Bortezomib synergizes with homoharringtonine in FLT3-ITD-relapsed/refractory acute myeloid leukemia by inducing FLT3-ITD protein degradation

Abstract

Mutations in Fms-like tyrosine kinase 3 (FLT3) are strongly associated with relapse and resistance in acute myeloid leukemia (AML) patients, and the treatment of relapsed or refractory AML (R/R AML) remains a major clinical challenge. We previously conducted a prospective clinical trial on R/R AML with chemotherapy regimen BHA (bortezomib, homoharringtonine and cytarabine), which demonstrated promising efficacy in patients with FLT3-mutated R/R AML. However, the therapeutic mechanism remains unclear. In this study, we aim to elucidate the therapeutic mechanism of BHA regimen on the basis of its efficacy for FLT3-mutated R/R AML. We retrospectively analyzed twenty-nine patients with R/R AML, after one course of therapy, patients harboring FLT3 mutations had a significantly higher complete remission/complete remission with incomplete hematologic recovery rate than those without FLT3 mutations (46.67% vs. 7.14%, respectively; P = 0.035). To further explore the underlying mechanisms, we conducted combination index analysis, inhibition of proliferation and apoptosis assays. Compared with 293 T-FLT3 cells, 293 T-FLT3-ITD cells were more sensitive to bortezomib, with significantly lower IC50 values. Bortezomib in combination with homoharringtonine had a synergistic effect on FLT3-ITD cells. Moreover, compared with monotherapy, the combination of bortezomib (4 nM) and homoharringtonine (1 nM) markedly increased total cell death in FLT3-ITD cell lines (MV4—11 and Molm-13). Mechanistically, bortezomib promoted the degradation of FLT3-ITD protein, and the degradation of FLT3-ITD protein was further enhanced by homoharringtonine. Notably, this degradation effect was partially reversed by chloroquine. These findings demonstrate that bortezomib and homoharringtonine have synergistic effects and lead to degradation of FLT3-ITD oncoprotein, potentially contributing to a higher complete remission rate in FLT3-ITD R/R AML.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10238-026-02182-8.

Introduction

Fms-like tyrosine kinase 3 internal tandem duplication mutation (FLT3-ITD) accounts for approximately 25% of newly diagnosed acute myeloid leukemia (AML) cases [1, 2]. Despite the availability of multiple therapeutic approaches, such as chemotherapy, targeted therapies, or combination of the both [1–5], the relapse rate remains high among newly diagnosed patients. Furthermore, patients with relapsed/refractory AML (R/R AML) with FLT3-ITD continue to have poor outcomes, even in the era of target therapy [6–11]. Therefore, this dilemma highlights the urgent need for an improved therapeutic strategy.

A chemotherapy regimen of homoharringtonine combined with cytarabine (HA regimen) usually serve as an induction therapy for newly diagnosed AML [12–14] in China but is rarely utilized for R/R AML because its efficacy has not been assessed by clinical trials However, some chemotherapy regimens evolved from HA regimen, which is composed of homoharringtonine, cytarabine (Ara-C), and supplementation with other chemotherapy agents, and have been designed to investigate the efficacy of R/R AML. For instance, the combination of homoharringtonine, Ara-C, with granulocyte colony-stimulating factor has been reported an improved the complete remission (CR) rate of 50% [15], indicating that the formation of a new chemotherapy regimen by incorporating other agents into HA regimen may be a promising route to improve the treatment response for R/R AML. The proteasome inhibitor bortezomib, which was initially developed for multiple myeloma, has also been shown to have antileukemic activity in AML [16, 17]. Studies indicate that when bortezomib is combined with other agents such as decitabine, anthracyclines, or lenalidomide, the rates of CR/CR with incomplete hematologic recovery (CR/CRi) can reach approximately 30%, and the CR/CRi rate can reach 19% in patients with relapsed AML or myelodysplastic syndromes after transplantation [17–20]. These findings suggest that bortezomib can be incorporated into new treatment regimens for R/R AML. We previously designed a new chemotherapy regimen in which bortezomib was incorporated with homoharringtonine and Ara-C to form BHA regimen; a single-arm, prospective clinical trial (ChiCTR2000029841) was performed to assess its efficacy for patients with R/R AML [21]. Although the overall response rate (ORR) and CR/CRi rate were 52.4% and 38.1%, respectively, both the ORR and the CR/CRi reached 66.7% in R/R AML patients harboring FLT3 mutations (including FLT3-ITD and FLT3-tyrosine kinase domain (FLT3‐TKD) mutations), indicating that BHA regimen is more effective in patients with FLT3-mutated R/R AML [21]. In one study, bortezomib was reported to be involved in the degradation of FLT3-ITD protein in AML cell lines [22], however, the therapeutic mechanism of BHA regimen remains unclear.

