ACC010, a novel BRD4 inhibitor, synergized with homoharringtonine in acute myeloid leukemia with FLT3 ‐ITD

Bromodomain‐containing protein 4 (BRD4) inhibitors have been clinically developed to treat acute myeloid leukemia (AML), but their application is limited by the possibility of drug resistance, which is reportedly associated with the activation of the WNT/β‐catenin pathway. Meanwhile, homoharringtonine (HHT), a classic antileukemia drug, possibly inhibits the WNT/β‐catenin pathway. In this study, we attempted to combine a novel BRD4 inhibitor (ACC010) and HHT to explore their synergistic lethal effects in treating AML. Here, we found that co‐treatment with ACC010 and HHT synergistically inhibited cell proliferation, induced apoptosis, and arrested the cell cycle in FMS‐like tyrosine kinase 3‐internal tandem duplication (FLT3‐ITD)–positive AML cells in vitro, and significantly inhibiting AML progression in vivo. Mechanistically, ACC010 and HHT cooperatively downregulated MYC and inhibited FLT3 activation. Further, when HHT was added, ACC010‐resistant cells demonstrated a good synergy. We also extended our study to the mouse BaF3 cell line with FLT3‐inhibitor‐resistant FLT3‐ITD/tyrosine kinase domain mutations and AML cells without FLT3‐ITD. Collectively, our results suggested that the combination treatment of ACC010 and HHT might be a promising strategy for AML patients, especially those carrying FLT3‐ITD.


Introduction
Acute myeloid leukemia (AML) is the most common form of acute leukemia in adults which is characterized by genetic heterogeneity [1]. Under the current therapeutic strategy, which mainly consists of chemotherapy and hematopoietic stem cell transplantation, the outcome remains poor with a higher recurrence rate and higher mortality. Despite the improvement in precision medicine, for new drugs targeting at individual molecular characteristics to benefit some molecular subtypes of AML, primary and secondary drug resistance remains a serious concern for most patients.
FMS-like tyrosine kinase-3 (FLT3) is a receptor tyrosine kinase normally expressed in hematopoietic stem and progenitor cells which plays a critical role in early hematopoiesis [2]. FLT3 internal tandem duplication (FLT3-ITD) is the most common mutational form in AML and occurs in approximately 25% of AML and 30% of cytogenetic normal AML [3,4]. AML patients with FLT3-ITD have a high incidence of relapse and relatively short survival duration. Current evidence suggests that FLT3-ITD constitutively activates FLT3 kinase activity to promote the proliferation of AML [5]. Although FLT3-ITD alone is not sufficient to generate AML, the role of a therapeutic target has broadened the treatment options for AML patients [6]. Consistently, several FLT3 kinase inhibitors (FLT3-TKIs), such as gilteritinib and quizartinib, have been clinically developed and exhibited a favorable response in patients with FLT3-ITD/TKD mutations. However, TKI resistance (TKI-R) may emerge in acquired FLT3-TKD mutations or in the case of high levels of FLT3 ligand [7,8]. Thus, an alternative therapy is highly needed under such circumstances.
BRD4 belongs to the family of BET (bromodomain and extraterminal) proteins, called chromatin "readers" that bind to the acetylated lysines on histone proteins, which has an important impact on transcriptional regulation of multiple important oncogenes, including MYC [9]. BRD4 has been reported to be crucial in AML maintenance through MYC activation and aberrant transcriptional elongation [10]. In AML with FLT3-ITD, BRD4 inhibitors, alone or in combination with FLT3 inhibitors, also exhibited excellent proliferation inhibition [11]. To date, several small-molecule inhibitors of BRD4, such as JQ1, OTX015, and Molibresib, have been developed and some of them have entered phase I/II clinical trials [12,13]. However, despite their high efficacy in in vitro experiments, their clinical outcome is unsatisfactory when administrated as single agents, along with the issue of the development of resistance. It has been reported that the mechanism of BRD4 inhibitor resistance in AML was mainly associated with WNT/b-catenin-mediated c-MYC reactivation after suppression, so targeting the WNT/ b-catenin-c-MYC axis possibly could help overcome BRD4 inhibitor resistance [14].
Homoharringtonine (HHT), a classic antileukemia drug in China, was originally isolated from the Cephalotaxus hainanensis. Over the past few years, our research team has devoted great efforts to understand its effect and mechanism in AML [15][16][17][18]. In our previous clinical trial, we demonstrated that HHT in combination with cytarabine and aclarubicin achieved a high complete remission rate of 73-83% in treating de novo AML [15,16]. Because this HHT-based regimen has high efficiency and is inexpensive, it has been the first-line choice for AML therapy in China. Furthermore, we found that HHT showed high efficiency in AML with FLT3-ITD via downregulating FLT3 expression and inhibiting FLT3-mediated downstream signaling activation [17]. Moreover, HHT inhibited c-MYC activation by directly binding the NF-jBrepressing factor, whereas HHT also reduced the protein expression of CTNNB1, which is known as an intracellular signal transducer in the WNT signaling pathway [19]. Therefore, HHT could be potentially combined with BRD4 inhibitors in FLT3-ITD-positive AML.
Because BRD4 inhibitors and HHT could downregulate MYC through different pathways, and the downregulation of WNT/b-catenin by HHT presumably could help overcome resistance to BRD4 inhibitors, we attempted to combine a novel BRD4 inhibitor ACC010 and HHT in AML cells. In this study, we demonstrated that the combination of ACC010 and HHT had significant synergistic effects in AML with FLT3-ITD in vitro and in vivo. The combination inhibited cell proliferation by inducing apoptosis and arrested the cell cycle at the G0/G1 phase. Mechanistically, downregulation of MYC and inhibition of the FLT3 pathway may account for the synergistic effects. Furthermore, synergistic effects were also shown in cells resistant to ACC010 or FLT3 inhibitors. Moreover, the synergistic effects of this combination could also be expanded to some subtypes of FLT3-ITD-negative AML. Therefore, our studies suggested that ACC010 plus HHT might be a promising combination regimen to treat AML, especially FLT3-ITD-positive AML.

