Sustained response to anakinra in ankylosing spondylitis

Sir, We write with regard to the use of the interleukin-1 (IL-1) receptor antagonist, anakinra, in ankylosing spondylitis (AS). We previously reported a 3-month open-label study of anakinra in nine patients with active AS where clinical, laboratory parameter and MRI improvements were described [1] (Table 1). Six of the nine patients achieved Assessment in AS 20% (ASAS20) improvement and 61% of the identified regions of enthesitis/osteitis determined by MRI improved or regressed, but the changes appeared less dramatic than those evident on anti-tumour necrosis factor (TNF) agents. Clinical and laboratory parameters EMS and CRP are shown at three timepoints: baseline, on therapy and following cessation. BASFI, Bath Ankylosing Spondylitis Functional Index; BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; ASQoL, Ankylosing spondylitis quality of life; EMS, early morning stiffness; CRP, C-reactive protein. (normal range <10 mg/l) Clinical and laboratory parameters EMS and CRP are shown at three timepoints: baseline, on therapy and following cessation. BASFI, Bath Ankylosing Spondylitis Functional Index; BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; ASQoL, Ankylosing spondylitis quality of life; EMS, early morning stiffness; CRP, C-reactive protein. (normal range <10 mg/l) Haibel et al. [2] subsequently reported a 24-week, open-label study of anakinra in 20 patients with AS. They reported non-significant decreases in Bath AS Disease Activity Index (BASDAI), an increase in C-reactive protein (CRP) and no improvement in MRI acute inflammatory lesions. However, 5 of the 13 patients who completed the study achieved an ASAS20 response and they concluded that a small subgroup had symptomatic improvement with anakinra. One possibility that could account for these apparent differences in these studies is that anakinra has only a short duration of efficacy beyond which there is no benefit. We, therefore, reviewed three patients from our cohort, who continued anakinra therapy on compassionate grounds for a mean of 23 months (range 19–30). Two continued to respond until the drug was withdrawn, because of lack of funding, and one (Patient 3) lost efficacy after 19 months. Clinical and laboratory responses demonstrate sustained efficacy beyond 6 months in these cases (Table 1). Following therapy cessation, CRP values rose in all three cases. Subsequently, two of the patients switched to the anti-TNF fusion protein, etanercept, with good clinical and acute-phase response. IL-1 is a pro-inflammatory cytokine that is central to inflammation and joint destruction, is up-regulated in spondyloarthropathies and claims for an IL-1 polymorphism in association with AS have been made [3–6]. Although there are conflicting results in open-label studies of the efficacy of anakinra in AS [1, 2], it is clear from both the studies that certain individuals respond to the drug for up to 24 weeks. In our three cases, the response was sustained for up to 30 months. It may be that patients with sustained response have a more IL-1-mediated disease than non-responders. Our cohort also had a 40% greater baseline CRP compared with the German cohort, indicating that higher serum markers of inflammation may be a predictor of response as has been suggested for anti-TNF response in AS [7]. Haibel et al. [2] reported that 21% of their cohort achieved ASAS40 response that is substantially greater than the reported placebo response of 5.1% in a previous AS anti-TNF study [8]. To summarize, we feel that the available data do not completely exclude a role for anakinra therapy in some cases of AS. It remains a strong possibility that the inflammation associated with AS may be amenable to IL-1 pathway modulation and it is possible that IL-1 blockade, either partially with anakinra or more completely with monoclonal antibodies, may have a role. Future trials will address these issues. Disclosure statement: The authors have declared no conflicts of interest.