Results
A total of 356 patients were randomly assigned to treatment (figure 2). Baseline disease characteristics of the study population and patient disposition through week 24 have been previously published. At baseline, 28 (35.9%) patients in group 1, 43 (29.1%) patients in group 2 and 44 (31.4%) patients in group 3 used concomitant DMARDs. At week 52, a total of 313 patients (group 1, n=70; group 2, n=118; group 3, n=125) were still receiving treatment, and 290 patients (group 1, n=65; group 2, n=109; group 3, n=116) completed the study through week 104.
(Enlarge Image)
Figure 2.
Patient disposition through week 104. AEs, adverse events.
At week 16, 41 (53%) patients in group 1 and 25 (18%) patients in group 2 had less than 20% improvement in total back pain and morning stiffness and entered early escape. Thirty-three (24%) patients in group 3 met the criteria for early escape at week 16, but as specified in the protocol, they did not have their dose adjusted.
Through week 104, 13 (17%) patients in group 1, 28 (20%) patients in group 2 and 24 (17%) patients in group 3 discontinued treatment. The most common reasons for discontinuation of the study agent were unsatisfactory therapeutic effect (27 patients: 8 in group 1, 11 in group 2 and 8 in group 3) and adverse events (19 patients: 2 in group 1, 7 in group 2 and 10 in group 3).
Efficacy
The intention to treat analysis revealed that 38.5% of patients in group 1, 60.1% in group 2 and 71.4% in group 3 had an ASAS20 response at week 104. In addition, 38.5% of patients in group 1, 55.8% in group 2 and 54.3% in group 3 had an ASAS40 response, and 21.8%, 31.9% and 30.7%, respectively, were in partial remission.
Actual observed efficacy outcomes at weeks 28, 52 and 104, without data imputation, are summarised in Table 1 according to early escape status. Patients who required early escape at week 16 generally had lower response rates than those in the same group who did not require early escape. However, of the patients in group 2 who met the early escape criteria, 42.9% and 43.8% achieved ASAS20 responses at weeks 52 and 104, respectively, after receiving a dose escalation from 50 to 100 mg (Table 1). Of the patients in group 3 who met the early escape criteria, 35.7% and 56.0% achieved ASAS20 responses at weeks 52 and 104, respectively, even though they did not receive dose adjustments.
The proportions of patients with ASDAS scores <1.1 or <2.1 at week 24, 52 or 104 are summarised according to the original randomised treatment groups (table S1). Significantly greater proportions of patients in group 2 and group 3 had ASDAS scores <1.1 or <2.1 at week 24 when compared with patients in group 1.
Observed ASAS20 and ASAS40 responses for each treatment regimen over time through week 104 are shown in figure 3. Mean BASDAI, BASMI (range of motion) and BASFI values over time through week 104 are shown in figure 4. By week 104, all treatment regimens led to mean BASDAI and BASFI scores that were less than 3. However, scores were reduced earlier in patients who were originally treated with golimumab from week 0 compared with those who initially received placebo and later crossed over to golimumab. Similarly, the mean BASMI score improved to less than 3 at week 104 for patients who received golimumab from week 0 through week 104, compared with a mean BASMI score of 3.17 for those patients who were assigned to placebo initially and who later entered early escape or crossed over to golimumab.
(Enlarge Image)
Figure 3.
The proportions of patients who had an ASAS20 or ASAS40 response through week 104. Treatment regimen 1 consisted of patients who were originally assigned to placebo at baseline and either entered early escape at week 16 to receive golimumab 50 mg through week 104 or crossed over at week 24 to receive golimumab 50 mg through week 104. Treatment regimen 2 consisted of patients who were originally assigned to golimumab 50 mg at baseline and either entered early escape at week 16 to receive golimumab 100 mg through week 104 or continued 50 mg through week 104. Treatment regimen 3 consisted of patients who were originally assigned to golimumab 100 mg at baseline and did not receive study medication adjustments. Observed data are presented without imputation. ASAS20, at least 20% improvement in the Assessment of SpondyloArthritis international Society (ASAS) criteria; ASAS40, at least 40% improvement in the criteria.
(Enlarge Image)
Figure 4.
