Practice Points
- In people with multiple sclerosis (MS), dalfampridine (DAL) may confer modest yet clinically meaningful improvements in cognitive function and may be considered a symptomatic treatment option for cognitive dysfunction.
- Current evidence does not support a consensus regarding the efficacy of DAL in improving fatigue, owing to conflicting findings in the literature;
however, its use may be individualized and cautiously trialed in people with MS for whom first-line fatigue treatments have not worked. - DAL demonstrates a favorable safety profile at the recommended dose of 10 mg twice daily; nevertheless, special precautions are warranted in people with renal impairment or a history of seizures.
Multiple sclerosis (MS) is an inflammatory, autoimmune, and chronic central nervous system (CNS) disease that causes demyelination of axons and impairment of nerve impulse conduction.1 MS stands as the foremost contributor to nontraumatic neurological disability among young adults.2,3 Cognitive functions, including attention, executive function, memory, and processing speed, are frequently compromised, with a prevalence of 34% to 65%.4-6 Fatigue is often observed as a co-occurring condition and has a prevalence of 59.1%.7 Fatigue in MS is a multidimensional concept that encompasses subjective fatigue and fatigability. Cognitive fatigability is characterized by a decline in cognitive function with sustained effort, whereas subjective cognitive fatigability mirrors one’s perception of mental exhaustion. Although these concepts can be related, they are distinct, and emerging evidence implies that fatigability may be associated with disability and functional outcomes.8,9 In addition, perceived mental and physical fatigability has been demonstrated to be higher in people with MS and is strongly associated with disease severity.10
Standard-of-care MS treatments aim to control disease activity, but effective therapies for cognitive impairment are a subject of debate.11 Cognitive rehabilitation has shown promise in improving cognitive abilities among people with MS. However, these improvements are often short-lived and may not translate into better everyday functioning, and efforts to replicate these findings have yielded mixed results.12,13 Therapeutic approaches targeting specific cognitive impairments often involve medications such as psychostimulants and acetylcholinesterase inhibitors, although their efficacy has been variably reported across studies.14-17 Rehabilitation strategies aimed at alleviating fatigue, such as exercise therapy and cognitive behavioral therapy, have shown limited to moderate short-term effectiveness in controlled clinical trials.18,19 Although disease-modifying treatments (DMTs) have demonstrated significant reductions in relapse rates and MRI-indicated inflammation in MS, their impact on cognitive function remains an area requiring further exploration. Despite the substantial impact of cognitive impairment on people with MS, there is currently a lack of consistent evidence regarding the effectiveness of symptomatic therapies and DMTs in improving cognitive function.20
Dalfampridine (DAL; also known as 4-aminopyridine or prolonged-release fampridine) is a potassium channel blocker that enhances signal transmission in demyelinated axons, thereby improving nerve impulse conduction.21 This medication was US Food and Drug Administration–approved as a symptomatic treatment for walking impairment associated with MS in both the United States and the European Union because it significantly improved walking speed by 25.2% compared with 4.7% in the placebo group.22 Given DAL’s presumed mechanism of action, which targets exposed potassium channels along demyelinated axons throughout the CNS, the compound may exert positive effects on neural functions beyond motor control, including cognitive function and fatigue. This systematic review seeks to evaluate the efficacy of DAL in enhancing cognitive function and fatigue-related outcomes, as well as its safety profile in people with MS. Although previous studies have primarily examined these outcomes as secondary to improvements in walking speed, this paper aims to investigate the efficacy and safety of DAL in addressing these outcomes as primary end points.
Methods
Study Protocol and Registration
This systematic review and meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and is registered in PROSPERO (CRD42024530710).
Search Strategy and Eligibility Criteria
A literature search for all available articles was performed in MEDLINE, ScienceDirect, Scopus, Embase, Cochrane Central Register of Controlled Trials, and Web of Science from database inception to April 1, 2024. The search was conducted using Medical Subject Headings terms and keywords that included terms pertaining to dalfampridine, multiple sclerosis, cognition, and fatigue. Randomized controlled trials (RCTs) were included only if they fulfilled the predefined eligibility criteria: (1) they had to study adults (≥ 18 years) with a clinical diagnosis of MS who were treated with DAL and include a comparator group of patients who received placebo (control group), and (2) they had to assess cognition or fatigue. Two authors (M.A., H.A.) independently screened the titles and abstracts; full manuscripts of potential studies were thoroughly reviewed to further assess their suitability for inclusion. Disagreements regarding eligibility were resolved through discussion.
