Background: An oral, fixed-dose sodium phenylbutyrate-ursodoxicoltaurine (PB-TURSO) coformulation was evaluated in a multicenter ALS trial (CENTAUR). Methods: Adults with definite ALS, ≤18 months from symptom onset, (N=137) were randomized 2:1 to PB-TURSO or placebo for 6 months. Completing participants were eligible to receive PB-TURSO in the open-label extension (OLE) (≤30 months). The primary efficacy endpoint in both periods was rate of ALS Functional Rating Scale–Revised (ALSFRS-R) total score decline. All-cause survival was analyzed July 2020 (longest follow-up, 35 months). Safety was assessed in both periods. Results: Over 6-month randomized treatment, mean ALSFRS-R total score decline was slower with PB-TURSO vs placebo (difference, 0.42 points/month; P=0.03). Participants receiving PB-TURSO in the OLE (continued or crossover from placebo) maintained or initiated functional benefit beyond 6 months of therapy. Mean hazard of death was 44% lower (P=0.02) in the original PB-TURSO group. Overall adverse event (AE) incidence was similar, though early (week ≤3) gastrointestinal AEs were more frequent during initial exposure to PB-TURSO (randomized period or OLE). Conclusions: PB-TURSO resulted in superior retention of function in the randomized period. Long-term OLE results support functional benefits of early vs delayed therapy and of sustained treatment. Survival was longer in the original PB-TURSO group after nearly 3 years.