Short-term blinatumomab for B-cell acute lymphoblastic leukemia pediatric patients results in measurable residual disease clearance

Abstract

Blinatumomab (Blin) effectively eradicates measurable residual disease (MRD) in pediatric B-cell acute lymphoblastic leukemia (B-ALL), but standard 28-day regimens face high costs and compliance challenges. A phase 2 trial assessed a shortened 14-day Blin protocol in pediatric patients with newly diagnosed B-ALL who were MRD-positive (≥ 0.01%, < 5%) at the end of induction therapy (ChiCTR2100053318). This non-randomized trial included 60 pediatric patients with CD19⁺ B-ALL. Post-induction (day 46), the Blin-14d group (n = 30) received Blin (14-day infusion) only, while the Non-Blin-14d group (n = 30) continued standard chemotherapy. Primary endpoint: MRD complete remission (CR) rate post-Blin. Secondary endpoints: safety and long-term outcomes. After Blin-14d-therapy, all Blin-14d group (100%) achieved MRD negativity (MFC-MRD-neg) within one week. Median time to MRD negativity was shorter in Blin-14d vs. Non-Blin-14d (39 days vs. 42.5 days, p = 0.0243). Predicted 3-year OS, but not EFS, favored Blin-14d (OS 100% vs. 85.9%, p = 0.0360; EFS 73.9% vs. 71.5%, p = 0.2334). Treatment deviations occurred in 10% (Blin-14d) vs. 50% (Non-Blin-14d). Adverse events (AEs) affected 77% of Blin-14d patients, mostly grade 1–2; two had grade 3 AEs. No grade 4 or higher Blin-related events occurred. With the limitations of a non-randomized study, short-term Blin was highly effective in inducing MRD remission with a favorable safety profile, representing a promising and efficient treatment strategy for MRD-positive B-ALL.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00277-026-07028-7.

Introduction

In recent years, the event-free survival (EFS) and overall survival (OS) of pediatric acute lymphoblastic leukemia (ALL) have been improved to 80% and 90% respectively due to the development of risk-adapted multiagent chemotherapy regimens. The cumulative recurrence rate has been reduced to less than 20% [1, 2]. Persistence of measurable residual disease (MRD) is one of the most important risk factor for ALL relapse [3–5]. Chinese Children’s Cancer Group Acute Lymphoblastic Leukemia-2015 Project (CCCG-ALL-2015) data showed that regardless of the initial risk grouping, patients with MRD ≥ 0.01% at day 46 had a higher cumulative risk of relapse than the MRD-neg ones (P < 0.001) [6, 7]. Therefore, achieving MRD remission is important for delaying and even preventing ALL relapse. However, St. Jude’s Total XVI study results suggest that the conventional intensity of chemotherapy might have reached its limit of efficacy. Data from China and international studies also show that the treatment-related toxicity of current chemotherapy is relatively high, which not only limits the efficacy but also affects the long-term quality of life. An alternative approach, based upon the application of newer immunotherapy agents with low toxicity profile, might improve efficacy and treatment outcome [8].

Blinatumomab is a bispecific T-cell engager (BiTE) targeted to CD19 and CD3, which promotes immune-mediated elimination of B-cell lymphoblasts by T cells. Studies have shown that blinatumomab is a safe and effective treatment for children and adults with MRD, relapsed or refractory (R/R) B-lineage ALL [9–12], and FDA has granted full approval for blinatumomab to treat MRD-pos B-ALL [13]. It was found that after 4 weeks of continuous intravenous blinatumomab treatment in the R/R B-ALL patient population, most patients had complete suppression of B cells in the circulation. The massive expansion and proliferation of T cells after blinatumomab treatment occurred in the second and third weeks of continuous infusion. The number of T cells continues to rise to more than two times the baseline, which facilitates further clearance of tumor cells. Notably, the pharmacodynamic effect of blinatumomab, specifically B-cell depletion, persists significantly longer than its short serum half-life of approximately 2 h. This is attributed to its mechanism of action, which involves the potent activation and proliferation of T cells during treatment, creating an expanded effector population, and the efficient sequential engagement of target cells [14, 15]. But T cells declining was also observed due to exhaustion one week after the end of administration [16, 17]. Blinatumomab has a median treatment period of 1–2 cycles in clinical trials and real-world studies for its unique mechanism of rapid onset of action. In the Phase II study (BLAST study) to validate the efficacy of MRD clearance, the median treatment period for blinatumomab was one cycle [11]. However, considering the above evidences and the high cost of blinatumomab therapy [18], the concept of shortened blinatumomab courses is emerging, with recent case reports and small series also demonstrating its efficacy in bridging to transplant or managing toxicity [19–21]. We propose to add one short-term course with 2 weeks of blinatumomab to the original treatment for MRD-pos patients at day 46 of initial induction therapy with the aim of inducing MRD remission and further improving the treatment outcome of pediatric B-ALL.