Cell proliferation assay
Acute myeloid leukemia cell lines were seeded in 96well plates with 1-2 9 10 4 cells/well, and primary AML cells were seeded in 96-well plates with 1 9 10 5 cells/well. Cells were treated with variable concentrations of ACC010 and/or HHT for 48 h. Cell viability was assayed by CellTiter-Lumi TM Luminescent Cell Viability Assay Kit (Beyotime, Shanghai, China) according to the manufacturer's instructions.

Growth curve assay
Cells were seeded in 96-well plates (5000 cells/well), which were treated with single agents or a combination regimen. MTS solution measuring 10 lL (CellTitre 96; Promega) (5 mgÁmL À1 ) was added to each well at every 24 h, varying from 0 to 96 h, and the cells were incubated for an additional 4 h at 37°C before absorbance was measured at 490 nm.

Cell apoptosis and cell cycle assay
Induction of apoptosis was assessed using a kit obtained from MULTI SCIENCES (Shanghai, China), following the manufacturer's instructions. Cells were collected after treatment with drugs for 48 or 72 h. Apoptotic cells were analyzed by flow cytometry after incubation with Annexin-V FITC and propidium iodide (PI) for 30 min at room temperature using FACScan TM flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA). Cell cycle assay was also detected by flow cytometry using PI DNA staining from MULTI SCIENCES. After treatment with drugs for 48 h, the cells were harvested, fixed overnight with 75% ethanol at 4°C, and then incubated with DNA staining for 30 min at room temperature. Analysis was conducted by FACScan TM flow cytometer (Becton Dickinson).

Western blot analysis
Cells were lysed in RIPA buffer (Thermo Fisher Scientific, Waltham, MA, USA) with protease inhibitor and phosphatase inhibitor cocktail (Thermo Fisher Scientific) on ice for 30 min. Next, the cell lysate was centrifuged at 12 000 g for 15 min at 4°C, and a BCA reagent (Thermo Fisher Scientific) was used to determine the protein concentration of the cellular supernatant. Western blotting was performed using 4-12% SDS/PAGE gel and cellular proteins were transferred onto a preactivated PVDF membrane (Millipore, Billerica, MA, USA). Then membranes were blocked with 5% nonfat milk for 1 h and incubated overnight with primary antibodies at 4°C. Subsequently, the membranes were washed thrice with TBST and incubated with secondary antibodies (Cell Signaling Technology) for 1 h at room temperature. The target proteins were visualized using FDbio-Femto ECL (Fudebio, Hangzhou, China) and analyzed using IMAGE LAB TM software (Bio-Rad, Hercules, CA, USA). The densitometry quantification was performed using the IMAGEJ software (NIH, Bethesda, MD, USA).

