MRD-driven HDACi-based therapy post-transplant: a real-world study in high-risk acute myeloid leukemia

Abstract

Background

Relapse remains a major challenge limiting long-term survival in patients with acute myeloid leukemia (AML) after allogeneic hematopoietic stem cell transplantation (allo-HSCT). Measurable residual disease (MRD) is a critical independent prognostic predictor in high-risk AML. While MRD-driven prophylactic or preemptive interventions hold promise for relapse prevention, optimal therapeutic strategies remain lacking. We investigated the efficacy and safety of chidamide, a selective histone deacetylase (HDAC) inhibitor with dual epigenetic and immunomodulatory properties, within MRD-adapted post-transplant strategies in high-risk AML.

Methods

This pilot study analyzed 43 consecutive AML patients with high-risk AML, including pre-transplant relapsed/refractory or MRD-positive, or post-transplant MRD-positive who received Chidamide-based therapy after allo-HSCT between December 2020 and July 2023. Patients who were MRD-negative post-transplant (n=17) received prophylactic therapy, while those who were MRD-positive or relapsed post-transplant (n=26) underwent preemptive intervention. Primary endpoints included relapse-free survival (RFS) and overall survival (OS). Secondary endpoints included MRD negativity rates, toxicity (CTCAE v5.0), and changes in peripheral blood lymphocyte subsets. Statistical analyses employed Kaplan-Meier methods and Log-rank tests for survival, Chi-square/Fisher's exact tests for categorical variables, and paired t-tests for lymphocytes subsets.

Results

At a median follow-up of 742 days, the overall sustained remission rate was 60.4%. In the preemptive group, Chidamide-based therapy achieved an 88.5% MRD conversion rate from positive to negative (median time to conversion: 50 days), with 65% maintaining sustained negativity. The prophylactic group demonstrated a 58.8% sustained MRD-negative rate. The 6-month and 1-year RFS and OS rates for the entire cohort were 79%, 69.8% and 100%, 72.1%, respectively. No statistically significant difference in RFS or OS was observed between the preemptive and prophylactic groups (P>0.05 for both). Grade 3/4 hematologic toxicities occurred in 32.6% (neutropenia) and 20.9% (thrombocytopenia) of patients, with no GVHD exacerbation observed. Chidamide treatment significantly increased peripheral blood CD8+ T cells (P=0.033) while reducing CD4+ T cells (P=0.007), suggesting enhanced cytotoxic potential.

Conclusion

Chidamide-based MRD-driven therapy demonstrates promising durable MRD eradication and clinically relevant survival rates in high-risk AML patients post transplant, potentially mediated by its dual epigenetic and immunomodulatory effects, including T-cell modulation. While limited by its retrospective nature and single-center design, these real-world findings support further investigation of HDAC inhibitor-based MRD-directed algorithms in prospective studies (NCT06066905).

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s00277-026-07117-7.

Introduction

Acute myeloid leukemia (AML) is an aggressive hematologic malignancy with high rates of relapse and significant mortality [1]. Allogeneic hematopoietic stem cell transplantation (allo-HSCT) represents the most powerful curative modality for many AML patients, offering the potential for long-term disease-free survival and improved overall outcomes, particularly for those with intermediate or high-risk disease features [2].

However, relapse following allo-HSCT continues to be the predominant cause of treatment failure and mortality, significantly undermining the potential benefits of transplantation [3]. The risk of post-transplant relapse is markedly higher in patients with acute myeloid leukemia (AML) who are at increased risk of relapse after allogeneic hematopoietic stem cell transplantation (allo-HSCT). This includes those who did not achieve complete remission (CR) or measurable residual disease (MRD) negativity prior to transplant, as well as those who develop detectable MRD after transplantation [4, 5]. Patients within these categories experience substantially inferior relapse-free survival (RFS) and overall survival (OS) rates compared to those with more favorable pre- and post-transplant disease status [6]. Measurable residual disease has been firmly established as a critical and independent prognostic factor for predicting impending relapse and guiding post-transplant management in AML [7, 8].

