Abstract
Background
Patients with unresectable locally advanced or metastatic solid tumors have limited therapeutic options. Antibody-guided allogeneic natural killer (NK) cell therapy represents a novel immunotherapeutic strategy to enhance tumor targeting and cytotoxicity by coupling NK cells with tumor-specific antibodies targeting antigens such as 5T4.
Objective
To evaluate the safety, tolerability, pharmacokinetics, immunogenicity, and preliminary efficacy of IBR854, a non-viral, non-genetically modified, 5T4 antibody-coupled allogeneic NK cell therapy, in patients with advanced solid tumors.
Patients and methods
This open-label, first-in-human phase 1 dose-escalation study enrolled 19 patients with unresectable locally advanced or metastatic solid tumors across five dose cohorts (3.0 × 109–12.0 × 109 cells per dose). Patients received intravenous IBR854 in a 3 + 3 escalation design. Safety, dose-limiting toxicities, adverse events, pharmacokinetics, anti-drug antibodies, and antitumor activity (RECIST v1.1) were assessed.
Results
No dose-limiting toxicities were observed. The most common treatment-emergent adverse events were elevated interleukin levels (42.1%), infusion-related reactions (31.6%), and fever (21.1%). No anti-drug antibodies were detected. The disease control rate was 43.8%. Median progression-free survival was 43 days. Pharmacokinetic analysis based on transgene copy number showed a dose-dependent prolongation of Tmax (0.867–3.433 h), with a relatively consistent half-life (1.390–2.648 h).
Conclusions
IBR854 demonstrated an acceptable safety and tolerability profile and achieved disease stabilization in a subset of heavily pretreated patients with advanced solid tumors. These findings support further clinical development of 5T4-targeted antibody-coupled allogeneic NK cell therapy.
Trial registration
This study was registered at ClinicalTrials.gov (registration number: NCT06001684).
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1007/s00262-026-04481-1.
Keywords: IBR854, 5T4 antigen, Allogeneic NK cell therapy, Solid tumors, First-in-human trial
Introduction
The management of patients with unresectable locally advanced or metastatic solid tumors remains a formidable clinical challenge, with many standard therapies failing to achieve sustained disease control. Adoptive cell‑based immunotherapies have emerged as a promising frontier in oncology, leveraging innate and adaptive immune mechanisms to mediate antitumor effects. In particular, natural killer (NK) cells have garnered increasing interest for cancer immunotherapy owing to their intrinsic cytotoxic capacity and ability to mediate cytotoxicity in a non‑major histocompatibility complex (MHC)-restricted recognition, and favorable safety profile. [1]
To improve tumor specificity, chimeric antigen receptor (CAR)-NK cells have been actively developed for solid tumors. At least 40 clinical studies of CAR-NK cells targeting solid cancers have been registered, most of which remain in early-phase testing [2, 3]. Current trials cover a broad range of targets and tumor types, including NKG2D ligands, mesothelin, CD70, MUC1, ROBO1, and 5T4 in colorectal, ovarian, pancreatic, renal, lung, brain, and other solid tumors. However, available clinical data remain limited, and reported efficacy has generally been modest, with disease stabilization more commonly observed than objective tumor regression [2, 3]. For example, a phase I study of NKG2D CAR-NK cells in advanced colorectal cancer showed acceptable safety but limited antitumor activity, with stable disease as the best response in one of the six heavily pretreated patients [4]. These findings suggest that CAR-mediated targeting can enhance NK cell specificity, but solid tumors continue to impose major barriers, including limited trafficking, short in vivo persistence, antigen heterogeneity, and an immunosuppressive tumor microenvironment.
Antibody-guided NK cell therapy represents an alternative strategy to enhance tumor targeting while avoiding genetic modification. By covalently conjugating tumor-specific monoclonal antibodies to the surface of NK cells, this modality achieves precise delivery of effector lymphocytes to antigen-expressing tumor sites, thereby enhancing tumor homing. Upon engagement of antibody-coated tumor cells, the Fc domain of the antibody interacts with Fcγ receptors on NK cells to trigger robust antibody-dependent cellular cytotoxicity (ADCC) in the context of therapeutic antibodies [5]. This dual mechanism of physical targeting and functional activation synergistically amplifies NK-mediated cytotoxicity without requiring genetic modification of the effector cell [6, 7]. Despite these advances, both CAR-NK and antibody-guided NK cell strategies share an important limitation: They primarily enhance tumor recognition without fully addressing the broader translational barriers of NK cell therapy in solid tumors. These include limited in vivo persistence, rapid cellular clearance, manufacturing complexity (particularly for genetically modified products), and the profoundly immunosuppressive tumor microenvironment.