RNA extraction and real-time PCR (qRT-PCR)
Total RNA was extracted from the cells using the TRIzol reagent (TaKaRa, Dalian, China). Reverse transcription was performed using RNA PCR core kit (TaKaRa). Quantitative real-time PCR was carried out using SYBR Green qPCR mastermix (TaKaRa). Analysis was performed with Bio-Rad CFX96 (Bio-Rad). GAPDH was used as internal control. The sequences of the primers were as follows:

In vivo studies
For AML xenografts, female NCG (NOD/ShiLtJGpt-Prkdcem26Cd52Il2rgem26Cd22/Gpt) mice, aged 5-6 weeks, were purchased from GemPharmatech Co., Ltd (Nanjing, China) and were raised in the Experimental Animal Center of Zhejiang Chinese Medicine University Laboratory Animal Research Center. Mice were exposed to a 10/14 h light-dark cycle, kept under normal room temperature and fed standard pellet food and tap water. All animal experiments were reviewed and approved by the Institutional Animal Care and Use Committee (Approval No: IACUC-20210517-11). And all animal experiments were performed following animal use guidelines and ethical approval. In this study, MOLM13-luciferase cells (1 9 10 6 ) were injected into each mouse via the tail vein. After 6 days, cell engraftment was assessed after intraperitoneal injection of luciferin (Promega). Then mice were randomly assigned to four groups according to the intensity of the luciferin signal by using the IVIS Imaging System (PerkinElmer, Waltham, MA, USA). Each group was treated as follows: 0.5% CMC-Na (days 6-26, PO), 20 mgÁkg À1 of ACC010 (days 6-26, PO), 0.5 mgÁkg À1 of HHT (days 6-12, intraperitoneal), or 20 mgÁkg À1 of ACC010 and 0.5 mgÁkg À1 of HHT (administered per the singleagent group). Mouse body weight was determined once a week. The growth of the leukemia cells was monitored every week by using an IVIS. The survival curve of the mice was analyzed by GRAPHPAD PRISM 9 (San Diego, CA, USA).

Cell transfection
To knock down BRD4, short hairpin RNAs (shRNAs) were designed and cloned into a modified pLKO.1puro-shRNA plasmid. The sequence of sh1 was GGAAGTGGAAGAGAATAAA. The sequence of sh2 was GATTACTATAAGATCATTA. To knock down APC, MV4-11 was transfected with mCherry protein containing shRNA or control lentiviruses cloned into pLVX-shRNA2 vector. The sequence of NC was AGCGTGTAGCTAGCAGAGG. The sequence of sh1 was CCCAGTTTGTTTCTCAAGAAA. The sequence of sh2 was TAATGAACACTACAGA-TAGAA. For MYC overexpression, the coding DNA sequence was cloned into the pCDH1-MSCV-MCS-EF1-GreenPuro vector and then transfected into MV4-11. BaF3 cells were stably expressing FLT3-ITD or FLT3-ITD/F691L or FLT3-ITD/D835V utilizing PCDH lentivirus vector containing those mutations.

Statistical analysis
Data were analyzed using GRAPHPAD PRISM 9 and expressed as mean AE SEM. Statistical significance was assessed using two-tailed Student's t-tests to compare means between two groups (P < 0.05 was considered statistically significant). CALCUSYN software (Biosoft, Cambridge, UK) was used to calculate combination index. Combination index (CI) values less than 1.0 indicate a synergistic interaction of the two agents in the combination. Assessment of compound synergy was also conducted by Bliss scoring utilizing Drug-Comb online web application tool (https://drugcomb. fimm.fi). The combination effect is additive if the Bliss score equals 0, whereas the combination effect is synergistic if the Bliss score is positive. Survival was analyzed using the Kaplan-Meier method and analyzed using a log rank test.