Materials and methods

Study design

We conducted a prospective, non-randomized controlled, open-label, single-center phase 2 study that enrolled patients who had received B-ALL standard chemotherapy at the Department of Pediatric Hematology and Oncology of Blood Diseases Hospital, Chinese Academy of Medical Sciences. The study was approved by the Institutional Ethics Committee of Blood Diseases Hospital and conducted in accordance with the Declaration of Helsinki. Informed consent was obtained from the parents, guardians, or patients, as appropriate (Clinical trial information: ChiCTR2100053318).

Patients

Patients aged 1 month to 18 years with CD19⁺, day46 MRD-pos and < 5% B-cell ALL were eligible. MRD-pos was defined as a level ≥ 0.01% as assessed by multiparameter flow cytometry (MFC). The diagnosis of ALL was based on morphology and immunophenotypic and genetic features of leukemic cells. Patients with B-cell ALL (B-ALL) aged between 1 year and < 10 years, and leukocyte count < 50 × 10⁹/L, hyperdiploidy > 50 chromosomes, or ETV6-RUNX1 oncogene fusion and without CNS3 or testicular involvement, MRD < 1% on day 19 of induction, and MRD < 0.01% on day 46 of induction were classified as having low-risk (LR) disease. Patients with MRD ≥ 1% (or ≥ 5% blasts morphologically without suitable markers for MRD) in bone marrow on day 46 of induction and infants younger than 6 months with KMT2A rearrangement and leukocyte count ≥ 300 × 10⁹/L were considered to have high-risk (HR) ALL. The remaining cases were classified as intermediate-risk (IR) ALL. This was a non-randomized, phase 2 trial, which enrolled I/HR patients. LR patients were not eligible. HR infants with KMT2A rearrangement were eligible only if they were MRD-pos at Day 46.

The allocation to the Blin-14d group was based on patient eligibility (newly diagnosed B-ALL, MRD-pos at end of induction) and contingent upon financial affordability for the drug and informed consent from the family. The control (Non-Blin-14d) group consisted of eligible patients who did not receive Blin-14d due to financial unsustainability or parental preference.

Therapy

All I/HR patients received dexamethasone for 4–5 days as upfront window therapy, followed by remission induction with vincristine, daunorubicin, pegaspargase and prednisone (VDLP) from day 5 to 28, and cyclophosphamide, cytarabine and mercaptopurine (CAM) from day 29 to 35 (Fig. 1 and Table S1). All I/HR patients who had MRD ≥ 0.01% on day 19 of remission induction received additional early intensification therapy (EIT) with cyclophosphamide, cytarabine, mercaptopurine, vincristine, and pegaspargase (CAM^plus) between day 50 and 57. Patients in the Non-Blin-14d group received standard of care therapy per institutional guidelines, which primarily consisted of follow-up chemotherapy.

Fig. 1.

