Real-world evidence of riluzole effectiveness in treating amyotrophic lateral sclerosis

Abstract

Objective

To compare the effect of riluzole on median survival in population studies of patients with amyotrophic lateral sclerosis (ALS) with that observed in clinical trials. Methods: Two independent PubMed searches were conducted, to identify population studies that reported median survival for ALS patients who were either treated with riluzole or remained riluzole-free. Results: We identified 14 studies that met the inclusion criteria of reporting median survival and an additional study that reported mean survival of both riluzole and riluzole-free patients. Analysis of the 15 studies found that a majority reported increased survival of riluzole vs. riluzole-free patients. In 8 studies, the median survival for patients treated with riluzole was 6–19 months longer compared with patients not treated with riluzole (p < 0.05). Three additional studies reported a clinically meaningful treatment effect (range 3–5.9 months) but did not meet statistical significance. The remaining 4 studies did not show a meaningful treatment effect between riluzole and riluzole-free groups (<3 months), and differences among the groups were not significant. Also, 5 of the studies used multivariate regression analysis to investigate the level of association between treatment with riluzole and survival; these analyses supported the positive effect of riluzole on survival. Conclusions: A majority of population studies that compared riluzole vs. riluzole-free ALS patients found significant differences in median survival between the two groups, ranging from 6 to 19 months. This is substantially longer than the 2- to 3-month survival benefit observed in the pivotal clinical trials of riluzole.

Keywords:

• Riluzole
• amyotrophic lateral sclerosis
• real-world
• neurodegenerative
• survival

Introduction

Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disorder that causes progressive paralysis and has no known cure. Riluzole, a glutamate antagonist approved by the Food and Drug Administration in 1995, is the only drug shown in clinical trials to increase survival among patients with ALS (1,2). Although riluzole is thought to protect motor neurons in the brain and spinal cord from glutamate damage by inhibiting presynaptic glutamate release (3) and enhancing glial and neuronal glutamate uptake (4,5)—thereby reducing extracellular glutamate concentrations and blocking persistent sodium currents (6,7)—the exact mechanism in patients with ALS is unknown (8,9).

It is well established that randomized clinical trials (RCTs) are the gold standard in clinical medicine and are conducted to evaluate the efficacy of treatment. They play a vital role in understanding the benefit and risk of treatment and are intended to answer the question, “Can a treatment work?” However, randomized clinical trials often have stringent inclusion/exclusion criteria. Consequently, one limitation is their generalizability to the broader patient population (10).

Compared with RCTs, population studies have less stringent inclusion/exclusion criteria and typically include a broader patient population. The study of patients in the general clinical population evaluates the effectiveness of a treatment to answer the question, “Does a treatment work?” Population studies also have limitations, one of which is the lack of randomization. Consequently, differences in survival and other outcomes may be a result of demographic or socioeconomic differences rather than of the treament itself. If appropriately designed, however, real-world evidence studies may provide value by including a wide range of patients representative of the general clinical patient population (11,12).

Two double-blind, placebo-controlled, RCTs led to the approval of riluzole 50 mg twice daily (bid) for the treatment of patients with ALS: a fixed-dose (100 mg/day) trial (n = 155) and a dose-ranging (50, 100, and 200 mg/day) trial (n = 959; 13,14). Although both found significant slowing in time to tracheostomy or death, riluzole’s clinical benefit in these trials was modest (2- to 3-month increase in median survival). A meta-analysis of patients receiving 100 mg riluzole (50 mg bid) in these 2 trials (n = 633) showed that riluzole significantly improved tracheostomy-free survival compared with placebo; median survival was 15.5 vs. 13.2 months (13–15). Examination of patient populations reveals that they tended to have a long-standing disease, which may be treatment-resistant, biasing these studies. The mean time from diagnosis to treatment was 2.2 years in the fixed-dose trial (13) and 1.8 years in the dose-ranging trial (14). This begs the question, “What is the clinical utility of these results given the fact that in clinical practice, patients are often prescribed riluzole closer to diagnosis, approximately 9–14 months after symptom onset?” (16,17).