In this study, we retrospectively analyzed 15 patients with FLT3-mutated R/R AML and 14 patients with FLT3-wild type (WT) R/R AML who received BHA regimen to further evaluate its clinical efficacy. Moreover, we performed in vitro studies to assess the synergistic effects of bortezomib and homoharringtonine and to elucidate the underlying mechanisms of this combination.

Materials & methods

Patients

Twenty-nine patients older than 18 years of age were treated with BHA regimen at the Second Hospital of Dalian Medical University in China between 2019 and 2025. Detailed information regarding the cytogenetic abnormalities and gene mutation profiles of the patients is provided in Supplementary Table 1. Eleven patients included were previously reported [21]. The study was reviewed and approved by the institutional review board and ethics committee of the Second Hospital of Dalian Medical University. It was conducted in accordance with Declaration of Helsinki.

Cell lines

MV4—11, Molm-13 and THP-1 cell lines were purchased from Wuhan Pricella Biotechnology Co., Ltd. and Nanjing COBIOER in China. Human leukemia cell line KG-1α was donated by Professor Ying Lu (Institute of Dermatology, Xinhua Hospital, Shanghai Jiao Tong University, China). K562 and HEK-293 T cell lines were purchased from American Type Culture Collection (ATCC) and from National Biomedical Experimental Cell Resource Bank, respectively. KG-1α and MV4—11 cell lines were maintained in IMDM supplemented with 20% FBS, K562 cell line in IMDM supplemented with 10% FBS, THP-1 cell line in RPMI 1640 supplemented with 10% FBS, Molm-13 cell line in RPMI 1640 supplemented with 20% FBS, 293 T cell line in DMEM supplemented with 10% FBS. These cell lines were all cultured in a humidified incubator at 37 °C containing 5% CO₂. FLT3-ITD cell lines: MOLM‑13, MV4‑11; FLT3-WT cell lines: THP‑1, KG‑1α, K562. All cells were routinely tested negative for mycoplasma contamination by PCR before experimentation.

Plasmid construction

Primers for FLT3-WT gene, FLT3-ITD gene and V5-Plvx-PURO vector were constructed by using Prime Design and corresponding tools (https://www.takarabio.com) and then were amplified using the following primers: forward, 5’-CGCGGGCCCGGGATCCATGCCGGCGTTGGCGCGC-3’, and reverse, 5’-TAGAGTCGCGGGATCCTACTTATCGTCGTCATCCTTGT-3’. A DNA fragment containing full-length FLT3-WT or FLT3-ITD sequence was inserted into plasmid vector V5-Plvx-Puro using an infusion kit (Takara, Cat. #638,943, Japan), produing plasmids separately named V5-Plvx-Puro-FLT3 and V5-Plvx-Puro-FLT3-ITD.

Establishment of 293 T stable cells

Plasmids including 1776 ng of target plasmid, 1333 ng of PSPAX2, and 889 ng of PMD2G were mixed with Lipofectamine 2000 (Thermo, Cat. #11,668,500, USA) and then added into 293 T cells incubated in serum-free medium for 6 h, then final concentration of FBS was supplemented. After 72 h, stable cells were selected using puromycin for over one week and then used for biological assays (Supplementary Fig. 2).

