A phase I, single-arm, open-label, dose-escalation, multicenter study of SAR445419, an off-the-shelf, ex vivo expanded allogeneic natural killer cell product, in participants with relapsed or refractory acute myeloid leukemia

Abstract

Background

SAR445419 is an investigational, off-the-shelf, allogeneic NK cell therapy derived from donor peripheral blood mononuclear cells and expanded ex vivo using PM21 particles.

Methods

This multicenter, Phase 1, dose-escalation study (NCT05712278) evaluated optimal dose(s), safety, and tolerability of SAR445419 in adults (aged ≥ 18 years) with relapsed/refractory acute myeloid leukemia (R/R AML). Following lymphodepleting chemotherapy (fludarabine 30 mg/m²/day and cytarabine 2 g/m²/day for 5 days), participants received six intravenous SAR445419 (1 × 10⁹ and 3 × 10⁹ NK cells/dose) doses, starting with the lower dose. The primary endpoint was the incidence of dose-limiting toxicities (DLTs), and key secondary endpoints included adverse events (AEs), hematological recovery, hematopoietic stem cell transplantation, and response rate.

Results

Of the 12 planned participants, 7 patients were enrolled, of whom 6 received SAR445419 before sponsor decided early study termination for reasons unrelated to safety or efficacy. No DLTs were observed. All participants experienced treatment-emergent adverse events (TEAEs); six had grade ≥ 3 TEAEs; and four had serious adverse events (SAEs). Two SAR445419-related SAEs (infusion-related reactions: both grade 2) were managed with supportive care. All participants experienced grade 3 anemia and grade 4 thrombocytopenia and neutropenia. Five deaths were reported, all due to disease progression, none related to SAR445419. No clinical responses were observed.

Conclusions

This study demonstrated the overall safety of off-the-shelf ex vivo expanded allogeneic NK cells in patients with R/R AML. The successful manufacturing and distribution support the feasibility of donor-derived off-the-shelf NK cells in a clinical setting. Further investigations are required to improve the clinical efficiency of NK cells.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00262-026-04333-y.

Introduction

Acute myeloid leukemia (AML) is the most common form of acute leukemia in adults and is the leading cause of leukemia-related fatality in the United States [1]. Relapse rate for elderly patients with acute myeloid leukemia (AML) after allogeneic hematopoietic stem cell transplantation (HSCT) generally ranges from 30 to 40%, depending on individual patient and disease factors [2, 3]. Available salvage therapies do not induce remission in many patients with relapsed/refractory (R/R) AML, and many patients are not eligible for potentially curative HSCT. This highlights the need for novel therapies to improve responses, overall survival, and quality of life in patients with R/R AML.

Natural killer (NK) cells are innate immune effector cells that exhibit potent antigen-independent cytotoxicity against virally infected or transformed cells, thereby performing both cytotoxic and immunoregulatory functions [4, 5]. Human cluster of differentiation (CD)56⁺ NK cells express various activating and inhibitory receptors, including the FcγRIII receptor CD16, human leukocyte antigen (HLA) Class 1, allele-specific killer-cell immunoglobulin-like receptors (KIRs), and NKG2D and NK receptors (NKRs), such as NKp30 and NKp46, but lack T-cell receptor and CD3 expression [4, 6–9]. Inhibitory KIRs bind with self-HLA Class 1 to protect healthy cells from NK cell-mediated lysis in the absence of activating signals. However, the downregulation of HLA Class 1 enhances tumor cell susceptibility to NK cell-mediated lysis, a phenomenon called “missing self.” Tumor cells often upregulate NK cell-activating ligands, including MHC class I polypeptide-related sequence A/B (MICA/B), Fas, trail receptor, B7-H6, and poliovirus receptor (PVR) [10, 11]. The balance between activating and inhibitory receptor ligands on a target cell regulates NK cell-mediated killing. Besides directly killing cells by releasing perforins or granzymes, NK cells also secrete immunoregulatory cytokines, such as interferon gamma (IFNγ), to support adaptive immune responses [4, 12].

