Despite availability of high-efficacy disease-modifying therapies (hDMTs), there is no cure for multiple sclerosis (MS). Productivity loss associated with MS can be substantial, yet whether a specific hDMT performs better than another in reducing the resultant indirect costs is not well understood.
To conduct a systematic literature review evaluating the impact of natalizumab and other hDMTs on productivity loss among patients with MS.
Embase, MEDLINE, Cochrane, EconLit, and NHS EED were searched (Jan 1, 2005 to Dec 4, 2019) using key terms; additional materials were identified from grey literature. Data from studies meeting all pre-defined inclusion and no exclusion criteria were collected.
Of 649 unique citations, 20 reported on productivity loss among relapsing remitting MS (RRMS) or unspecified/mixed MS patients receiving hDMTs. Work productivity loss was the most commonly reported outcome (k=14), followed by disability/activity impairment (6), sick leave (k=6), household productivity loss (k=2), and retirement (k=1). Most studies were on natalizumab (k=13), followed by fingolimod (k=5), alemtuzumab (k=4), cladribine (k=1), and other DMTs (k=2). Only two studies provided a direct comparison between natalizumab and other DMTs. A cohort study showed natalizumab led to significantly greater reduction in productivity loss than fingolimod, beta-interferons, and glatiramer acetate (k=1 cohort study). A separate cost-minimization analysis, modeling indirect costs, estimated fingolimod led to greater cost savings than natalizumab, though significance was not reported. Across non-comparative studies, natalizumab (k=7 cohort studies, 1 cost analysis, 1 single arm trial) and other DMTs (k=3 cohort studies) consistently reduced patients’ productivity loss burden.
Natalizumab was consistently reported to improve productivity. However, a paucity of evidence and heterogeneity in study design limit comparisons to other hDMTs. Future research should evaluate the impact of specific hDMTs on attenuating productivity loss using methods reflecting real-world evidence, rather than modeling techniques, and productivity loss definitions already commonly used in the literature.