The oncofetal antigen 5T4 (trophoblast glycoprotein, TPBG) is a transmembrane glycoprotein that is broadly overexpressed in a variety of solid malignancies, including colorectal, ovarian, lung, and other carcinomas, while displaying limited expression in normal adult tissues. As an oncofetal antigen, 5T4 was originally identified by its shared expression between trophoblast and cancer cells, and its tumor-restricted distribution confers a favorable therapeutic index by minimizing on-target off-tumor toxicity [8]. The N-glycosylated cell surface localization and stable expression of 5T4 make it particularly amenable to antibody engagement and subsequent induction of ADCC via NK cells. Owing to its prevalence across diverse tumor types and restricted normal tissue expression, 5T4 continues to be investigated as a promising target for antibody-based and immune-engaging therapeutic strategies. [9]
IBR854 was developed to integrate tumor targeting specificity with simplified manufacturing and preserved NK cell function, aiming to overcome key translational limitations of both CAR-engineered and conventional antibody-guided NK cell therapies. IBR854 is a novel, non‑genetically modified, antibody‑guided allogeneic NK cell therapy designed to target 5T4‑expressing solid tumors. It leverages healthy donor‑derived NK cells conjugated with a monoclonal anti‑5T4 antibody, enabling selective binding to tumor cells and triggering cytotoxicity, while avoiding the risks associated with viral gene transfer or autologous T‑cell manufacturing. Preclinical studies demonstrated that IBR854 exerts potent cytotoxic activity against multiple 5T4-positive solid tumor cell lines and induces dose-dependent tumor growth inhibition in xenograft models, with favorable safety characteristics and minimal off-target binding.
Here, we report the detailed safety, pharmacokinetics, and exploratory efficacy outcomes of IBR854, together with correlative biomarker analyses of donor–recipient HLA/KIR and clinical outcomes. In this patient cohort, no dose‑limiting toxicity was observed, and the regimen was well tolerated, with a disease control rate (DCR) of 43.8%. Overall, this study supports the further development of IBR854 as a novel “off‑the‑shelf” NK cell therapy platform targeting 5T4‑positive solid tumors.
Methods
Study design
This open-label, multicenter phase 1 study (approvals: 2023LP00207, 2023LP00208, 2024LB00345, and 2024LB00346) evaluated IBR854 in patients with advanced solid tumors. The primary objectives were to assess safety, tolerability, and determine the maximum tolerated dose (MTD) and recommended phase 2 dose (RP2D); secondary objectives included preliminary efficacy, pharmacokinetics (PK), and immunogenicity; exploratory objectives assessed donor–recipient HLA/KIR matching and 5T4 expression in relation to clinical outcomes.
The study included screening (≤ 28 days), treatment, and follow-up phases, as illustrated in Fig. 1. A standard 3 + 3 dose-escalation design was used across five dose levels. The starting dose of 3.0 × 109 cells was selected based on preclinical safety evaluations and prior experience with NK cell-based therapies. IBR854 was administered on days 1 and 8 of each 21-day cycle. Dose escalation proceeded after completion of a 21-day dose-limiting toxicity (DLT) evaluation period in the preceding cohort.
Fig. 1.
Flow diagram of the clinical trial workflow. This flowchart depicts the three core phases of the IBR854 phase 1 trial: (1) screening period (days -28 to enrollment), during which eligibility confirmation, laboratory assessments, and baseline tumor imaging were completed after ICF signing; (2) treatment period (21-day cycles, C1-Cn), with IBR854 administered on cycle day 1 (C1D1) and cycle day 8 (C1D8) per cohort, and the first cycle designated as the dose-limiting toxicity (DLT) observation window; (3) follow-up period, including end-of-treatment (EOT) assessment within 7 days of last infusion, 30-day safety follow-up, and survival surveillance every 4 weeks until 3 months post-first dose. Adverse events (AEs) and concomitant medications/treatments were recorded throughout the study, with follow-up terminated at 30 days post-last cycle or initiation of new antitumor therapy (whichever occurred first). Abbreviations: DLT, dose-limiting toxicity; ICF, informed consent form; EOT, end of treatment; and AE, adverse event
Considering the potential risk of on-target off-tumor toxicity due to 5T4 expression in certain normal tissues, comprehensive safety monitoring and predefined stopping criteria were incorporated into the study design. DLTs were predefined (Table S2), and safety was continuously reviewed by a Safety Review Committee to guide dose escalation and determine MTD/RP2D, with routine safety monitoring including potential on-target off-tumor effects related to 5T4 expression such as epithelial toxicity. Cytokine release syndrome (CRS) was graded according to the 2018 consensus criteria from the American Society for Transplantation and Cellular Therapy (ASTCT) for immune effector cell-associated CRS and neurotoxicity syndromes.