However, a consensus on the optimal strategy for preventing relapse or preemptively treating MRD positivity in these patients after allo-HSCT is currently lacking [9]. While various post-transplant interventions are being investigated, including donor lymphocyte infusions (DLI), targeted therapies, and epigenetic modifiers, their efficacy, safety, and optimal application in the MRD-positive or high-risk setting remain areas of active research and debate [10].

Histone deacetylase inhibitors (HDACi) are a class of epigenetic agents with demonstrated anti-leukemic activity. Chidamide, a selective benzamide-type HDAC inhibitor, has shown promise in preclinical and clinical studies in AML. In vitro and in vivo studies have illustrated that Chidamide induces apoptosis and inhibits the proliferation of leukemia cells [11]. Furthermore, Chidamide possesses favorable immunomodulatory properties that could be beneficial in the post-transplant setting, including enhancing cellular immunity by upregulating co-stimulatory molecules like CD86 and boosting the anti-tumor activity of natural killer (NK) cells and cytotoxic T lymphocytes (CTLs) against AML cells [9]. Experimental evidence also supports a synergistic effect when Chidamide is combined with hypomethylating agents (HMAs) in inducing leukemia cell apoptosis [12], a combination that has demonstrated clinical efficacy in treating relapsed or refractory AML patients [13]. These multifaceted mechanisms suggest that Chidamide, potentially as part of a combination regimen, could serve as a valuable tool for eradicating residual leukemia cells and augmenting graft-versus-leukemia (GvL) effects to prevent relapse after allo-HSCT.

However, published data specifically evaluating the efficacy and safety of Chidamide-based therapy utilized in an MRD-driven prophylactic or preemptive manner for preventing relapse in high-risk AML patients following allo-HSCT are limited. In this context, we conducted a study to evaluate the efficacy and safety of a Chidamide-based regimen for these high-risk relpased AML patients driven by MDR after transplant.

Methods

Patient inclusion and treatment schedule

Between December 2020 and July 2023, 43 consecutive patients diagnosed with acute myeloid leukemia (AML) who underwent allogeneic hematopoietic stem cell transplantation (allo-HSCT) and subsequently received Chidamide-based therapy at our center were enrolled in this study. All included patients represented a high-risk cohort, Classified as high‑risk AML, defined by at least one of the following:

• Relapsed or refractory disease prior to allogeneic hematopoietic stem cell transplantation (allo‑HSCT).

• Presence of measurable residual disease (MRD) before allo‑HSCT (MRD‑positive pre‑HSCT).

• Development of MRD positivity after allo‑HSCT (MRD‑positive post‑HSCT).

Detailed baseline disease characteristics, including the distinction between primary (n = 40) and secondary (n = 3) AML, are summarized in Table 1.

Table 1.

Patient baseline demographics and disease characteristics

	N = 43	Prophylactic cohort(n = 17)	Pre-emptive/Salvage cohort(n = 26)
Sex, n (%)
Male	19 (44.2)	7(41.2)	12(46.2)
Female	24 (55.8)	10(58.8)	14(53.8)
Median age, years (range)	38 (17–62)	34 (17–59)	38 (19–62)
Disease Type, n (%)
De novo AML	40 (93.0)	16(94.2)	24(92.4)
Secondary AML	3 (7.0)	1(5.8)	2(7.6)
ELN 2022 Risk, n (%)
Favorable	10 (23.3)	4(23.5)	6(23.0)
Intermediate	19 (44.2)	8(47.1)	11(42.4)
Adverse	14 (32.6)	5(29.4)	9(34.6)
HSCT Type, n (%)
Haploidentical	18 (41.9)	8(47.0)	10(38.4)
Sibling	11 (25.6)	4(23.5)	7(27.0)
UBT	3 (6.9)	2(11.8)	1(3.8)
Unrelated	11(25.6)	3(17.7)	8(30.8)
Transplant Number, n (%)
1st	40 (93.0)	16(94.2)	24(92.4)
2nd	3 (7.0)	1(5.8)	2(7.6)
Status Before HSCT, n (%)
NR/Relapse	19(44.2)	9(52.9)	10(38.5)
MRD positive	24(55.8)	8(47.1)	16(61.5)
Status Before Chidamide, n (%)
MRD negative	17 (39.5)	17(100%)	0
NR/Relapse	6 (14.0)	0	6(19.3)
MRD positive	20 (46.5)	0	20(80.7), FCM-MDR=3, Mol-MRD=17
Treatment Regimen, n (%)
Chidamide monotherapy	23 (53.5)	14(82.3)	9(34.6)
Chidamide + HMAs	15 (34.9)	3(17.7)	12(46.2)
Chidamide + other drugs	5 (11.6)	0	5(19.2)