The safety and efficacy of adoptive NK cell therapies have been explored for more than two decades in multiple clinical settings [13]. Several prospective studies on various adoptive allogeneic NK cell therapies have demonstrated their clinical feasibility in treating patients with R/R AML and high-risk patients with AML [14–18]. These studies suggest that the adoptive transfer of ex vivo expanded NK cells is safe and generally not associated with severe infusion related reactions (IRR), cytokine release syndrome (CRS), immune effector cell-associated neurotoxicity syndrome (ICANS), or graft-versus-host disease (GVHD); in the context of HSCT, no graft rejection was noted [19–25]. Many adoptive NK cell therapy products are manufactured for each patient by utilizing haploidentical donors. Ex vivo expansion of NK cells is required to manufacture sufficient NK cells from one donor apheresis collection to support the development of a scalable, off-the-shelf product. Most methods use feeder cells (FCs) (i.e., the K562 cell line), cytokines (i.e., interleukin [IL]-2, IL-15, IL-18, or IL-21), or both to stimulate NK cell activation and proliferation [26–28]. Ciurea et al. used the FC21 cell line that expresses membrane-bound (mb) IL-21 and the co-stimulatory molecule 4-1BB ligand (4-1BBL) to expand NK cells (FC21-NK cells). These cells display a highly consistent, yet unique phenotypic and functional profile characterized by potent inflammatory cytokine production and cytotoxicity [14]. FC21-NK cells exhibit a unique receptor expression characterized by high expression of CD56 (CD56 superbright), CD16, and NKG2D [29, 30]. A non-cellular, membrane-particle preparation (PM21) was developed for feeder-free expansion by lysing the plasma membranes of FC21 cells. Similar to FC21, PM21 plasma membrane particles express mbIL-21 and 4-1BBL. FC21-NK and PM21-NK cells are functionally and phenotypically similar in terms of surface receptor expression, in vitro cytotoxicity, cytokine production, and in vivo antitumor activity [30–32].

SAR445419 is an investigational, off-the-shelf allogeneic donor peripheral blood mononuclear cell (PBMC)-derived ex vivo expanded NK cells cancer immunotherapy that was being developed for use in hematologic malignancies. The NK cells in SAR445419 are expanded ex vivo using PM21 particles (Figs. 1 and 2). Ex vivo, PM21 expansion of NK cells facilitates the generation and administration of a large number of hyperactive NK cells, thereby maximizing the therapeutic potential for patients [30, 31, 33]. It was hypothesized that the administration of mature, fully functional NK cells expanded ex vivo will have anti-tumor activity, induce deep remission and allows patients to undergo curative allogenic HSCT or have improved survival outcomes when allogenic HSCT is not a suitable option. This study aimed to determine the optimal dose(s) and assess the safety and tolerability of SAR445419, following fludarabine and cytarabine lymphodepleting therapy in adult participants with R/R AML (NCT05712278).

Fig. 1.

Off-the-shelf NK cell manufacturing. CD3 cluster of differentiation, CRF corticotropin-releasing factor, PM21 plasma membrane particles

Fig. 2.

Off-the-shelf NK cell treatment. AML acute myeloid lymphoma, DL dose level, HCT hematopoietic cell transplantation

Methods

Study design and participants

This multicenter, Phase 1, single-arm, dose-escalation study enrolled adult participants (aged ≥ 18 years) who had relapsed or primary refractory AML according to the World Health Organization classification 2022. Eligible participants were those who had relapsed AML with or without prior allogeneic stem cell transplantation, including isolated central nervous system or extramedullary disease, and ≥ 1 prior line of therapy for AML that included chemotherapy, hypomethylating agents, venetoclax, or targeted therapy. Participants with an Eastern Cooperative Oncology Group (ECOG) performance status (PS) score of > 2, a history of active or chronic autoimmune conditions, or a secondary primary malignancy that required active therapy within the past 5 years were excluded from the study. Additional inclusion and exclusion criteria are provided in Supplementary Table S1.

The study was conducted in compliance with the international ethics guidelines, including the Declaration of Helsinki and the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) guidelines for Good Clinical Practice (GCP), as well as all applicable laws, rules, and regulations. The study was approved by the Institutional Review Board at each site, and written informed consent was obtained from all study participants prior to their enrollment.