Western blotting

The cells were harvested and washed in PBS, after which they were lysed in RIPA buffer (Beyotime, Cat. #P0013B) supplemented with protease and phosphatase inhibitors (PMSF, Beyotime, Cat. #ST506; CocktailⅡ, MedChemExpress, Cat. #HY-K0022) on ice for 30 min. Afterward, the cells were centrifuged at 12,000 rpm for 30 min at 4 °C. The supernatants were harvested, and a Protein standard (5 mg/ml BSA) (Beyotime, Cat. #P0007) was used to determine the protein concentration. Protein (30 µg) was separated via 10% SDS-PAGE and transferred to a PVDF membrane. The membrane was blocked in 5% BSA in TBS-T for 1 h and incubated with primary antibody overnight at 4 °C. Afterward, the membrane was washed three times with TBS-T for 20 min per wash and incubated with secondary HRP-conjugated antibody for 1 h at room temperature. Finally, the membrane was washed three times with TBS-T for 20 min per wash. The proteins were visualized by using Pierce ECL Western (Thermo Fisher, Cat. #32,106). Protein levels were normalized against β-actin as internal loading controls when quantifying relative expression. Each experiment was repeated three times. The primary antibodies used as follows: FLT3 (abcam, Cat. #245,116, 1:1000), β-actin (Proteintech, Cat. #66,009—1, 1:20,0000), p-FLT3 (CST, Cat. #3461, 1:1000), STAT5 (CST, Cat. #25,656, 1:1000), p-STAT5 (CST, Cat. #4322, 1:1000).

Cell viability assay

A total of 7.5 × 10³ cells were seeded in a 96-well plate and treated with different concentrations of the designated drugs for 24, 48 or 72 h. Afterward, 10 μL of CCK8 solution was added to the plate for 2 h, and the absorbance of the plate was measured at 490 nm. Half maximal inhibitory concentration (IC50) values were calculated by using GraphPad Prism. Drug combination effects were assessed after 48 h of treatment using the CCK8 viability assay across multiple dose combinations. Based on the calculated 25% inhibitory concentration results [23, 24], different concentration gradients were set to conduct drug synergy experiments. As for assessing the synergistic effect between bortezomib and homoharringtonine, the combination index (CI) was calculated using CompuSyn software. Based on the Chou-Talalay theory [25], the inhibition rate fraction affected (Fa) values of the combined drugs at different concentrations was detected by the CCK8 method, and the combination indexs for three drug combinations, including bortezomib with homoharringtonine, bortezomib with Ara-C, and homoharringtonine with Ara-C were calculated using CompuSyn software (ComboSyn Inc., USA) to evaluate the synergistic effect in cell lines. The CI value in this paper is the synergistic index when Fa is 25%. Each experiment was repeated three times.

CI < 1: synergistic effect; CI = 1: additive effect; CI > 1: antagonistic effect.

Flow cytometry analysis of apoptosis

5 × 10⁵ cells were seeded in a 6-well plate for 12, 24 or 48 h. Cells were harvested and washed in PBS. Then cells were incubated with Annexin V-FITC (Annexin-V) and propidium iodide (PI) (Elabscience, Cat.#E-CK-A211) for 15 min at room temperature in the dark. Fluorescence intensities of FITC and PI were determined by flow cytometric analysis using a FACSCantoⅡ flow cytometer (Becton Dickinson, San Diego, California, USA). Each experiment was repeated three times.

RNA extraction, reverse transcription, and PCR

First, 1 × 10⁶ cells were harvested and washed in PBS and then lysed in 1 mL of TRIzol (Invitrogen, Cat. #AG21101), after which 200 μL of chloroform was added. The mixtures were left standing for 10 min and then centrifuged at 12,000 rpm for 15 min. The supernatants were collected, and an equal volume of isopropanol was added; the samples were then left standing for 10 min and centrifuged at 12,000 rpm for 15 min. The supernatant was removed, and absolute ethanol was added. After centrifugation, the supernatant was removed, and DEPC-water was added to dissolve the sample. RNA was reverse transcribed by using an M-MLV Reverse Transcription Kit (AG, Cat. #AG11711). cDNA (50 ng) was amplified via PCR by using 2 × Accurate Taq Master Mix (dye plus) (AG, Cat. #11,009). The amplified products were analyzed via agarose gel electrophoresis.

Statistical analysis

Data analysis included the use of Fisher’s exact test, Kaplan–Meier survival analysis, the log-rank test, Two-way ANOVA analysis and One-way ANOVA analysis. Statistical analysis was conducted by using the statistical software GraphPad Prism 9. Median overall survival (OS) and corresponding 95% confidence intervals (CIs) were estimated using the Kaplan–Meier method implemented in R software (version 4.5.2; R Foundation for Statistical Computing, Vienna, Austria) with the “survival” package. P < 0.05 was considered to indicate statistical significance.