Patient eligibility
Eligible patients were adults aged 18–75 years with unresectable locally advanced or metastatic solid tumors without standard treatment options, including at least one measurable target lesion per Response Evaluation Criteria in Solid Tumors (RECIST) V1.1, an Eastern Cooperative Oncology Group (ECOG) performance status of ≤ 2, and an expected survival of > 3 months. Adequate organ function was required, including an absolute neutrophil count (ANC) of ≥ 1.5 × 109/L, a platelet count (PLT) of ≥ 75 × 109/L, a hemoglobin (HGB) of ≥ 80 g/L, and normal renal, hepatic, and coagulation function as specified. Fertile patients and their partners agreed to use effective contraception for 6 months after the last dose of study drug. Detailed inclusion and exclusion criteria are listed in Table S1.
NK cell preparation
IBR854 cell injection is a non-genetically edited, antibody-guided biochemically activated NK cell product. Its core structure consists of a full-length monoclonal anti-5T4 antibody conjugated to allogeneic NK cells derived from peripheral blood of healthy donors via an L2 linker (based on click chemistry), enabling specific recognition of the 5T4 antigen and tumor cell killing. The product was stored at ≤ -150℃, with strict cold chain management during transportation and use to maintain cell viability. The manufacturing cycle was approximately 18–23 days, and products were delivered to study centers only after passing quality release tests (including cell viability, purity, and sterility).
Immunological response assessment
Immunogenicity was assessed by detecting ADA against 5T4 in peripheral blood, with blood samples collected at pre-dose on cycle 1 day 1 (C1D1), C1D16 ± 1, C1D21 ± 1, and at EOT. Samples were sent to MEDx Translational Medicine Research (Suzhou) Co., Ltd. for quantitative analysis via enzyme-linked immunosorbent assay (ELISA). The immunogenicity analysis set included patients who received at least one dose of study drug and had post-treatment immunogenicity data.
Exploratory analyses included: retrospective immunohistochemistry testing of 3–5 unstained tumor pathological sections collected at enrollment to assess the percentage, intensity, and distribution of 5T4 antigen expression, and its correlation with objective response rate (ORR) and disease control rate (DCR); and genotyping of donor–recipient HLA (HLA-A, B, and C) and killer cell immunoglobulin-like receptor (KIR) via polymerase chain reaction-sequence-specific primer (PCR-SSP) to analyze the association between donor–recipient matching and efficacy.
Statistics
This phase 1 study was not powered for formal hypothesis testing, and the sample size was determined based on dose-escalation requirements. Safety and efficacy analyses were descriptive. Adverse events were coded according to the Medical Dictionary for Regulatory Activities (MedDRA, version 25.0) and graded per the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI CTCAE), version 5.0. Efficacy was assessed according to Response Evaluation Criteria in Solid Tumors (RECIST), version 1.1. ORR and DCR were summarized with two-sided 95% confidence intervals (CIs). As a sensitivity analysis, efficacy was also assessed in the intent-to-treat (ITT) population, defined as all treated patients. Patients who discontinued the study before the first post-baseline tumor assessment and, therefore, had no evaluable RECIST v1.1 response were conservatively imputed as having progressive disease. Progression-free survival (PFS) and overall survival (OS) were estimated using the Kaplan–Meier method. Pharmacokinetic parameters were derived using non-compartmental analysis. Exploratory analyses of biomarker associations with clinical outcomes were descriptive.