The study included a screening period within 21 days before chemotherapy, a treatment period from the first chemotherapy administration to Day 56, and a survival follow-up period wherein adverse events (AEs) and serious adverse events (SAEs) were monitored every 30 days ± 7 days until resolution, with survival status being collected every 2 months ± 14 days. The study was completed when all participants had been followed up for 12 months or until death, loss of follow-up, consent withdrawal, or study termination, to evaluate remission duration and the 1-year survival rate. The dose escalation was guided by the modified toxicity probability interval-2 (mTPI-2) design [34]. The graphical representation of the study design/treatment plan is illustrated in Fig. 3.

Fig. 3.

Study design. EOT, end-of-treatment; DLT, dose-limiting toxicity; D, day; DL, dose level; NK, natural killer

Study treatment

Eligible participants were administered fludarabine (30 mg/m²/day) and cytarabine (2 g/m²/day) for 5 days, followed by six doses of intravenous (IV) SAR445419 administered three times weekly over 2 weeks. The first administration of SAR445419 was at least 36 h following the completion of chemotherapy. Two escalating dose levels of SAR445419 were explored: 1 × 10⁹ and 3 × 10⁹ NK cells/administration. Before the administration of SAR445419, participants were pre-medicated with diphenhydramine hydrochloride 25 to 50 mg intravenously or orally or equivalent and acetaminophen 650 to 1000 mg intravenously or orally.

Manufacturing of SAR445419

All products used in this study were manufactured by Sanofi in Framingham (45 New York Avenue), MA, United States using phase appropriate processes and analyses. Universal donors were selected using a Sanofi proprietary algorithm based on activating KIR, that enhances cancer targeting by exploiting KIR-HLA mismatches, boosting NK cell activation, and minimizing the risk of GVHD [35, 36]. Donors were selected for NK-specific characteristics, including the number of licensed KIRs and the highest likelihood of KIR–ligand mismatch, and for enhanced responsiveness to targets lacking complementary HLA. After apheresis and freezing of the leukopack from the selected donors at the collection sites, NK cells were selected by depletion of the CD3⁺ cells and then activated by the addition of the PM21 preparation at the beginning of the culture in the Prodigy (Miltenyi Biotec). On day 7, the cells were restimulated with another addition of PM21 and then transferred to Sartorius wave bags. They were grown for approximately 6 more days until culture volume reached up to 15 L. The ex vivo expanded cells were then harvested, washed, and concentrated in a LOVO and filled in bags containing 23 mL of cell suspension at 5 × 10⁷ cell/mL. Those bags were immediately frozen at − 140 C in a CryoStore bag and shipped to various clinical centers (Figs. 1 and 2). This process is reported in another publication (manuscript under preparation). The lot-to-lot variability was carefully monitored against specifications for cell viability (all batches above 80%), CD3⁺ cell count maximum (all batches below 5 × 10⁵ CD3⁺ cells/mL), purity (CD56⁺CD3⁻ all batches above 90% of all viable cells), and safety (mycoplasm spp, sterility endotoxin, and residual impurities), all within specifications and monitored for cytotoxicity (using K562 cell model), cell markers (CD19, CD16, CD335, and CD314), and cytokine production (TNFα, INFγ). All batches harvested were within specifications and each batch produced an average of 22 bags of 1e9 cells. (Supplementary Table S2).

Study assessments

The primary objective of this study was to determine the candidate dose(s) of SAR445419 for further development by assessing the occurrence of dose-limiting toxicities (DLTs). The DLT evaluation period commenced from the initiation of chemotherapy to 28 days after the first administration of SAR445419, starting at the lowest dose, after chemotherapy with fludarabine and cytarabine. Key secondary objectives included the overall safety and tolerability of SAR445419, time to neutrophil and platelet recovery, the rate of HSCT prior to additional AML treatment, and measures of preliminary antileukemic activity, including the composite complete remission (CRc) rate and the duration of response per the European LeukemiaNet (ELN) response criteria (Supplementary Table S3). CRc was defined as complete remission (CR) plus CR with incomplete hematologic recovery (CRi) plus CR with partial hematologic recovery (CRh).