Note: Some patients may have both NR/Relapse and MRD positive status before HSCT/Chidamide initiation

Patients were retrospectively categorized into two groups based on their post-transplant MRD status and clinical presentation at the time of initiating Chidamide-based therapy: 1) Prophylactic Group (n = 17): This group comprised patients who were MRD-negative following allo-HSCT. Chidamide-based therapy was initiated as a prophylactic measure aimed at preventing future relapse and maintaining sustained remission.2) Preemptive/Salvage Group (n = 26): This group included patients who were either MRD-positive without overt morphological relapse (receiving preemptive therapy) or who had confirmed morphological relapse (receiving salvage therapy) after allo-HSCT. Chidamide-based therapy was initiated in these patients to target detectable disease.

The core treatment regimen consisted of oral Chidamide administered at a dose of 5 mg daily for six days per week. The planned duration of Chidamide therapy was a minimum of one year post-transplantation, or until disease progression, unacceptable toxicity, or patient withdrawal. Concomitant or adjunctive therapies, such as hypomethylating agents (HMAs), were permitted at the discretion of the treating physician, particularly in patients with detectable disease (i.e., those in the Preemptive/Salvage Group).

Definition and assessment of MRD

Measurable residual disease (MRD) status was a primary factor in patient stratification and outcome assessment. MRD was evaluated using multiparameter flow cytometry (MFC) for leukemia-associated immunophenotypes (LAIPs), and quantitative or digital droplet polymerase chain reaction (qPCR/ddPCR) or next-generation sequencing (NGS) for specific fusion genes (e.g., RUNX1::RUNX1T1, CBFB::MYH11, PML::RARA, BCR::ABL1) or somatic mutations. Samples for MRD assessment were typically obtained from bone marrow.

MRD positivity was defined based on the specific method used: MFC: Detection of a leukemia-associated immunophenotype at a level of ≥ 0.1% of nucleated cells, consistent with the recommendations from the European LeukemiaNet (ELN) MRD Working Party guidelines. Molecular (qPCR/ddPCR/NGS): Detection of the specific molecular marker above the established limit of detection or quantification for that assay, following published guidelines or laboratory standards.

MRD response and the duration of MRD negativity were key endpoints evaluated during the study period. For patients in the Preemptive/Salvage Group who were initially MRD-positive post-transplant, an MRD response (defined as conversion to negativity) required achieving MRD negativity on at least two consecutive assessments performed while receiving Chidamide-based therapy.

The duration of MRD negativity was calculated differently for the two treatment groups: 1)Prophylactic Group: Calculated from the initiation date of Chidamide-based therapy until the first documented detection of MRD positivity, morphological relapse, death, or treatment discontinuation for other reasons. 2)Preemptive/Salvage Group (for patients who achieved negativity): Calculated from the date of confirmed MRD negativity (defined as the first of the two consecutive negative assessments) until the first documented detection of MRD positivity, morphological relapse, death, or treatment discontinuation for other reasons.

Statistical analysis

Safety and tolerability were assessed by evaluating both hematological and non-hematological toxicities, which were graded according to the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0.

The primary outcomes analyzed were relapse-free survival (RFS) and overall survival (OS). RFS was defined as the time from the transplantation to the date of relapse, death from any cause, or the date of last follow-up for patients alive and relapse-free. OS was defined as the time from the transplantation to the date of death from any cause or the date of last follow-up for living patients. Secondary outcomes included measurable residual disease (MRD) conversion rates (for initially MRD-positive patients), duration of MRD negativity, and changes in peripheral blood immune cell subsets (including NK cells and T cell subsets) during the study period.

