Clinical outcomes according to treatment arms and the presence of cytogenetic abnormalities. (A) Overall survival in all patients (left panel) and in subset with gene expression profiling (GEP)–defined low-risk myeloma. Survival is superior on the thalidomide arm only in patients presenting with cytogenetic abnormalities (CAs; P = .02), which also pertained to the subset with low-risk disease defined by GEP (P = .01). (B) Event-free survival. Event-free survival is superior in patients randomized to thalidomide versus the control arm, regardless of the presence of CA. (C) Cumulative CR rates. The proportion of patients achieving CR status was higher among those randomized to thalidomide in comparison to patients treated on the control arm, regardless of CA. (D) Duration of CR from its onset. The duration of CR is superior among patients randomized to thalidomide only among those presenting with CA (P = .05). (E) Postrelapse survival. Postrelapse survival was superior among patients initially randomized to the control arm in the absence of CA at baseline (P = .04), whereas such difference was not observed in case patients presented with CA (P = .99).