Relocalization of PU.1 Underlies the On- and Off-Tumor Effects of SWI/SNF Blockade in AML

SWI/SNF ATP-dependent chromatin remodeling activity has been identified as a strong dependency of acute myeloid leukemia (AML). In AML cell lines, genetic knockdown or chemical inhibition of the SWI/SNF ATPases SMARCA4 (BRG1) or SMARCA2 (BRM) by the selective small-molecule inhibitor BRM014 reduces proliferation and induces leukemic differentiation. These effects arise because SWI/SNF inhibition reduces DNA accessibility at binding sites of PU.1, a factor essential for driving high levels of MYC in leukemic cells.

However, given that both SWI/SNF and PU.1 are essential in hematopoietic development, important questions regarding the use of SWI/SNF inhibitors against AML remain. First, the chromatin features that influence SWI/SNF-dependent PU.1 regulation within healthy hematologic and leukemic cells have not been established. Furthermore, the ability of SWI/SNF inhibitors to target leukemic cells within the bone marrow has remained uncertain, and the effects of these drugs in an immunocompetent setting have not been reported, despite the essential functions of SWI/SNF and PU.1 in normal hematopoiesis. Finally, the sensitivity of primary human AML samples and normal hematopoietic cells has also not been established. As a result, many questions regarding the nature of the therapeutic window for SWI/SNF inhibition against AML have remained unresolved.

Here we evaluate SWI/SNF inhibition using BRM014 in clinically relevant models and examine the consequences of the deregulation of SWI/SNF-dependent PU.1 sites. We demonstrate that inhibition of SWI/SNF shifts PU.1 occupancy from enhancers to promoters in both leukemic and healthy hematopoietic cells. In AML, the shift of PU.1 away from enhancers to the promoters of differentiation-related genes results in loss of MYC expression and differentiation of leukemic cells. In non-leukemic mice, SWI/SNF inhibition suppresses PU.1-dependent B cell development and bone marrow-derived monocyte populations, reflecting off-tumor effects caused by SWI/SNF-dependent PU.1 relocalization. These effects influence peripheral immune cells, as well lineage-committed hematopoietic progenitors within the bone marrow. Moreover, we observe profound therapeutic response to SWI/SNF inhibition in an immunocompetent AML mouse model (Figure 1). In vivo, BRM014 induces differentiation of peripheral leukemic blasts and reduces the leukemic stem cell burden in bone marrow but also induces lymphocytopenia due to its off-tumor effects on PU.1. Nevertheless, the robust regression of leukemic burden seen in vivo over a short two-week treatment period provides a compelling justification for continued study of SWI/SNF inhibitors in the treatment of AML.

Further, response of primary human CD34+ hematopoietic stem and progenitor cells (HSPCs) and AML specimens highlight that normal hematopoiesis is affected at concentrations similar to those needed for therapeutic response in AML (Figure 2). Hence, hematopoietic and immune-related side effects, particularly at higher doses and over the duration of treatment, will be important considerations for clinical use of SWI/SNF inhibitors against AML and other tumors. Overall, our results reveal a variable therapeutic window for SWI/SNF blockade in AML and highlight important off-tumor effects of such therapies in immunocompetent settings.