We undertook the present comprehensive literature search to determine whether the net increase in median survival associated with riluzole in population studies of patients with ALS differs from that demonstrated in placebo-controlled RCTs. The 2 objectives of this study were to: (1) identify all real-world studies that include both riluzole and a riluzole-free treatment group and provide median survival data (months) in order to calculate the net change in median survival for these two groups, and (2) perform a critical review of the clinical study designs of the 2 pivotal trials to understand limitations that may have affected trial outcomes, including endpoints or patient inclusion/exclusion criteria.

Materials and methods

In June 2019, 2 independent PubMed searches were conducted to identify peer-reviewed, retrospective, and prospective studies that investigated the treatment effect of riluzole in patients with ALS in real-world settings. For a study to be included, results had to report the net change in median survival for patients treated with riluzole compared with a riluzole-free control group. This predetermined outcome measure aligns with the primary outcome measure of riluzole pivotal RCTs. Single-arm studies were not included because treatment-effect could not be determined in the absence of a comparator group.

Search 1 used the following terms: “riluzole” AND “amyotrophic lateral sclerosis” AND [“population” OR “retrospective” OR “prospective”] and identified 120 articles. These articles were reviewed to determine if they met prespecified criteria for inclusion. The second independent search (Search 2) took a broader approach and used only the term “riluzole” and delivered 1582 articles with abstracts, which were reviewed to determine if they met prespecified criteria. Results from Search 1 and Search 2 were cross-compared. Reference lists of identified studies were also examined.

Results

A cross-comparison of Searches 1 and 2 and a review of the reference lists identified 14 studies that met prespecified inclusion criteria (18–31). Another study that reported mean survival was identified and included in this report (32).

Most of the studies reported that riluzole treatment increased median survival compared with patients who were managed without riluzole (Table 1; 18–32). Eight of the 15 studies reported a statistically significant increase in median survival ranging from 6 to 19 months (p < 0.05; 18–20,22,25,26,30,32). Three studies reported a clinically meaningful treatment effect (net change in median survival ≥3 months), but comparisons were not statistically significant (21,24,28). Four studies did not show a meaningful treatment effect among riluzole and riluzole-free groups (<3 months; 23,27,29,31). While no significant difference in overall survival between riluzole and riluzole-free groups was detected by Traynor et al. (21,33) using the logrank test (p = 0.32), early portions of survival curves were statistically different using the Peto logrank test (p = 0.015; 21,33).

Table 1 Median survival data from population studies—riluzole vs. riluzole-free.