Synergistic anti-FLT3-ITD AML effects of ACC010 and HHT in vitro
We first evaluated the antileukemia effect of ACC010 and HHT in AML cell lines. We found that both MV4-11 and MOLM13 cells (carrying FLT3-ITD) were both of the most sensitive to ACC010 and HHT, whereas U937 was relatively resistant to both drugs in AML cell lines (Fig. 1A). Next, we demonstrated that ACC010 and HHT had significant synergistic effects on MV4-11 and MOLM13, which turned out significant synergistic effect (Fig. 1B,C; Fig. S2A,C). Moreover, both compounds and their combination inhibited the growth of MV4-11 and MOLM13 cells in a time-dependent manner (Fig. 1D). These findings showed the synergistic inhibiting effects of ACC010 and HHT in FLT3-ITD AML cell lines. Then we observed the impact of ACC010 and HHT on the viability of primary AML cells in five patients with FLT3-ITD. Primary AML cells demonstrated a good synergy for the combination of ACC010 and HHT (Fig. 1E,F; Fig. S2B,C). We also tested this combination regimen in normal samples from healthy donors or CD34+ hematopoietic stem cells from the umbilical cord blood, and we found limited inhibiting effects in comparison to those in primary AML samples (Fig. S3A). Taken together, the combination of ACC010 with HHT showed a prominent synergistic and highly selective antiproliferative effect in FLT3-ITD-positive AML cells.
3.2. ACC010 and HHT synergistically induced the apoptosis and arrested the cell cycle at G0/G1 phase in FLT3-ITD-positive AML cells To further explore their synergy, we investigated the apoptotic phenotype of ACC010 combined with HHT in FLT3-ITD-positive AML cells. We treated both MV4-11 and MOLM13 cells with 2 lM of ACC010 or 4 nM of HHT or their combination for 48 or 72 h. Annexin-V and PI staining showed that ACC010 or HHT alone induced cell apoptosis, and the effect was enhanced significantly when ACC010 and HHT were combined ( Fig. 2A,B). Besides, the combination of ACC010 and HHT also induced more apoptosis in primary AML cells (Fig. S3B). To confirm the effect of drug-induced apoptosis, we measured the levels of apoptosis-related proteins. After treatment with single drugs or a combination regimen for 48 h in MV4-11 and MOLM13, we extracted the proteins, and the levels of cleaved-PARP, caspase-3, caspase-7 increased more apparently in the combination regimen (Fig. 2C).
Moreover, we also investigated the cell cycle distribution with PI DNA staining under treatment with 2 lM of ACC010, 4 nM of HHT and their combination for 48 h. We observed that 2 lM of ACC010 could induce G0/G1 cell cycle arrest, whereas 4 nM of HHT did not influence the cell cycle. Meanwhile, their combination significantly arrested the cell cycle at the G0/ G1 phase compared with the single drug or control group (Fig. 2D,E). Similar results were also found in primary AML cells (Fig. S3C). Further, we found the cell cycle-related proteins to be associated with the phenotype. The protein levels of CDKs (CDK2, CDK4, and CDK6) regulating G1 phase progression decreased to varying degrees in the combination groups. Both RB and pRB protein levels had a greater attenuation in the combined treatments, which implied inhibition of G1/S cell cycle progression. Meanwhile, the levels of P21 and P27 (CDK inhibitors) ameliorated, which also indicated G0/G1 arrest (Fig. 2F). Collectively, ACC010 and HHT inhibited the proliferation of FLT3-ITD-positive AML by the apoptosis induction and G0/G1 cell cycle arrest.

