN, etanercept; HAQ, Health Assessment Questionnaire; LDA, low disease activity; LOCF, last observation carried forward; MTX, methotrexate. Analyses included patients who received at least one dose of study drug in the extension phase; LOCF.

Because the majority of patients in the extension phase were treated with etanercept, safety findings are reported for all patients exposed to etanercept in the study. The total exposure to etanercept during the 24-week randomized phase and the 104-week extension (when treatment assigned in the randomized phase could be modified to include any combination of etanercept, methotrexate, hydroxychloroquine, and/or sulfasalazine) was 798.11 PYs. Among 365 patients exposed to etanercept, 1354 adverse events were reported, which represents 1.70 events per PY (Table 2). The most frequently reported adverse events (excluding injection site reactions) were bronchitis (0.07 events per PY), nasopharyngitis (0.07 events per PY), influenza (0.06 events per PY), urinary tract infection (0.06 events per PY), and pharyngitis (0.05 events per PY).

Fifty-seven serious adverse events were reported in patients exposed to etanercept during the study (0.07 events per PY). Fifteen serious infections (0.02 events per PY) were reported. Six patients had malignancies (0.01 events per PY), which all resulted in withdrawal from the study; two of the six malignancies were considered by the investigators to be treatment-related (i.e. basal cell and cervical cancers). No cases of demyelinating disease or tuberculosis were reported during the study. Adverse events resulted in death in three patients during the study extension, including one death caused by septic shock, respiratory acidosis, and respiratory failure (considered to be treatment-related), one caused by myocardial infarction, and one caused by malignant lung neoplasm (the latter two deaths were not considered treatment-related).

Safety findings in the etanercept-plus-methotrexate and conventional DMARD-plus-methotrexate treatment groups in the second study phase (see Table 2, Supplementary Content 3) were similar to those reported in previous studies of etanercept in RA, with no new safety signals observed.

In patients who had been initially randomized to the etanercept-plus-methotrexate group, rates of ACR response and DAS28 LDA and remission were maintained from week 24 to week 128; in those initially randomized to the conventional DMARD-plus-methotrexate group, these rates increased after treatment modification was allowed at week 24 to week 76 and subsequently remained stable through week 128 (Fig. 1A-E). Patients in the randomized etanercept-plus-methotrexate group achieved improvement in DAS28 from baseline to week 24 (mean change [SD], -3.2 [1.3]), which was maintained to week 76 (-3.4 [1.3]) and week 128 (-3.4 [1.4]) (Fig. 2A). Patients in the randomized conventional DMARD-plus-methotrexate group had corresponding DAS28 changes of -1.8 (1.4), -3.4 (1.3), and -3.5 (1.3). Similarly, from week 24 to week 76, relatively small changes were seen in other clinical outcomes in the etanercept-plus-methotrexate group and greater improvement in the DMARD-plus-methotrexate group, followed by stabilization to week 128 (Table 3).

Comparable trends were observed in PROs in these treatment groups in the extension, including the proportions of patients achieving a normal HAQ-DI score (Fig. 1F). Minimal changes were seen in the proportions of patients in the etanercept-plus-methotrexate group who achieved MCID improvement in the SF-36 MCS (59%-58%), PCS (76%-77%), and vitality domain (44%-42%) from week 24 to week 76, whereas larger changes were observed in the proportions of patients in the conventional DMARD-plus-methotrexate group achieving these endpoints (46%-59%; 59%-73%; and 27%-38%, respectively). Rates of MCID improvement in these scores subsequently remained at stable levels to week 128 in both groups. As for clinical outcomes, minimal changes from week 24 to week 76 were reported for PROs, including HAQ-DI, the SF-36 MCS and PCS, and vitality domain scores, among patients initially randomized to receive etanercept plus methotrexate, with greater improvement after week 24 seen among those randomized to receive hydroxychloroquine or sulfasalazine plus methotrexate (Fig. 2B-E; Table 3). Improvements were maintained from week 76 to week 128 in PROs in both treatment groups.

Fig. (2). Mean scores (SD) for (A) DAS28, (B) total HAQ Disability Index, (C) SF-36 PCS, (D) SF-36 MCS, and (E) SF-36 vitality domain. Treatment modification was permitted at the start of the extension (i.e., after week 24); during the extension, 259/260 patients in the ETN + MTX group continued to receive etanercept and 105/126 patients in the DMARD + MTX group received etanercept. DAS28, Disease Activity Score in 28 joints; DMARD, disease-modifying anti-rheumatic drug; ETN, etanercept; HAQ, health assessment questionnaire; MCS, mental component summary; MTX, methotrexate; PCS, physical component summary; SF-36, 36-item short-form health survey. Analyses included patients who received at least 1 dose of study drug in the extension phase; LOCF.

The proportion of physicians who were satisfied with control of their patients’ RA remained stable (at 94%-96%) for patients in the etanercept-plus-methotrexate group from week 24 to week 128; the physician satisfaction rate increased from 42% to 98% for patients in the conventional DMARD-plus-methotrexate group between these time points. No change was seen in the proportion of patients in the etanercept-plus-methotrexate group who were willing or very willing to retake the RA treatment received in the past 6 months from week 24 to week 76 (99%), whereas an increase was reported in the proportion of willing or very willing patients in the DMARD-plus-methotrexate group in this timeframe (82%-98%). Minimal change was seen in patient satisfaction after week 76 in either group.

Based on responses to the WPAI:RA questionnaire, patients in both treatment groups had improvement from baseline to week 24 in the percentage of work time missed and the overall work impairment due to RA in the previous 7 days, and this improvement was maintained through the 104-week extension (Fig. 3A and B). Minimal changes were seen in current employment, according to responses to the CBRU questionnaire (Fig. 3C). In both treatment groups, reductions were seen from baseline to week 24 in the proportion of patients who required visits to a rheumatologist or the emergency department in the past 6 months, and in those who required caregiver assistance in the past month (Fig. 3D-F); these rates generally remained stable through the extension.