Investigator, year, Country	Methodology	Inclusion (I)/exclusion (E)	Number of patients	Definition of survival	Median survival (months)	Net change in median survival (p value)
Meinenger, 2000, France (18)	Retrospective historical database 1989–1991 (riluzole-free) 1995–1997 (riluzole treated)	I: Patients deceased during each time period	Riluzole (n = 356) Riluzole-free (n = 161)	Time from diagnosis to death	18.4 12.4	6 months (p < 0.001)
Brooks, 2001, USA (19)	Retrospective ALS onset <1996 and ALS onset ≥1996	I: ALS CARE–Wisconsin Database	Onset <1996 Riluzole (n = 51) Riluzole-free (n = 241)	Time from symptom onset to death	58.4 47.7	10.7 months (p = 0.13)
			Onset > 1996 Riluzole (n = 112) Riluzole-free (n = 65)		>67 49.1	>17.9 months p = 0.044)
Turner, 2002, UK (20)	Retrospective Database from a specialist MND clinic in a tertiary referral center 1990–2000	I: ALS diagnosis by a consultant neurologist E: All patients given a diagnosis other than ALS at presentation, or shown later to not have ALS Patients for whom survival data was unavailable	Any El Escorial ALS category Riluzole (n = 299) Riluzole-free (n = 349)	Time from symptom onset to death	51 32	19 months (p < 0.0001)
			El Escorial “suspected” or “possible” at diagnosis Riluzole (n = 98) Riluzole-free (n = 128)		66 49	17 months (p < 0.035)
Traynor, 2003, Ireland (21)	Retrospective Complete registry of all patients diagnosed with ALS in the Republic of Ireland 1996–2001	I: Irish residents attending a general neurology clinic and diagnosed as suspected, possible, probable, or definite ALS E: Patients attending the sole multi-disciplinary ALS clinic in Ireland Patients for whom data concerning riluzole prescription were missing	Riluzole (n = 149) Riluzole-free (n = 97)	Time from diagnosis to death or tracheostomy*	14.3 10.1	4.2 months (p = 0.32, analysis of overall survival curve)
Mitchell, 2006, UK (22)	Retrospective Patient records from the Preston MND Care and Research Center 1980–2003	I: Patients diagnosed as probable or definite ALS according to the original EEC with the addition of clinically probable laboratory supported ALS under the revised Airlie House criteria Patients for whom details of dates of first symptom and death/censoring were available	Riluzole (n = 148) Riluzole-free (n = 327)	Time from symptom onset to death or date of censoring	36.8 27	9.8 months p < 0.05
Sívori, 2007, Argentina (23)	Prospective Patient records from the Hospital J.M. Ramos Mejia División Neurologia 1999–2004	I: Patients with a diagnosis of sporadic ALS based on the EEC	Riluzole (n = 26) Riluzole + NIV (n = 18) NIV† only (n = 11) No Riluzole or NIV (n = 42)	Time from diagnosis to death	11.2 16.6 13.5 10.7	Riluzole only vs No Riluzole or NIV 0.5 months (p = 0.798)
Zoccolella, 2007, Italy (24)	Prospective National health system registry death certificates from National Death Data Base Registry 1998–1999	I: Diagnosis of ALS based on EEC and the Airlie House revised version of 1988	Riluzole (n = 73) Riluzole-free (n = 53)	Time from diagnosis to death	18.3 12.4	5.9 months (p = not significant)
Georgoupoulou, 2013, Italy (25)	Prospective Patient records from MND center 2000–2009	I: Definite or probable ALS	Riluzole (n = 133) Riluzole-free (n = 60)	Time from symptom onset to death	43 31	12 months (p < 0.01)
			Riluzole (n = 133) Riluzole-free (n = 60)	Time from symptom onset to death or tracheostomy	38 31	7 months (p = 0.11)
Rooney, 2013, Ireland (26)	Retrospective Irish ALS registry 1995–2010	I: Patients with ALS diagnosis logged in Irish ALS registry with riluzole prescription information taken from Health Services Executive (HSE) of Ireland	Riluzole (n = 857) Riluzole-free (n = 410)	Time from diagnosis to death	17.5 10.2	7.3 months (p < 0.001)
Cetin, 2015, Austria (27)	Retrospective Registries with invoice data of the 9 regional sickness funds of Austria (GKK) 2008–2012	I: Patients discharged from a hospital with main diagnosis of ALS or who were prescribed riluzole during the study period E: Age <20-years-old	Riluzole (n = 528) Riluzole-free (n = 358)	Time from diagnosis to death	22.2 27.6	‒5.4 months (p = 0.61)
Chen, 2016, China (28)	Prospective Clinic-based cohort study 2007–2013	I: ALS diagnosis according to the Airlie House diagnostic criteria	Riluzole (n = 415) Riluzole-free (n = 1125)	Time from diagnosis to death or tracheostomy	67 64	3 (p = 0.780)
Stevic, 2016, Serbia (32)	Retrospective Inpatient and outpatient records from all five Belgrade hospitals with neurology departments 1992–2009	I: Belgrade resident ≥5 years before ALS diagnosis Diagnosis of probable or definite ALS according to the EEC	Riluzole (n = 61) Riluzole-free (n = 264)	Time from symptom onset to death	54.0† 43.2†	10.8 (p < 0.05)
Calvo, 2017, Italy (29)	Retrospective Database included 13 ALS Italian referral centers, located in 10 Italian regions. Data also collected from Emilia Romagna Registry for ALS (ERRALS) 2009–2013	I: ALS diagnosis in the time period according to R-EEC	Riluzole (n = 2053) Riluzole-free (n = 595)	Time from symptom onset to death or tracheostomy	43 43	0 (p = 0.552)
Favero 2017, Brazil (30)	Prospective cohort study Patient records from MND clinic (Clinical investigation on Neuromuscular Diseases Section) 1999–2011	I: ALS diagnosis according to EEC	Patients who survive ≥19 months (A19): Riluzole (n = 53) Riluzole-free: (n = 34)	Time from diagnosis to death	36.6 27.4	9.2 p = 0.001
			Patients who survived <19 months (B19): Riluzole (n = 82) Riluzole-free (n = 62)		9.0 6.8	2.2 p = 0.176
Mandrioli, 2018, Italy (31)	Prospective Emilia Romagna Registry for ALS–ERRALS 2009–2014	I: ALS diagnosis according to R-EEC E: Patients who died <6 months from diagnosis	Riluzole (n = 573) Riluzole-free (n = 108)	Time from symptom onset to death or tracheostomy or date of censoring	26 26	0