Data Collection and Main Outcomes
Extraction of data from each study was done using a data collection sheet that included study country, inclusion and exclusion criteria, baseline demographics (age and sex), dosages of DAL, duration of intervention, and scores used to measure outcomes. Further data related to MS were gathered, including the types of MS, diagnostic criteria, baseline Expanded Disability Status Scale (EDSS) scores, and DMTs used. In addition, data pertaining to the outcomes were extracted to assess the influence of DAL on fatigue and cognition. Included studies reported data on subjective fatigue using the Modified Fatigue Impact Scale (MFIS), Würzburg Fatigue Inventory for Multiple Sclerosis, Brief Fatigue Inventory (BFI), and Fatigue Severity Scale (FSS). Studies that assessed cognition utilized different measures, including the Paced Auditory Serial Addition Test (PASAT), administered at 3-second (PASAT-3) or 2-second (PASAT-2) digit presentation intervals; the Symbol Digit Modalities Test (SDMT); and Word List Generation (WLG), a semantic verbal fluency task that is part of the Brief Repeatable Battery of Neuropsychological Tests, used to evaluate cognitive function in people with MS.
Risk of Bias Assessment
The risk of bias for each RCT was independently assessed by 2 authors (M.H.H., R.A.) using the RoB 2 assessment tool (Cochrane Collaboration). The tool evaluated the risk of bias in the domains of randomization and concealment of allocation, blinding of participants and outcome assessors, missing outcome data, selective reporting, and other potential sources, resulting in a classification of low risk, some concerns, or high risk of bias.23 Discrepancies were resolved by consensus among the authors.
Data Analysis
RCTs were required to report their findings in a manner that facilitated meta-analytic synthesis. Specifically, they needed to provide appropriate data, such as posttreatment means and SDs, for inclusion in the meta-analysis. Outcomes were pooled using random-effects models. Continuous variables and their effect estimates were calculated using either standardized mean differences (SMD) or mean differences (MD), depending on the similarity of the scores reported. Dichotomous outcomes were reported as risk ratios (RR). Heterogeneity among studies was assessed using the I² statistic. If I² exceeded 50%, indicating substantial heterogeneity, a sensitivity analysis was conducted to explore the potential impact of this variability on the overall findings. Additionally, subgroup analyses were performed to evaluate whether the treatment effect varied across different time points. RevMan software (Cochrane Collaboration) was used to perform all statistical analyses. A P value of less than .05 was considered statistically significant. Only studies reporting quantitative data, such as mean and SD or outcome effect that allow for meta-analytic synthesis, were included in the meta-analysis; trials not reporting extractable data were qualitatively synthesized only.
Results
Study Selection
Across the 5 databases and 1 register, 1329 studies were identified. After 773 duplicate records were removed, the remaining 556 were screened by title and abstract, followed by a full-text screening. Subsequently, 543 records were excluded, and 13 were retrieved by the authors and assessed for eligibility. All were included in this systematic review and meta-analysis.24-36 The full details of the study selection can be seen in Figure 1, and the detailed search strategy is in Table S1 (all supplemental materials are available in a PDF at the end of the online article).
Study Characteristics
Table 1 contains the study characteristics. The RCTs included a total of 827 people with MS. Their mean age ranged from 39.3 to 53.5 years. All 13 studies were RCTs, with 10 being double-blinded,24-28,30-33,36 1 being single-blinded,35 1 being a triple-blinded study in which participants, providers, and assessors were blinded,34 and 1 with 2 distinct phases, an open-label phase and a double-blinded phase.29 Nine studies focused on the relapsing-remitting, primary progressive, and secondary progressive subtypes of MS,25,26,28-30,32-34,36 while 1 examined the primary progressive and secondary progressive subtypes,24 and 2 focused on the relapsing-remitting subtype.31,35 One study included people with a single MS attack,36 and another did not specify disease subtype.27 Baseline EDSS ranged from 2.0 to 5.8 in the DAL groups and from 2.0 to 5.5 in the placebo groups. However, 3 studies did not differentiate between the DAL and placebo groups in their baseline EDSS reporting,24,29,34 and 2 studies did not report EDSS altogether.26,32 Further details on the interventions in each study are provided in Table 2, while the inclusion and exclusion criteria are outlined in Table S2.
Risk of Bias
The risk of bias assessment results are presented in Figures 2 and 3. Four papers were identified as high risk24,33,35,36 due to either a lack of information on whether an intention-to-treat analysis or a per-protocol analysis method was used, which resulted in a substantial percentage of missing data. Other high-risk features were a single-blinded study design (Rocca et al35) and no prespecified analysis plan before unblinded outcome data were available (Rossini et al24). Four papers were judged as having some concerns of bias.27,29,31,32 For Marion et al, these were due to missing more than 5% of participant data without sufficient analysis methods to correct for such a bias.32 The remaining 3 papers were concerning due to the absence of information on whether a prespecified analysis plan was established before unblinded outcome data became available.27,29,31 Five papers had a low risk of bias.25,26,28,30,34 A summary of the risk of bias assessment can be found in Table 3, while detailed justifications are provided in Table S3.