Results
Patient characteristics
The first patient signed the ICF on August 31, 2023, and the last patient completed the final visit on December 26, 2024, with data cutoff on August 08, 2025. A total of 31 patients were screened, 12 (38.7%) of whom failed screening (11 due to ineligibility and 1 due to ICF withdrawal). Nineteen patients were successfully enrolled and received at least one infusion of IBR854 (Fig. 2), with cohort enrollment as follows: 3.0 × 109 viable cells (n = 4), 5.0 × 109 viable cells (n = 3), 7.0 × 109 viable cells (n = 3), 9.0 × 109 viable cells (n = 6), and 12.0 × 109 viable cells (n = 3).
Fig. 2.
CONSORT flow diagram of subject disposition. This CONSORT-compliant flowchart summarizes the full subject trajectory from screening to study completion: of 31 screened individuals, 12 failed enrollments (11 due to non-meeting eligibility criteria and 1 due to ICF retraction), and 19 were enrolled and received at least one IBR854 infusion. Enrolled subjects were stratified into five dose cohorts: 3.0 × 109 viable cells (n = 4), 5.0 × 109 viable cells (n = 3), 7.0 × 109 viable cells (n = 3), 9.0 × 109 viable cells (n = 6), and 12.0 × 109 viable cells (n = 3). The diagram further delineates post-enrollment outcomes (study completion, withdrawal due to death/loss to follow-up/other causes) for each cohort, with numbers of subjects at each stage explicitly quantified
The study population comprised predominantly older adults, with a median age in the early 60 s, and a slight predominance of female patients. Most participants had a good performance status (ECOG 0-1), indicating relatively preserved functional capacity at baseline. Tumor involvement was heterogeneous, with metastases most frequently observed in lymph nodes and lungs, while other sites such as the liver, colon, and peritoneum were less common. Pathologically, adenocarcinoma represented the largest subgroup, although a range of other histological types was also included, reflecting a diverse tumor spectrum. The majority of patients had advanced disease, with nearly all presenting at stage IV. Patients were heavily pretreated (median, 3 prior lines), with a notable proportion having received ≥ 4 lines of therapy, consistent with a refractory population. In this setting, disease stabilization may represent a clinically meaningful outcome despite the absence of objective responses (Table 1).
Table 1.
Subject demographic and baseline characteristics
| Characteristics | Number of patients, % |
|---|---|
| Age (years) | |
| Median (range) | 62 (20–75) |
| Mean (SD) | 57.3 (15.60) |
| Sex | |
| Male | 8 (42.1) |
| Female | 11 (57.9) |
| ECOG score | |
| 0 | 1 (5.3) |
| 1 | 18 (94.7) |
| Tumor site | |
| Lungs | 8 (23.5) |
| Mediastinum | 2 (5.9) |
| Liver | 2 (5.9) |
| Colon | 3 (8.8) |
| Lymph nodes | 9 (26.5) |
| Peritoneum | 2 (5.9) |
| Neck | 1 (2.9) |
| Others | 7 (20.6) |
| Pathological type | |
| Adenocarcinoma | 9 (47.4) |
| Mucinous adenocarcinoma | 1 (5.3) |
| Squamous cell carcinoma | 2 (10.5) |
| Others | 7 (36.8) |
| Latest TNM stage | |
| III | 1 (5.3) |
| IV | 17 (89.4) |
| Number of previous lines | |
| 1–2 | 7 (43.76) |
| ≥ 3 | 9 (56.3) |
| Mean (SD) | 3.25 (1.69) |
ECOG, Eastern Cooperative Oncology Group
Favorable safety profile
Treatment-emergent adverse events (TEAEs) occurred in all 19 patients (100%), with treatment-related adverse events (TRAEs) reported in 16 patients (84.2%). Grade ≥ 3 TEAEs were observed in 5 patients (26.3%), while only one patient (5.3%) experienced a grade ≥ 3 TRAE. A total of four serious adverse events (SAEs) were reported in three patients (15.8%), including one patient (5.3%) with a drug-related SAE. No treatment-related deaths occurred (Table S3).
Notably, no DLTs were observed across all dose cohorts, and the MTD was not reached, indicating a favorable safety margin and supporting further dose exploration within the evaluated range. The incidence of grade ≥ 3 TEAEs was relatively low (26.3%), and treatment-related high-grade events were infrequent (5.3%), suggesting that severe toxicity attributable to the study treatment was limited despite the heavily pretreated nature of the study population. All CRS events were grades 1–2, and no grade ≥ 3 CRS was observed. As CRS represents a critical safety concern in cellular immunotherapy, these findings further support the favorable tolerability profile of IBR854 (Table 2).
