Article Text

This article has a correction. Please see:

Download PDFPDF

Extended report
Golimumab in patients with active rheumatoid arthritis despite methotrexate therapy: 52-week results of the GO-FORWARD study
  1. Edward Keystone1,
  2. Mark C Genovese2,
  3. Lars Klareskog3,
  4. Elizabeth C Hsia4,5,
  5. Stephen Hall6,
  6. Pedro C Miranda7,
  7. Jacek Pazdur8,
  8. Sang-Cheol Bae9,
  9. William Palmer10,
  10. Stephen Xu4,
  11. Mahboob U Rahman4,5
  1. 1University of Toronto and Mount Sinai Hospital, Toronto, Ontario, Canada
  2. 2Stanford University, Palo Alto, California, USA
  3. 3Karolinska Institute and Karolinska University Hospital, Stockholm, Sweden
  4. 4Centocor Research and Development, Inc, Malvern, Pennsylvania, USA
  5. 5University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania, USA
  6. 6Cabrini Medical Centre, Malvern, Victoria, Australia
  7. 7Universidad de Chile and Hospital San Juan de Dios, Santiago, Chile
  8. 8Instytut Reumatologii, Warszawa, Poland
  9. 9Department of Rheumatology, The Hospital for Rheumatic Diseases, Hanyang University, Seoul, Korea
  10. 10Westroads Medical Group, Omaha, Nebraska, USA
  1. Correspondence to Dr Edward Keystone, University of Toronto and Mount Sinai Hospital, 60 Murray Street, 2nd Floor, Room 2-006, Toronto, Ontario M5G 1X5, Canada; edkeystone{at}mtsinai.on.ca

Abstract

Objective To evaluate the efficacy and safety of golimumab to 52 weeks in patients with active rheumatoid arthritis despite methotrexate.

Methods Patients were randomly assigned to receive placebo plus methotrexate (group 1), golimumab 100 mg plus placebo (group 2), golimumab 50 mg plus methotrexate (group 3) and golimumab 100 mg plus methotrexate (group 4). At week 16, patients in groups 1, 2 and 3 who had less than 20% improvement in tender and swollen joints entered early escape. At week 24, patients in group 1 who had not entered early escape crossed over to 50 mg golimumab plus methotrexate.

Results At week 16, 31%, 27% and 17% of patients in groups 1, 2 and 3, respectively, entered early escape. At week 52, 44%, 45%, 64% and 58% of patients in groups 1, 2, 3 and 4, respectively, achieved 20% improvement in the American College of Rheumatology criteria; and 34%, 31%, 42% and 53%, respectively, achieved low disease activity (≤3.2) according to the 28-joint disease activity score. Patients in group 4 appeared to have an increased risk of serious adverse events and serious infections.

Conclusion The results of various outcome measures showed that the response rates achieved by patients receiving golimumab to 24 weeks were sustained to 52 weeks. The safety profile appeared to be consistent with the known safety profile of tumour necrosis factor inhibitors.

View Full Text

Statistics from Altmetric.com

Golimumab is a human antibody to tumour necrosis factor (TNF) alpha that is administered subcutaneously every 4 weeks. A previous phase II study of patients with rheumatoid arthritis (RA) despite methotrexate showed that the addition of golimumab to methotrexate was effective at reducing the signs and symptoms of RA after 16 weeks and the response rate was maintained over 1 year.1 The effect of golimumab in this population over 6 months was confirmed in a phase III, randomised, placebo-controlled study called GO-FORWARD.2 Here we report the 1-year results of the GO-FORWARD study.

Methods

The details of the GO-FORWARD study, along with complete patient inclusion criteria, have been published previously.2 Briefly, patients with active RA who had been receiving a stable dose of methotrexate of 15–25 mg/week for at least 4 weeks were randomly assigned in a 3:3:2:2 ratio to receive placebo injections with methotrexate (group 1), golimumab 100 mg injections with placebo capsules (group 2), golimumab 50 mg injections with methotrexate (group 3) or golimumab 100 mg injections with methotrexate (group 4). All injections were administered subcutaneously every 4 weeks. Patients in groups 1, 3 and 4 continued their initial stable doses of methotrexate, while patients in group 2 immediately initiated sham methotrexate capsules at randomisation.

At week 16, patients in groups 1, 2 and 3 with less than 20% improvement from baseline in both tender and swollen joint counts entered early escape in which their study medications were automatically adjusted in a double-blind fashion. Patients in group 1 initiated active golimumab 50 mg instead of placebo injections while continuing methotrexate, patients in group 2 initiated active methotrexate (at the stable dose they were receiving during screening) instead of placebo capsules while continuing golimumab 100 mg injections, and patients in group 3 had their golimumab dose increased from 50 mg to 100 mg while continuing methotrexate. Patients in group 4 did not receive study medication adjustments even if they met the criteria for early escape.