Statistical analyses were performed using SPSS software, version 27.0 (IBM Corp., Armonk, NY). Differences in categorical variables, such as MRD conversion rates and adverse event rates between groups, were analyzed using Pearson’s Chi-square test or Fisher’s exact test, as appropriate. Changes in peripheral blood NK and T cell percentages from baseline (pre-treatment) to post-treatment were evaluated using paired sample t-tests. Survival probabilities were estimated using the Kaplan-Meier method, and differences between groups were compared using the log-rank test. A two-sided P-value of ≤ 0.05 was considered statistically significant for all analyses.

Results

Patient characteristics

The study enrolled a cohort of 43 adult AML patients who presented with relapsed/refractory disease or measurable residual disease (MRD) positivity prior to receiving allo-HSCT and subsequently received Chidamide-based therapy post-transplant. The median age of enrolled patients were 38 years (range, 17–62 years), with 19 males and 24 females. Pre-transplant risk stratification according to the 2022 European LeukemiaNet (ELN) guidelines identified 10 patients with favorable risk, 19 with intermediate risk, and 14 with adverse risk disease.

Following allo-HSCT, patients were categorized into two groups based on their post-transplant disease status for the purpose of Chidamide-based intervention: (1) Prophylactic Group (n = 17): Patients who achieved MRD negativity post-transplant and received Chidamide as prophylactic therapy. (2) Preemptive/Salvage Group (n = 26): Patients who were either MRD-positive or in a relapsed/refractory state post-transplant and received Chidamide as preemptive or salvage therapy.

The Chidamide-based treatment regimens administered were diverse, including Chidamide monotherapy (n = 23), combination therapy with hypomethylating agents (HMAs) (n = 15), and combination with other therapeutic interventions (n = 5). Comprehensive baseline characteristics for the entire cohort are summarized in Table 1. The median time to chidamide-based initiation post-transplantation was 60 days(range,17-1108d), with a median treatment duration of 360 days(range,37-892d) Table 2.

Table 2.

Patient baseline demographics and disease characteristics

	N = 43
Sex, n (%)
Male	19 (44.2)
Female	24 (55.8)
Median age, years (range)	38 (17–62)
Disease Type, n (%)
De novo AML	40 (93.0)
Secondary AML	3 (7.0)
ELN 2022 Risk, n (%)
Favorable	10 (23.3)
Intermediate	19 (44.2)
Adverse	14 (32.6)
HSCT Type, n (%)
Haploidentical	20 (46.5)
Sibling	12 (27.9)
Unrelated	11 (25.6)
Transplant Number, n (%)
1st	40 (93.0)
2nd	3 (7.0)
Status Before HSCT, n (%)
MRD negative	5 (11.6)
NR/Relapse	19 (44.2)
MRD positive	25 (58.1)
Status Before Chidamide, n (%)
MRD negative	17 (39.5)
NR/Relapse	6 (14.0)
MRD positive	20 (46.5)
Treatment Regimen, n (%)
Chidamide monotherapy	23 (53.5)
Chidamide + HMAs	15 (34.9)
Chidamide + other drugs	5 (11.6)

Note: Some patients may have both NR/Relapse and MRD positive status before HSCT/Chidamide initiation

Impact of chidamide-based treatment on MRD response

This section details the impact of Chidamide-based therapy on measurable residual disease (MRD) status and subsequent outcomes in the cohort of 43 AML patients post-allo-HSCT(Fig. 1).

Fig. 1.

The impact of Chidamide-based therapy on measurable residual disease (MRD) status and subsequent outcomes in the cohort of 43 AML patients post-allo-HSCT

Preemptive/Salvage Group (n = 26)

In the Preemptive/Salvage Group, comprising patients who were MRD-positive or in overt relapse post-transplant, Chidamide-based treatment resulted in an MRD conversion rate from positive/relapsed status to negativity of 88.5% (23/26). The median time to achieve MRD negativity was 50 days (range: 11–224 days). Among these 23 responders, 65% (15/23) maintained sustained MRD negativity, with a median duration of sustained negativity of 563 days (range: 24-1178 days). The incidence of subsequent MRD recurrent(or relapse) among these 23 responders was 30.4%. Stratifying the Preemptive/Salvage Group by baseline disease status revealed differential initial responses. Patients just with MRD positivity at baseline (n = 20) achieved a significantly higher initial MRD conversion rate compared to those with overt morphological relapse (n = 6) (95% [19/20] vs. 33.3% [2/6], P = 0.005). However, among those who achieved MRD negativity, the rates of sustained MRD negativity were comparable between the two subgroups(65% vs. 50%, P = 0.644; Fig. 2A.

