Dimethyl fumarate is associated with lower rates of infection and lower infection-related healthcare costs when compared with ocrelizumab.

BACKGROUND

Infections in people with multiple sclerosis (PwMS) may have a detrimental effect on disease progression, risk of hospitalization, and healthcare resource utilization (HRU). The infection risk and HRU costs may vary between disease-modifying therapies (DMTs); however, the individual risks and differences associated with DMTs are not well characterized. Some DMTs may increase the risk for infections in PwMS; however, previous studies have reported an intact humoral immune response in dimethyl fumarate (DMF)-treated patients. The objective was to compare infection-related HRU and healthcare costs (HCCs) between PwMS treated with DMF or ocrelizumab (OCR).

METHODS

Eligible patients were identified from the Optum US claims database between April 2017 and September 2020 (DMF n = 1429; OCR n = 3170). Patients were followed from index date to first occurrence of: (1) end of study, (2) end of insurance eligibility, (3) discontinuation of index DMT, or (4) switch from index DMT to another DMT. Outcomes were annualized rate of infection encounters (defined as infection encounters [n] during follow-up window / days followed [n] × 365); annualized infection-related HCCs (defined as aggregated costs of infection encounters during follow-up window / days followed [n] × 365); location-specific infections, and overall infection-related events. Propensity score matching (PSM) 1:1 method was used; PS was calculated via logistic regression for probability of DMF treatment conditional on demographics and comorbidities. Mean differences (MD) were reported for infection encounter measures.

RESULTS

After PSM, DMF and OCR cohorts (n = 1094 in each cohort) were balanced based on baseline characteristics (standardized MD of adjusted baseline characteristics <0.1). Mean (standard deviation) follow-up was 296 (244) days for DMF patients and 297 (243) for OCR patients. DMF patients experienced lower annualized rates of overall infection encounters vs OCR patients (MD -0.51 [95% confidence interval (CI): -0.92 to -0.11], p = 0.01). When stratified by type of infection encounter, DMF patients experienced significantly lower annualized rates of outpatient (MD [95% CI]: -0.44 [-0.80 to -0.08], p = 0.02) and inpatient/hospitalization infection encounters (-0.08 [-0.14 to -0.02], p<0.01) vs OCR patients. A trend towards a shorter duration of infection-related hospitalization in the DMF vs the OCR group was observed (MD [95% CI]: -2.20 [-4.73 to 0.26] days, p = 0.08). The most common infection types in both DMT groups were urinary tract infections, sepsis, and pneumonia. DMF patients experienced lower annualized infection-related HCCs (MD [95% CI]: -$3642 [-$6380 to -$904], p < 0.01) vs OCR patients, which were driven largely by infection-related hospitalization costs (-$3639 [-$6019 to -$1259], p < 0.01).

CONCLUSION

DMF-treated patients PS-matched with OCR patients experienced lower annualized rates of infection encounters and lower infection-related HCCs.