Golimumab plus methotrexate and golimumab plus placebo were compared with placebo plus methotrexate for the first 24 weeks of the study. At week 24, all remaining patients in group 1 who had been receiving placebo injections began receiving golimumab 50 mg injections in a blinded fashion. Patients in groups 2, 3 and 4 continued to receive their originally assigned treatment unless they entered early escape, in which case they continued to receive the modified dosage. Injections continued to be administered subcutaneously every 4 weeks to week 48, with final study assessments at week 52. To maintain the blind, each patient was administered two injections every 4 weeks. Even though all patients were receiving golimumab injections after week 24, placebo injections and placebo methotrexate tablets were continued as appropriate to week 52.

We evaluated the response to treatment using the American College of Rheumatology (ACR) and European League Against Rheumatism response criteria.3,,9 Physical function was assessed using the health assessment questionnaire disability index.10 Low disease activity was defined as a 28-count disease activity score (DAS28) of 3.2 or less.11 DAS28 remission was defined as a score less than 2.6. A patient was considered to have achieved sustained DAS28 remission if they were in DAS28 remission for six consecutive visits between weeks 4 and 52. A patient was considered to have achieved a sustained clinical response if they had an ACR response for six consecutive visits between weeks 4 and 52. Visits were once every 4 weeks except for the primary endpoint at week 14, which was halfway between the week 12 and week 16 visits.

Data are presented using two approaches:

  1. Intention-to-treat (ITT). Patients who met the treatment failure criteria at weeks 14 or 242 and those who entered early escape at week 16 were considered to be non-responders at week 52. In addition, patients who discontinued treatment because of an unsatisfactory therapeutic effect after week 24 or who were missing all ACR components were considered to be non-responders at week 52. For patients who did not meet treatment failure criteria, missing ACR components at week 52 were replaced with the last non-missing value.

  2. By treatment regimen. Treatment regimen 1 consisted of patients who were originally assigned to placebo plus methotrexate and received golimumab 50 mg at week 16 if they met the criteria for early escape or crossed over to 50 mg at week 24. Treatment regimen 2 consisted of patients who were initially assigned to golimumab 100 mg plus placebo capsules and received methotrexate at week 16 if they met the criteria for early escape. Treatment regimen 3 consisted of patients who were initially assigned to golimumab 50 mg plus methotrexate and received 100 mg plus methotrexate at week 16 if they met the criteria for early escape. Treatment regimen 4 consisted of patients who were initially assigned to golimumab 100 mg plus methotrexate and received that dose throughout the study. Actual observed ACR20, ACR50 and ACR70 response rates over time to week 52 are presented without data imputation based on the treatment regimen assigned at baseline. Patients who entered early escape were counted in the treatment regimen to which they were initially assigned. Efficacy data are also presented by actual treatment received based on the early escape status (see table 1 for categories).

Table 1

Efficacy measures at week 52 (except where otherwise indicated) by early escape status

Results

Overall, 444 patients were randomly assigned to treatment. At week 16, 41 patients (30.8%) in group 1, 36 patients (27.1%) in group 2 and 15 patients (16.9%) in group 3 had less than 20% improvement in both the tender and swollen joint counts and entered early escape. Fourteen patients (15.7%) in group 4 met the criteria for early escape at week 16, but the protocol did not allow for dose adjustments in this group. To week 52, 17 patients (12.8%) in group 1, 17 patients (12.8%) in group 2, five patients (5.6%) in group 3 and 13 patients (14.6%) in group 4 discontinued treatment (supplemental material 1).

Baseline disease characteristics of the study population were published previously.2 Patients who entered early escape generally had greater disease activity at baseline than those who did not enter early escape, but the trend was not consistent among the groups (see supplemental file 2, available online only).

Efficacy

The ITT analysis revealed that 44% of patients in group 1, 45% of patients in group 2, 64% of patients in group 3 and 58% of patients in group 4 achieved an ACR20 response at week 52 (figure 1A). ACR50 and ACR70 values are also shown (figure 1C,E).

