Objectives To compare the efficacy of golimumab (GLM) and pamidronate (PAM) in the treatment of SpA.
Methods Study design: an open randomized controlled trial Patients and methods: Inclusion criteria: (1) ≥18 years of age; (2) fulfills 2009 ASAS classification criteria for axial SpA; (3) Active spondylitis as defined by a BASDAI score of ≥4 (with spinal pain score ≥4), despite treatment with NSAIDs for ≥3 months. Exclusion criteria: (1) Hepatitis B/C carriers; (2) Major surgery within 8 weeks; (3) Active infection; (4) Pregnancy/lactation; (5) Contraindications to anti-TNF or bisphosphonates. Patients were randomized to receive GLM (50mg subcutaneously monthly) or PAM (60mg intravenously monthly) in a 2:1 ratio on top of existing therapies. Latent tuberculosis was screened and treated in the GLM arm. Assessment for clinical efficacy (BASDAI, BASFI, BASMI, ESR, CRP, ASDAS, VAS pain, global assessment, SF36) was performed at week 0,2,4,8,12,16,20,24,32,40 and 48. MRI of the spine and SIJ was performed at week 0, 24 and 48 and graded by the Spondyloarthritis Research Consortium of Canada (SPARCC) scoring system (SIJ score 0-72; spinal score 0-108). The primary efficacy end-point was the proportion of patients who achieved the ASAS20 response at week 48. Intra-group paired data over time were compared by the paired Students' t-test whereas inter-group differences were compared by ANCOVA with adjustment for baseline values.
Results 30 patients were recruited (83% men; age 33.4±10.9 years; disease duration 4.4±3.4 years) – 20 assigned to GLM and 10 assigned to PAM. Baseline demographic and clinical characteristics were not significant different between the two arms, except for a non-significantly higher mean ASDAS (CRP) (4.07±0.77 vs 3.70±0.65) and SIJ SPARCC (15.8±17.7 vs 7.8±5.93) score in GLM-treated patients. At week 48, a higher proportion of patients achieved ASAS20 (50% vs 20%; p=0.23) and ASAS40 (35% vs 0%; p=0.04) responses in the GLM compared to the PAM group. The ASDAS, BASDAI, BASFI, CRP and ESR levels significantly improved with GLM treatment but not with PAM. Interestingly, patient reported outcomes such as pain score and SF36 improved significantly in both treatment groups. In patients treated with GLM, the SPARCC SIJ (15.8±17.7 to 3.80±5.19; p<0.01) and spine (11.4±10.8 to 3.56±5.65; p<0.01) scores at week 48 decreased significantly compared to baseline. However, there was only a modest but non-significant reduction in the corresponding MRI scores observed in PAM-treated patients. There was no serious adverse events (SAEs) reported and the frequency of any adverse events (AEs) was not significantly different between the two arms. Minor upper respiratory infection (URI) was the commonest AE (30%), followed by dyspepsia (10%) and deranged liver function (10%) in GLM-treated patients. In patients treated with PAM, the commonest AE was post-infusion fever/myalgia/headache (30%), followed by dyspepsia (10%), phlebitis (10%) and minor URI symptoms (10%).
Conclusions In patients with axial SpA, GLM was more effective than PAM in reducing clinical disease activity as well as inflammation of the spine and SIJ. PAM led to improvement in subjective pain and quality of life but did not have significant positive effects on MRI inflammation, CRP or ESR.
Disclosure of Interest None declared