Table 2.
Summary of treatment-emergent adverse events
| Adverse event | Any grade n (%) | Grade ≥ 3 n (%) |
|---|---|---|
| Cytokine elevation | 8 (42.1) | 1 (5.3) |
| Infusion-related reaction | 6 (31.6) | 1 (5.3) |
| Pyrexia | 4 (21.1) | 0 |
| Cytokine release syndrome | 2 (10.5) | 0 |
| Anemia | 1 (5.3) | 0 |
| Gastrointestinal hemorrhage | 1 (5.3) | 0 |
| Oxygen saturation decreased | 1 (5.3) | 0 |
| Chills | 1 (5.3) | 0 |
Overall, the safety profile observed in this first-in-human study compares favorably with that reported for many T-cell-based therapies, which are often associated with higher rates of severe immune-mediated toxicities, and is consistent with or potentially improved relative to existing NK cell-based approaches. This favorable tolerability may be attributable to the non-genetically modified design, the use of allogeneic NK cells with inherently lower toxicity risk, and the antibody-coupled strategy that enables targeted activation without excessive immune stimulation.
Evidence of disease stabilization
Among 16 evaluable patients, no objective responses were observed. Disease control was achieved in 43.8% of patients, all of whom had stable disease (Fig. 3). In the ITT sensitivity analysis including all 19 treated patients, the DCR was 36.8% (7/19). Given that the study population was heavily pretreated (median, 3 prior lines), with a substantial proportion having received ≥ 4 lines of therapy, these findings are indicative of a refractory clinical setting in which disease stabilization may represent a clinically meaningful outcome.
Fig. 3.
Waterfall plot of best percentage change in target lesion size. This waterfall plot illustrates the maximum percentage change in target lesion diameter for all evaluable subjects (n = 16) per RECIST v1.1 criteria, stratified by treatment cohort (C1 = 3.0 × 109, C2 = 5.0 × 109, C3 = 7.0 × 109, C4 = 9.0 × 109, and C5 = 12.0 × 109 viable cells per infusion; horizontal axis). The vertical axis denotes the percentage change from baseline tumor size (negative values = reduction and positive values = enlargement). Each bar is color-coded by best overall response (BOR): complete response (CR), partial response (PR), stable disease (SD), progressive disease (PD), not evaluable (NE), or study limitation (SL). Subjects with ongoing treatment at data cutoff are marked with “Ongoing,” enabling direct visualization of treatment-induced tumor size dynamics across cohorts
At the data cutoff, disease progression had occurred in 9 patients (56.3%), with a median PFS of 43 days (95% CI, 35.0-NE) (Fig. S1a). No apparent dose–response relationship was observed. Three patients (18.8%) had died, and the median OS was not reached (range, 61–282 days) (Fig. S1a).
Pharmacokinetics
Mean concentration–time relationship is characterized by DNA copy number of NK cells (Fig. 4). Across all dose cohorts, both parameters showed a similar kinetic profile, with peak levels observed shortly after the end of infusion, followed by a rapid decline to near-baseline levels within hours. A comparable pattern was observed after the second infusion, with transient re-elevation and no evidence of accumulation between dosing days. Notably, exposure was consistently achieved across dose levels without a clear dose-dependent trend. The lack of accumulation and limited persistence may support a favorable safety profile and allow for flexible repeat dosing.
Fig. 4.
Concentration time in peripheral blood after multiple infusion. Concentration quantified as NK cell-specific gene copy number (copies/μg gDNA) at identical at key time points: 0.5 h, 2 h, 4 h, 8 h, and 24 h post-infusion on C1D1 and C1D8, plus C1D16 ± 1 and C1D21 ± 1. Curves are color-coded by dose cohort
Pharmacokinetic analysis further indicated that, despite the absence of a clear dose–exposure relationship in terms of peak concentration or overall exposure, there was a trend toward prolonged time to peak concentration (0.867–3.433 h) with increasing dose. In contrast, the elimination half-life remained relatively consistent across dose cohorts (1.390–2.648 h), suggesting similar clearance kinetics.
