Effect of TNF‐inhibitor therapy on spinal structural progression in ankylosing spondylitis patients: A systematic review and meta‐analysis

To review the effect of tumor necrosis factor‐alpha inhibitor (TNFi) therapies on radiographic progression in ankylosing spondylitis (AS) patients as evaluated by the modified Stoke Ankylosing Spondylitis Spine Score (mSASSS). Pubmed, MEDLINE, EMBASE, and the Cochrane Library databases were searched from inception to August 2019. All comparative and non‐comparative studies that evaluated the clinical effectiveness of TNFi on radiographic progression as assessed by mSASSS change at a minimum follow‐up of 1 year were included. The Newcastle–Ottawa Scale and Cochrane Collaboration Risk of Bias Tool were utilized to assess the methodological quality. Pooled analysis was performed for continuous and binomial variables where appropriate. Inter‐rater reliability of mSASSS status and change scores were assessed with intra‐class coefficients (ICC). Twenty‐one studies were identified with a total of 4460 patients (mean age: 40.4 years [range 25.3‐50 years]; 76% male; mean baseline mSASSS: 12.7 units [range 5.5‐19.8 units]). All studies (3 randomized and 18 observational studies) were considered to have moderate‐to‐high methodological quality. The inter‐rater reliability of mSASSS status and change scores from 14 of the 21 studies were excellent (ICC ranges, 0.91‐0.99) and moderate‐to‐excellent (ICC ranges, 0.58‐0.90), respectively. From the 21 studies, 11/21 (50%) demonstrated a delayed effect in mSASSS in AS patient administered TNFi. When stratifying these studies into those with ≤4 years of follow‐up and >4 years follow‐up, 3/11 (27%) and 8/10 (80%) studies respectively indicated a delayed effect of mSASSS with TNFi in AS patients. Pooling for meta‐analysis from 3 studies (1159 patients) with study durations ranging 4‐8 years, indicated that TNFi‐treated patients had reduced odds of structural progression (odds ratio 0.81; 95% CI 0.68‐0.96; P = .01; I2 = 0%). Mean rate of mSASSS change from 16 studies ranged from −0.15 to 7.3 mSASSS units for all AS patients. Meta‐analysis indicated a numerical, but statistically non‐significant, reduction in the rate of mSASSS change with TNFi treatment (7 studies [1438 patients]; mean difference, −0.24; 95% CI, −0.49‐0.01; P = .06; I2 = 0%). This systematic review and meta‐analysis indicated that >4 years of TNFi usage was associated with delayed structural progression by mSASSS. The narrative analysis of the data from 21 studies further confirmed that studies with >4 years of follow‐up had delayed structural progression with T