Statistical analysis

The DLT-evaluable population included participants who had received chemotherapy, were observed at least until Day 28 (the DLT evaluation period), and had received at least five of the six scheduled doses of SAR445419 within 21 days. Safety analyses were performed based on the dose level, chemotherapy only, and overall; focused on descriptive statistics. The safety population included all participants who received chemotherapy. AEs were coded as per the Medical Dictionary for Regulatory Activities (MedDRA) and graded per the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI-CTCAE version 5.0). Hematology and clinical chemistry results were graded by NCI-CTCAE v5.0., when applicable. The number and percentage of participants with laboratory abnormalities were reported for baseline and treatment-emergent periods. The efficacy population included participants who received ≥ 4 out of the six planned administrations of SAR445419 within 21 days and had ≥ 1 evaluable post-baseline tumor assessment of response or those who discontinued early. Key endpoints (CRc rate, HSCT rate, and survival metrics) were summarized descriptively.

Results

Patient disposition and baseline characteristics

This study was conducted between June 16, 2023, and March 12, 2024, at three centers in the United States. Approximately 12 DLT-evaluable participants were planned to be enrolled to the study, but due to early termination of the study based on the sponsor decision, for reasons unrelated to the safety or efficacy of SAR445419, only seven participants were enrolled and analyzed. Patient disposition and a summary of the analysis population are depicted in Fig. 4. The median (range) age of the participants was 62 (35–72) years. Four (57.1%) participants were male. Most of the participants were White (6 [85.7%]). A total of five participants had an ECOG PS score of 1 (3 [42.9%]) and 2 (2 [28.6%]). The median (range) number of prior AML treatments was 2 (1–4), and the median (range) percentage of bone marrow blasts at the baseline was 23% (10%–88%). Two participants had previously undergone HSCT. The median (range) time from the initial diagnosis of AML to the first dose of SAR445419 was approximately 6.4 (2.4–38.4) months (Table 1).

Fig. 4.

Patient disposition and population analysis

Table 1.

Patient demographics and disease characteristics

	Chemo + dose level 1 1 × 109 NK cells/administration (n = 3)	Chemo + dose level 2 3 × 109 NK cells/administration (n = 3)	Chemotherapy only (n = 1)	All (N = 7)
Age, years median (range)	62.0 (59–69)	43.0 (35–72)	68.0 (68–68)	62.0 (35–72)
Age group, n (%)
< 65	2 (66.7)	2 (66.7)	0	4 (57.1)
65 to < 75	1 (33.3)	1 (33.3)	1 (100)	3 (42.9)
Sex, n (%)
Male	2 (66.7)	1 (33.3)	1 (100)	4 (57.1)
Female	1 (33.3)	2 (66.7)	0	3 (42.9)
Race, n (%)
White	3 (100)	2 (66.7)	1 (100)	6 (85.7)
Asian	0	1 (33.3)	0	1 (14.3)
Ethnicity, n (%)
Hispanic or Latino	1 (33.3)	0	0	1 (14.3)
Not Hispanic or Latino	2 (66.7)	3 (100)	1 (100)	6 (85.7)
ECOG PS, n (%)*
1	1 (33.3)	1 (33.3)	1 (100)	3 (42.9)
2	1 (33.3)	1 (33.3)	0	2 (28.6)
Karnofsky PS, n (%)
90	1 (33.3)	1 (33.3)	0	2 (28.6)
80	0	0	0	0
70	0	1 (33.3)	0	1 (14.3)
Total number of prior treatment lines, median (range)	2 (1–3)	2 (1–4)	4 (4–4)	2 (1–4)
Prior HSCT, n (%)
Yes	1 (33.3)	0	1 (100)	2 (28.6)
No	1 (33.3)	3 (100)	0	4 (57.1)
Disease characteristics at baseline
Time from initial diagnosis to first dose of SAR445419, months median (range)	5.64 (2.4–38.4)	7.2 (4.8–25.2)	NC	6.4 (2.4–38.4)
Blasts/Total cells (%) at baseline, median (range)	20.52 (10–53.3)	37 (23–88)	10 (10–10)	23 (10–88)
Category, n (%)
< 30%	2 (66.7)	1 (33.3)	1 (100)	4 (57.1)
30% to 50%	0	1 (33.3)	0	1 (14.3)
> 50%	1 (33.3)	1 (33.3)	0	2 (28.6)
White blood cell counts (× 109/L) at baseline, mean (SD)	3.83 (5.51)	3.97 (2.30)	0.80 (NC)	3.46 (3.64)
White blood cell counts (× 109/L) at baseline by category, n (%)
0 to ≤ 5	2 (66.7)	2 (66.7)	1 (100)	5 (71.4)
5 to ≤ 10	0	1 (33.3)	0	1 (14.3)
10 to ≤ 20	1 (33.3)	0	0	1 (14.3)
Normal	1 (33.3)	1 (33.3)	0	2 (28.6)
Abnormal	2 (66.7)	2 (66.7)	1 (100)	5 (71.4)
Extramedullary leukemic, n (%)
Normal	0	1 (100)	1 (100)	2 (100)
Abnormal	0	0	0	0