Fig. 2.

The efficacy of Chidamide-based therapy on MRD response and relapse. (A) MRD conversion rate and sustained negativity rate in preemptive group stratified by status before Chidamide (MRD + vs. Relapse). (B) Relapse incidence in the preemptive vs. prophylactic intervention groups

Prophylactic Group (n=17)

In the Prophylactic Group, consisting of patients who were MRD-negative post-transplant and received Chidamide preventatively, the rate of sustained MRD negativity was 58.8%. The incidence of MRD recurrence or relapse in this initially MRD-negative group was 41.2% (7/17), occurring at a median time of 285 days (range: 72–857 days) from Chidamide initiation.

Overall Relapse Incidence

In this study, with a median follow-up of 742 days (range, 184–2349 days), the cumulative relapse rates at 6 months, 12 months, and overall were 4.7%(2/43), 11.6%(5/43) and 23.3%(10/43). In the Preemptive/Salvage Group, among MRD-negative achievers, 5 cases of MRD recurrence occurred within 12 months of conversion. Comparing the overall incidence of relapse between the two intervention groups, the Preemptive/Salvage Group had a relapse incidence or MRD recurrence of 38.5% (10/26), while the Prophylactic Group had an incidence of 41.2% (7/17). No statistically significant difference in overall relapse incidence was observed between the two groups (P = 0.859; Fig. 2B).

In summary, Chidamide-based treatments demonstrated considerable efficacy in achieving MRD negativity, particularly in the preemptive setting for MRD-positive patients after transplant. While initial response rates differed based on baseline disease burden (MRD vs. overt relapse), sustained negativity rates among responders were similar.And the overall relapse incidence did not differ significantly between the preemptive/salvage and prophylactic strategies in this cohort.

Survival outcomes

Survival outcomes were assessed for the entire cohort of 43 AML patients. The median follow-up duration was 742 days (range, 184–2349 days) calculated from the date of allo-HSCT. The cumulative MRD-free survival rate, defined as the time from treatment initiation to MRD conversion, hematologic relapse, or death from any cause (whichever occurred first), was 48.8%. Relapse-free survival (RFS) was 67.4%, and overall survival (OS) was 72.0%.The MRD-free survival was 86.0% and 53.5% at 6 months and 1 year, respectively. And RFS was 79.0% and 69.8%, OS was 100% and 72.1%. Kaplan-Meier survival analysis revealed no statistically significant differences in either relapse-free survival or overall survival (P > 0.05 for both comparisons) between preemptive/salvage and prophylactic using Chidamide-based intervention. Survival curves for the two groups are presented in Fig. 3.

Fig. 3.

Kaplan-Meier survival curves for RFS and OS. (A) Relapse-Free Survival (RFS) for the entire cohort and by intervention group (Preemptive vs. Prophylactic). (B) Overall Survival (OS) for the entire cohort and by intervention group (Preemptive vs. Prophylactic)

Tolerability of chidamide-based treatment

The safety profile of Chidamide-based treatment was assessed. During the initial treatment cycles (first four cycles were specifically mentioned in the original data description, although not explicitly in the text), reversible Grade 3/4 neutropenia was observed in 32.6% (14/43) of patients, and Grade 3/4 thrombocytopenia occurred in 20.9% (9/43) of patients. Severe non-hematologic adverse events were reported in one patients experienced Grade 3/4 infections, and the other patient had Grade 3/4 diarrhea. All severe adverse events were managed and resolved. In addition, Grade 1/2 non-hematologic adverse events were reported in five patients, including fatigue, gastrointestinal disturbances, pneumonia, mild liver dysfunction, and erythema. Analysis of the overall rate of adverse events (any grade, any type) across the three main treatment regimen groups showed no statistically significant differences: Chidamide monotherapy (52.2%, 12/23), combination with hypomethylating agents (HMAs) (66.7%, 10/15), and combination with other drugs (80%, 4/5) (P = 0.427). Importantly, no new-onset graft-versus-host disease (GVHD) or worsening of pre-existing GVHD exceeding one grade was observed during Chidamide-based treatment.

