CD71 as an independent prognostic factor in pediatric B-ALL: a regional multicenter retrospective study in Southern China (SCCLG-ALL-2016)

Abstract

Background:

Despite high cure rates in pediatric B-cell acute lymphoblastic leukemia (B-ALL), relapse remains a major challenge. Current risk stratification systems fail to identify a subset of occult high-risk (HR) patients. Transferrin receptor 1 (CD71), a key regulator of iron uptake, is often overexpressed in aggressive tumors; however, its independent prognostic value in pediatric B-ALL has not been fully elucidated.

Objectives:

To evaluate the independent prognostic significance of CD71 expression in a large cohort of pediatric B-ALL patients and to determine its utility in identifying HR subgroups.

Design:

A retrospective, multicenter cohort study.

Methods:

We analyzed 1207 pediatric B-ALL patients treated under the South China Children’s Leukemia Group-ALL-2016 protocol between January 2016 and June 2023 across 20 centers in Southern China. CD71 expression was assessed by flow cytometry at diagnosis, with positivity defined as ⩾20% expression on blast cells. We analyzed associations between CD71 status, clinical characteristics, and survival outcomes using Kaplan–Meier methods and multivariate Cox regression models.

Results:

CD71 positivity was rare in B-ALL (1.0%) compared to T-ALL (15.6%). Clinically, CD71-positive B-ALL patients presented with significantly lower initial white blood cell (WBC) counts (median 4.0 vs 10.0 × 10⁹/L, p = 0.015). Despite this lower peripheral tumor burden, CD71 positivity was associated with extremely poor outcomes. The 5-year cumulative incidence of relapse (CIR) was significantly higher in the CD71-positive group compared to the negative group (39.4% vs 10.1%, p = 0.004). Multivariate analysis confirmed that CD71 positivity is an independent risk factor for relapse (hazard ratio (HR) = 5.94, p = 0.003) and adverse events (HR = 4.868, 95% confidence interval: 1.512–15.671, p = 0.008), irrespective of age, WBC, genetic subtype, or minimal residual disease status.

Conclusion:

CD71 expression identifies a rare but biologically distinct B-ALL subgroup characterized by aggressive behavior despite low initial WBC counts. These findings indicate that the inclusion of CD71 in diagnostic panels may help to optimize current risk stratification systems, serving as a critical tool to recognize occult HR patients who might otherwise be under-stratified by conventional prognostic markers.

Clinical trial registration:

The trial is registered with the Chinese Clinical Trial Registry (Chi-CTR; https://www.chictr.org.cn/; number ChiCTR2000030357; Registration date: February 29, 2020).

Plain language summary

A multicenter study investigating how the protein CD71 helps identify children with B-cell acute lymphoblastic leukemia who are at high risk of relapse despite having “good” initial clinical features

Acute lymphoblastic leukemia (B-ALL) is the most common childhood cancer. While survival rates have improved, relapse remains a leading cause of death. Currently, doctors assess risk using factors like age and white blood cell (WBC) counts. However, some children classified as “low risk” still experience unexpected relapses. This study aimed to determine if measuring a protein called CD71 on leukemia cells could serve as an independent warning sign to identify these hidden high-risk patients. The team conducted a large-scale study involving 1,207 children with B-ALL treated across 16 hospitals in Southern China between 2016 and 2023. They analyzed CD71 expression using flow cytometry at diagnosis. Researchers then tracked long-term health outcomes to see if high CD71 levels were linked to relapse, even after accounting for other known risk factors. High CD71 expression was rare, found in only 1.0% of patients. Crucially, these patients typically had low initial WBC counts, a feature usually associated with a good prognosis. Despite this “favorable” appearance, children with high CD71 were nearly six times more likely to relapse compared to those with low levels. CD71 was confirmed as an independent risk factor for poor outcomes. These findings identify an aggressive leukemia subgroup easily missed by current risk systems. A low WBC count does not always guarantee low risk. The researchers suggest routinely including CD71 in diagnostic screening. Identifying these patients early could allow medical teams to use more intensive or novel treatments to prevent relapse.