ECOG, Eastern Cooperative Oncology Group; HSCT, hematopoietic stem cell transplantation; NC, not calculated; NK, natural killer; PS, performance status; SD, standard deviation

^*Performance status was assessed using either ECOG or KPS according to site preference; four patients were evaluated by ECOG only, two by KPS only, and one patient had both assessments

SAR445419 treatment

A complete regimen of SAR445419, included a total of six doses with three doses given weekly over 2 weeks, allowing for a maximum of 21 days after the first administration to complete all six doses. Dosing beyond Day 21 was allowed with permission of the sponsor. The median number of administrations per participant was 6 (range: 6–6) for dose level 1 and 6 (range: 4–6) for dose level 2, respectively. A total of five participants received ≥ 5 administrations, and the median (range) exposure was 15 (0–23) days (Supplementary Table S4).

Safety

No DLT was reported during the study. All seven (100%) participants in the safety population experienced TEAEs, of whom six (85.7%) had grade ≥ 3 TEAEs. SAR445419-related TEAEs were reported in four (57.1%) participants and SAR445419-related SAEs were reported in two (28.6%) participants (infusion-related reaction, each grade 2). One (14.3%) participant treated at dose level 2 experienced grade 3 pneumonia unrelated to SAR445419, resulting in permanent discontinuation of SAR445419. Another participant (14.3%) had a grade 5 TEAE of respiratory failure, unrelated to SAR445419. Five (71.5%) deaths were reported due to disease progression; all were observed after treatment discontinuation and were not related to SAR445419. Except for IRR, no other AESIs, such as CRS, ICANS, or tumor lysis syndrome of any grade, were observed during the study. The overall safety profile as per the dose levels is provided in Table 2. As commonly seen in R/R AML studies, during the treatment period, all seven (100%) participants experienced grade 3 anemia and grade 4 thrombocytopenia and neutropenia. One (14.3%) and six (85.7%) participants experienced grade 3 and grade 4 lymphocytopenia, respectively. Further details on the hematology abnormalities observed at baseline and during treatment are provided in Supplementary Table S5.

Table 2.

Overview of adverse event profile – safety population

n (%)	Chemo + dose level 1 1 × 109 NK cells/administration (n = 3)	Chemo + dose level 2 3 × 109 NK cells/administration (n = 3)	Chemotherapy only (n = 1)	All (N = 7)
DLTs	0	0	0	0
Any TEAE	3 (100)	3 (100)	1 (100)	7 (100)
Grade ≥ 3 TEAE	3 (100)	2 (66.7)	1 (100)	6 (85.7)
Pneumonia	2 (66.7)	1 (33.3)	0	3 (42.9)
Sepsis	2 (66.7)	0	1 (100)	3 (42.9)
Urinary tract infection	0	1 (33.3)	0	1 (14.3)
Insomnia	1 (33.3)	0	0	1 (14.3)
Hypotension	1 (33.3)	0	0	1 (14.3)
Epistaxis	1 (33.3)	0	0	1 (14.3)
Stomatitis	0	1 (33.3)	0	1 (14.3)
Mouth hemorrhage	1 (33.3)	0	0	1 (14.3)
Pain	1 (33.3)	0	0	1 (14.3)
Grade 5 TEAEa
Respiratory failure	0	1 (33.3)	0	1 (14.3)
Any treatment-emergent SAE	2 (66.7)	1 (33.3)	1 (100)	4 (57.1)
Any TEAE leading to permanent discontinuation of SAR445419
Pneumonia	0	1 (33.3)	0	1 (14.3)
Any treatment-emergent AESI
Infusion-related reaction	1 (33.3)	2 (66.7)	0	3 (42.9)
Any SAR445419-related TEAE
Infusion-related reaction	2 (66.7)	2 (66.7)	0	4 (57.1)
Any SAR445419-related treatment-emergent SAE
Infusion-related reaction	1 (33.3)	1 (33.3)	0	2 (28.6)
Any SAR445419-related treatment-emergent AESI
Infusion-related reaction	1 (33.3)	2 (66.7)	0	3 (42.9)
Deathsb	3 (100)	1 (33.3)	1 (100)	5 (71.4)