EEC: El Escorial Criteria; R-EEC: Revised El Escorial Criteria; NIV: noninvasive ventilation.

*Although Traynor et al. (21) reported survival as measured from symptom onset, examination of a subsequent analysis of the same patients (33) determined that survival times were reported from diagnosis. †Mean survival.

Survival was analyzed by univariate and/or multivariate regression analyses in 5 studies (Table 2; 22,24,28,29,32,33). Using a Cox model with both forward and backward selection, Mitchell et al. (22) found that risk of death significantly decreased in riluzole-treated patients compared with those not treated (HR 0.203; 95% confidence interval [CI]: 0.085–0.482; p < 0.001; 22). Similarly, Stevic et al. (32) conducted a multivariate regression analysis and found that riluzole treatment positively impacted survival in patients with ALS (risk of death without riluzole treatment; HR 1.76; 95% CI: 1.26–2.45; p < 0.01; 32). Although Calvo et al. (29) was among the 4 studies not showing meaningful treatment effect because no net difference was observed in median survival from symptom onset to tracheostomy-free survival, multivariate Cox analysis found a significant association between riluzole treatment and improved survival (risk of death with riluzole treatment; HR 0.79; 95% CI: 0.64–0.98; p = 0.03; 29). Chen et al. (28) reported no difference in median survival and no significant impact on survival by multivariate analysis. However, they observed a significantly better prognosis for the quartile of patients who were treated with riluzole 100 mg/day for the longest period of time compared with riluzole-free patients (risk of death with riluzole treatment; HR 0.488; 95% CI: 0.320–0.746; p = 0.001), indicating that the long-term use of riluzole could improve patient survival (28). While observing a nonsignificant increase in net survival, Zoccollela et al. (24) used a Cox proportional model to show that treatment with riluzole was a predictor of increased survival at 12 months from diagnosis. Significance was marginal (p = 0.06), after adjustment for age, gender, site of symptom onset, and onset-diagnosis interval (risk of death with riluzole treatment; HR 0.51; 95% CI: 0.25–1.03; p = 0.06; 24).

Table 2 Survival by univariate/multivariate analyses—riluzole vs. riluzole-free.

Investigator, year, Country	Number of patients	Data
Mitchell, 2006, UK (22)	Riluzole (n = 148) Riluzole-free (n = 327)	Multivariate Cox analysis HR (95% CI) No = 1 0.20 (0.089–0.46) p < 0.001
Zoccolella, 2007, Italy (24)	Riluzole (n = 73) Riluzole-free (n = 53)	Cox proportional model HR (95% CI) No = 1 0.51 (0.25–1.03) p = 0.06
Calvo, 2017, Italy (29)	Riluzole (n = 2053) Riluzole-free (n = 595)	Multivariate Cox analysis HR (95% CI) No = 1 0.79 (0.64–0.98) p = 0.030
Chen, 2016, China (28)	Riluzole (n = 415) Riluzole-free (n = 1125)	Multivariate Cox analysis HR (95% CI) No = 1 0.86 (0.685–1.068) p = 0.167
Stevic, 2016, Serbia (32)	Riluzole (n = 61) Riluzole-free (n = 264)	Cox regression analysis HR (95 % CI) Yes = 1 1.76 (1.26–2.45) p < 0.01

HR: hazard ratio.

Discussion

This analysis indicates that the actual survival benefit of riluzole may be substantially greater than the 2–3 months indicated in the pivotal RCTs. Most of the studies (8 of 15) included in this analysis found a statistically significant net increase in median survival with 6-to 19-month riluzole treatment (18–20,22,25,26,30,32). Three additional studies showed that riluzole extended survival by 3–6 months over riluzole-free management (21,24,28,33). Univariate and multivariate analyses confirmed the strong association of riluzole with decreased risk of death (22,29,32) and supported early initiation (24) and long-term management (28). Two additional studies identified during our search that did not report median survival used univariate and/or multivariate analyses and reported that riluzole had a significant impact on survival (34,35).