Overall, these findings indicate that Chidamide-based treatments were generally well tolerated in this post-transplant setting, with a safety profile primarily characterized by manageable and reversible hematologic toxicities.

Correlation of T cells and outcomes

To investigate the potential impact of Chidamide-based treatment on immune cell, we evaluated changes in peripheral blood NK, T, and B cell percentages before and after treatment in the entire cohort. The percentages of NK cells did not show a statistically significant change post-treatment compared to pre-treatment levels (19.23 ± 17.26% vs. 22.10 ± 16.33%, P = 0.118,Fig. 4A). Similarly, the percentages of total CD3 + T cells (72.82 ± 21.10% vs. 70.78 ± 19.23%, P = 0.279,Fig. 4B) and CD19 + B cells (6.05 ± 10.02% vs. 6.88 ± 10.87%, P = 0.215, Fig. 4C) remained statistically unchanged after treatment. However, significant alterations were observed within the T cell subsets. The percentage of CD4 + T cells significantly decreased following Chidamide treatment (26.13 ± 16.45% vs. 17.04 ± 9.86%, P = 0.007, Fig. 5A), while the percentage of CD8 + T cells increased (48.83 ± 20.23% vs. 57.21 ± 19.64%, P = 0.033, Fig. 5B). Further analysis exploring the correlation between these observed changes in T cell subsets and clinical outcomes(e.g. MRD response and recurrence ) did not reveal statistically significant associations.

Fig. 4.

The changes in peripheral blood NK, T and B cell percentages before and after Chidamide-based treatment. A. The changes in peripheral blood NK cell.B.The changes in peripheral blood CD3 + T cell C.The changes in peripheral blood CD19 + B cell

Fig. 5.

Significant alterations within CD4 + and CD8 + T cell subsets. A. The changes in peripheral blood CD4 + T cell B.The changes in peripheral blood CD8 + T cell

Discussion

Relapse remains the leading cause of treatment failure and mortality in acute myeloid leukemia (AML) patients following allogeneic hematopoietic stem cell transplantation (allo-HSCT), particularly in high-risk subsets characterized by pre-transplant relapsed/refractory disease or persistent measurable residual disease (MRD) [1, 2]. Effective post-transplant strategies are urgently needed to reduce relapse risk and improve long-term outcomes in this challenging population. The present study evaluated the feasibility, safety, and efficacy of Chidamide-based therapy as a prophylactic or preemptive approach in a cohort of high-risk AML patients post-allo-HSCT. Our findings offer valuable insights into the potential role of Chidamide in this setting.

Current approaches to prevent relapse following allogeneic hematopoietic cell transplantation (allo-HCT) are transitioning from conventional chemotherapy to molecularly targeted agents and epigenetic modulators. FLT3 inhibitors (FLT3i) and MENIN inhibitors exhibit high on-target efficacy but are clinically actionable only in genetically defined subsets (e.g., FLT3-ITD or KMT2A-rearranged disease). While broadly active agents like venetoclax (BCL-2 inhibition) demonstrate wider applicability, its clinical utility is constrained by myelosuppression and acquired resistance mechanisms.Hypomethylating agents (HMAs, including decitabine and azacitidine) have shown some efficacy in delaying relapse, their impact on long-term outcomes in high-risk or MRD-positive patients is limited [13, 14]. For instance, the RELAZA study demonstrated only a median 8-month postponement of recurrence with azacitidine maintenance [13], and the RELAZA2 study reported a 12-month relapse-free survival of 46% in MRD-positive patients receiving preemptive azacitidine [15]. These results highlight the significant unmet need for more effective strategies, particularly for patients with detectable MRD, who face a substantially higher risk of overt relapse.