Results

BHA regimen benefits patients with FLT3-mutated R/R AML

To further evaluate the efficacy of BHA regimen in treating FLT3-mutated R/R AML, 31 patients with R/R AML who received BHA regimen were retrospectively analyzed. Among all of the participants, 17 harbored FLT3 mutations; specifically, 11 had FLT3-ITD mutations, 2 had FLT3‐TKD mutations, 2 had both FLT3-ITD and FLT3-TKD mutations, and 2 presented with other subtypes of FLT3 mutations. Given that FLT3-ITD and FLT3-TKD represent the predominant clinically relevant FLT3 mutations in R/R AML [26], the two patients with other types of FLT3 mutation were excluded from the subsequent analyses (Table 1). The cohort consisted of 15 males and 14 females, with 16 patients who had relapsed AML and 13 who had primarily refractory AML. The median patient age was 48 (range, 20–69) years. In accordance with the French-American-British (FAB) classification, 13 patients were classified as AML-M2, 8 as AML-M5, and the remaining patients as M0, M1, and M4. According to risk stratification, 20 patients were classified as high risk, 5 as intermediate risk, and 4 as low risk. According to ELN2022 [27], a cytogenetic assay revealed that 2 patients had favorable karyotypes, 21 patients had intermediate karyotypes and 6 adverse karyotypes. The median OS was 8 months (95% CI, 5 months to not reached) in the FLT3-mutated group and 4 months (95% CI, 2 months to not reached) in the FLT3-WT group. The 1-year OS of patients in the FLT3-mutated group was 33.33%, while that of patients in the FLT3-WT group was 21.42%, with no statistically significant difference (Supplementary Fig. 1). The CR/CRi rates were determined to be 46.67% (7/15) in the FLT3-mutated group and 7.14% (1/14) in the FLT3-WT group. The CR/CRi rate in the FLT3-mutated group was significantly better than that in the control group (P = 0.035). These findings suggest that BHA regimen is more effective at achieving complete remission in patients with FLT3-mutated R/R AML (Table 1).

Table 1.

Patients’ characteristics. (N = 29)

Characteristics		Patients with FLT3 mutation (N = 15)	Patients with FLT3 wild type (N = 14)	P value
Sex, n (%)
	Male	5(33.33)	10(71.43)	0.07
	Female	10(66.67)	4(28.57)
Age (years)	Median (range)	46(20-69)	51.5(34-65)	-
FAB classification, n(%)				0.3
	M0	0	2(14.29)
	M1	1(6.67)	0
	M2	6(40)	7(50)
	M4	2(13.33)	3(21.43)
	M5	6(40)	2(14.29)
Risk stratification, n(%)				0.63
	Low risk	1(6.67)	3(21.43)
	Medium risk	3(20)	2(14.29)
	High risk	11(73.33)	9(64.29)
Disease status, n (%)				0.14
	Primary refractory	9(60)	4(28.57)
	Relapse	6(40)	10(71.43)
Karyotype, n (%)				0.12
	Favorable	1(6.67)	1(7.14)
	Intermediate	13(86.67)	8(57.14)
	Complex/poor	1(6.67)	5(35.71)
Disease, n (%)				0.99
	De novo	13(86.67)	13(92.86)
	Secondary to MDSa	2(13.33)	1(7.14)
CR/CRi, n (%)				0.035
	CR/CRi	7(46.67)	1(7.14)

^aMDS: myelodysplastic syndrome

Bortezomib and homoharringtonine synergistically inhibit proliferation of stable 293 T-FLT3-ITD cells

