Meta-analysis results derived from a substantial and, generally, high-quality evidence base on the efficacy of anti-TNFs in patients with AS (considered either as individual treatments or as a common class) show statistically significant and clinically important benefits in terms of improved function and reduced disease activity following around 3 months of treatment with an anti-TNF. Smaller benefits were seen across outcomes in patients with nr-AxSpA, being most noticeably smaller for the function and disease-activity outcomes. However, in the light of the clinical and statistical heterogeneity seen across the nr-AxSpA trials, both the reliability of the nr-AxSpA-pooled estimates and their true relevance to patients seen in clinical practice are questionable. Data from (less robust) observational studies suggest that good levels of treatment response are maintain in around 50% of patients after around 2 years of treatment. Evidence for an effect of anti-TNFs on radiographic disease progression is limited, although results from ongoing studies should clarify whether or not progression rates are reduced in the longer term. The results from studies based on registry data demonstrated that sequential treatment with anti-TNFs can be worthwhile in patients with AS, although drug survival, response rates and benefits were reduced with second and third anti-TNFs. Data from large systematic reviews, which included patients with a wide range of diseases, suggested that, in the short term, anti-TNFs as a group were associated with significantly higher rates of serious infections, tuberculosis reactivation, non-melanoma skin cancer, total AEs and withdrawals because of AEs than control treatments. Longer-term data on AEs were limited.
Implications for service provision
From our review of natural history a key study on flares suggested that the 12-week period required to confirm sustained active spinal disease in AS patients commencing an anti-TNF may be too long. The findings suggest that shorter time periods might therefore be considered in future guidance, which would minimise the delay in starting treatment and the discomfort experienced by patients.
Suggested research priorities
Randomised trials are needed to identify the nr-AxSpA population that will benefit the most from TNF-inhibitors; trials using stratified randomisation and pre-planned analyses by stratified group should inform this issue. Groups could be stratified according to their imaging status (i.e. MRI positive or not) and their CRP level; both the cut-off points to be used for CRP level elevation, and the eligibility criteria used for CRP level elevation, should be given careful consideration, given the variation evident in previous trials. These studies should help to inform clearer guidance as to what ASAS and the anti-TNF licences mean when referring to ‘elevated CRP level’ in patients with nr-AxSpA. There is also a clear need for more accurate biomarkers, or other measures of disease activity, to be developed. In the previous nr-AxSpA trials the placebo-controlled phases typically lasted around 3 months; a placebo-controlled follow-up period of at least 6 months in future trials would therefore be useful for studying persistence of response.
Long-term longitudinal studies are needed on the natural history of nr-AxSpA to help clarify the characteristics of patients who do (or do not) eventually develop AS. Similar to the RCT recommendations, these studies should include analyses stratified by how patients were diagnosed; a comparison of patients with imaging (MRI) evidence of nr-AxSpA versus patients who are diagnosed with only clinical criteria evidence, would be particularly useful, albeit difficult to perform.
Large, long-term longitudinal, cohort studies are needed to clarify the effect of anti-TNFs on the progression of structural damage in AS. In the absence of a gold standard imaging tool across the spectrum from nr-AxSpA to AS, sequential MRI and radiography assessment should be used at pre-defined end points to ascertain the true sensitivity and specificity of these tools in the diagnosis and assessment of neo-formation, and ankyloses characteristic of structural progression in the spine and sacroiliac joints of these patients.
Studies are also needed to better inform the efficacy estimates relating to sequential use of anti-TNFs. An ongoing study is looking at comparing the effect of intermittent versus standard use of anti-TNFs in patients with stable (low-active) disease.182