Examining the design of the pivotal RCTs reveals possible reasons why the impact of riluzole on survival might be different in these trials than in real-world use. The RCTs may have underestimated riluzole’s benefit due to short reporting time, limiting the survival data available. Survival was measured from the time of enrollment to tracheostomy, death, or the end of the placebo-controlled period (18 and 21 months for the dose-ranging and fixed-dose trials, respectively; 13,14). Patients still alive and tracheostomy free at the end of the placebo-controlled phase were censored and not included in the analysis (13). In the dose-ranging trial, over half of the patients were still alive and tracheostomy free at study end (53% receiving riluzole 50 mg bid and 50.4% receiving placebo; 14). Consequently, it is unknown how much longer riluzole-treated patients would have survived or been tracheostomy-free since they were censored (13,14). Results from the general ALS population do not reflect those of patients in the pivotal RCTs, since population studies can follow patients throughout the course of the disease. Riluzole may have different mechanisms of action (MOAs) at different stages of disease (36). At early stages, riluzole appears to reduce neuronal hyperexcitability. In animal studies, it suppresses persistent Na⁺ current and repetitive firing; enhances calcium-dependent K⁺ current and slows firing frequency; and reduces voltage-dependent Ca⁺² current, causing presynaptic reduction of transmitter release (37). Preclinical studies identified MOAs also supporting the long-term use of riluzole. Intracellular aggregates of misfolded and hyperphosphorylated transactive response DNA binding protein 43 kDa (TDP-43) are a major component of the ubiquitin-positive inclusions detected in the CNS of patients with ALS (38,39). Protein kinase CK1 isoform δ (CK1δ) has been identified as the major enzyme that hyperphosphorylates TDP-43, allowing its mislocalization to the cytoplasm. In turn, a decrease in nuclear TDP-43 results in aberrant mRNA maturation with reduced EAAT2 expression on astrocytes, subsequent glutamate accumulation, and toxicity. Riluzole has been shown to inhibit CK1δ catalytic activity, thus preventing the aforementioned events leading to neuronal death from aberrant TDP-43 proteinopathy and glutamate excitotoxicity (39).

In light of the possible MOAs, another notable feature of the RCTs was the extended length of time from diagnosis to initiation of riluzole. Patients were eligible for the pivotal RCTs if they had clinically “probable” or “definite” ALS of no more than 5 years’ duration. At enrollment, mean disease duration was 2.2 ± 1.7 years (riluzole) and 2.3 ± 1.8 years (placebo) in the fixed-dose trial and 1.7 ± 1.2 years (riluzole 50 mg bid) and 2.3 ± 1.8 years (placebo) in the dose-ranging trial (13,14). Consequently, the trials were not designed to capture data regarding riluzole’s effect on survival during the early phases of the disease. A retrospective analysis of data from the dose-ranging RCT addressed this issue, comparing the effect of introducing riluzole therapy at different stages of ALS (40). Staging information was not reported in the original database, so it was calculated based on the King’s clinical staging system. Because the original trial included only patients with probable or definite ALS, the earliest stage of patients enrolled was Stage 2. Fang et al. (40) found that riluzole was associated with longer time spent in Stage 4 but not with a longer time from Stages 2 or 3 to subsequent stages or death. However, no data were available for Stage 1 due to eligibility criteria (40). Lack of Stage 1 data may have masked any early benefit of riluzole. In support of early initiation of therapy, several population studies found that the greatest benefit of riluzole occurs early in the course of disease, with little to no benefit when riluzole is initiated more than 18–24 months after diagnosis (18,21,24).

