Fig. 1: The genetic landscape at relapse in KMT2A-r ALL.

a Illustration of the cohort divided into ALL and AML, age, relapse time, and specific KMT2A-r. ALL cases were divided into very early relapse (n = 11) if relapse occurred less than 9 months after diagnosis (D) or if the patient had resistant disease, and early relapse (n = 14) if relapse occurred after 9 months from diagnosis. The AML cohort was divided into early or late relapse if relapse occurred before (n = 5) or after 1 year after diagnosis (n = 6). b Schematic illustration of the study, including the number of AML and ALL patients and samples (D=Diagnosis, G=Germline, R=Relapse) analysed by each technology (whole genome, WGS; whole exome sequencing, WES). c Pathways mutated in very early and early relapse ALL. The colours indicate mutation type (CN-LOH, copy number loss of heterozygosity; protein del, protein deletion; protein ins, protein insertion), and the width of the slice, the number of patients with a mutation in each pathway. d Heatmap of nonsynonymous mutations in genes within enriched pathways at relapse, with genes in rows and patients in columns. On the right, a bar plot indicating the fraction of patients with mutations in the specific gene, and on the left, the pathways, are shown. Yellow indicates diagnose-specific alterations, grey shared between diagnosis and relapse, and lilac relapse-specific. Note the PTPN11^{F285_E8splice} (P17), is outside of the hot-spot sites, and although deleterious according to predictions, it is of unclear significance. D-R shows that no germline sample was available, R indicates that only a relapse sample was analysed. e The frequency of mutations in a certain pathway at diagnosis and relapse in very early and early relapse ALL. f Illustration of relapse-specific TP53-mutations, with the x-axis indicating the amino acid position, with our mutations at the top.