Cognition
Nine of the included studies assessed cognition.24-31,33 Of those, only 4 could be selected for meta-analysis.27,28,30,31 Pooled analysis demonstrated an overall significant difference favoring DAL over placebo in improving cognitive measures (SMD = 0.21; 95% CI, 0.04-0.38; P = .01; I2 = 6%). Subgroup analysis of specific cognitive measures failed to demonstrate a significant difference in SDMT (SMD = –0.05; 95% CI, –0.64 to 0.55; P = .88; I2 = 65%), PASAT-3 (SMD = 0.27; 95% CI, –0.02 to 0.56; P = .07; I2 = 0%), PASAT-2 (SMD = 0.18; 95% CI, –0.30 to 0.65; P = .46; I2 = 26%), WLG (SMD = 0.23; 95% CI, –0.12 to 0.58; P = .19; I2 = 0%) (Figure S1). Rossini et al reported that although a cognitive improvement was noticed from early evaluation to final testing, it was due to a learning effect and that the placebo vs DAL effect was not significant.24 Goodman et al reported that the PASAT-3 scores did not show a significant difference between the placebo and the DAL treatment groups (P = .46).25 Another study by Goodman et al similarly reported no significant difference between the treatment groups with different DAL doses and the placebo group.26 Satchidanand et al corroborated these findings by also reporting no significant difference between DAL and placebo in SDMT (P = .395) and PASAT (P = .441) scores.33 Broicher et al observed significant differences favoring DAL across multiple cognitive measures. During the open-label phase of the study, improvements were noted in tonic alertness (P = .0244), phasic alertness (P = .0428), psychomotor speed (P = .0140), verbal fluency (P = .0002), and cognitive fatigue (P = .0003).29 These findings were further validated in the double-blind RCT phase.29
Fatigue
Nine of the included studies assessed fatigue,24,25,29,30,32-36 but only 2 were eligible for inclusion in the meta-analysis.35,36 Pooled analysis failed to demonstrate an overall significant difference favoring DAL over placebo (MD = 3.70; 95% CI, –4.37 to 11.77; P = .37;I2 = 19%) (Figure S2). Rossini et al reported a significant reduction in fatigue compared with baseline following DAL administration (P < .001); however, the effect did not differ significantly from that of the placebo (P = .19).24 Goodman et al similarly reported a decrease in reported fatigability on the modified BFI for both the placebo and DAL groups without a significant difference (P = .13), which also correlated with improvement in fatigability as per the MFIS scale.25 In the open-label phase of the study, Broicher et al reported that a significant difference in fatigue could be found (P = .0126), and these results were corroborated in the double-blind RCT phase of the study, which also resulted in a significant difference in fatigue (P = .0005).29 De Giglio et al similarly reported an improvement in MFIS fatigue scores with a mean improvement of –7.84 (95% CI, –11.7 to –3.9) in the DAL group compared with –0.2 (95% CI, –4.6 to 4.9) in the placebo group (P = .0085; d = –0.47), but the difference between the physical and psychological subscales did not reach statistical significance.30 Marion et al reported a nonsignificant difference between DAL and placebo at both 4 and 8 weeks of treatment (P = .167; P = .582, respectively).32 Satchidanand et al similarly reported no significant difference in treatment end points between DAL and placebo (P = .900).33 Finally, Valet et al also reported that DAL did not lead to significant improvements in fatigue compared with placebo.34
Adverse Events
Adverse events (AEs) caused by DAL use were reported in 8 of the 13 included studies.25,26,30-33,35,36 Pooled analysis of the AEs did not demonstrate a significant difference between DAL and placebo in terms of causing AEs (RR = 1.11; 95% CI, 0.99-1.25; P = .08; I2 = 0%) (Figure S3). The details of the reported AEs are presented in Table 4.