Figure 1

American College of Rheumatology (ACR)20, ACR50 and ACR70 responses. Panels A, C and E show the results of the intention-to-treat analysis at week 52. Data are displayed according to the patients' originally assigned treatment group: group 1 (n=133), placebo plus methotrexate; group 2 (n=133), golimumab 100 mg plus placebo; group 3 (n=89), golimumab 50 mg plus methotrexate; and group 4 (n=89), golimumab 100 mg plus methotrexate. In this analysis, patients who entered early escape and met other treatment failure rules (see Methods section) were considered to be non-responders. Panels B, D and F show the results by treatment regimen: treatment regimen 1, patients who were originally assigned to placebo plus methotrexate and entered early escape at week 16 to golimumab 50 mg plus methotrexate if they had less than 20% improvement in tender and swollen joints or crossed over to golimumab 50 mg plus methotrexate at week 24; treatment regimen 2, patients who were initially assigned to golimumab 100 mg plus placebo and entered early escape at week 16 to golimumab 100 mg plus methotrexate if they had less than 20% improvement in tender and swollen joints; treatment regimen 3, patients who were initially assigned to golimumab 50 mg plus methotrexate and entered early escape at week 16 to golimumab 100 mg plus methotrexate if they had less than 20% improvement in tender and swollen joints; treatment regimen 4, patients who were initially assigned to golimumab 100 mg plus methotrexate and continued to receive that dose to week 52.

Actual observed ACR20, ACR50 and ACR70 responses for each treatment regimen over time to week 52 are shown in figure 1B,D,F, respectively.

Actual observed efficacy outcomes at week 52 for patients in each group according to their early escape status are summarised in table 1. Patients who required early escape at week 16 generally had lower response rates at week 52 than those in the same group who did not require early escape. Six of the 15 patients (40.0%) in group 3 who entered early escape at week 16 achieved an ACR20 response at week 52, but a similar number (seven of 14, 50.0%) of patients in group 4 who also met the criteria for early escape (but did not receive treatment adjustment) achieved an ACR20 response at week 52.

The proportions of patients who achieved DAS28 low disease activity at week 52 or had no tender and swollen joints at weeks 24 or 52 are summarised in table 1. The proportions of patients who achieved sustained clinical response or sustained DAS28 remission are summarised in table 2.

Table 2

Sustained efficacy outcomes

Safety

Safety results to week 24 were previously reported.2 Adverse events from baseline to week 52 are summarised in tables 3 and 4. The exposure-adjusted rate of adverse events in each of the golimumab groups appeared to be similar to the placebo group. However, patients who received the 100 mg dose of golimumab with or without methotrexate appeared to have greater rates of serious adverse events and serious infections compared with patients who received 50 mg with methotrexate.

Table 3

Summary of adverse events to week 52 for patients who received golimumab by early escape status

Table 4

Incidence of adverse events per 100 patient years to week 52 summarised by the actual treatment received at the time of the event

One patient in group 2 died of sepsis during the 24-week portion of the study,2 and one other patient in the same group died after week 24. The latter patient had acute hepatic failure and died after major haemorrhaging during a liver biopsy. The patient had had a prestudy history of liver function test abnormalities and started methimazole 1 month before the event.

Serious infections were similar in nature and frequency to those observed in the first 24 weeks of the study. Serious infections that occurred over the entire 52-week study are summarised in table 4. Events that occurred after week 24 include tuberculosis pleurisy, pneumonia, fever/pneumonia, sinusitis, bronchitis, cellulitis, lung disorder (acute pneumopathy) and sepsis.

In addition to the three golimumab-treated patients with malignancies observed to week 24, four additional patients had malignancies to week 52. One patient in group 2 with basal cell carcinoma before week 24 had squamous cell carcinoma after week 24, making a total of five patients with malignancies after week 24. Other malignancies after week 24 included a patient in group 1 who had crossed over to 50 mg plus methotrexate at week 24 with squamous cell carcinoma and basal cell carcinoma, a patient in group 4 with basal cell carcinoma and two patients with breast cancer (one each in groups 3 and 4).

Injection-site reactions were uncommon (table 3) and consisted mostly of injection-site erythema and bruising. No injection-site reactions were considered to be severe or serious. One patient in group 3 discontinued study treatment because of moderate injection-site erythema.

Discussion

The 1-year results of the GO-FORWARD trial indicate that the response rates achieved at week 242 by patients who received the combination of golimumab and methotrexate were sustained to week 52. This was demonstrated even by the conservative ITT analysis, in which patients who met various treatment failure criteria at week 16, including early escape, were considered to be non-responders throughout the remainder of the study period. The response rates of the two golimumab doses in combination with methotrexate appeared to be similar.

In addition to the original treatment assignments, the study protocol included treatment adjustments based on patients' improvement in tender and swollen joints (ie, early escape) and crossover from placebo to active treatment. This allowed us to evaluate four different treatment regimens of various combinations of golimumab with and without methotrexate. These results also showed that the response rates achieved at week 24 by patients receiving the combination of golimumab plus methotrexate were maintained to week 52. Patients who were initially assigned to golimumab 50 mg plus methotrexate and had the opportunity for early escape to 100 mg (treatment regimen 3) had similar response rates to those who were initially assigned to golimumab 100 mg plus methotrexate and received that dose for the entire study (treatment regimen 4). The response rates at week 52 for patients who were initially assigned to placebo plus methotrexate and entered early escape at week 16 or crossed over at week 24 approached those of patients who were initially assigned to golimumab plus methotrexate.

