Figure 5
Figure 5. Survival signature predicts survival of patients with NK-AML. (A-C) Survival analysis based on 3 survival signatures derived from genes differentially expressed in patient cluster_2 and cluster_4. The effect of the expression of individual probe sets on survival was tested by dividing the entire data set into low- and high-scoring samples (median). Probe sets associated with poor and good OS (P < .05, moderated t test) and the ability to separate the data set (P < .05, log-rank test) were used to generate a poor outcome signature (A), good outcome signature (B), and combined survival signature (C). (D-E) Testing of the combined survival signature on 2 independent NK-AML patient data sets.13,15 (F) Testing of a previously published HSC signature32 revealed its inability to predict survival in the NK-AML patient data set in D.

Survival signature predicts survival of patients with NK-AML. (A-C) Survival analysis based on 3 survival signatures derived from genes differentially expressed in patient cluster_2 and cluster_4. The effect of the expression of individual probe sets on survival was tested by dividing the entire data set into low- and high-scoring samples (median). Probe sets associated with poor and good OS (P < .05, moderated t test) and the ability to separate the data set (P < .05, log-rank test) were used to generate a poor outcome signature (A), good outcome signature (B), and combined survival signature (C). (D-E) Testing of the combined survival signature on 2 independent NK-AML patient data sets.13,15  (F) Testing of a previously published HSC signature32  revealed its inability to predict survival in the NK-AML patient data set in D.

Close Modal

or Create an Account

Close Modal
Close Modal