Disease Activity Is Associated with Non-Adherence in Dimethyl Fumarate Compared to Fingolimod in Clinical Practice

Background: Dimethyl fumarate (DMF) and fingolimod (FTY) are frequently used oral disease modifying therapies for relapsing multiple sclerosis (MS). Phase 3 trials established these agents as effective and generally well tolerated. Using propensity score (PS) analysis, our previous 12-month experience showed comparable efficacy and adherence, though DMF had greater radiographic disease activity and side effects early after treatment initiation. In our cohort, the direct relationship between non-adherence and disease activity on these two oral agents remained unclear.

Objectives: To assess the effect of non-adherence on disease activity in DMF compared to FTY in patients with MS in clinical practice.

Methods: Patients followed in a large academic MS center who were prescribed DMF (n=458) or FTY (n=317) within 12 months of respective FDA approval and had 12-month follow-up data available were identified. We also identified a subgroup of DMF (n=99) and FTY (n=64) patients with disease activity, defined as clinical relapse and/or new radiographic disease activity (e.g. new T2 and/or gadolinium-enhancing lesions). Effects of non-adherence in patients with disease activity were compared between treatment groups using PS analysis and an adjusted linear regression model. Degree of covariate balance was determined by comparing standardized differences before and after propensity adjustment. PS weighting showed favorable covariate balance between DMF and FTY groups. Reason for non-adherence, e.g. intolerability, forgetfulness, and cost/insurance issues; was also collected.

Results: In our subgroup population of patients with disease activity, 49 DMF and 24 FTY patients were non-adherent on respective therapies during initial 12 months of treatment. PS weighting showed higher risk of non-adherence in DMF vs. FTY patients [difference estimate= 8.12, 95% CI (2.88, 13.37), p=0.003]. Most common reason for adherence issues in this cohort was forgetfulness (DMF n= 33/49; FTY n=13/24). Non-adherence due to intolerability was unexpectedly low in the DMF group (n= 6/49).  

Conclusions: In our patient cohort, disease activity was associated with increased non-adherence in DMF patients compared to their FTY counterparts, the most common reason being forgetfulness. Counseling measures on importance of adequate adherence with a twice-daily medication should be stressed in clinical practice.