Fig. 7: Relevance of the cell of origin for the influence of trisomy 12 on the CLL proteome and clinical outcome.

a Validation strategy for Tris12M-PG, Tris12U-PG, M-PG, and U-PG. b Plot of loadings on LF1 and LF2 of individual patients. LF1 and LF2 could classify patients into 4 groups according to IGHV status and trisomy 12. c CLL protein correlation network analysis of HiRIEF dataset based on 1047 high-variance proteins. Protein groups were defined and color-coded based on modularity clustering (N1-N6) and enrichments detailed in Supplementary Fig. 9. A Heatmap of log2 mean relative protein abundances for all PGs for each modularity cluster (N1-N6) and node-cluster mean protein abundances for Tris12M-PG, Tris12U-PG, M-PG, and U-PG are shown. d Time to first treatment of patients from Validation1_DIA cohort, stratified into groups by IGHV mutation status and trisomy 12 (tris12). Trisomy 12 M-CLL patients had significantly faster disease progression than M-CLL patients without trisomy 12 (log-rank test, p = 0.05). e Time to first treatment of untreated patients from Validation4_untreated cohort, stratified into groups by IGHV mutation status and trisomy 12 (tris12). Trisomy 12 M-CLL patients had significantly faster disease progression than M-CLL patients without trisomy 12 (log-rank test, p = 0.01). f Time to progression of patients uniformly treated with ibrutinib (Validation5_ibrutinib), stratified into groups by IGHV mutation status and trisomy 12 (tris12). Trisomy 12 M-CLL patients did not have significantly faster disease progression than M-CLL patients without trisomy 12 (log-rank test, p = 0.69). Source data are provided as a Source Data file.