Acute Myeloid Leukemia (AML) presents a major therapeutic challenge. While initial response to standard 7+3 chemotherapy is observed in most patients, relapse with chemo-resistant disease is a frequent event, leading to a dismal overall survival rate of <40%. Emerging evidence points towards a population of rare AML cells, termed persistent blasts, that survive chemotherapy and fuel relapse. However, the mechanisms by which these blasts evade the cytotoxic effects of treatment remain unclear. The prevailing hypothesis implicates quiescent leukemia stem cells (LSCs), since are found within the persistent blast population and their dormant state inherently renders them resistant to chemotherapy. However, why chemoresistance persists in the progeny of LSCs as they emerge from quiescence to regenerate the leukemic population is unclear. Genomic analyses of paired primary/relapse AML samples showed persistence of multiple genetic clones at relapse (our unpublished results) suggesting that treatment resistance is not due to genetic mechanisms. Alternatively, persistent blasts may acquire resistance through phenotypic adaptation to chemotherapy. We developed novel mouse models of AML that faithfully recapitulate the features of chemo-resistant disease observed in patients: initial response to a treatment regimen adapted from the clinic (5+3) followed by inevitable relapse. These models incorporate a GFP-based label-retaining system, enabling isolation of quiescent and chemotherapy-persistent cells. Persistent and relapsing cells exhibit true chemoresistance, both in vitro and in vivo. Resistance is not associated with the emergence of new genetic mutations, suggesting a non-genetic mechanism at play. We observed that chemo-persistent blasts encompass both proliferating and quiescent populations, including quiescent LSCs. While not specifically selected by treatment, LSCs remain however the only cells with the ability to regenerate the leukemia population. Single-cell transcriptomic analyses revealed unique transcriptional features in chemotherapy-persistent blasts. These blasts were characterized by activation of interferon (IFN) signaling in both quiescent and proliferating populations. Interestingly, the chemoresistance-associated IFN gene signature was able to stratify AML patients treated with a 7+3 chemotherapy regimen. Silencing the most highly upregulated IFN target gene, IFI6, did not affect growth of leukemia cells. However, it completely reversed the chemoresistant phenotype in both in vitro and in vivo experiments. Notably, IFI6 is known to negatively regulate innate immunity, suggesting that AML cells adapt to chemotherapy by preventing excessive activation of interferon signaling after treatment, thus protecting LSCs from cell death or senescence induced by IFN. The molecular mechanisms underlying this chemosensitizing activity of IFI6 will be presented and discussed.
No relevant conflicts of interest to declare.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal