Prophylactic infusion of allogeneic double-negative T cells as immune modulators to prevent relapse in high-risk AML patients post-Allo-HSCT: a phase I trial

Abstract

Relapse remains a major challenge for high-risk acute myeloid leukemia (AML) patients following allogeneic hematopoietic stem cell transplantation (allo-HSCT). In our first-in-human Phase I trial (ChiCTR-1900022795), we have demonstrated that third-party donor-derived double-negative T cells (DNTs) are safe and effective for treating relapsed AML. This Phase I study aims to further evaluate the safety and efficacy of allo-DNTs in preventing relapse in AML patients post-allo-HSCT. Six high-risk AML patients received three infusions of off-the-shelf allo-DNTs at one-month intervals, administered 60 to 100 days post-allo-HSCT without lymphodepleting chemotherapy. No dose-limiting toxicity, DNT-related graft-versus-host disease (GvHD), or severe cytokine release syndrome (CRS) occurred. With a median follow-up of 20.9 months (range: 11.4–24.6), four patients (66.7%) remained in minimal residual disease (MRD)-negative complete remission (CR), with recurrence-free survival exceeding 24 months. Patients in remission showed increased CD8⁺ and CD4⁺ T cells, total DNTs, and higher frequencies of granzyme-secreting T cells, which were absent in relapsed patients. In vitro, co-culturing AML patient CD8⁺ T cells with allo-DNTs upregulated granzyme B and interferon-γ expression, indicating CD8⁺ T cell activation. These findings suggest that allogeneic DNT immunotherapy is a safe, promising strategy to prevent relapse in high-risk AML patients post-allo-HSCT by combining intrinsic antitumor activity with immune modulation.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40164-025-00680-1.

To the editor

Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is a standard treatment for high-risk acute myeloid leukemia (AML), yet about 40% of patients experience relapse [1–5]. Post-transplant interventions offer limited efficacy and carry high risks, including graft-versus-host disease (GvHD) and high treatment-related mortality [6]. Allogeneic double-negative T cells (allo-DNTs, CD3⁺CD4⁻CD8⁻), though rare (1–5% of peripheral leukocytes), show potent graft-versus-leukemia (GVL) activity without GvHD [7]. Our Phase I clinical trial (ChiCTR-1900022795) demonstrated that DNT therapy induced 50% complete remission (CR) rate in relapsed AML patients post-HSCT [8]. Here, we report a prophylactic trial of allo-DNT infusion in six AML patients post-allo-HSCT, demonstrating a favorable safety profile and early evidence supporting the potential of allogeneic DNT cell therapy as an immune-modulating approach to prevent relapse in high-risk AML patients following allo-HSCT.

Characterization of patients and DNTs

Six patients (median age 44 years; range 27–62) were enrolled (Additional file 1). The median prior treatment lines were 3 (range 1–8) before allo-HSCT (Table 1). Three patients were refractory/relapsed AML, one of which also possessed pre-transplant progressive disease and high-risk genetic mutations (IDH2 and TP53), and the other one had disease progression pre-transplantation. Two patients suffered history of myelodysplastic syndrome (MDS), one of which also had IDH2 mutation-positive pre-transplantation. The remaining one was multiparameter flow cytometry (MFC) MRD-positive pre-transplantation. One patient received haplo-HSCT, others underwent umbilical cord blood transplantation. Third-party allo-DNTs were administered without lymphodepletion, starting at a median of 92.5 days (range 64–100) post-HSCT. Each patient received three monthly infusions at escalating doses of 1 × 10⁸ and 1.5 × 10⁸ cells/kg.

Table 1.