Synergistic antileukemia effect of ACC010 plus HHT on FLT3-ITD-positive AML in vivo
Owing to their significant synergistic antileukemia effects in vitro, we further investigated the effects of ACC010 and HHT in vivo in MOLM13-luc-NCG xenograft mice. We divided 24 mice into four groups: control group (0.5% CMC-Na), ACC010 group (20 mgÁkg À1 ), HHT group (0.5 mgÁkg À1 ), and combination group (20 mgÁkg À1 of ACC010 and 0.5 mgÁkg À1 of HHT). We injected each mouse with 1 9 10 6 MOLM13-luciferase cells. Six days after injection, when leukemia cells were engrafted in the bone marrow, we started the aforementioned drug administrations. All mice had equal tumor burdens before being administered any treatment. Mice treated with a combination of ACC010 (20 mgÁkg À1 ) and HHT (0.5 mgÁkg À1 ) had significantly lower leukemia tumor burdens on day 13 and day 20 than mice treated with vehicle or ACC010 or HHT alone (Fig. 3A,B). After 27 days, only mice in the ACC010 group and combination group survived. The latter showed more reduction in tumor burden. All treatment groups had prolonged survival, with the combination group exhibiting the best survival (Fig. 3C). To assess treatment tolerance, we determined the weights of the mice every week. We found the drug doses to be well-tolerated because there was no obvious effect on body weight (Fig. 3D). Hence, ACC010 and HHT exhibited a synergistic antileukemia effect on FLT3-ITD-positive AML in vivo.   The synergistic effects in MV4-11 and MOLM13, which express FLT3-ITD, were notable. As reported, ITD mutations lead to the constitutive activation of FLT3 and aberrant activation of multiple downstream pathways such as STAT5, AKT, and ERK1/2 [20].
Thus, we detected the FLT3 and FLT3-activated signaling pathways after ACC010 and HHT treatments. We observed that the co-treatment of ACC010 and HHT mediated greater attenuation of FLT3 as well as its activated form p-FLT3; at the same time, the downstream signaling of FLT3 was also inhibited by this combination, which was revealed by the downregulation of p-STAT5, p-AKT and p-ERK1/2 (Fig. 4A). Thus, ACC010 combined with HHT inhibited the activation of the FLT3-related pathway.  Data are presented as mean AE SEM (n = 3). Statistical analyses were performed using two-tailed Student's t-tests. *P < 0.05, **P < 0.01, ***P < 0.001.
Meanwhile, previous studies reported that BET inhibitors could deplete the binding of BRD4 to superenhancers, resulting in transcription elongation defects of some oncogenes, including MYC [21]. Recently, it has been reported that HHT also targeted MYC via binding NF-jB-repressing factor. MYC was also critical for the maintenance of FLT3-ITD-positive AML. Thus, we investigated whether BET inhibitors could cooperate to inhibit MYC to repress FLT3-ITD-positive AML. In our study, we observed that both ACC010 and HHT decreased the MYC level in a dose-dependent manner and that their combination at a lower concentration also decreased the expression of MYC (Fig. 4B). Therefore, we consider that MYC downregulation also accounted for the synergistic effects of ACC010 and HHT in FLT3-ITD-positive AML.

ACC010 and HHT synergistically inhibited the proliferation of FLT3-ITD-positive AML resistant to BRD4 inhibitors
The drug resistance to BRD4 inhibitors remains a critical problem. Previous studies documented that increased expression levels of MYC were related to the resistance in AML. Moreover, whereas BRD4 inhibitors transitorily repressed MYC transcription in types of human leukemias regardless of their sensitivity, resistant cells exhibited a rapid restoration of MYC transcription [22]. In the present study, we selected five cell lines, MV4-11 and MOLM13 as the cells sensitive to ACC010, and THP1, JURKAT, and K562 as the resistant cells. We observed similar results of finding a rebound of MYC transcription in resistant cells, but not in sensitive cells (Fig. S4A,B). Then we used the MYC-PCDH lentivirus vector to overexpress MYC in MV4-11, and we discovered that after MYC overexpression, MV4-11 appeared to be more resistant to ACC010 (Fig. 5A,B). The synergistic effect of ACC010 and HHT still existed, and it was more notable than that of the cells infected with overexpression control lentiviral vector (Fig. 5C,D; Fig. S4C).
As has been reported, the activation of the WNT/bcatenin was the main cause of the restoration of MYC transcription, which contributed to the BRD4 inhibitor resistance in AML. The resistant leukemia cell lines showed increased transcription levels of WNT/bcatenin target genes DVL1, GSK3B, and FZD5 [23]. Consistently, we conducted a qRT-PCR analysis of those target genes in AML cell lines and observed that relatively resistant cell lines (THP1, U937, and KG1-a) were of particularly high expression (Fig. S4D). Literature also supported that knockdown of APC, a key molecule that negatively regulates the WNT/b-catenin, could lead to WNT/b-catenin activation and resistance to BRD4 inhibitors. Then we transfected MV4-11 with APC shRNA and established APC knockdown MV4-11 cells. Undoubtedly, the APC knockdown cells exhibited ACC010 resistance, but an excellent synergistic effect of ACC010 and HHT was still found (Fig. 5E-H; Fig. S4E). Additionally, JURKAT and K562 as ACC010-resistant leukemia cell lines in our study were also tested by the combination therapy and demonstrated noticeable synergistic effects (Fig. S4F,  G). Briefly, the co-treatment of ACC010 and HHT exhibited excellent synergy in ACC010-resistant cells.
3.6. Co-treatment of ACC010 and HHT inhibited the proliferation of leukemia cells with FLT3-TKD mutations or without FLT3-ITD ACC010 combined with HHT showed an excellent synergy in FLT3-ITD-positive AML cells, which were also sensitive to FLT3-TKIs, but whether this combination works in FLT3-TKI-resistant AML remains undetermined. Previous studies reported that the acquisition of secondary FLT3-TKD mutations including F691L and D835V might account for the resistance to FLT3-TKIs [8]. As we found that cells with FLT3-ITD mutation had good synergy for ACC010 and HHT, we transfected FLT3-ITD, FLT3-ITD/ F691L, and FLT3-ITD/D835V mutations separately into mouse BaF3 cells to further study how FLT3-ITD alone or with FLT3-TKD mutations is affected by the combination therapy. The synergistic effects were shown on both FLT3-ITD-and FLT3-TKI-resistant cells (Fig. 6A,B; Fig. S4H). Furthermore, we speculated whether this combination could be expanded to AML without FLT3-ITD, so we tested the combination regimen in AML cell lines and AML patients without FLT3-ITD, which resulted in varying degrees of synergistic effects (Fig. 6C-F; Fig. S2A-C). These findings suggest that the co-treatment of ACC010 and HHT is feasible in treating AML with not only FLT3-ITD but also TKD mutations or without ITD mutation.