Consistent with this observation, exploratory analyses of dose–exposure relationships based on normalized PK parameters for both single and multiple infusions (Figure S2) did not demonstrate a clear association between dose and exposure (Cmax or AUC), likely reflecting the limited number of evaluable PK samples and substantial inter-individual variability.
Immunological responses
No anti-drug antibodies (ADAs) were detected at any scheduled timepoint across all dose cohorts. Interpretation of cytokine analyses was limited by the high proportion of missing data and a small number of patients with evaluable post-treatment samples. Nevertheless, among the available samples, a temporal increase in circulating cytokine levels relative to baseline was observed following IBR854 administration. IL-6 concentrations showed a gradual elevation over time, with a more pronounced increase at later timepoints, while IFN-γ levels demonstrated a modest and transient rise, peaking at C2D1 before declining thereafter (Figure S3).
Importantly, these cytokine changes were exploratory in nature and not systematically linked to clinically adjudicated CRS events due to incomplete time-matched sampling. In addition, several instances of elevated cytokines were also deemed clinically significant clinical events most of which were attributed to infusion-related reactions or inter-current conditions such as infection, instead of treatment-induced CRS itself.
Association between donor–recipient HLA/KIR matching and disease control
An exploratory analysis was performed to evaluate the association between NK cell donor–recipient HLA/KIR matching and disease control. Patients were stratified according to HLA/KIR matching scores (Table 3 and Table S4). Disease control was observed more frequently in patients with intermediate matching scores. Specifically, stable disease was achieved in two of five patients with matching scores of 1–2 (DCR 40.0%, 95% CI 5.3–85.3) and in five of seven patients with scores of 2–3 (DCR 71.4%, 95% CI 29.0–96.3). In contrast, no disease control was observed in patients with either low (0–1) or high (≥ 3) matching scores.
Table 3.
Exploratory analysis of donor–recipient HLA/KIR matching, dose level, and disease control
| HLA/KIR matching score | Dose (live cells) | No. of patients (n) | Efficacy | Disease control rate, n (%) [95% CI] |
|---|---|---|---|---|
| 0–1 | 3.0 × 10⁹ | 0 | – | |
| 5.0 × 10⁹ | 0 | – | ||
| 7.0 × 10⁹ | 1 | PD | ||
| 9.0 × 10⁹ | 0 | – | ||
| 12.0 × 10⁹ | 0 | – | ||
| Total | 1 | PD | 100 (0.0, 97.5) | |
| 1–2 | 3.0 × 10⁹ | 1 | PD | |
| 5.0 × 10⁹ | 2 | SD | ||
| 7.0 × 10⁹ | 1 | PD | ||
| 9.0 × 10⁹ | 0 | – | ||
| 12.0 × 10⁹ | 1 | PD | ||
| Total | 5 | SD | 40 (5.3, 85.3) | |
| 2–3 | 3.0 × 10⁹ | 2 | SD | |
| 5.0 × 10⁹ | 1 | SD | ||
| 7.0 × 10⁹ | 1 | SD | ||
| 9.0 × 10⁹ | 2 | SD, PD | ||
| 12.0 × 10⁹ | 1 | PD | ||
| Total | 7 | SD | 71.4 (29.0, 96.3) | |
| ≥ 3 | 3.0 × 10⁹ | 0 | – | |
| 5.0 × 10⁹ | 0 | – | ||
| 7.0 × 10⁹ | 0 | – | ||
| 9.0 × 10⁹ | 1 | PD | ||
| 12.0 × 10⁹ | 1 | PD | ||
| Total | 3 | PD | 0 (0.0, 70.8) |
HLA, human leukocyte antigen; KIR, killer cell immunoglobulin-like receptor; BOR, best overall response; SD, stable disease; PD, progressive disease; DCR, disease control rate; and CI, confidence interval
These findings suggest a potential nonlinear relationship between HLA/KIR matching and disease control; however, given the very small sample size within each subgroup and the overall heterogeneity of the cohort, these findings should be interpreted as strictly hypothesis generating. The apparent trend may be influenced by statistical noise and unmeasured confounding factors, and, therefore, should not be considered definitive evidence of a biological relationship. Larger, prospectively designed studies are required to validate this observation.
Discussion
This first-in-human study demonstrates that IBR854, an antibody-coupled allogeneic NK cell therapy targeting 5T4, exhibits a favorable safety and tolerability profile across the evaluated dose range. No DLTs were observed, and the MTD was not reached. Treatment-related adverse events were generally low-grade, with a relatively low incidence of grade ≥ 3 TRAEs (26.3%) and few serious adverse events.