Treatment protocol. Black arrows indicate time points at which minimal residual disease (MRD) was measured: during the induction chemotherapy (Day 19 and Day 46), before and after the treatment of blinatumomab/CAM^plus. Dex, dexamethasone; VDLP consists of vincristine, daunorubicin, pegaspargase and prednisone; CAM consists of cyclophosphamide, mercaptopurine, and cytarabine; CAM^plus consists of cyclophosphamide, cytarabine, mercaptopurine, vincristine, and pegaspargase. For exact treatment days, refer to the Methods section and Table S1

For Blin-14d group, blinatumomab was added to CAM^plus during EIT. Blinatumomab was administered by continuous intravenous infusion over 14 days, at a dose of 15 µg/m²/d (weight≤45 kg) or 28 µg/d (weight>45 kg) before or after CAM^plus. For the prevention of severe cytokine release syndrome (CRS) and neurologic events, patients were treated with 10 mg/m² dexamethasone 6 h before the start of the blinatumomab infusion, followed by a second dose of 5 mg/m² 30 min before the start of the infusion. Levetiracetam was administrated in prevention of seizure related to blinatumomab treatment. Other side effects were managed with standard of care.

The timing of blinatumomab administration relative to the CAM^plus chemotherapy block was not pre-specified. The treatment sequence was determined by family preference and physician discretion, with the latter based primarily on the patient’s response to and tolerance of conventional chemotherapy. The decision to pursue treatment with blinatumomab itself was subject to family consent and financial considerations, reflecting the real-world context of this study.

For both groups, treatment could be modified based on disease response, including escalation to alternative therapies such as chimeric antigen receptor T-cell therapy (CAR-T) or hematopoietic stem cell transplantation (HSCT) at the clinician’s discretion.

Objectives

The primary objective was MRD complete remission (MRD-CR, MRD < 0.01%) response rate after Blin-14d therapy. Secondary objectives included cumulative incidence of MRD-CR before consolidation therapy, incidence of adverse event and EFS, for which events were relapse, death, second malignancy, or failure to achieve CR. OS (time from induction therapy to death) was an additional key secondary efficacy end point. The exploratory objectives were measures of pharmacodynamics of Blin-14d therapy, including changes of interleukin-6 (IL-6) and lymphocyte subpopulation.

Statistical analyses

This study employs a concurrent parallel group design, comparing Blin-14d therapy (added sequentially during early intensification) in the experimental group (n = 30) versus conventional chemotherapy in the control group (n = 30). Based on prior data, MRD-CR rates are anticipated to be 90% (experimental) and 55% (control) [22, 23]. A 1:1 enrolment ratio (total N = 60) was determined using a two-group rate difference test (α = 0.05, β = 0.2, power = 80%).

Demographic and clinical characteristics were analyzed descriptively, with categorical variables summarized as counts (%) and continuous variables as means or medians. Survival curves and cumulative incidence of MRD-CR were analyzed using Kaplan–Meier curves and log-rank tests. Data were censored at last follow-up if no event occurred. Statistical analyses utilized SPSS 25.0 and were plotted using GraphPad Prism 9.3.0.

Outcome measures

Hematologic and MRD assessments were performed of bone marrow aspirates obtained at screening (day 1, day 19 and day 46), before and one week after the end of blinatumomab administration, and before consolidation therapy. MRD was assessed in all patients using multiparameter flow cytometry (MFC). For the detection of MRD after blinatumomab therapy, the MFC analysis was specifically designed to account for potential CD19 downregulation or loss. The post-Blin gate for identifying residual B-lineage blasts was defined using a combination of CD22 and cytoplasmic CD79a expression to ensure the detection of leukemic cells even in the context of CD19 antigen escape. MFC-MRD remission was defined as the absence of a detectable leukemia-associated immunophenotype (LAIP) with a minimum sensitivity of 0.01% (10^{− 4}). Next-generation sequencing (NGS) was employed only in a subset of the Blin-14d group to evaluate depth of response in exploratory studies. NGS-MRD remission was defined as the absence of clonal immunoglobulin (Ig) gene rearrangements identified at diagnosis, below a minimum sensitivity of 0.0001% (10^{− 6}). Genetic abnormalities including mutated or fused genes were measured by reverse transcription quantitative polymerase chain reaction (RT-qPCR). Molecular complete remission (mCR) refers to the achievement of negativity for the trackable specific genetic lesions (a fusion transcript such as BCR-ABL1 or a sequence mutation) by RT-qPCR or targeted sequencing. Peripheral lymphocyte subsets and peripheral blood levels of cytokines were evaluated by MFC analyses. Cytokines release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) were graded according to the ASTCT consensus guideline [24]. Individual organ toxicities were graded in accordance with the National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE) Version 5.0. NGS Ig/TCR MRD detection was performed using the LymScan Assay (Kindstar Globalgene Technology, Inc.).