Population studies provide evidence both for early and prolonged riluzole therapy. In a population study at a Chinese medical center that assessed the validity of the King’s College staging system and examined the association between stage and prognosis, Chen et al. (41) observed that patients with ALS who initiated long-term treatment with riluzole in Stage 1 or 2 had a significantly longer duration of Stage 2 (11.9 vs. 7.1 months, p = 0.019) and significantly greater median survival (45.3 vs. 35.1 months, p = 0.035) compared with patients who started treatment in Stage 3 or 4 (41). Thakore et al. (42), using the Fine’til 9 (FT9) and King’s staging systems on patients from the PRO-ACT dataset, observed that riluzole treatment was associated with prolonged survival at multiple stages (42). After adjusting for age and Revised ALS Functional Rating Scale (ALSFRS-R) slope at first visit, riluzole significantly reduced the risk of the following transitions: (1) King’s stages: 1: >2 (HR 0.81), and 2: >3 (HR 0.82), 4: >death (HR 0.57), and (2) FT9 stages: 1: >2 (HR 0.84), 3: >4 (HR 0.71), and 4: >death (HR 0.67). In contrast, the survival benefit of riluzole treatment in patients with bulbar onset was more pronounced in early rather than late King’s stages (42). Mandrioli et al. (31) reported that increased time on riluzole was significantly associated with improved survival (HR 0.98 for each 1% increase in days treated; p < 0.001; 31). Median survival was 46 months for patients treated with riluzole for ≥90% of disease duration compared with 15 months for patients who were treated for <90% of disease duration (31).

These observations of increased survival with early initiation and long-term treatment are consistent with clinical recommendations to initiate riluzole “as early as possible after diagnosis” to “slow disease progression in patients with ALS,” with the rationale that delaying motoneuron and, therefore, muscle degeneration at earlier stages of the disease will prolong those stages associated with the highest (best) levels of functioning (43,44). Experts recommend beginning riluzole “as soon as possible following a diagnosis of definite, probable, suspected, or possible ALS by World Federation of Neurology (WFN) criteria” (16). In addition, continued use of riluzole through the course of the disease is supported by observations that survival benefits persist in later King’s and FT9 stages (40,42).

Real-world evidence studies have limitations, the most prominent one being the lack of randomization. Which patients receive riluzole therapy in the real-world and are they the ones most likely to have prolonged survival even without drug therapy? Several of the studies we identified did in fact report that riluzole was prescribed significantly more frequently in younger patients (21,25,29), patients with NIV (25), or percutaneous endoscopic gastrostomy (PEG; 21,25), and those who had most recently accessed healthcare (22). In contrast, other studies found no significant difference between the riluzole and riluzole-free groups with respect to age (22,24) or with respect to NIV (21,24) or PEG (24) use, and in one study, those receiving riluzole were significantly older than the riluzole group (28). Several studies used multivariate analysis to address these potential biases (Table 2).

The setting for a given study may affect the generalisability of its results. Most population studies analyzed used data from patients at tertiary medical centers. However, many used data from regional (24,27,31) or national (Irish; 26) databases. Traynor et al. (21) limited study population to patients at general neurology clinics but not at multidisciplinary ALS clinics. Calvo et al. (29) included data both from tertiary referral hospitals and a population registry (29). Studies from tertiary care centers may be affected by selection bias similar to that of clinical trials. In general, patients at tertiary medical centers are more likely to be younger and have familial ALS and/or limb-onset ALS compared with the general ALS population (10,11). A study’s location can also bias results, predominantly in countries where free medical care is provided. Socioeconomic factors could also represent an important bias. In countries such as China, where ALS treatment is not covered by insurance and riluzole is expensive, only wealthier patients take riluzole (28).

Other limitations to the individual studies, as well as to our ability to draw conclusions from the aggregate results, include low internal validity, less rigorous data collection, and susceptibility to sources of bias for comparing outcomes. Survival is not defined identically in all studies. Approximately one-third of the 15 studies in this review measured survival from time of symptom onset (6 studies) and two-thirds from time of diagnosis (9 studies). Similarly, studies used different endpoints for survival, measuring either to death (11) or death/tracheostomy (4). Even “riluzole use” had more than 1 definition: typically as ≥1 dose but also as >2 weeks of riluzole use (28).

An additional limitation of our analysis was the quality of some of the studies. While this review included all population studies that compared survival in riluzole-treated vs. riluzole-free cohorts, they were not necessarily designed for this purpose; some had methodological flaws. For example, studying benefits associated with noninvasive ventilation (NIV) and riluzole, the subpopulation used for the riluzole/riluzole-free median survival comparison consisted of only 68 patients (26 riluzole, 42 riluzole-free; 23). Stevic et al. (32) reported a mean rather than median survival data. Medians are standard for clinical studies because 1 or 2 outliers can greatly skew results of mean calculations (32).

