Amyotrophic lateral sclerosis (ALS) is a rare neurodegenerative disease. Riluzole may increase survival and delay the need for mechanical ventilation. The CAESAR project ('Comparative evaluation of the efficacy and safety of drugs used in rare neuromuscular and neurodegenerative diseases', FV AIFA project 2012-2013-2014) involves evaluating prescribing patterns, and analysing effectiveness and comparative safety of drugs, in patients with neurodegenerative diseases. The aim of this study is to evaluate adherence to riluzole in patients with ALS during the first year of use, identifying adherence clusters. A retrospective cohort study was conducted using administrative data from Latium, Tuscany, and Umbria. We identified subjects with a new diagnosis of ALS between 2014 and 2019, with the first dispensation of riluzole within 180 days of diagnosis. We considered a two-year look-back period for the characterization of patients, and we followed them from the date of first dispensing of riluzole for 1 year. We calculated 12 monthly adherence measures, through a modified version of the Medication Possession Ratio, estimating drug coverage with Defined Daily Dose. Adherence trajectories were identified using a three-step method: (1) calculation of statistical measures; (2) principal component analysis; (3) cluster analysis. Patient characteristics at baseline and during follow-up were described and compared between adherence groups identified. We included 264 ALS patients as new users of riluzole in Latium, 344 in Tuscany, and 63 in Umbria. We observed a higher frequency of males (56.2%) and a mean age of 67.4 (standard deviation, SD, 10.4) in the overall population. We identified two clusters in all regions: one more numerous, including adherent patients (60%, 74%, 88%, respectively), and another one including patients who discontinued therapy (40%, 26%, 12%, respectively). In Tuscany patients discontinuing riluzole more frequently died (28.6% vs. 15.4%, p-value <0.01). Additionally, low-adherers had a higher frequency of central nervous system disorders (69.0% vs. 52.5%, p-value 0.01), and a greater use of non-pharmacological treatments (p-values ≤0.01 for invasive ventilation and tracheostomy). We did not observe any differences in Lazio, whereas in Umbria we observed a higher use of drugs for dementia-related psychiatric problems among low-adherers (57.1% vs. 7.8%, respectively, p-value <0.01), although with small numbers. Most ALS patients who start riluzole adhere to therapy during the first year. Patients who discontinue therapy early show greater fragility and mortality.