Development of the HSCT-64 model and its performance on internal test sets from the discovery cohorts. (IMAGE)
Caption
(A) Model coefficients for aHSCT-64 and nHSCT-64. The 64 core prognostic genes were classified into four categories based on their associations with risk: for example, the expression levels of 21 genes (red) were positively associated with increased mortality risk in both models; the remaining three categories follow similar interpretations. (B, C) Allo-HSCT (B) and non-HSCT (C) cases from the discovery cohorts were randomly split into training (70%) and testing (30%) sets. The aHSCT-64 and nHSCT-64 models were trained on the training sets, and their predictive performance was evaluated on the test sets using the concordance index (C-index) with respect to overall survival (OS). (D, E) As in (B, C), but evaluated using the time-dependent area under the ROC curve (tAUC) for predicting death/alive within 1–6 years respectively. (F, G) Kaplan–Meier survival analysis of high-risk and low-risk subgroups defined by aHSCT-64-rank (F) and nHSCT-64-rank (G) in the respective test sets. Significant survival differences were observed (log-rank test). HR: hazard ratio.
Credit
Yance Feng, Yali Shen, Ke Huang, Qian Li, Yu Tao, Rongqiu Liu, Liping Zhan, Hua Yang, Yang Xun , Yichao Xu, Wenli Tang, Binjun Xiong, Hui Shi, Liting Cheng, Li Wei, Hua You
Usage Restrictions
Credit must be given to the creator. Only noncommercial uses of the work are permitted. No derivatives or adaptations of the work are permitted.
License
CC BY-NC-ND