Clinical characteristics of the patients

Pt No.	Dose Level (/kg)	Age	Sex	Risk Profile		Prior treatment	Transplant type	Time between transplant and DNT infusion (days)	Pre-infusion level (count/ml)
				2022 ELN Risk Category	High risk factors for recurrence				CD4+T	CD8+ T	DNT
01002	1 × 108	62	F	Adverse	Refractory AML; Disease progression before allo-HSCT; Genetic high-risk factors (IDH2+ & TP53+)	Aza + Venetoclax→CR(MRD+)→Aza+ Venetoclax ×2→relapse/NR→haplo-HSCT	haplo-HSCT	92	111	0	4550
01003	1 × 108	52	M	Adverse	History of MDS	MDS(IB-1)→secondary AML→Cladribine+ Ara-C + Venclexta→CR→UCBT	UCBT	100	245,386	690,142	26,433
01005	1 × 108	27	F	Favorable	Refractory AML	IA→CR1→IA→Ara-C×2→relapse→M + CLAG→CR→UCBT	UCBT	64	221,964	477,453	44,604
01007	1.5 × 108	35	M	Adverse	MRD positive before allo-HSCT	IA→CR→IA→HHT + Ara-C→Ara-C×2→UCBT (CBFβ-MYH11+)	UCBT	93	3195	3901	111
01008	1.5 × 108	35	M	Adverse	Refractory AML; Disease progression before allo-HSCT;	IA→CR1→IA→Ara-C×2→HHT + Ara-C→relapse→IDA + CLAG→NR→IDA + CLAG→NR→UCBT	UCBT	96	159,730	389,614	2285
01009	1.5 × 108	53	F	Adverse	History of MDS; MRD positive before allo-HSCT	Venetoclax (IDH2 0.64%)→CR (IDH2+)→IA→Ara-C→MRD+ (IDH2+)→UCBT	UCBT	85	192,616	510,483	6472

Abbreviations: IDH2, Isocitrate Dehydrogenase 2; haplo-HSCT, haploidentical hematopoietic stem cell transplantation; UCBT, umbilical cord blood transplantation; r/r AML, relapsed/refractory Acute Myeloid Leukemia; Aza, Azacitidine; Ara-C, Cytarabine; IA, Idarubicin (IDA) & Cytarabine; M + CLAG, Mitoxantrone & Cladribine & Cytarabine & G-CSF (granulocyte colony stimulating factor); CR, Complete remission; MDS-IB, Myelodysplastic syndromes with increased blasts; M, Male; F, Famale; NR, Non-remission; MRD, minimal residual disease; Pt, patient; ELN, European leukemia net; DNT, double negative T cell

Safety profiles

Mild infusion reactions (fever, headache, hypertension) occurred in 72.2% (13/18) of infusions, resolving within 24 h. Most patients experienced transient IL-6 and IL-10 elevation (Fig. 1B, Additional file 2: Figure S1). One patient developed mild cytokine release syndrome six days post-first infusion and resolved spontaneously within a week; another reported mild joint pain and resolved without intervention. A third developed toxic erythema on the lower limbs 34 days post-infusion, improving with topical glucocorticoids. Patient #01003 experienced mild chronic GvHD during the second infusion, which resolved without intervention and was thought to be unrelated to DNT infusion. No cases of DNT-related acute GvHD or neurotoxicity were observed (Additional file 2: Table S1).

Fig. 1.

Clinical outcome and kinetics of peripheral blood biomarkers following prophylactic infusion of allo-DNTs. (A) Swimmer plots showing the clinical outcomes of six patients after three doses of allo-DNTs treatment. CRi: incomplete CR. (B) Measurement of plasma IL-6. (C) Kaplan–Meier curves showing predicted 1-year OS and 1-year PFS post HSCT. (D) Total absolute number of GZMB⁺ CD4⁺ T and CD8⁺ T cells. (E-H) Absolute counts of donor-DNTs (E), total DNTs (F), recipient-CD4⁺ T cells (G), and recipient -CD8⁺ T cells (H)

Clinical outcomes

As of May 1, 2025, with a median follow-up of 20.9 months (range: 11.4–24.6), four patients (66.7%) maintained MRD-negative CR without further treatment (Additional file 2: Table S2), with the longest recurrence-free survival exceeding 24 months (Fig. 1A). The 1-year OS and progression-free survival (PFS) were 83.3% and 66.7%, respectively, and the 1-year relapse incidence was 33.3% (95% CI: 9.6–80.5%) (Fig. 1C), with no non-relapse mortality. Patient #01007 experienced molecular relapse 28 days post-final infusion, achieved MRD-negativity with interferon α2b, but relapsed and died 14.2 months post-HSCT. Patient #01002, with TP53 and IDH2 mutations, complex karyotype (43–46, XX, -3, del (5) (q11q31)), del(7q10), -17), experienced pre-transplant disease progression, relapsed 8.1 months post-HSCT and died.

PK and immunomodulatory functions of allo-DNTs in patients

Donor-derived DNTs were detected in peripheral blood 1–4 days post-infusion, persisting up to 28 days and, in two non-recurrent patients, up to 360 days (Fig. 1E, Additional file 2: Table S3). Non-recurrent patients exhibited higher cytokines levels and greater expansion of total and recipient-derived DNTs, CD4⁺, and CD8⁺ T cells post-infusion (Fig. 1F-H, Additional file 2: Figure S1). They had higher effector memory T cells (T_em) proportions and granzyme expression (Fig. 1D, Additional file 2: Figure S2). However, cytokine and immune cells levels were relatively low in recurrent patients. In vitro, co-culturing patient-derived CD8⁺ T cells with allo-DNTs upregulated IFN-γ and GZMB expression, immune activation-associated cytokines and pathways (Additional file 2: Figure S3, Table S4).

In conclusion, despite our small sample size and limited follow-up, this study highlights unique advantages of allo-DNT therapy over existing post-HSCT interventions. Allo-DNT infusions, administered without lymphodepleting preconditioning, minimize treatment-related toxicity and, as an off-the-shelf product, offer a ready-to-use option for relapse prevention. Additionally, DNTs possess a dual immune-regulatory function by enhancing the indirect GVL effect through CD8⁺ T cell activation while mitigating GvHD risk. Our findings underscore the promising potential of DNT cell infusions to prevent relapse in high-risk AML patients post-transplant.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

AML

Acute myeloid leukemia

allo-HSCT

Allogeneic hematopoietic stem cell transplantation

UCBT

Umbilical cord blood transplantation

DNTs

Double negative T cells

DLT

Dose-limiting toxicity

GvHD

Graft-versus-host disease

CRS

Cytokine release syndrome

CR

Complete remission

MRD

Measurable residual disease

OS

Overall survival

PFS

Progression-free survival

DLI

Donor lymphocyte infusion

GVL

Graft-versus-leukemia

PK

Pharmacokinetics

T_em

Effector memory T cells

IFN-γ

Interferon-gamma

GZM

Granzyme

MDS

Myelodysplastic syndrome

IDH2

Isocitrate Dehydrogenase 2

ELN

European leukemia net

Author contributions

Guangyu Sun was responsible for patient enrollment and treatment in the clinical trial. Xingchi Chen conducted data analysis and drafted the manuscript. Tianzhong Pan performed the detection and analysis of patient specimens. Kaidi Song analyzed and interpreted patient data related to hematologic malignancies and transplantation. Haicun Xie was responsible for cell preparation. Meijuan Tu, Wen Yao, Yaxin Cheng, Ziwei Zhou, and Dongyao Wang assisted in patient treatment. Yongsheng Han, Baolin Tang, Liming Yang, and Xiaoyu Zhu designed the experiments and revised the manuscript. All authors have read and approved the final manuscript.

Funding

This work was supported by the National Natural Science Foundation of China (grants # U23A20453, 82270223 and 82170209), Anhui Provincial Key Research and Development Project (grant # 2022e07020015), Anhui Health Research Project (grant # AHWJ2022a011), Anhui Provincial Department of Education Scientific Research Project (2023AH010079, 2022AH030128), Anhui Provincial Natural Science Foundation (2308085J09), International Cooperation Projects in Anhui Province (2023h11020005) and Ruichuang Biotechnology Co., Ltd., Shaoxing, China.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

This study was performed in accordance with the Declaration of Helsinki and was approved by the ethics committee of the First Affiliated Hospital of the University of Science and Technology of China.

Consent for publication

All authors are consent for publication.

Competing interests

Liming Yang & Haicun Xie are employed by Ruichuang Biotechnology Co., Ltd., Shaoxing, China. Liming Yang is inventor of several DNT technology-related patents and intellectual properties. The remaining authors declare no competing interests.