Results

Patients

Between September 2021 and May 2024, sixty eligible patients were included in this study, of which 30 (50%) received Blin-14d therapy, constituting the Blin-14d group (pt1 to pt30), while the remaining constituted the Non-Blin-14d group (ptN1 to ptN30). Prior to the Blin-14d therapy, twenty-three patients (77%) received Blin-14d therapy before CAM^plus, and 7 (23%) after CAM^plus. Baseline characteristics and pre-treatment MRD levels are shown in Table 1. The MRD level for the Blin-14d group corresponds to the assessment immediately prior to blinatumomab initiation. There were no significant differences in age, sex and level of MRD pre-treatment between the patients in Blin-14d and Non-Blin-14d group. More patients in the Non-Blin-14d group had high risk features (7 vs. 2) although this did not reach a statistical significance (p = 0.064).

Table 1.

Patient characteristics

	Blin-14d	Non-Blin-14d
Characteristics	No. of pts	No. of pts	p-value
Sex
Male	18	17
Female	12	13	0.793
Age (Years)
Average	6.97	9.03
Median	6.38	9.21	0.082
Range	1.58–15.43	2.25–15.93
Risk Group
Intermediate	28	23
High	2	7	0.064
MRD (%)*	Prior to Blin Initiation	At D46
Average	0.289	0.233
Median	0.065	0.020	0.690
Range	0.01–2.50	0.01–1.79
CNS Status			0.514+
CNS1	23	26
CNS2	4	1
CNS3	0	0
Traumatic LP	3	3
Trackable Genetic Lesions	23	23
BCR-ABL1	2	5
ETV6-RUNX1	8	4
KMT2A-r	1	0

^*MRD levels shown are from multiparameter flow cytometry (MFC) at the end of induction therapy. NGS-MRD was also performed but is not the data presented in this table. For patients who received blinatumomab FIRST (n = 23): MRD level at D46; for patients who received CAM^plus FIRST (n = 7): MRD level prior to blinatumomab initiation

⁺For CNS status, categories were dichotomized as CNS1 vs. non-CNS1 (including CNS2, CNS3, and traumatic LP) for analysis. Comparison performed using Fisher's exact test (two-sided)

MRD response, event-free and overall survival

One week after Blin-14d therapy, all patients (100%) achieved MFC-MRD-neg remission. Among the 23 patients with a trackable specific genetic marker at diagnosis, 14 remained positive after induction. After Blin-14d therapy, 12 of these 14 (86%) achieved molecular complete remission (mCR), defined as negativity for their specific genetic marker. Patient 9 and 12 were BCR-ABL1(p190) fusion gene positive, and continued to receive dasatinib during the infusion. In Non-Blin-14d group, the MRD-neg remission rate was 63% (17/27) among pediatric patients who received early intensification therapy (CAM^plus regimen) following MRD-pos on day 46. Given the imbalance in baseline risk stratification between the groups, a multivariable logistic regression was performed. After adjusting for risk group (HR vs. IR), the Blin-14d group showed a strong and independent predictor of achieving MFC-MRD-neg response [adjusted odds ratio (aOR) = 28.92, 95% CI: 5.77-144.95, p < 0.001].

In the exploratory part of this study, 11 of 30 patients were retrospectively tested for NGS-MRD (Table S2), among whom pt1 failed to obtain pre/post-Blin-14d NGS-MRD data due to poor data quality and pt7 was not tested for NGS-MRD at primary diagnosis. For pre-Blin-14d MRD monitoring, NGS-MRD was undetectable for pt8 with MFC-MRD positive status (0.09%). As noted above, all patients achieved hematological CR with MFC-MRD level < 0.01% after blinatumomab administration. However, the NGS-MRD was detectable in 4 of 10 patients, with a median level of 0.0133% (0.0005%-0.7170%).