Given the observed clinical benefit of BHA regimen in patients with FLT3-mutated R/R AML, we next explored its potential molecular mechanism underlying the targeted efficacy against FLT3 mutations. To further investigate the targeting effect of the drug combination in BHA regimen on FLT3 mutations, we established stable 293 T cell models expressing FLT3-WT or FLT3-ITD (Supplementary Fig. 2). Bortezomib treatment resulted in consistently lower IC50 values in FLT3-ITD cells compared with FLT3-WT cells across all examined time points, with statistically significant differences between all time points (P < 0.05), demonstrating enhanced sensitivity of FLT3-ITD cells. These results revealed that FLT3-ITD cells were more sensitive to bortezomib than FLT3-WT cells (Fig. 1a-c), while there was no difference in homoharringtonine and Ara-C. To determine which drug combination could synergize with bortezomib, we performed combination assays with bortezomib plus homoharringtonine (BH) or bortezomib plus Ara-C (BA) using the established cell models. The results showed a synergistic effect of bortezomib and homoharringtonine in 293 T-FLT3-ITD cell lines (FLT3-WT-CI = 1.26, FLT3-ITD-CI = 0.84), while bortezomib combined with Ara-C show synergistic effects both in FLT3-ITD and FLT3-WT cell lines (FLT3-WT-CI = 0.85, FLT3-ITD-CI = 0.97). In contrast, homoharringtonine combined with Ara-C displayed strong synergy in FLT3-WT cell lines (FLT3-WT-CI = 0.66, FLT3-ITD-CI = 2.43), but show an antagonistic effect in FLT3-ITD cell lines (Fig. 1d-e). These results suggest that bortezomib plays a key role in BHA combination therapy, and homoharringtonine can enhance the proliferation inhibitory effect of bortezomib on FLT3-ITD cell lines.

Fig. 1.

Bortezomib and homoharringtonine synergistically inhibit the proliferation of stable 293 T-FLT3-ITD cells. (a-c) 293 T-FLT3-ITD stable cell lines are more sensitive to bortezomib-induced proliferation inhibition. 293 T-Plvx, 293 T-FLT3 and 293 T-FLT3-ITD stable cell lines were treated with bortezomib (a), homoharringtonine (abbreviation as HHT in figures) (b), and Ara-C (c) for 24, 48 and 72 h, and cell viability was determined by using the CCK8 method. IC50 values are mean values from three independent experiments ± SEM. Dose–response curves of Ara-C in 293 T stable cell lines in 24 and 48 h; IC50 values were determined by nonlinear regression (GraphPad Prism), with ND (not determined) indicating that 50% inhibition was not achieved within the tested range. (d-e) Bortezomib combined with homoharringtonine synergistically inhibited the proliferation of stable 293 T-FLT3-ITD cell lines; however, the synergistic effect of homoharringtonine combined with Ara-C was not detected. 293 T-Plvx, 293 T-FLT3 and 293 T-FLT3-ITD stable cell lines were treated with homoharringtonine combined with Ara-C, bortezomib combined with homoharringtonine, and the combination of homoharringtonine, Ara-C and bortezomib for 48 h, after which cell viability was determined by using the CCK8 method. The CI was calculated using CompuSyn software. The CI value is the synergistic index when Fa is 25%. Each experiment was repeated three times. CI values are mean values from three independent experiments ± SEM.

To further confirm the synergistic effect of bortezomib and homoharringtonine (BH) in leukemia cell lines, we performed combined drug sensitivity tests in FLT3-ITD cell lines (Molm-13 and MV4—11) and FLT3-WT cell lines (K562, KG-1α, THP-1) (Fig. 2a-b, Supplementary Fig. 3). Experimental results showed that bortezomib combined with homoharringtonine exhibited consistent synergistic effects in both FLT3-ITD cell lines MV4—11 and Molm-13 (MV4—11-CI = 0.59, Molm-13-CI = 0.36, both < 1). However, the synergistic effects varied among FLT3-WT cell lines, showing synergy in K562 and KG-1α (K562-CI = 0.65, KG-1α-CI = 0.9, both < 1), but antagonism in THP-1 (THP-1-CI = 4.54, > 1). These results suggest that the synergistic effect of bortezomib combined with homoharringtonine is consistent across different FLT3-ITD cell lines (Fig. 2a). However, the synergistic effect of homoharringtonine combined with Ara-C was inconsistent across different cell lines, showing synergy in MV4—11, K562, and KG-1α (MV4—11-CI = 0.52, K562-CI = 0.79, KG-1α-CI = 0.16, all < 1), but antagonism in Molm-13 and THP-1 (Molm-13-CI = 1.06, THP-1-CI = 1.37, both > 1) (Fig. 2b). In summary, the synergistic effect of bortezomib combined with homoharringtonine in the two FLT3-ITD cell lines was generally stronger than in the three FLT3-WT cell lines.

Fig. 2.