Discussion
This systematic review and meta-analysis demonstrates a significant improvement in cognition associated with DAL compared with placebo in people with MS. No significant differences were observed in terms of fatigue or AEs. Our findings on cognition contradict those of an earlier systematic review and meta-analysis by Valet et al, which did not demonstrate a significant difference between DAL and placebo in improving higher-order cognitive functions.37 A different systematic review and meta-analysis by Zhang et al showed that DAL significantly improved performance on the SDMT aspect of cognition.38 Notably, the methodology in Zhang et al differs from ours, as they analyzed their results based on changes from baseline.38
Demyelinated nerve fibers expose more potassium channels, leading to elevated potassium currents that delay or inhibit action potential propagation by interfering with hyperpolarization and impairing the transmission of generated impulses.39-42 By blocking voltage-gated potassium channels, DAL delays repolarization, enhances neurotransmitter release, increases action potential duration, and subsequently improves nerve conductivity.39,40,43 Although DAL’s ability to improve axonal conduction could hypothetically enhance cognitive functioning, evidence regarding its effect on fatigue remains conflicting and inconsistent across the literature. Guidelines recommend DAL as a complementary treatment to DMTs to improve walking impairment in people with MS.44 However, no definitive recommendations have been published regarding its use in the symptomatic treatment of cognitive impairment or fatigue.
The overall effect of DAL appeared to favor improvements in cognitive function across various domains: There was a general tendency favoring DAL in most measures, even though there were no significant differences observed in individual cognitive scores. This suggests that the improvement in cognition may have been modest across multiple domains, contributing to an overall effect. Still, the small observed differences should not be dismissed. They suggest that DAL could be considered as a symptomatic treatment for cognitive impairment, and the literature supports its efficacy across several studies.29,30,45-49
DAL showed no significant difference from placebo in reducing fatigue, consistent with findings from several studies.24,25,28,35,48 However, other studies have suggested a potential benefit of DAL in alleviating fatigue.29,30,47,50,51 The lack of significant effect of DAL on fatigue may be explained by conceptual and methodological factors. First, significant heterogeneity observed in the fatigue outcome may be due to different fatigue assessment tools, each capturing distinct and different angles of fatigue and limiting comparability. Second, most included studies evaluated subjective fatigue rather than fatigability, a related yet distinct concept that might demonstrate functional impairment more efficiently.8 Data suggest that people with MS experience significant physical and mental fatigability and that both are correlated with disease disability.10 Notably, cognitive fatigability has been found to be a distinct and quantifiable aspect of MS, underscoring the need to differentiate it from subjective fatigue in clinical studies.52 The quantitative and qualitative discrepancies revealed in our review suggest that DAL may be effective in certain individuals, likely due to variations in demographics and comorbidities. Interestingly, although amantadine is widely recognized as a first-line treatment for fatigue in people with MS,53-55 Sadeqi et al reported that DAL and amantadine demonstrate comparable efficacy in managing fatigue.56 Therefore, clinicians should carefully evaluate the potential role of DAL in managing MS-related fatigue.
DAL is generally well tolerated at the recommended dose of 10 mg twice daily.57 Our analysis did not reveal a substantial increase in the risk of AEs in people with MS receiving DAL, suggesting a favorable safety profile. The most commonly reported AEs in our study were nausea, insomnia, and paresthesia. Other AEs reported in the literature include urinary tract infections and falls.57,58 One notable concern is DAL’s association with an increased risk of seizures due to its potential to lower the seizure threshold; however, this risk is dose-dependent, primarily occurring at doses exceeding the recommended level.59-62 Another rare but serious AE is QT interval prolongation, which can predispose patients to torsades de pointes.63,64 DAL is contraindicated in individuals with a history of seizures or moderate to severe renal impairment (creatinine clearance ≤ 50 mL/min) and should be used cautiously in people with mild renal impairment and avoided in pregnant people, as, prior to 2015, it was classified as pregnancy category C.42,58 Despite these considerations, DAL remains a relatively safe therapeutic option for people with MS, provided careful monitoring is in place for these specific populations.
This systematic review and meta-analysis has several limitations. First, some studies exhibited a high risk of bias,24,33,35,36 while others had a suspected risk of bias,27,29,31,32 which may affect the validity and robustness of the conclusions drawn. Second, there was variability in the diagnostic criteria used across the included studies, which may have led to inconsistencies in inclusion. Third, most of the included studies failed to report outcomes in a manner that allowed for meta-analytic synthesis, particularly for fatigue, thereby limiting our ability to incorporate these findings into the analysis. Fourth, the sample sizes of the included studies were small, potentially limiting the generalizability of the findings and necessitating cautious interpretation of the results. Lastly, discrepancies in the scoring systems used to report outcomes hindered a comprehensive assessment of all aspects of the outcomes.
Further high-quality studies with larger sample sizes, utilizing the latest validated diagnostic criteria and comprehensively assessing all aspects of cognition and fatigue, are essential to validate our findings and strengthen the evidence supporting DAL’s role in the symptomatic management of MS.
Conclusions
This systematic review and meta-analysis demonstrated that DAL significantly improves cognitive function in people with MS. These findings highlight its potential role in the symptomatic management of cognitive impairment and support its integration into MS treatment strategies.