The final follow-up was on October 1st, 2025. For patients in both group who chose other treatment regimens, follow-up continued despite being off-protocol. The median time from D46 MRD-pos to first detection of MRD-neg in bone marrow was significantly earlier and consistent in Blin-14d group compared to Non-Blin-14d [39 (IQR: 33.5-53.75) days vs. 42.5 (IQR: 30-99.5) days, p = 0.0243] (Fig. 2A). The median follow-up from D46 MRD-pos was 27.6 (7.9–41.6) months in Blin-14d group and 30.2 (2.4–46.4) months in Non-Blin-14d group. The event-free survival (EFS) between the Blin-14d and Non-Blin-14d group was 96.7% vs. 86.5% for 1-year EFS and 93.3% vs. 75.7% for 2-year. The 3-year EFS was 73.9 ± 14.5% vs. 71.5 ± 8.6% [hazard ratio for disease progression or mortality, 0.50 (95% CI, 0.16–1.6); p = 0.2334]. The 3-year overall survival (OS) was 100% for the Blin-14d group vs. 85.9% ± 6.6% for the Non-Blin-14d group [hazard ratio for mortality, 0.12 (95% CI, 0.02–0.87); p = 0.0360] (Fig. 2).

Fig. 2.

MRD response, event-free survival and overall survival. Kaplan-Meler curves for (A) Cumulative incidence of MRD conversion, (B) Event-free survival and (C) Overall survival, Blin-14d group versus Non-Blin-14d group

Figure 3 shows the MFC-MRD response and outcomes in individual patients at different time points throughout the study. In Blin-14d group, pt3 experienced an MFC-MRD relapse 14.9 months after Blin-14d, subsequently achieved a second MFC-MRD complete remission (MFC-MRD-CR2) following higher-dose chemotherapy by the last follow-up (35.5 months). Pt4 developed a CNS relapse 33.4 months after Blin-14d and received CD19⁺ CAR-T therapy, with allo-HSCT planned at the last follow-up (37.4 months). Pt7 relapsed 22.5 months after Blin-14d, received CD19⁺ CAR-T and experienced a second relapse at 35.5 months. Pt9 underwent allo-HSCT off-protocol while in CR due to persistent BCR-ABL1(p190)-pos and family preference, and remained alive at 34.4 months. In Non-Blin-14d group, nine patients experienced disease progression, including two HR patients died (one from early disease progression and the other from a severe treatment-related infection), six with bone marrow relapse and one with MFC-MRD relapse. Treatment deviations occurred in 16 patients (53%) in this group, of whom 3 received Blin-14d after consolidation therapy, three received Blin-28d for persistent MRD > 0.01%, and three received CAR-T for bone marrow relapse. PtN27 received a standard 28-day course before CAT^plus therapy and showed MFC-MRD-neg at 14-day assessment. Among later-relapsing patients, two died from disease progression, while 4 underwent allo-HSCT and survived.

Fig. 3.

Swimmer plot of MRD response and outcomes in individual patients. The time line is synchronized for different chemotherapy blocks according to protocol time lines. The actual time of starting a block for individual patients may differ. The red square indicates the conversion of minimal residual disease (MRD). HSCT, hematopoietic stem cell transplantation; Blin, blinatumomab; CAR-T, Chimeric Antigen Receptor T-Cell Immunotherapy; CAM^plus consists of cyclophosphamide, cytarabine, mercaptopurine, vincristine, and pegaspargase

Adverse events of blinatumomab

During the Blin-14d infusion, twenty-three (77%) adverse events were reported in 22 patients (Table 2). Three (10%) experiencing grade 3 AEs, including severe numbness of limbs, ALT increasing and fungal infection. No grade 4 or higher events related to blinatumomab was reported. One patient had postponed treatment due to COVID-19 infection. The most frequent manifestation of CRS was fever, particularly a moderate fever of > 38.3 °C. The median time to CRS onset from blinatumomab infusion was 1.25 (1–4) days, and the median time to resolution was 2 days. All adverse events related to blinatumomab were managed including early intervention before infusion and when fever of ≥ 37.3 °C occurred. As the study design did not require stepping dosing, during the Blin-14d infusion, 14 (47%) patients were dose escalated and 16 (53%) patients received full dose. The median level of IL-6 at day2 was 17.78 pg/ml in the dose escalated vs. 57.29 pg/ml in the full dose group, respectively (p = 0.0130) (Fig. 4A) and the incidence of AEs was 57% vs. 94% (p = 0.0309).

