In acute myeloid leukemia (AML), bromodomain and extraterminal protein inhibitors (BETi) have shown modest response rates, primarily due to non-genetic resistance. Resistance development requires cell survival and adaptation under therapeutic pressure. However, the precise mechanisms driving these processes, as well as the relationship between the initial adaptation and subsequent resistance evolution, remain elusive. In this study, we aimed to discover and therapeutically explore chromatin adaptation patterns to BETi at an early treatment stage and follow the evolutionary routes of this adaptation until the development of established resistance.

We assessed early transcriptional responses via SLAMSeq or nuclear RNASeq and chromatin binding dynamics of key transcription factors/co-regulators involved in BRD4 recruitment 24 hr post-BETi. ChIPSeq was performed for RUNX1 (wild type and fused to RUNX1T1), FLI1, PU.1, LMO2, ERG, p300, CDK9 and the histone mark H3K27ac as well as ATACSeq in AML models driven by RUNX1-RUNX1T1 rearrangement, a subtype sensitive to BETi. These revealed that within 24 hr of treatment, BETi-induced release of BRD4 from chromatin initiated an acute compensatory feedback that mitigated downregulation or even increasedthe expression of specific transcriptional modules critical for AML maintenance.

Notably, p300, a direct recruiter of BRD4 to chromatin, was redistributed to the same regions where BRD4 was lost, suggesting an acute compensatory role for p300. Moreover, a similar p300-mediated “rescue” of BETi was observed in another AML model - with mutated NPM1. We functionally tested this “rescue” using matched SLAMSeq and ChIP in a model of sequential BET inhibition for 24 hr followed by PROTAC-based degradation of p300 during the final 4 hr of BETi. Here, the initially rescued critical transcriptional modules were abrogated through p300 degradation, validating our mechanism.

Using RNASeq following pharmacological inhibition across cell lines representative of different AML subtypes, we found that BRD4 and p300 control similar transcriptional programs. Sequential BETi followed by inhibition of p300 lysine acetyltransferase activity resulted in robust downregulation of key transcriptional modules across various AML subtypes, including those rescued by p300 after single BETi. This sequential treatment exhibited optimal synergistic efficacy, leading to enhanced cell death in vitro in cell lines, murine AML models and patient samples ex vivo and also in vivo in a primary murine t(8;21) AML model, without affecting normal hematopoiesis.

We further tracked transcriptional changes and the dynamics of p300 binding at chromatin across resistance evolution to BETi in models representative for RUNX1-RUNX1T1-rearranged (SKNO1) and NPM1-mutated AML (OCI-AML3). p300 activity was critical throughout all resistance adaptation stages, although the specific transcriptional programs it regulated varied between the two AML subtypes. In OCI-AML3 cells, p300 was most essential during early stages of resistance, regulating transitional transcriptional patterns for AML homeostasis such as an inflammatory signature centered around S100A8/9 and an Interferon-responsive program (ISG). p300 inhibition led to a downregulation of S100A8/9 and a strong induction of ISG. Confirmatory assays, such as inducible RNAi knockdown or pharmacologic inhibition of mediators of these programs, demonstrated a pro-resistance role for S100A8/9 and an apoptosis-inducing role for ISG induction.

In contrast, in SKNO1 cells, p300 drove a linear resistance evolution, and was necessary at all stages, albeit that only a few central targets mediated established BETi resistance. The most significant inducer of resistance here was NCAM1, which we and others have previously associated with resistance in AML. Notably, p300 inhibition significantly reduced NCAM1 expression, and inducible knockdown of NCAM1 re-sensitized BETi-resistant SKNO1 cells to BETi.Altogether, our study is one of the first to elucidate mechanisms underpinning acute resistance, in this case to BETi through p300 activity with in vivo corroboration. Moreover, we demonstrate how these mechanisms transition to chronic resistance. Importantly, our data suggest that sequential treatment with BET and p300 inhibition may effectively prevent resistance development, thereby improving therapeutic outcomes.

Disclosures

Lugo:Pharmaron: Current Employment; GSK: Ended employment in the past 24 months. Kühn:Kura-Oncology: Research Funding; Servier: Consultancy, Other: Speaker activity; Syndax: Research Funding; Pfizer: Consultancy; Giliead: Other: Speaker Activity; Jazz: Consultancy, Other: Speaker activity; Janssen: Consultancy; BMS/Celgene: Consultancy, Other: Speaker Activity; Abbvie: Consultancy, Other: Speaker activity. Gallipoli:GSK: Other: Spouse works for GSK; Astellas: Honoraria. Yeh:Astellas: Honoraria; Novartis: Honoraria. Prinjha:GSK: Ended employment in the past 24 months, Other: Holds shares in GSK. Huntly:Amphista: Consultancy; Menerini: Consultancy; Novartis: Honoraria, Membership on an entity's Board of Directors or advisory committees; Istesso: Consultancy; Janpix: Membership on an entity's Board of Directors or advisory committees; AstraZeneca: Research Funding; Pfizer: Honoraria, Membership on an entity's Board of Directors or advisory committees. Sasca:Astellas: Honoraria; Abbvie: Honoraria; AstraZeneca: Honoraria; Gilead: Honoraria; BMS: Honoraria.

This content is only available as a PDF.
Sign in via your Institution