Future population studies would ideally include functional data as well because survival is not the only outcome considered when choosing treatment. Real-world data evaluating the impact of riluzole treatment on function, along the disease continuum, would be valuable to physicians and patients.

Conclusion

This review indicates that riluzole treatment may extend survival by 6–19 months, which is far greater than the 2- to 3-month survival benefit reported in original pivotal RCTs. Real-world evidence studies provide valuable information concerning characteristics among the broad patient population observed in clinical practice and physician treatment practices outside clinical trials. This information is necessary to guide treatment decisions and for reimbursement and payment decisions (43,44). This analysis of population studies of patients with ALS was conducted to determine if the observed survival benefit of riluzole exceeds what was demonstrated in pivotal RCTs.

Real-world studies provide additional insights to clinicians treating patients with ALS to better understand the benefits of early use and long-term effectiveness of riluzole. Observations from these real-world studies provide evidence to positively influence the conversation that clinicians have with their patients on the benefit of riluzole treatment.

Acknowledgments

Medical writing assistance (funded by ITF Pharma, a subsidiary of Italfarmaco S.p.A.) was provided by ECIR Medical Communications (Benjamin J Epstein, PharmD; President). ECIR developed the first draft based on an author-approved outline and assisted in implementing author revisions. Editorial assistance in formatting, proofreading, copyediting, fact-checking, and publication submission was also provided by ECIR. All authors contributed equally to the article and approved the final version for submission.

Declaration of interest

No author received payment from a third party for any aspect of the submitted work. Dr. Andrews has served as a consultant for Avexis, Inc., AL-S Pharma, and Cytokinetics, Inc.; and has received research grants from Neuraltus, Roche, Orion, Biogen, and Novartis. Dr. Jackson reports receiving personal fees from BrainStorm Cell Therapeutics, Inc., Mallinckrodt Pharmaceuticals, Anelixis Therapeutics, and Alexion Pharmaceuticals; consulting and Speakers Bureau fees from ITF Pharma (outside the submitted work), Avanir Pharmaceuticals, Inc., Strongbridge Biopharma, CSL Behring, Mitsubishi Tanabe Pharma America, Cytokinetics, Inc., Orion Corporation, and Alexion Pharmaceuticals; and research grants from Mitsubishi Tanabe Pharma America, the Muscular Dystrophy Association, Cytokinetics, Inc., and the Amyotrophic Lateral Sclerosis Association. Dr. Heiman-Patterson reports receiving personal fees as an advisory board/consultant from Mitsubishi Tanabe Pharma America, Cytokinetics, Inc., ITF Pharma (outside the submitted work), and Biohaven Pharmaceuticals and as advisory board for Alnylam Pharmaceuticals Inc. Paolo Bettica, MD, PhD is a full-time employee of Italfarmaco S.p.A., and as such, reports personal fees from Italfarmaco S.p.A. during the conduct of the study. Dr. Brooks reports receiving research support from the Muscular Dystrophy Association, RTI International, MediciNova, Mitsubishi Tanabe Pharma America, Biohaven Pharmaceuticals, Orion, Neuraltus Pharmaceuticals, Cytokinetics, Inc., Santhera Pharmaceuticals, Biogen, ITF Pharma (outside the submitted work), Acceleron Pharma, Philips Respironics, and the Centers for Disease Control and Prevention (CDC). Dr. Pioro reports receiving personal fees from Avanir Pharmaceuticals, Inc., Biohaven Pharmaceuticals. Inc., Cytokinetics, Inc., ITF Pharma (outside the submitted work), Mitsubishi Tanabe Pharma America, and Otsuka America, Inc., and research grants from Centers for Disease Control and Prevention/National Institutes of Health (CDC/NIH) and the Amyotrophic Lateral Sclerosis Association. The authors have indicated that they have no other conflicts of interest regarding the content of this article.

Additional information

Funding

This work was supported by ITF Pharma, Inc., a subsidiary of Italfarmaco S.p.A.