Table 2.

Adverse events

	Total (N = 30)
	No. of pts (%)
Adverse events
Any*	23(77)
Fever	22(73)
Abdominal pain	3(10)
Neurologic symptom	3(10)
Infection	3(10)
Nausea	2(7)
Hypotension	2(7)
Edema	2(7)
Alanine aminotransferase increased	1(3)
Hypokalemia	1(3)
Telangiectasia	1(3)
Headache	1(3)

*Among all patients. Multiple events have occurred in some patients

Fig. 4.

Safety and pharmacodynamics of blinatumomab. (A) The change of IL-6 in peripheral blood during blinatumomab infusion by stepping-up or not; (B) The course of CD4 + and CD8 + T cells as well as NK cells in the peripheral blood in all assessable patients and (C-F) patients who received Blin-14d before/after CAM^plus. Data are median and interquartile range (IQR)

At follow-up, immunoglobulin levels were reviewed in 13 patients, 12 of whom showed hypogammaglobulinemia. Serum immunoglobulin G (IgG) levels were monitored as an indication for B-cell function. A significant decrease in IgG was observed immediately after blinatumomab treatment. Levels continued a progressive decline throughout subsequent chemotherapy phases, with recovery beginning only after therapy cessation. Prophylactic intravenous immunoglobulin (IVIG) was administered for IgG levels <3 g/L. Most infection events coincided with neutropenic periods from chemotherapy rather than with hypogammaglobulinemia.

Pharmacodynamics of blinatumomab

Data for pharmacodynamics of blinatumomab were obtained from all patients in the Blin-14d group. Count of lymphocyte subpopulations were calculated for the indicated number of evaluable patients per detected point and the mean count is shown for each point (Fig. 4B). Following the initiation of blinatumomab infusion, detectable circulating B cells quickly decreased from the peripheral blood, accompanied by a rapid decline in T cell counts. Longitudinal assessment revealed post-treatment recovery and expansion of total T cells, CD4⁺, and CD8⁺ subsets above baseline levels. Total T cells increased from a baseline median of 893/µL to 1228/µL at day 15 (end of infusion) and 1857/µL at day 22. Corresponding changes in counts of CD4⁺ T cells were 425/µL to 465/µL and 721/µL later, and counts of CD8⁺ T cells were 374/µL to 612/µL and 909/µL later.

Twenty-three patients (77%) received Blin-14d therapy before, and 7 (23%) after CAM^plus. Figure 4 C-F selectively show the proliferative trends of total T cells, CD4⁺ and CD8⁺ T cells, and NK cells respectively during and one week after the infusion. Comparative analysis between these two groups revealed that total T-cell median count was 1287/µL vs. 1308/µL (p > 0.99) on day 15 and 1961/µL vs. 1165/µL (p = 0.007) at day 22. Correspondingly, CD4⁺ T cells were 583/µL vs. 503/µL (p > 0.99) at day 15 and 1961/µL vs. 1165/µL (p = 0.007) at day 22, and CD8⁺ T cells counts were 644/µL vs. 673/µL (p > 0.95) at day 15 and 997/µL vs. 673/µL (p = 0.112) at day 22.

Discussion

This prospective study explored the combination of short-term blinatumomab (Blin-14d) into early chemotherapy for pediatric B-ALL patients with MRD-pos. The design of our study was based on three hypotheses. First, although most treatment failures manifest as later relapses, early treatment response such as MRD level is a powerful prognostic factor. Therefore, early interventions designed to improve MRD clearance are an attractive strategy for mitigating long-term relapse risk. Second, we hypothesized that starting blinatumomab infusion with low tumor burden would probably improve response rates with lower incidence of adverse events [25]. Third, we hypothesized that shortening of blinatumomab infusion to 14, instead of 28 days, would better allow persistence of tumor-directed T-cell cytotoxicity.

