Background: Efficacy and safety of BOS vs imatinib (IMA) in patients (pts) with newly diagnosed chronic phase CML was assessed in the phase 3 BFORE trial. Here we characterize the safety profile of BOS after 5 yrs follow-up, with a focus on GI, liver, effusion and renal treatment-emergent adverse events (TEAEs). Methods: Pts who received ≥1 dose of BOS (n=268) or IMA (n=265) 400 mg/d in BFORE were included. Adverse events (AEs) of special interest were analyzed by selecting prespecified MedDRA terms to generate TEAE clusters. Final database lock: June 12, 2020. Results: Median duration of treatment (Tx) was 55 mo for pts receiving BOS or IMA; respective median (range) dose intensity was 393.6 (39–583) vs 400.0 (189–765) mg/d. Any grade TEAEs occurred in 98.9% and 98.9% of BOS- vs IMA-treated pts. Most common newly occurring TEAEs (any grade) after 12 mos were increased lipase (9.0%) with BOS, and diarrhea (8.3%) with IMA. In BOS- vs IMA-treated pts, 25.4% vs 14.3% had AEs leading to permanent Tx discontinuation; the majority discontinued in yr 1 (14.2% vs 10.6%). Most frequent AEs leading to discontinuation were increased ALT (overall, 4.9%; yr 1, 4.5%) with BOS vs thrombocytopenia (overall, 1.5%; yr 1, 1.5%) with IMA. GI, liver, effusion and renal TEAEs, respectively, occurred in 79.9%, 44.0%, 6.0% and 10.4% (maximum grade 3/4 [G3/4]: 9.0%, 26.9%, 1.1% and 2.2%) of BOS- vs 61.5%, 15.5%, 2.3% and 9.8% (G3/4: 1.1%, 4.2%, 0.4% and 0.8%) IMA-treated pts. One grade 5 renal TEAE occurred in the BOS arm and was not considered related to Tx. Cumulative rates per Tx yr are shown in the Table. Most common GI TEAEs were diarrhea (BOS vs IMA: 75.0% vs 40.4% [G3/4: 9.0% vs 1.1%]) with BOS, and nausea (37.3% vs 42.3% [G3/4: 0% vs 0%]) with IMA. In both arms, the most common liver, effusion and renal TEAEs, respectively, were increased ALT and/or AST (34.0% vs 8.3% [G3/4: 22.0% vs 2.3%]), pleural effusion (5.2% vs 1.9% [G3/4: 0.7% vs 0.4%]) and increased blood creatinine (6.7% vs 8.3% [G3/4: 0.4% vs 0.4%]). GI, liver, effusion and renal TEAEs infrequently led to Tx discontinuation (1.9%, 7.8%, 0.7% and 0.7% vs 1.1%, 0.8%, 0% and 0.4%). Conclusions: The safety profiles of BOS and IMA in BFORE were distinct, with no new safety signals identified after 5 yrs follow-up. Onset of TEAEs occurred primarily during yr 1 (eg, GI and liver), with an increased incidence of some TEAEs (eg, effusion and renal) in later yrs. Discontinuations due to AEs generally occurred early into Tx, with few due to GI, liver, effusion and renal AEs. These safety results support the use of first-line BOS as a standard of care in pts with CP CML. Clinical trial information: NCT02130557.