(57) Survival Analysis of Riluzole for Patients with Motor Neuron Disease in Hong Kong: A Population-Based Cohort Study

Publication Number: 731

Background: Motor neuron diseases (MND) are a group of neurodegenerative disorders that affect the motor neurons in the brain and spine. Riluzole is a supportive therapy for amyotrophic lateral sclerosis (ALS), a subtype of MND. Some evidence suggests that it can prolong patients' survival time. However, a few articles cannot detect a clinical or statistical significance on its effectiveness.

Objectives: This study aims to compare the survival between riluzole users and non-users with MND.

Methods: This population-based cohort study includes incident MND patients from 1993 to 2018 in Hong Kong. The data was collected from a clinical electronic database, the Clinical Data Analysis and Report System, covering all Hong Kong public hospitals. The observation period was from the onset of MND to their death date, study end date (December 31st, 2018) or first use of edaravone, whichever comes first. We censored the observation period when the patient is exposed to edavarone because, as the second approved drug for MND, its inclusion may affect the association. A multivariable Cox regression model was conducted for estimating hazard ratios (HRs) for all-cause mortality among riluzole users and non-users. The covariates included for adjustment were the Charlson Comorbidity Index, sex, disease history of hypertension, Parkinson's disease, and depression were the covariates included for adjustment. A time-depending exposure model was adapted to reduce the immortal time bias.

Results: In the study period, 1 939 incident cases with MND diagnosis were identified. 82.3% of the participants were diagnosed with ALS. About 10.5% of patients (n=204) received at least one prescription of riluzole. The age of disease onset was 60 years old [Standard deviation (SD): 15]. The Charlson Comorbidity score between riluzole users and non-users are 1.77 and 2.17 (p=0.001), with median survival time from onset to death of 1.43 (SD: 1.94) and 0.84 (SD: 3.12) years. From the time-varying Cox regression, the HR for riluzole users was 1.27 (95% confidence interval: 0.99-1.64; p-value: 0.06).

Conclusions: Riluzole users had a longer median survival than non-users by a few months. However, no statistically significant difference was found. Further prospective studies with a larger sample size are needed to inform the risk and benefit of riluzole use.