n (%) = number and percentage of participants with at least one TEAE

^aGrade 5 TEAE occurring during the treatment period and unrelated to SAR445419. ^bAll deaths were due to disease progression unrelated to SAR445419

AESI, adverse event of special interest; SAE, serious adverse event; TEAE, treatment-emergent adverse event

Efficacy

Of the seven enrolled participants, six received ≥ 4 doses of SAR445419 and were efficacy evaluable per protocol definition. One discontinued the study after chemotherapy before receiving SAR445419 and was not evaluable for efficacy. Another participant was not evaluable for efficacy, because they did not undergo the post-baseline tumor assessment. No clinical responses were observed in the five efficacy-evaluable participants (Table 3). Of these, four participants were evaluated with post-baseline disease for disease response bone marrow evaluations, yielding no response (NR), whereas one was reported to have clinical progression of AML based on peripheral blood counts and was transferred to another institution without undergoing a tumor response assessment. That participant was reported to have died of progressive disease 80 days after receiving their last dose of SAR445419. One participant treated with 3 × 10⁹ NK cells/administration had residual blasts after treatment but did proceed to HSCT and was alive as of the last contact date.

Table 3.

Efficacy results – efficacy population

	Chemo + dose level 1 1 × 109 NK cells/administration (n = 2)	Chemo + dose level 2 3 × 109 NK cells/administration (n = 3)	All (N = 5)
Best overall responsea, n (%)
No Response (NR)	2 (100)	2 (66.7)	4 (80.0)
Non evaluable for Response (NE)	0	1 (33.3)	1 (20.0)
Refractory Disease (RD)	0	0	0
Composite complete remission rate
(CR, CRh, or CRi), n (%; [90% CIb])	0	0	0
Alternative complete remission rate (CR or CRh), n (%; [90% CIb])	0	0	0
Overall response rate (CR, CRh, CRi, MLFS or PR), n (%; [90% CIb])	0	0	0

^aAssessed by the investigator according to the modified ELN 2022 criteria for AML; ^bEstimated using Clopper–Pearson method

CR, complete response; CRh, complete remission with partial hematologic recovery; CRi, complete remission with incomplete hematologic recovery; MLFS, Morphologic Leukemia-Free State; NE, non-evaluable for response; NR, no response; RD, refractory disease

Pharmacokinetics and immunogenicity

The persistence of SAR445419 cells was assessed with a next-generation sequencing (NGS)-based chimerism assay that differentiates product donor and participant cells based on unique single nucleotide polymorphisms. In the three participants with available baseline and on treatment samples (two at 1 × 10⁹ and one at 3 × 10⁹ NK cells/administration), product cells were detected within 1 h after the first infusion and reached a peak after fourth infusion, consisting of up to 18% of circulating PBMCs (Fig. 5). The total number of donor cells reached up to 4 cells/µL, with a peak between Day 8 and 12, depending on the participant. The participant who received a higher dose (3 × 10⁹ NK cells/administration) of SAR445419 exhibited a higher absolute number of NK product cells, which could be detected on Day 28 of treatment.

Fig. 5.

Persistence of donor cells. ^aPatient 1 received SAR445419 from different donor during the last (6th) infusion; however, it was poorly detected, and thus, the dashed line almost completely aligns with the solid line. Sample for Patient 1 was not available at DL2 on Day 1. DL, dose level; PBMC, peripheral blood mononuclear cell

To assess the potential immunogenicity of SAR445419, the presence of donor-specific anti-HLA antibodies was analyzed. In participants with available pre- and post-treatment samples, no treatment-emergent antibodies directed toward donor-specific HLAs were detected (data not shown).