We also evaluated the data according to each patient's early escape status. As expected, patients in group 1 who received placebo plus methotrexate during the first part of the study were more likely to require early escape at week 16 than those in the other groups. These patients also showed the greatest efficacy responses at week 52 compared with patients in other groups who entered early escape. Of the few patients who required early escape in group 3, almost half achieved an ACR20 response at week 52. However, a similar proportion of patients in group 4 met the criteria for early escape at week 16 and achieved an ACR20 response at week 52 without dose modification. The proportion of patients in group 2 who required early escape at week 16 was between that of group 1 and group 3; however, the ACR20 response rate at week 52 was also similar to that of group 4. Therefore, for the patients in groups 2 and 3 who responded after early escape, it is not clear whether this effect was the result of the modified dosage or a longer duration of treatment. In addition, the response rates of patients who received early escape were still lower than those who did not receive early escape, indicating that patients who required early escape may have had more refractory disease. Patients who required early escape also tended to have greater disease activity at baseline. However, the interpretation of these data is limited by the small number of patients who met the criteria for early escape, particularly in groups 3 and 4.

Whereas these analyses demonstrated the sustained response of the study population to golimumab, other outcomes showed that some patients achieved and maintained extremely low levels of disease activity by week 52. Nearly 40% of patients receiving golimumab plus methotrexate achieved sustained DAS28 remission and approximately 14% achieved a sustained clinical response. Of the patients who did not require early escape in groups 3 and 4, approximately 23% had zero tender and swollen joints at week 52.

Safety results to week 52 were similar to those to week 24.2 The results of previous studies have suggested that patients who receive high doses of TNF inhibitors may have an increased risk of serious infections.12 13 Our results appear to be consistent with those findings because patients in group 4 had greater rates of serious adverse events and serious infections than those in the other groups. However, the results of a phase I study of golimumab in patients who had previously received a TNF inhibitor showed that the incidence of serious infections among patients who received golimumab 100 mg was not greater than placebo.14 In each of these studies, the overall number of serious adverse events and serious infections was small and neither of the individual studies was powered to detect differences between the treatment groups in uncommon safety events.

Several limitations of the study must be considered while interpreting the results. First, the placebo-comparator portion of the study was only 24 weeks and included an early escape option at week 16. This design was implemented for ethical reasons to limit the time patients were receiving ineffective therapies. However, the design also limits the interpretation of the study results. Second, the treatment regimen determined by the protocol only allowed for dose escalation at one time point and did not allow for decreasing the dose of study agent. The applicability to actual clinical practice is thus limited.

The results of this study suggest that subcutaneous injections of golimumab every 4 weeks were effective at reducing the signs and symptoms in patients with active RA despite methotrexate, and this effect was maintained to 52 weeks. The combination of golimumab and methotrexate was more effective than either golimumab or methotrexate alone. There was no difference in efficacy of the 50 or 100 mg dose of golimumab when used in combination with methotrexate. The safety profile of golimumab appeared to be consistent with the known safety profile of TNF inhibitors. Patients who completed the 1-year study will continue to be followed for up to 5 years.

Acknowledgments

The authors would like to thank the patients, investigators and study personnel who made this trial possible. A full list of investigators was published previously.2 The authors also thank Scott Newcomer, MS of Centocor Ortho Biotech Inc, a wholly owned subsidiary of Johnson & Johnson, Inc, who helped prepare the manuscript, but did not contribute to the study design, acquisition of data, or analysis and interpretation of the data.

References

View Abstract

Supplementary materials

  • Web Only Data ard.2009.116319

    Files in this Data Supplement:

Footnotes

  • Funding This study was funded by Centocor Research and Development, Inc, a wholly owned subsidiary of Johnson & Johnson, Inc and Schering Plough Research Institute, Inc.

  • Competing interests EK, MCG, LK, SH, PCM, JP, S-CB and WP (or their institutions) have received research grants from Centocor and/or Schering-Plough. LK and EK have received consulting fees from Centocor and/or Schering-Plough. ECH, MUR and SX are employees of Centocor Research and Development, Inc and own stock and/or stock options in Johnson & Johnson, Inc.

  • Ethics approval This study protocol was conducted with the approval of the ethics committee at each study site.

  • Patient consent Obtained.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles

  • Miscellaneous
    BMJ Publishing Group Ltd and European League Against Rheumatism