Advances in understanding AML pathogenesis and the interplay between leukemia cells and the immune system have spurred the development of novel targeted and immunomodulatory therapies. Combinatorial strategies are increasingly explored to leverage synergistic mechanisms [10, 11]. Beyond HMAs, combinations with donor lymphocyte infusion (DLI) [16], cytokines [17, 18], or immunomodulatory drugs like lenalidomide [19, 20] have been investigated post-HSCT. Chidamide, as a selective histone deacetylase (HDAC) inhibitor, modulates gene expression by inhibiting HDAC activity and altering chromatin structure. Hypomethylating agents (HMAs), such as azacitidine and decitabine, reduce DNA methylation by inhibiting DNA methyltransferases (DNMTs), thereby restoring the expression of tumor suppressor genes. The combination of these agents synergistically regulates the epigenetic state of acute myeloid leukemia (AML) cells, influencing tumor cell growth and immune responses [21–23]. Dual epigenetic therapy, targeting both DNA methylation and histone acetylation, represents a promising avenue [9, 12].

Emerging data suggest that combinatorial epigenetic therapy (azacitidine + chidamide, an HDAC inhibitor) may enhance graft-versus-leukemia (GVL) effects through synergistic chromatin remodeling, yet optimal dosing and scheduling require further validation.Histone deacetylase (HDAC) inhibitors, including Chidamide, are particularly attractive due to their multifaceted mechanisms: they can directly inhibit tumor growth, modulate the immune microenvironment, enhance T-cell cytotoxicity, and potentially balance the graft-versus-leukemia (GvL) effect with graft-versus-host disease (GvHD) [24].

Potential mechanisms of chidamide combined with HMAs in combating GVHD. Chidamide and hypomethylating agents (HMAs) may exert anti-GVHD effects through immune cell regulation, inflammatory factor suppression, and epigenetic modulation. In terms of immune cell regulation, they potentially modulate T-cell subset balance by increasing the proportion of regulatory T cells (Tregs), thereby suppressing effector T-cell activation. Tregs play a critical role in maintaining immune tolerance and inhibiting autoimmune responses [25]. Myeloid-derived suppressor cells (MDSCs) suppress GVHD through an interleukin-13-regulated, arginase-1-dependent mechanism [26]. Regarding inflammatory factor suppression, Chidamide may inhibit the release of pro-inflammatory cytokines, alleviating GVHD-associated inflammation. Acute GVHD is linked to cytokine release syndrome (CRS) [27]. Epigenetically, Chidamide and HMAs regulate immune-related gene expression through epigenetic mechanisms, influencing immune cell function and interactions. Epigenetic therapy may target Th1/Th17 polarization, thereby reversing immune escape and treating relapse of non-APL acute myeloid leukemia (AML) after allogeneic stem cell transplantation [28].

Preclinical studies have further demonstrated synergistic anti-AML effects when Chidamide is combined with DNA methyltransferase inhibitors [9, 29], providing a strong rationale for such combinations in the clinical setting. Clinical evidence has demonstrated the safety and efficacy of azacitidine combined with chidamide as maintenance therapy in high-risk acute myeloid leukemia (AML) patients following allogeneic hematopoietic stem cell transplantation (allo-HSCT). Preliminary results indicate that this regimen exhibits a favorable safety profile and preliminary efficacy [30, 31]. An interim analysis of a phase II clinical trial further suggests that maintenance therapy with azacitidine plus chidamide can reduce the relapse rate in high-risk AML patients after transplantation [30]. Additionally, a chidamide-based triple-drug combination regimen shows potential utility in reversing molecular relapse in AML1-ETO-positive acute myeloid leukemia patients following transplantation [32].

In this pilot study, we observed encouraging clinical outcomes with Chidamide-based therapy in a high-risk post-transplant AML cohort. In the preemptive setting, Chidamide-based treatment was associated with a high MRD negativity conversion rate (88.5%) and a notable proportion achieving sustained MRD negativity (65% among responders). Prophylactic treatment also yielded a favorable sustained MRD-negative rate (58.8%). Importantly, the 1-year relapse-free survival (RFS) and overall survival (OS) rates of 69.8% and 72.1%, respectively, in this high-risk patient population appear favorable when compared to historical data from similar cohorts treated with HMAs alone or other single agents [3, 15]. These survival outcomes suggest that Chidamide-based approaches may offer a more robust control of residual disease and improve long-term prognosis in this challenging setting, aligning with preclinical and clinical evidence supporting the augmentation of anti-leukemic effects by HDAC inhibitors in combination therapies [9–11].