HHT synergized with JQ1 or OTX015 to inhibit the proliferation of AML cells
Based on the above results, we conjectured that HHT might have synergistic effects with BRD4 inhibitors. Therefore, we conducted the combination regimen of two other BRD4 inhibitors, JQ1 and OTX015, to combine with HHT in AML cell lines. We first evaluated the antileukemia effect of JQ1 and OTX015 in AML cell lines. We found that among these AML cell lines, THP1 and U937 resistant to ACC010, were also relatively resistant to JQ1 and OTX015 (Fig. S5A,B).
Then the cell proliferation was tested after treated with single agents or in combination for 48 h. Results confirmed our conjecture that both JQ1 and OTX015 had synergistic inhibiting effects with HHT in AML cells (Fig. 7A-D; Fig. S5C,D). These findings support that the co-treatment of BRD4 inhibitors and HHT has a great potential for clinical application.

Discussion
It is noteworthy that targeting BRD4 is a promising strategy in AML therapy. Several structure-based inhibitors are being developed afterward. Although BRD4 inhibitors showed remarkable antileukemia activity in in vitro experiments, the clinical trials were not proceeding as expected owing to limited efficacy and high pharmacokinetics (PK) variability in patients. BRD4 inhibitor OTX015 is the first compound that has The results demonstrated that although the clinical response to OTX015 was observed in AML patients, OTX015's therapeutic index (the ratio between its toxicity and therapeutic dose) was narrow. Patients who achieved remissions ultimately relapsed after 2-5 months [12]. Other reports described that BRD4 inhibitor monotherapy may gain resistance due to the restoration of c-MYC expression induced by WNT/b-catenin activation. Thus, developing BRD4 inhibitors with a broader therapeutic index and their combinations with other targeted therapy to improve efficacy is of much interest in the future. To date, co-treatment of BRD4 inhibitors with histone deacetylase inhibitors has exhibited strong synergistic effects in AML with or without FLT3-ITD [24]. Another combination therapy with DOT1L inhibitors in MLL-driven AML models has also displayed marked synergistic activity [25]. Our study presented a novel combination regimen of a BRD4 inhibitor and HHT with substantial antileukemia effects in vitro and in vivo. Furthermore, evidence suggested that MYC provided a possible inherent mechanism for this combination, and HHT might help overcome resistance to BRD4 inhibitors via downregulation of the WNT/b-catenin pathway. Our results demonstrated that BRD4 inhibitor-resistant AML cells with MYC overexpression or APC knockdown were also sensitive to BRD4 inhibitors when combined with HHT.
In this article, we focused on the FLT3 mutated subtype in AML, which is associated with poor prognosis. Current guidelines recommend earlier incorporation of targeted agents to achieve deeper remissions after detecting FLT3 mut at diagnosis. The clinical application of FLT3 inhibitors has improved outcomes in AML patients with FLT3 mutations. However, as mentioned before, FLT3 inhibitors are faced with a complication that drug resistance compromises their efficacy in clinical therapy. Previous studies reported that resistance mechanisms include the acquisition of secondary FLT3-TKD mutations. Among those mutations, F691 and D835 in FLT3-ITD were found to be substantial barriers to disease control in AML patients treated with FLT3-TKIs [8]. Those mutations hinder the drug binding sites, making them unfavorable for interaction with FLT3 inhibitors. These findings underscore the need to develop novel agents or to test combination therapies with other agents to improve the outcome of AML with FLT3-ITD in addition to FLT3-TKD mutations.
Currently, some clinical trials involving combination regimens with FLT3-TKIs in AML have completed, including the combination of sorafenib and 5azacytidine, which is proven to be effective in untreated FLT3-ITD-positive AML patients, with an overall response rate of 78% [26]. Another clinical trial to combine sorafenib with histone deacetylase inhibitor vorinostat and proteasome inhibitor bortezomib, and it also observed efficiency in poor-risk AML [27]. Meanwhile, preclinical studies made several attempts to combine FLT3-TKIs with other agents to help overcome drug resistance. Among those agents combined with FLT3-TKIs, BCL2 inhibitor venetoclax exhibited strong synergistic effects on inhibiting the proliferation of FLT3-ITD-positive cells and re-sensitized FLT3-TKI-resistant cells; BRD4 inhibitor JQ1 had synergistic lethal effects in AML and could help overcome FLT3-TKI resistance; HHT could help to inhibit the growth of drug-resistant clones and increase the antileukemia effects in FLT3-ITD AML [28][29][30]. Our study combined BRD4 inhibitor and HHT to treat FLT3-ITD AML. This regimen did not include FLT3-TKIs but also still presented significant killing effects. In addition, observations suggest that the combination can be utilized in AML cells with FLT3-TKD mutations. Further studies are needed to confirm whether the combination of BRD4 inhibitors and HHT can enhance the efficacy of FLT3-TKIs to help overcome resistance. Another FLT3-TKI resistance mechanism was considered to be related to high levels of FLT3 ligand, which could lead to the activation of the FLT3-MAPK pathway and thus confer resistance to FLT3 inhibitors [7]. This mechanism was not discussed in our article.

