Cellular senescence is a stress-responsive cell-cycle arrest program that terminates the further expansion of (pre-)malignant cells. Key signalling components of the senescence machinery, such as p16, p21 and p53, as well as trimethylation of lysine 9 at histone H3 (H3K9me3), also operate as critical regulators of stem-cell functions (which are collectively termed 'stemness'). In cancer cells, a gain of stemness may have profound implications for tumour aggressiveness and clinical outcome. Here we investigated whether chemotherapy-induced senescence could change stem-cell-related properties of malignant cells. Gene expression and functional analyses comparing senescent and non-senescent B-cell lymphomas from Eμ-Myc transgenic mice revealed substantial upregulation of an adult tissue stem-cell signature, activated Wnt signalling, and distinct stem-cell markers in senescence. Using genetically switchable models of senescence targeting H3K9me3 or p53 to mimic spontaneous escape from the arrested condition, we found that cells released from senescence re-entered the cell cycle with strongly enhanced and Wnt-dependent clonogenic growth potential compared to virtually identical populations that had been equally exposed to chemotherapy but had never been senescent. In vivo, these previously senescent cells presented with a much higher tumour initiation potential. Notably, the temporary enforcement of senescence in p53-regulatable models of acute lymphoblastic leukaemia and acute myeloid leukaemia was found to reprogram non-stem bulk leukaemia cells into self-renewing, leukaemia-initiating stem cells. Our data, which are further supported by consistent results in human cancer cell lines and primary samples of human haematological malignancies, reveal that senescence-associated stemness is an unexpected, cell-autonomous feature that exerts its detrimental, highly aggressive growth potential upon escape from cell-cycle blockade, and is enriched in relapse tumours. These findings have profound implications for cancer therapy, and provide new mechanistic insights into the plasticity of cancer cells.

Highlights d Genetic and functional analyses of myeloid preleukemia and leukemia in Down syndrome d Non-GATA1 preleukemic mutations are often not required for preleukemia d Previously undescribed transforming hotspot mutation in CSF2RB identified d Loss of function of 18 genes validated in transformation of preleukemia to leukemia

Mutations in the gene encoding isocitrate dehydrogenase 2 (IDH2) occur in several types of cancer, including acute myeloid leukemia (AML). In model systems, mutant IDH2 causes hematopoietic differentiation arrest. Enasidenib, a selective small-molecule inhibitor of mutant IDH2, produces a clinical response in 40% of treated patients with relapsed/refractory AML by promoting leukemic cell differentiation. Here, we studied the clonal basis of response and acquired resistance to enasidenib treatment. Using sequential patient samples, we determined the clonal structure of hematopoietic cell populations at different stages of differentiation. Before therapy, IDH2-mutant clones showed variable differentiation arrest. Enasidenib treatment promoted hematopoietic differentiation from either terminal or ancestral mutant clones; less frequently, treatment promoted differentiation of nonmutant cells. Analysis of paired diagnosis/relapse samples did not identify second-site mutations in IDH2 at relapse. Instead, relapse arose by clonal evolution or selection of terminal or ancestral clones, thus highlighting multiple bypass pathways that could potentially be targeted to restore differentiation arrest. These results show how mapping of clonal structure in cell populations at different stages of differentiation can reveal the response and evolution of clones during treatment response and relapse.