Discussion

This Phase 1, single-arm, dose-escalation study assessed the manufacturing feasibility, safety and potential efficacy of SAR445419, an allogeneic, ex vivo expanded off-the-shelf NK cell therapy derived from universal donor NK cells in participants with R/R AML. The findings of this study demonstrated a manageable SAR445419 safety profile; however, no clinical responses were observed in the five efficacy-evaluable participants who were enrolled before the study was terminated by the sponsor for business portfolio reprioritization. Adoptive NK cell immunotherapy has been explored as a potential treatment for leukemia, based on the known inherent biological activity of NK cells against AML [4, 37, 38]. These previously reported studies of adoptively transferred haploidentical NK cells have demonstrated safety and feasibility of this approach with remissions observed in 20% to 50% of patients with R/R AML [17]. Haploidentical NK cell infusions outside the post-HSCT setting resulted in complete remissions in five out of 19 patients with adverse-risk AML. A high remission rate (CR 75%) was observed in KIR-ligand mismatched donors [18]. Additional studies of ex vivo expanded NK cells (FC21-NK) or PM21-NK demonstrated safety (no CRS- or GVHD-related AEs), with an overall marrow remission rate of 85% in NCT01787474, and two participants achieved CR (NCT02809092) [39, 40]. The SAR445419 product was previously tested following the administration of fludarabine and cytarabine in an investigator-initiated study (NCT04220684) wherein participants tolerated the treatment well, with CR observed in one participant who received a dose of 1 × 10⁷ cells/kg [41]. Overall, these findings support the hypothesis that mature, fully functional NK cells expanded ex vivo have a manageable safety profile.

In this study, the safety profile was consistent with that seen in other NK cell therapy studies. Two participants had SAEs attributed to SAR445419: both were IRRs (grade 2) that were considered manageable with supportive care measures. Except for IRR, no other AESIs, such as CRS, ICANS, or tumor lysis syndrome, were observed during the study. The hematologic cytopenia events observed during the treatment were consistent with the expected events in the R/R AML setting in patients receiving preparative chemotherapy with fludarabine and cytarabine.

Nevertheless, none of the five efficacy-evaluable participants experienced a clinical response. The PK evaluation demonstrated detectable SAR445419 cells in all participants with available samples, with peak levels in days 8 to 12 and longer persistence was observed in a participant with highest dose. However, the absolute NK cell counts were lower than those measured in trials using other NK cell products [27, 42]. Thus, the exposure level or the NK:AML ratio might have been insufficient to achieve a clinical response. Although no anti-product HLA antibody formation was detected, the SAR445419 cells might have been rejected by patient immune cells prior to adequate in vivo expansion. In contrast to other published studies using exogenous IL-2, no external cytokine support was provided in this study. Further, the fludarabine/cytarabine or cytarabine/fludarabine lymphodepletion regimen differed from that in other NK studies that had used more intense cyclophosphamide/fludarabine dosing [18, 42]. The patient characteristics, such as the number and the type of prior therapies, might also have influenced the likelihood of response, but this is difficult to assess in such a small cohort. Therefore, it is difficult to attribute the lack of clinical response to a specific cause.

Conclusion

In conclusion, the findings from this study demonstrated the successful manufacturing and overall safety of SAR445419, an off-the-shelf ex vivo expanded allogeneic NK cells in patients with R/R AML. The manufacturing and distribution processes were successfully executed, which supported the feasibility of using these donor-derived off-the-shelf NK cells in a clinical setting. However, further studies are warranted to explore adequate NK-cell number and frequency of dosing, lymphodepletion modifications, exogeneous cytokine support, and NK cell engineering to improve therapeutic outcomes.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contribution

