CLL mutational landscape. Comparison of CLL driver alterations in different cohorts of patients. (A) Mutational frequencies for a subset of 28 genes identified as frequent driver alterations in 2 whole genome/exome studies (International Cancer Genome Consortium [ICGC],⁶⁶  Dana-Farber [DF]⁶⁷ ) and 1 target sequencing at high coverage by Nadeu et al⁷³  (Deep-Seq). Red, ICGC-CLL (n = 452); purple, ICGC-MBL (n = 54); blue, DF-pretreated CLL (DF-pre) (n = 123); green, DF-CLL8 cases enrolled in the clinical trial CLL8 (n = 278); and yellow, Deep-Seq (n = 406). The distribution of genomic alterations in MBL is similar to those of population-based CLL. Mutations in SF3B1, POT1, ATM, TP53, and RPS15 are more frequent in CLL8. MYD88 mutations are uncommon in clinical trial patients but are frequent in the DF-untreated, a cohort a decade younger that the ICGC. Detection of subclonal mutations by Deep-seq increases the global frequency of virtually all mutated genes. (B) Frequency of 11 CNA identified as putative driver alterations by Puente et al⁶⁶  using single-nucleotide polymorphism (SNP) arrays (red, ICGC-CLL; purple, ICGC-MBL) and Landau et al⁶⁷  using whole-exome sequencing data (blue, DF-pre; green, DF-CLL8). These CNAs were also analyzed by Nadeu et al⁷³  by SNP arrays (yellow).