The ability of Blin-14d to achieve universal MFC-MRD-neg and NGS-MRD-neg in 60% of evaluable patients proves its potency in eradicating MRD. The deeper remission observed with NGS, which detects leukemic clones at sensitivity up to 10^{− 6}~ 10^{− 8}, aligns with evidence linking NGS-MRD-neg to superior long-term outcomes [26]. Notably, the discordance in one patient (NGS-MRD-neg before treatment but MFC-MRD-pos) highlights technical limitations of NGS, including variability due to sample quality or clonal evolution [27]. Despite using a MFC panel designed to detect blasts independent of CD19 expression (gating on CD22 and cyCD79a), discrepancies with NGS-MRD were observed in some patients. This likely arises from their distinct detection principles that MFC identifies cells with a specific, viable immunophenotype, whereas NGS detects persistent malignant clones based on their DNA signature, independent of cell viability or phenotypic expression.

Although the non-randomized design led to an imbalance with more high-risk patients in the control arm, a multivariable analysis confirmed that the superior MRD-CR rate of the Blin-14d group was independent of baseline risk stratification. The EIT achieved a 63% MRD-CR rate in the Non-Blin-14d group, as the non-randomized controlled design allowed patients who failed to achieve MRD-CR after CAM^plus to subsequently enroll in the Blin-14d group. This crossover led to an overestimation of the purely EIT-induced MRD-CR rate in the control group. After recalculation, the corrected rate is 53%, which aligns with the assumed EIT-induced MRD-CR rate used in our initial sample size calculation. However, the overall data suggest that earlier MRD conversion in the Blin-14d group relieved the need for intensive interventions such as HSCT or cellular immunotherapy compared to the Non-Blin-14d group, where 53% required treatment escalation for MRD persistent or disease relapse. Furthermore, the treatment deviations highlight the aggressive nature of MRD-pos B-ALL and the importance of developing effective early interventions, like the Blin-14d regimen, to alter the disease course and avoid the need for such intensive salvage therapies. It is important to note that achieving NGS-MRD-neg did not universally guarantee cure in our cohort. Three of seven patients who attained this deep molecular remission subsequently relapsed (Table S1). This observation underscores that even remission detectable below the threshold of 10⁻⁶ may not eradicate all residual disease, potentially due to the persistence of a reservoir undetectable by current methods or the evolution of resistant clones. This residual risk necessitates continued MRD surveillance and may guide the exploration of extended or combination immunotherapies for long-term remission.

Two patients (pt9 and pt12) with BCR-ABL1 (p190)-pos B-ALL in our cohort exhibited a “CML-like” phenotype, as described by Zuna et al. [28], characterized by persistent low level BCR-ABL1 transcript despite morphological remission and MFC-MRD-neg. Consistent with observations by Jun Li et al. [29], in such cases, the risk of B-ALL relapse may be better reflected by the immunophenotypic disease burden (MFC-MRD) rather than the quantitative level of the fusion transcript alone. Therefore, in line with standard ALL response criteria, we considered these patients as remission based on sustained MFC-MRD-neg, while monitoring their BCR-ABL1 levels as a separate and chronic marker akin to CML. This approach underscores the importance of distinguishing between the leukemic clone and the underlying tyrosine kinase-driven “CML-like” clone, which may require long-term TKI management.

Prior studies of 28-day courses, including the work by Philipp et al. [16], have established that while blinatumomab induces durable responses, its prolonged duration can lead to T-cell exhaustion, as evidenced by upregulation of markers such as PD-1 and TIM-3. In our study, we observed a different pattern of T-cell kinetics. Following the 14-day infusion, we noted a recovery in CD4⁺ and CD8⁺ T-cell counts. While this quantitative recovery does not directly assess functional status, it is consistent with the hypothesis that a shorter treatment duration may mitigate the prolonged activation that drives exhaustion, as demonstrated mechanistically in prior studies. This favorable kinetic profile warrants further investigation with functional assays in future trials. An interesting observation was the compromised T-cell expansion in subgroup patients receiving CAM^plus chemotherapy before Blin-14d. This may be related to the known lymphodepleting effects of multi-agent chemotherapy regimens. Components of the CAM^plus regimen, such as cyclophosphamide and cytarabine, are well-documented to induce transient but significant depletion of circulating lymphocytes, including T-cell subsets [30, 31]. This creates an immunosuppressive environment that may delay the recovery of a functional T-cell population necessary for an effective blinatumomab response.