Bortezomib combined with homoharringtonine synergistically inhibits the proliferation of FLT3-ITD cells lines. (a, b) Bortezomib combined with homoharringtonine synergistically inhibited the proliferation of FLT3-ITD cell lines. The synergistic effect of homoharringtonine combined with Ara-C and bortezomib combined with homoharringtonine was detected in the K562, KG-1α, THP-1, MV4-11 and Molm-13 AML cell lines. After 48 h of treatment, cell viability was determined by using the CCK8 method. The CI was calculated using CompuSyn software. The CI value in this paper is the synergistic index when Fa is 25%. Each experiment was repeated three times. CI values are mean values from three independent experiments ± SEM

CI < 1: Synergistic Effect. The inhibitory effect produced by the combination of drugs is greater than the sum of the effects of the two drugs used individually.

CI = 1: Additive Effect. The effect of the combined drugs is equal to the theoretical additive effect of the two drugs used individually.

CI > 1: Antagonistic Effect. The effect of the combined drugs is less than the theoretical additive effect of the two drugs used individually.

Combination treatment exerts synergistic antileukemic activity via apoptosis induction

To evaluate the synergistic antileukemic effect of bortezomib and homoharringtonine on FLT3-ITD cell lines, we investigated whether the combination treatment could synergistically induce apoptosis in AML cell lines. K562, KG-1α, MV4—11, and Molm-13 cell lines were treated with bortezomib (1, 2, and 4 nM) alone for 12, 24, and 48 h (Fig. 3a-b). Annexin-V/PI staining demonstrated that bortezomib promoted apoptosis in FLT3-ITD cell lines (MV4—11 and Molm-13 cell lines), with the strongest proapoptotic effect occurring at 4 nM (P < 0.01). In contrast, bortezomib did not induce apoptosis in FLT3-WT cell lines (K562 and KG-1α cell lines). When bortezomib (4 nM) was combined with homoharringtonine (1 nM) for 48 h, the apoptotic response in FLT3-ITD cell lines was superior to either of the two agents alone (P < 0.01) (Fig. 3c-d). However, the combination therapy did not induce apoptosis in FLT3-WT cell lines. These results indicate that the combination of bortezomib and homoharringtonine synergistically induces apoptosis in FLT3-ITD cell lines.

Fig. 3.

Bortezomib combined with homoharringtonine synergistically promotes apoptosis in FLT3-ITD cells lines. (a) Compared with FLT3-WT cell lines (KG-1α and K562), FLT3-ITD cell lines (MV4—11 and Molm-13) are more sensitive to bortezomib-induced cell death. KG-1α, K562, MV4-11 and Molm-13 AML cell lines were treated with 1, 2 or 4 nM bortezomib for 48 h, after which they were subjected to Annexin-V/PI staining and flow cytometry analysis to assess apoptosis. (b) Bortezomib promotes apoptosis in FLT3-ITD cell lines in a concentration-dependent manner (but not in FLT3-WT cells lines). Apoptosis was assessed at 0, 12, 24 and 48 h after treatment with 1, 2 or 4 nM bortezomib, and the results are plotted as a line graph. (c, d) 1 nM homoharringtonine combined with different concentrations of bortezomib synergistically promotes apoptosis in FLT3-ITD cell line lines. Cells were treated with 1 nM homoharringtonine alone or in combination with different concentrations of bortezomib for 48 h, after which they were subjected to Annexin-V/PI staining and flow cytometry analysis to assess apoptosis. The apoptosis results are plotted as a bar graph. In the Annexin-V/PI scatter plot, we statistically analyzed the apoptosis data of Q1-UR plus Q1-LR, that is, the values of early apoptosis plus late apoptosis. Each experiment was repeated three times. (*P < 0.05, **P < 0.01)

Bortezomib is involved in the induction of FLT3-ITD degradation, and the combination of bortezomib with homoharringtonine enhances FLT3-ITD downregulation

The direct degradation of proteins that drive disease onset or progression is an important targeted therapeutic strategy [28–30]. Previous studies have reported that bortezomib can degrade the FLT3-ITD protein [22]. To clarify the synergistic effect of bortezomib and homoharringtonine on the FLT3-ITD protein, stable 293 T cell lines were treated with bortezomib for 0, 12, and 24 h. Bortezomib significantly enabled the degradation of the FLT3-ITD protein after 12 h; however, the FLT3-WT protein accumulated (Fig. 4a). Consistently, in FLT3-ITD AML cell lines (MV4—11 and Molm-13), bortezomib promoted FLT3-ITD degradation after 12 h, with a more rapid effect being observed in MV4—11 cell lines. In contrast, no significant degradation of the FLT3 protein was detected in FLT3-WT cell lines (THP-1) following 24 h of bortezomib treatment (Fig. 4b). These results suggest that bortezomib specifically promotes the degradation of the FLT3-ITD protein.