Marina Konopleva, Naval Daver, Jolie Schafer, Kyle Jensen, and Rao Saleem contributed to the conceptualization of the study. Yi Li, Agata Drewniak, Pierre Bay, Winston Huh, Ozlem Yildirim, and Sarah Cooley were responsible for data curation and formal analysis. Investigation was conducted by Marina Konopleva, Vijaya Raj Bhatt, Ioannis Mantzaris, Abhishek Maiti, Krishna Gundabolu, and Naval Daver. Methodology was developed by Yi Li, Agata Drewniak, Winston Huh, Ozlem Yildirim, and Sarah Cooley. Project administration was managed by Jolie Schafer, Kyle Jensen, Rao Saleem, Ozlem Yildirim, and Sarah Cooley. Software development was carried out by Yi Li and Agata Drewniak. All authors contributed to the writing of the original draft and participated in the review and editing of the manuscript.

Funding

This study was funded by Sanofi.

Data availability

Qualified researchers may request access to patient-level data and related study documents, including the clinical study report, study protocol with any amendments, blank case report form, statistical analysis plan, and dataset specifications. Patient-level data will be anonymized, and study documents will be redacted to protect the privacy of the trial participants. Further details on Sanofi’s data-sharing criteria, eligible studies, and process for requesting access can be found at https://www.vivli.org/.

Declarations

Ethics approval and consent to participate

The study was conducted in compliance with the international ethics guidelines, including the Declaration of Helsinki and the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) guidelines for Good Clinical Practice (GCP), as well as all applicable laws, rules, and regulations. The study was approved by the Institutional Review Board at each site, and written informed consent was obtained from all study participants prior to their enrollment.

Consent for publication

Not applicable.

Competing interests

Marina Konopleva reports receiving consultancy fees from AbbVie, Forty Seven, Precision BioSciences, Gilead Sciences, Genentech, Janssen, Sanofi, MEI Pharma, Daiichi Sankyo Pharmaceutical, AstraZeneca Co., and Menarini; research funding from AbbVie, Allogene Therapeutics, Cellectis, Forty Seven, Gilead Sciences, Genentech, Sanofi, MEI Pharma, Rafael Pharmaceuticals, Daiichi Sankyo Pharmaceutical, AstraZeneca Co., Menarini, and Precision BioSciences; and holds shares/stocks in Reata Pharmaceuticals. Ioannis Mantzaris reports having received honoraria from Kite Pharma. Reports participation in advisory board for Syndax. Abhishek Maiti reports receiving research funding from Celgene, Lin BioScience. Krishna Gundabolu reports receiving consultancy fees from Bristol-Myers Squibb, CTI BioPharma Corp, Blueprint Medicines, Sanofi Aventis GmbH, Spark Therapeutics. Reports participation advisory bords for Incyte and SOBI. Vijaya Raj Bhatt reports participating in the Safety Monitoring Committee for Protagonist, serving as a member of the National Comprehensive Cancer Network Acute Myeloid Leukemia Panel, as an associate editor for the journal Current Problems in Cancer, and as a contributor for BMJ Best Practice and receiving consulting fees from Jazz, research funding (institutional) from Cynata Therapeutics, MEI Pharma, Actinium Pharmaceutical, Sanofi U.S. Services, AbbVie, Pfizer, Incyte, Jazz, and National Marrow Donor Program, and drug support (institutional) from Chimerix for a trial. Naval Daver reports receiving consultancy fees from Bristol-Myers Squibb, Daiichi, Pfizer, Gilead, Servier, Genentech, Astellas, AbbVie, ImmunoGen, Amgen, Trillium, Jazz, Syndax, Sumitomo, Kura, AROG, Servier, Shattuck labs, Sanofi, Arcellx, Caribou, Debiopharm, Aptos, Astra-Zeneca, StemLine, VincerX; research funding from Hanmi, Trovagene, FATE, Novimmune, Glycomimetics, BMS, Astellas, Daiichi, Abbvie, Immunogen, Shattuck labs, Sanofi, Arcellx, Caribou, Debiopharm, Aptos, Astra-Zeneca, StemLine, VincerX, Nerviano, FATE therapeutics, Sumitomo, Kura Jolie Schafer, Kyle Jensen, Rao Saleem, Yi Li, Agata Drewniak, Pierre Bay, Winston Huh, Ozlem Yildirim, and Sarah Cooley are employee of Sanofi and may hold stocks and/or stock options.