Outcome by CBF-AML subtype and treatment arm. (A-B) At 36 months, RFS was estimated at 64% (95% CI, 53-73) in arm A patients as compared with 64% (95% CI, 52-73) in arm B patients (Figure 2A; HR, 0.97 [95% CI, 0.59-1.57]; P = .89 by the log-rank test). At 36 months, RFS was estimated at 68% (95% CI, 57-76) in t(8;21) AML patients as compared with 61% (95% CI, 49-70) in inv(16)/t(16;16) AML patients (Figure 2B; HR, 1.17 [95% CI, 0.72-1.91]; P = .53 by the log-rank test). (C-D) At 36 months, CIR was estimated at 32% (95% CI, 23-42) in arm A patients as compared with 32% (95% CI, 23-43) in arm B patients (Figure 2C; SHR, 0.97 [95% CI, 0.58-1.64]; P = .92 by cause-specific hazard Cox model). At 36 months, CIR was estimated at 29% (95% CI, 21-40) in t(8;21) AML patients as compared with 34% (95% CI, 25-45) in inv(16)/t(16;16) AML patients (Figure 2D; SHR, 1.10 [95% CI, 0.65-1.86]; P = .71 by cause-specific hazard Cox model). (E-F) Overall survival from CR. At 36 months, OS from CR was estimated at 87% (95% CI, 77-93) in arm A patients as compared with 83% (95% CI, 71-90) in arm B patients (Figure 2E; HR, 1.03 [95% CI, 0.46-2.29]; P = .95 by the log-rank test). At 36 months, OS from CR was estimated at 84% (95% CI, 74-91) in t(8;21) AML patients as compared with 86% (95% CI, 75-93) in inv(16)/t(16;16) AML patients (Figure 2F; HR, 0.77 [95% CI, 0.34-1.72]; P = .52 by the log-rank test).