Fig. 4.

Bortezomib degrades the FLT3-ITD protein, and the combination of bortezomib and homoharringtonine enhances the degradation of the FLT3-ITD protein. (a) Bortezomib decreases the level of FLT3-ITD protein, but no significant effect on the level of FLT3 protein was detected. 293 T stable cell lines were treated with 8 nM bortezomib for 0, 12 and 24 h. Cell lysates were collected, subjected to SDS-PAGE, and analyzed via Western blotting with appropriate antibodies. (b) Bortezomib decreases the level of FLT3-ITD protein but does not significantly affect FLT3 protein. MV4-11, Molm-13 and THP-1 AML cell lines were treated with 4 nM bortezomib for 0, 12 and 24 h. (c) Bortezomib degrades the FLT3-ITD protein, and homoharringtonine combined with bortezomib accelerates this degradation. MV4-11 AML cell lines were treated with 4 nM bortezomib, 1 nM homoharringtonine, or a combination of bortezomib and homoharringtonine for 0, 3, 6 and 9 h. Each experiment was repeated three times. Protein levels were normalized against β-actin as internal loading controls when quantifying relative expression

To assess whether homoharringtonine could influence bortezomib-induced protein degradation, MV4—11 cell lines were treated with bortezomib or homoharringtonine alone or with the two drugs in combination for 0,  3,  6 or  9 h. In MV4‑11 cell lines treated with homoharringtonine alone for up to 9 h, no appreciable change was observed in FLT3-ITD protein levels compared with untreated controls. By contrast, treatment with bortezomib alone or in combination with homoharringtonine resulted in a reduction of FLT3-ITD protein levels after 6 h, while the combination group exhibiting a rapid degradation rate (Fig. 4c). We examined the expression of FLT3-ITD protein in FLT3-ITD AML cell lines MV4—11 and Molm-13 after 9 h of treatment with bortezomib alone and in combination with homoharringtonine. We also added the autophagy inhibitor Chloroquine (CQ) to investigate whether FLT3-ITD protein was degraded via the autophagy pathway. The results indicated that the combination with the autophagy inhibitor inhibited protein degradation to some extent (Supplementary Fig. 4).

Discussion

FLT3 mutations are among the most common genetic alterations in AML, with 25—30% of AML patients harboring FLT3 mutations [31]. Although several induction remission regimens (including standard salvage chemotherapy and FLT3 inhibitors) are currently available, issues such as drug resistance and a lack of durable remission remain major challenges [10]. Therefore, we sought to identify a new reinduction regimen for R/R AML. We retrospectively analyzed 15 patients with FLT3-mutated R/R AML and 14 patients with FLT3-WT R/R AML treated with BHA regimen. After one course of therapy, the CR/CRi rate in FLT3-mutated R/R AML patients (including those with FLT3-ITD and FLT3-TKD mutations) was higher than that in patients without FLT3 mutations (46.67% vs. 7.14%; P = 0.035), suggesting the potential molecular selectivity of this regimen. Consistently, in vitro experiments confirmed that compared with FLT3-WT cell lines, FLT3-ITD cell lines demonstrated increased drug sensitivity to bortezomib. Moreover, bortezomib induced marked apoptosis and promoted degradation of the FLT3-ITD protein, whereas homoharringtonine enhanced these effects, supporting synergistic antileukemic activity.

Currently, reinduction therapy for FLT3 mutated RR/AML usually incorporates FLT3 inhibitors such as quizartinib or gilterinib. In the ADMIRAL trial [32], patients treated with gilteritinib achieved a CR/CRi of 34%, with a median OS of 9.3 months, whereas those in the salvage chemotherapy arm achieved a CR/CRi of 15.3% and an OS of 5.6 months. Notably, the survival benefit was attributed mainly to the longer duration of remission gained by gilteritinib. Like gilteritinib, quizartinib also resulted in a median OS of 6.2 months, whereas salvage chemotherapy resulted in a median OS of 4.7 months [33]. Furthermore, studies have recently shown that therapeutic strategies involving the combination of FLT3 inhibitors with azacitidine and venetoclax have improved outcomes, with composite remission rates of approximately 70—75% and a median OS of 10—15 months [8, 34]. In our study, BHA regimen yielded a CR/CRi rate of 46.67% and a median OS of 8 months in patients with FLT3-mutated R/R AML, indicating that the efficacy of BHA regimen is superior to that of FLT3 inhibitor–containing chemotherapy regimen, which gained a CR/CRi of 20—28.5% [8, 34, 35].