This study has several limitations. First, the principal limitation is the absence of a comparator group receiving standard 28-day blinatumomab. A randomized trial comparing the 14-day and 28-day regimens is necessary to establish non-inferiority. Second, given the non-randomized allocation and frequent treatment deviations (especially in the Non-Blin-14d group), survival comparisons are vulnerable to confounding factors. Therefore, these survival signals should be considered exploratory. An unexpected finding was the favorable OS trend in the Blin-14d group without a corresponding improvement in EFS. This pattern may be explained by the time-dependent nature of the treatment effect. Specifically, early EFS favored the Blin-14d group, reflecting rapid MRD clearance. However, this initial advantage diminished over time. Given the updated median follow-up of 31.8 months, the small sample size and relatively short observation period limit definitive conclusions about EFS, as relapses in pediatric ALL often occur beyond this timeframe. Extended follow-up is needed to determine whether early MRD conversion translates into long-term survival benefits. The favorable OS trend likely reflects differential access to effective salvage therapies (e.g., immunotherapy, HSCT) following events. Patients in the Blin-14d group had better access to such options, whereas financial constraints limited patients in the Non-Blin-14d arm.

Several mechanistic and clinical questions remain unresolved. The impact of TKI coadministration such as dasatinib in Ph⁺ B-ALL on blinatumomab efficacy and T-cell function remains unclear due to limited data. Preclinical work suggests TKIs modulate T-cell activity, but clinical correlations are needed [32]. Furthermore, the pharmacodynamic relationship between chemotherapy-induced lymphodepletion and blinatumomab requires basic studies, for example, tracking T-cell receptor clonality or exhaustion markers post-therapy. Further trials could explore risk-adapted blinatumomab durations such as 14-day versus 28-day based on MRD thresholds, as well as combinations with novel immunotherapies like PD-1 inhibitors to sustain T-cell responses. Additionally, integrating NGS-MRD into clinical decision-making could refine prognostication, particularly for patients with IGK/L clones, whose significance remains ambiguous.

Conclusions

In conclusion, our study shows encouraging clinical activity and manageable safety profile for Blin-14d therapy, which also demonstrated earlier responses and therapeutic consistency compared to Non-Blin-14d group. For newly diagnosed pediatric B-ALL patients with MRD-pos, early chemotherapy followed by short-term blinatumomab would be a promising option. Further research must address optimal timing of dosing, long-term immune effects, and combined approaches to prolong remission durability.

Electronic Supplementary Material

Below is the link to the electronic supplementary material.

Author contributions

L.Z. and X.Z. designed and supervised the clinical trial. Y.M. designed the sample size for this study. W.Y., Y.Z., Y.G., Y.C. and X.C. executed the research and collected the data. C.X., C.W. and Y.Z. analyzed the data and drafted the original manuscript. All authors reviewed and approved the final version of this manuscript. C.X., C.W. and Y.Z. contributed equally to this article. All authors have approved the article.

Funding

The project was supported by the National Key Research and Development Program of China (2021YFE0106900), the CAMS Innovation Fund for Medical Sciences (CIFMS 2022-I2M-1-022, 2021-I2M-1-041), the National Natural Science Foundation of China (82270144, 82270189), and the Haihe Laboratory of Cell Ecosystem Innovation Fund (22HHXBSS00039).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics

Ethical Committee approval code: IIT2021056-EC-3.

Patient consent

Informed consent was obtained from the parents, guardians, or patients, as appropriate.

Clinical trial information

ChiCTR2100053318.

Competing interests

The authors declare no competing interests.