Importantly, in contrast with published clinical studies of CAR-NK and CAR-T cell therapies, where CRS and neurotoxicity remain key safety concerns, particularly in early-phase solid tumor trials, IBR854 showed a more favorable immune-related safety profile. CRS occurred at a low rate (10.5%) with no grade ≥ 3 CRS or ICANS observed. These findings suggest a potentially wider therapeutic window compared with existing engineered or antibody-mediated NK cell platforms.
Despite the absence of objective responses, a subset of patients achieved disease stabilization, indicating limited clinical activity in a heavily pretreated population. However, no partial or complete responses were achieved, and the duration of disease control was short, underscoring the limited antitumor efficacy observed in this phase 1 study. This pattern of outcome should be interpreted in the context of the primary safety-focused design of the trial. The clinical activity observed is consistent with the known biological characteristics of NK cells, which typically mediate rapid but transient cytotoxic effects and may preferentially delay tumor progression rather than induce deep tumor regression, particularly in the absence of sustained in vivo persistence or expansion. In addition, the immunosuppressive tumor microenvironment in advanced solid tumors likely limits the durability and magnitude of NK cell-mediated responses.
Pharmacokinetic analyses indicated measurable but transient in vivo persistence of infused NK cells, characterized by rapid post-infusion peaks, lack of accumulation upon repeat dosing, and no clear dose–exposure relationship, with a trend toward delayed peak concentration and relatively consistent elimination kinetics. These findings suggest that increasing cell dose may influence early distribution or trafficking dynamics without substantially prolonging systemic persistence. The lack of clear dose–exposure proportionality further highlights the complexity of cellular pharmacokinetics, which is influenced by host factors, immune clearance, and tissue distribution. Importantly, the observed rapid decline in circulating NK cell-associated signals is consistent with previously reported pharmacokinetic profiles of adoptively transferred NK cell therapies, which similarly show limited peripheral persistence despite clinical activity in early-phase studies. In this context, the observed median progression-free survival of 43 days may reflect early on-target cytotoxic effects during the initial post-infusion period, together with downstream immunologic modulation within the tumor microenvironment, rather than prolonged systemic persistence of infused cells.
The absence of detectable anti-drug antibodies suggests a low risk of classical humoral immunogenicity. However, the observed cytokine elevations, although limited by small sample size and missing data, suggest a treatment-associated innate immune activation occurs following infusion, likely reflecting NK cell activation and immune crosstalk rather than antigen-specific adaptive responses. This distinction is important for interpreting immunogenicity and safety in cell-based therapies, as it separates innate immune pharmacodynamic effects from true humoral immunogenicity. In addition, the potential contribution of alloimmune responses, including anti-HLA antibodies, to NK cell persistence remains an important consideration for future studies and warrants systematic evaluation in larger cohorts.
Exploratory analyses suggest that donor–recipient HLA/KIR interactions may influence clinical outcomes, with intermediate matching associated with potentially favorable disease control. This observation is consistent with current understanding of NK cell education and activation, in which a balance between inhibitory and activating signals is required for optimal function [10, 11]. Excessive inhibitory signaling may suppress cytotoxicity, whereas excessive mismatch may limit persistence or promote premature clearance. These findings warrant further investigation in larger, biomarker-driven studies.
The therapeutic rationale of IBR854 is based on targeting 5T4-positive tumors; however, exploratory 5T4 immunohistochemistry analyses were available only in a limited subset of patients and did not suggest a clear association with clinical outcomes (Table S5). Importantly, clinical activity of NK cell-based therapies is influenced by multiple factors beyond target antigen expression alone, including effector cell fitness, tumor microenvironment, and host immune regulatory mechanisms, which may collectively determine treatment response. In addition, the very small sample size and retrospective nature of tissue evaluation further limit the ability to draw robust conclusions regarding the predictive value of 5T4.
Several limitations should be acknowledged. The small sample size, tumor heterogeneity, and short follow-up inherent to a phase 1 design limit definitive conclusions regarding efficacy. In addition, the incomplete nature of the immune monitoring dataset and incomplete biomarker data restricted the ability to identify predictors of response or resistance. The absence of a control arm further limits interpretation of disease stabilization in this population.