Bortezomib has been reported to regulate autophagy-related transcription factors, promoting lysosomal trafficking and degradation of proteins aberrantly expressed in AML [36]. Homoharringtonine, an alkaloid extracted from Cephalotaxus mannii, has long been used for AML for a long time in China [37, 38]. In addition, homoharringtonine inhibits protein synthesis by targeting ribosomal function, thereby reducing the levels of short-lived oncogenic proteins critical for leukemia cell survival [39–41]. Consistent with the findings from a previous study in which bortezomib promoted the degradation of the FLT3-ITD protein through autophagy pathways [22], our research also verified that the degradation of the FLT3-ITD protein can be induced by bortezomib; in contrast, this degradation was partially abolished by the addition of an autophagy inhibitor (chloroquine). Notably, bortezomib did not affect the degradation of the wild-type FLT3 protein in our study. Thus, the higher CR/CRi rate in response to BHA regimen is attributed mainly to the degradation of the mutated FLT3-ITD oncoprotein triggered by bortezomib and the synergistic effects of bortezomib combined with homoharringtonine; thus, the therapeutic mechanism is distinct from the inhibition of FLT3 activity.

The duration of remission could not be reliably assessed in our investigation because among the 7 patients who achieved CR/CRi, only 5 did not undergo allogeneic blood stem cell transplantation, and 2 underwent transplantation. Unfortunately, the small sample size of 5 is likely too small to provide a reliable estimate of remission durability; hence, a study with a larger sample size is needed to validate the effects of the BHA regimen on R/R AML patients carrying FLT3-ITD.

In summary, this study systematically supports the therapeutic potential of BHA regimen for FLT3-mutated R/R AML at both the clinical and cellular levels. Nevertheless, due to the limited sample size in this retrospective analysis, future research should involve larger multicenter cohorts with long follow-up periods. identification of the molecular mechanism underlying FLT3-ITD protein degradation triggered by bortezomib plus homoharringtonine is needed. Therefore, the BHA regimen may be a promising therapy for FLT3-ITD AML in the future.

Conclusion

This retrospective clinical study demonstrated that BHA regimen resulted in significantly higher CR/CRi rates in patients with R/R AML harboring FLT3 mutations than in those without FLT3 mutations. Mechanistically, bortezomib promoted apoptosis in FLT3-ITD cell lines and induced degradation of the FLT3-ITD protein, and this effect was further enhanced by concomitant treatment with homoharringtonine. In summary, clinical and in vitro evidence supports that the combination of bortezomib and homoharringtonine exhibits molecular selectivity in targeting FLT3-mutated R/R AML and thus represents a promising therapeutic option for this subgroup of patients.

Supplementary Information

Below is the link to the electronic supplementary material.

Authors contributions

WM and WFR performed the experiments，analyzed data, interpreted data, and drafted the manuscript; ZCT and HD collected data; YY and KZJ organized data; YJS and LJC designed the research and supervised the manuscript. All authors read and approved the final manuscript.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by grants from the Science and Technology Program Project of Liaoning Province (No.2025JH2/101800154), Scientific Research Public Welfare Project Fund for High-Quality Development of Public Hospitals, China Health Promotion Foundation (No.HXLX20240028) and Dalian Municipal Medical Science Research Program Project, 2022 (No. 2212018). "1 + X" Program Clinical Research Peak Project, The Second Affiliated Hospital of Dalian Medical University, 2024 (LCGF2024001).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.

Ethical approval

This study involved retrospective analysis of human subjects with written informed consent obtained from all patients in accordance with institutional ethical standards and the Declaration of Helsinki.

Consent to participate

Informed consent was obtained from all individual participants included in the study.

Consent to publication

The participants have consented to the submission of the article to the journal.