Regarding tolerability, Chidamide-based treatment exhibited a generally manageable safety profile in the post-transplant setting. The most common Grade 3/4 adverse events were reversible hematologic toxicities, specifically neutropenia (32.6%) and thrombocytopenia (20.9%), which were effectively managed with supportive care. Severe non-hematologic events were infrequent and also resolved. Notably, the overall adverse event rates did not differ significantly among patients receiving Chidamide monotherapy versus combination regimens, suggesting that Chidamide can be safely integrated into combination strategies, particularly with HMAs. A critical finding in the post-transplant context is the absence of new-onset GVHD or significant worsening (> Grade 1) of pre-existing GVHD. This is highly favorable, indicating that Chidamide may selectively enhance GvL activity without exacerbating detrimental alloreactivity, a major challenge in post-transplant management.

The immune microenvironment post-HSCT significantly influences relapse risk; limited and exhausted T-cell immunity is a known contributor to immune evasion and relapse [33]. We observed a significant increase in the percentage of CD8 + T cells concurrently with a significant decrease in CD4 + T cells. This shift in the CD4/CD8 ratio is consistent with the known immunomodulatory effects of HDAC inhibitors, including Chidamide’s reported ability to enhance cytotoxic T-cell activity and potentially promote NK/CTL-mediated killing in preclinical models [34]. Other studies have similarly reported increases in cytotoxic T cell populations with Chidamide-containing regimens [35]. While we did not observe significant changes in NK cell percentages in this cohort, the prominent effect on T cell subsets highlights a potential mechanism by which Chidamide contributes to anti-leukemic activity in vivo, possibly by reinforcing the GvL effect primarily mediated by CD8 + T cells. The observed decrease in CD4 + T cells without an increase in GVHD is particularly intriguing and warrants further investigation into specific CD4 + helper T cell subsets (e.g., Treg, Th1, Th17) and their functional status post-treatment.

Despite these promising findings, our study has several limitations inherent to its retrospective, single-center design. The heterogeneity of treatment regimens (monotherapy vs. various combinations) and the relatively small sample size, particularly within specific subgroups (e.g., prophylactic vs. preemptive, different combination partners), limit the ability to draw definitive conclusions regarding optimal strategies or perform robust subgroup analyses. While we identified significant changes in T cell subsets, a more comprehensive analysis of T cell function, exhaustion markers, and other immune cell populations (e.g., B cells, NK subsets, myeloid-derived suppressor cells) would provide a deeper understanding of Chidamide’s immunomodulatory effects in this complex setting. Furthermore, longer follow-up is needed to confirm the durability of MRD negativity and remission, fully assess long-term toxicities, and characterize late relapse patterns.

In conclusion, this study provides compelling real-world evidence supporting the feasibility, manageable safety profile, and encouraging efficacy of Chidamide-based therapy for relapse prevention and preemptive treatment in high-risk AML patients post-allo-HSCT. The high rates of MRD eradication and favorable survival outcomes, coupled with the observed increase in cytotoxic CD8 + T cells and lack of GVHD exacerbation, suggest a potentially potent and selective anti-leukemic effect. These findings underscore the potential value of Chidamide as a component of MRD-directed post-transplant strategies and strongly support the need for larger, prospective, multi-center studies to validate these results, define optimal Chidamide-based regimens and treatment durations, and ultimately integrate this approach into standard clinical practice for high-risk AML patients after transplantation.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

JYW and XD designed the study. WL and PW analyzed the data and wrote the initial manuscript draft. CWL, LSH, XMC, and SJW conducted patient follow-up visits. XLL, JQT, and PLL were involved in evaluating and collecting clinical data. All authors read and approved the final manuscript.

Funding

This work was supported by the National Natural Science Foundation of China (Nos. 82200148, 82070128, 82070176); Guangdong Natural Science Foundation (2021A1515011436); the National Key R&D Program of China (No. 2017YFE0131600); and NSFC Incubation Project of Guangdong Provincial People’s Hospital (KY0120220026).

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics statement

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of Guangdong Provincial People’s Hospital. Informed consent was obtained from all individual participants included in the study.

Competing interests

The authors declare no competing interests.