Future studies should incorporate larger, more homogeneous patient cohorts and prospective biomarker strategies, including standardized assessment of 5T4 expression and comprehensive immune profiling. Given the favorable safety profile observed in this study, combination approaches with immune checkpoint inhibitors or agents that modulate the tumor microenvironment may further enhance NK cell persistence and antitumor activity. Such strategies may be critical to translating disease stabilization into more durable and clinically meaningful responses.
Supplementary Information
Below is the link to the electronic supplementary material.
Authors contributions
L.S analyzed data, wrote and revised the manuscript. P.M performed research and collected data. Z.M, N.S, and J.B contributed products. S.W. and N.L. conceived of or designed study; funding acquisition. All authors read and approved the final manuscript.
Funding
This study was also supported by the National Key Research and Development Program of China (2023YFC2508500), the National Natural Science Foundation of China (82272951 and 82272953), the National High Level Hospital Clinical Research Funding (2025-LYZX-C-A03), and the Innovation Project for Medical and Health Sciences and Technology of the Chinese Academy of Medical Sciences (2024-I2M-TS-005).
Availability of data and material
This study was registered at ClinicalTrials.gov (registration number: NCT06001684).
Declarations
Conflict of interest
The authors declare no competing interests.
Ethics approval statement
This FIH study was performed in accordance with the Declaration of Helsinki and was approved by the Ethics Committee and Institutional Review Boards of National Cancer Center/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College (approvals No. 2023LP00207/2023LP00208 and updated approvals No. 2024LB00345/2024LB00346).
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Liangjie Sun and Peiwen Ma contributed equally to this work.
Contributor Information
Shuhang Wang, Email: wangshuhang@cicams.ac.cn.
Ning Li, Email: lining@cicams.ac.cn.
References
- 1.Chu J, Gao F, Yan M et al (2022) Natural killer cells: a promising immunotherapy for cancer. J Transl Med 20(1):240. 10.1186/s12967-022-03437-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Jørgensen LV, Christensen EB, Barnkob MB et al (2025) The clinical landscape of CAR NK cells. Exp Hematol Oncol 14(1):46 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Yu Y, Ho M (2025) CAR NK cell therapy for solid tumors: potential and challenges. Antibody Ther 8(4):275–289 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Wang D, Li B, Shen G et al (2025) NKG2D CAR-NK adoptive cellular immunotherapy combined with or without PD-1 blockade in the treatment of patients with metastatic colorectal cancer: an exploratory study. Cancer Immunol Immunother 74(11):341 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Tuhin IJ, Zhu HJ, Monty MA et al (2025) From innate power to intelligent design: the evolution of NK cell-based cancer immunotherapy. Crit Rev Oncol Hematol 216:104972. 10.1016/j.critrevonc.2025.104972 [DOI] [PubMed] [Google Scholar]
- 6.Terrén I, Orrantia A, Vitallé J et al (2020) Modulating NK cell metabolism for cancer immunotherapy. Semin Hematol 57(4):213–224. 10.1053/j.seminhematol.2020.10.003 [DOI] [PubMed] [Google Scholar]
- 7.Wang W, Erbe AK, Hank JA (2015) NK cell-mediated antibody-dependent cellular cytotoxicity in cancer immunotherapy. Front Immunol 6:368. 10.3389/fimmu.2015.00368 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Stern PL, Harrop R (2017) 5T4 oncofoetal antigen: an attractive target for immune intervention in cancer. Cancer Immunol Immunother 66(4):415–426. 10.1007/s00262-016-1917-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Wang R, Lai Q, Tang L et al (2018) A novel 5T4-targeting antibody-drug conjugate H6-DM4 exhibits potent therapeutic efficacy in gastrointestinal tumor xenograft models. Am J Cancer Res 8(4):610–623 [PMC free article] [PubMed] [Google Scholar]
- 10.Mehta RS, Rezvani K (2016) Can we make a better match or mismatch with KIR genotyping? Hematol Am Soc Hematol Educ Program 2016(1):106–118. 10.1182/asheducation-2016.1.106 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Anfossi N, André P, Guia S et al (2006) Human NK cell education by inhibitory receptors for MHC class I. Immunity 25(2):331–342. 10.1016/j.immuni.2006.06.013 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
This study was registered at ClinicalTrials.gov (registration number: NCT06001684).