AB0750 Back pain and morning stiffness as mediators of tofacitinib treatment effect on fatigue in patients with ankylosing spondylitis: a mediation analysis

Abstract

BackgroundFatigue is a prevalent symptom of ankylosing spondylitis (AS)1 and can contribute to higher levels of disease, disability and poor health-related quality of life.2 Back pain and morning stiffness are also commonly reported symptoms.3 Tofacitinib is an oral Janus kinase inhibitor for the treatment of adult patients (pts) with AS. In randomised studies, pts with active AS treated with tofacitinib experienced greater improvements in fatigue, back pain and morning stiffness at Week 12 and 16 of treatment compared with placebo (PBO).4,5 Treatment of these symptoms is a priority for pts with AS and their healthcare providers, however, the mechanisms underlying the interrelationships between fatigue, back pain, morning stiffness and treatment are unclear.ObjectivesTo describe the interrelationships between fatigue, back pain, morning stiffness and tofacitinib treatment in pts with AS, using mediation modelling.MethodsData from Phase 2 (NCT01786668)4 and Phase 3 (NCT03502616)5 studies of pts with active AS treated with tofacitinib 5 mg twice daily (BID) or PBO were used. Mediation modelling, a statistical method to assess the extent to which the effect of an independent variable on a dependent variable is indirect, via identified mediators, or direct, capturing all other (unmeasured) effects, was applied.6,7 The initial model included: treatment as the independent binary variable (tofacitinib 5 mg BID vs PBO); fatigue (measured by Functional Assessment of Chronic Illness Therapy-Fatigue) as the dependent variable; mediators included back pain (measured by total back pain/nocturnal spinal pain [numerical rating scale, 0–10]) and morning stiffness (represented by the mean of Bath Ankylosing Spondylitis Disease Activity Index questions 5 and 6).ResultsPooled data from 370 pts were included in the analysis. The initial model showed that 57.5% (p<0.001) of the tofacitinib treatment effect on fatigue was mediated via back pain and morning stiffness (indirect effect); mediation via morning stiffness alone was 49.7% (p<0.01), and 21.2% (p=0.02) via back pain alone. The effect of treatment attributable to factors other than back pain and morning stiffness (ie, direct effect) was not statistically significant (-28.4%; p=0.33). As a result, the initial model was re-specified to exclude the direct treatment effect on fatigue. In the re-specified model (Figure 1), 44.0% (p<0.0001) of the indirect effect of tofacitinib on fatigue was mediated via back pain and morning stiffness; mediation via morning stiffness alone was 40.0% (p<0.001), and 16.0% (p<0.01) via back pain alone. Analyses of the individual study data gave results generally consistent with those from the pooled data.ConclusionOverall, indirect pathways via morning stiffness accounted for ~84% of the effect of tofacitinib treatment on fatigue: (1) treatment affects morning stiffness, which impacts fatigue; and (2) treatment affects morning stiffness, which affects back pain and, ultimately, back pain affects fatigue. The indirect pathway via back pain alone accounted for ~16% of treatment effect on fatigue. These results suggest that in tofacitinib-treated pts with AS, improvements in fatigue are fully mediated through combined treatment effects on morning stiffness and back pain.