Conclusions
Our study showed a significant synergistic effect of the combination of BRD4 inhibitors and HHT in AML with FLT3-ITD mutation. With MYC overexpression or APC knockdown to generate cells resistant to BRD4 inhibitor, we found that HHT could enhance the efficacy of the BRD4 inhibitor to help overcome resistance. Furthermore, we extended this regimen to cells with FLT3-TKD mutation and cells without FLT3-ITD. The experimental results revealed the potential of ACC010 combination with HHT in the therapy of AML patients harboring FLT3-TKD who did not respond well to classic FLT3 inhibitors. Meanwhile, the combination regimen could treat AML patients without FLT3-ITD mutation. Moreover, our results support that other BRD4 inhibitors to combine with HHT for AML therapy.

Supporting information
Additional supporting information may be found online in the Supporting Information section at the end of the article. Fig. S1. Chemical structure of ACC010 and the target protein in AML cell line. A. Chemical structure of ACC010 described in the patent CN106132968B. B. Expression of BRD4, MYC, CDK6 and BCL2 were analyzed by Western blot after treated with ACC010 or JQ1 in MV4-11. C-D. The IC50 of ACC010 was analyzed in MV4-11 BRD4 knockdown cell. ** for p < 0.01, *** for p < 0.001. Fig. S2. Algebraic estimate and Bliss score analysis in AML cell lines and patients. A. The combination index (CI) of ACC010 and HHT in AML cell lines was calculated using CalcuSyn software after 48 hours of drugs' treatment. B. The combination index (CI) of ACC010 and HHT in AML patients was calculated using CalcuSyn software after 48 hours of drugs' treatment. C. The Bliss score analysis of combinations in AML cell lines and patients was conducted utilizing DrugComb online web application tool (https:// drugcomb.fimm.fi). Fig. S3. Lethal effects of ACC010 and HHT against normal samples and effects on apoptosis and cell cycle in primary AML cells. A. normal samples which from healthy donors or CD34+ hematopoietic stem cells were treated with variable concentrations of ACC010 or HHT for 48 hours and cell viability was analyzed. B. Apoptosis in primary AML cells induced by two drugs or their combination utilizing FCM after incubation with Annexin-V and PI. C. Cell cycle analysis by flow cytometry in primary AML cells cultured with drugs for 48 h.