Mean BASDAI, BASMI and BASFI scores through week 104. Treatment regimen 1 consisted of patients who were originally assigned to placebo at baseline and who either entered early escape at week 16 to receive golimumab 50 mg through week 104 or crossed over at week 24 to receive golimumab 50 mg through week 104. Treatment regimen 2 consisted of patients who were originally assigned to golimumab 50 mg at baseline and who either entered early escape at week 16 to receive golimumab 100 mg through week 104 or continued 50 mg through week 104. Treatment regimen 3 consisted of patients who were originally assigned to golimumab 100 mg at baseline and who did not receive study medication adjustments. Observed data are presented without imputation. BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; BASFI, Bath Ankylosing Spondylitis Functional Index; BASMI, Bath Ankylosing Spondylitis Metrology Index.
Observed proportions of patients achieving a BASDAI50 response, as well as actual changes from baseline in BASFI, BASMI (linear method) and PCS and MCS scores of the SF-36 (health-related quality of life) are summarised in Table 1 for patients in each group according to their early escape status. Consistent with clinical efficacy as assessed by ASAS response criteria, patients who required early escape at week 16 generally had lower response rates or less improvement from baseline than those in the same group who did not require early escape.
Safety
Safety results through week 24 were previously reported. Adverse events from baseline through week 104 for patients who received at least one dose of golimumab are summarised in Table 2. Adverse events were attributed to the dose that the patients were receiving at the time of the event. Thus, patients are counted in the treatment group in which they were assigned at the time of the event. The proportions of patients with at least one adverse event generally increased with longer average duration of follow-up. More patients who received the 100 mg dose of golimumab at any point during the trial had one or more serious adverse events (24 [14.5%] of 165) and serious infections (7 [4.2%] of 165) compared with patients who received golimumab 50 mg (16 [8.0%] of 213 and 3 [1.4%] of 213, respectively).
Through week 104, infections were reported in 37.7% of patients who had received placebo, 62.0% of those who had received golimumab 50 mg and 70.9% of those who had received golimumab 100 mg; serious infections were reported in 1.3%, 1.4% and 4.2% of patients, respectively (table S2). The incidence of infections per 100 patient-years in patients receiving golimumab 50 mg was similar to that for patients receiving placebo. A dose-dependent trend in the rate of infections and serious infections was noted for the two golimumab doses. However, when comparing the incidence per 100 patient-years of infection, serious infection or upper respiratory tract infection, the 95% CIs for the two golimumab doses overlapped and thus showed no apparent dose effect.
No patients died through week 104 of the study. Serious infections were similar in nature and frequency to those observed in the first 24 weeks of the study. Serious infections that occurred after week 24 included the following events: urosepsis and clostridial infection in group 1 (one patient each); tonsillitis (one patient) and anal abscess (two events in one patient) in group 2; and infectious bursitis, cellulitis, Lyme disease and pulmonary tuberculosis (one patient for each event) and pelvic inflammatory disease and endometriosis of the appendix (both in the same patient) in group 3. The patient with tuberculosis was negative at screening according to QuantiFERON and tuberculin testing. No opportunistic infections were reported through week 104. However, one patient in group 2 had coccidiomycosis after week 104.
As previously noted, two patients, one each from groups 1 and 3, had basal cell carcinoma reported prior to week 24. No additional malignancies were reported past week 24.
Injection-site reactions occurred in 7.3% of patients in group 1 who entered early escape, 8.8% of patients in group 1 who crossed over to active treatment at week 24, 24.0% of patients in group 2 who entered early escape, 8.0% of patients in group 2 who did not enter early escape and 11.4% of patients in group 3 (Table 2) and mostly comprised injection-site erythema and swelling. No injection-site reactions were considered to be serious or resulted in discontinuation of study agent.
Liver enzyme elevations were more common in patients who received golimumab 100 mg than in those who received the 50 mg dose. While 2.8% of all patients had markedly abnormal elevations of alanine aminotransferase levels, no patient had concomitant markedly abnormal postbaseline elevations of alanine/aspartate aminotransferase (at least three times the upper limit of normal) and total bilirubin (at least two times the upper limit of normal).
Ten patients were found to be positive for antibodies to golimumab through week 24. Of these, five (50%) patients changed from positive to negative between weeks 24 and 104. It is unclear if antibodies to golimumab had an effect on treatment response.