Keywords

acute lymphoblastic leukemia B-ALL CD71 pediatric oncology prognostic biomarkers

Introduction

Acute lymphoblastic leukemia (ALL) represents the most common malignancy in children.¹ With the optimization of contemporary risk-stratified therapy, the 5-year overall survival (OS) rate for pediatric ALL now approaches or exceeds 90%.^1–3 However, despite these significant advancements, relapse remains the leading cause of treatment failure and cancer-related mortality in children.^4,5 For instance, a recent report from the South China Children’s Leukemia Group (SCCLG) indicated that while the 3-year cumulative incidence of relapse (CIR) was 4.7% in patients without central nervous system involvement, it rose to 12.8% in high-risk (HR) subgroups,⁶ underscoring the persistent challenge of relapse. Although current risk stratification systems—based on initial white blood cell (WBC) count, age, cytogenetic features, and minimal residual disease (MRD)—have substantially improved outcomes,⁷ clinical outcomes remain heterogeneous; notably, a subset of patients classified as “standard” or “intermediate” risk (IR) continues to experience unexpected relapse.^8,9 Therefore, identifying novel biomarkers to further refine the risk stratification system is of great clinical significance for recognizing occult HR patients.

CD71 (Transferrin Receptor 1, TfR1) serves as a critical transmembrane glycoprotein regulating cellular iron uptake.¹⁰ Given that iron is an essential cofactor for DNA synthesis and cell proliferation, CD71 is frequently overexpressed in malignant cells with high proliferative potential to meet their metabolic demands.^11,12 Indeed, high CD71 expression has been documented as an adverse prognostic indicator in various solid tumors, such as ovarian, breast, and pancreatic cancers, where it is linked to “iron addiction” and ferroptosis sensitivity.^13–15 In the context of acute leukemias, CD71 expression exhibits distinct lineage specificity. Multiple studies have consistently demonstrated that CD71 expression levels in B-cell ALL (B-ALL) are significantly lower than those in acute myeloid leukemia (AML) and T-cell ALL, as well as mixed phenotype acute leukemia (MPAL) and normal bone marrow blasts.^11,16,17 These findings suggest that while B-ALL blasts typically maintain low CD71 levels, aberrant overexpression may signify a distinct pathological state. Although prior single-center studies conducted in Egypt and Poland suggested that CD71 positivity confers an adverse prognosis in ALL,^18,19 specifically, its independent prognostic value within the context of large-scale cohorts, particularly in the Southern Chinese population, has not been fully elucidated. To address this gap, we analyzed a large-scale, multicenter retrospective cohort to evaluate whether CD71 serves as an independent biomarker for identifying occult HR patients overlooked by current risk stratification, thereby potentially guiding more precise therapeutic strategies.

Methods

Study design

This retrospective, multicenter cohort study enrolled patients from 20 tertiary hospitals affiliated with the SCCLG (the complete list of participating centers is provided in Supplemental Table 1). The inclusion criteria were as follows: (1) a diagnosis of ALL between January 2016 and June 2023; (2) age <18 years at diagnosis; and (3) uniform treatment according to the SCCLG-ALL-2016 protocol. Patients were excluded if they met any of the following criteria: (1) mature B-cell leukemia (Burkitt leukemia); (2) MPAL; (3) leukemia secondary to immunodeficiency or occurring as a second malignancy; (4) Down syndrome; (5) prior receipt of corticosteroids for more than 1 week or any anthracyclines before enrollment; or (6) missing flow cytometric CD71 expression data or key clinical follow-up information. The initial cohort comprised 1348 patients (1207 B-ALL and 141 T-ALL) to explore lineage-specific expression patterns of CD71. While T-ALL patients were included in preliminary comparative analyses, they were not included in the subsequent survival and risk-stratification studies because the adverse prognostic impact of CD71 was found to be strictly B-cell lineage-specific. Therefore, the core prognostic analysis was conducted on the final cohort of 1207 pediatric B-ALL patients.

The trial was registered with the Chinese Clinical Trial Registry (ChiCTR) under the number ChiCTR2000030357. This study was conducted in accordance with the Declaration of Helsinki and approved by the Medical Ethics Committee of Sun Yat-sen Memorial Hospital (Number 2020-KY-004). The ethical approval, initially granted in January 2020 and maintained through annual renewals until January 2023, encompassed the retrospective analysis of the entire SCCLG-ALL-2016 cohort, including patients diagnosed from 2016 through 2023. This approval process ensured continuous institutional oversight throughout the data collection period for all participants.

All patients were treated according to the SCCLG-ALL-2016 protocol, which employs a risk-directed stratification strategy described in detail previously.^20,21 Briefly, patients were dynamically stratified into low-risk, IR, and HR groups based on presentation features (age, initial WBC), immunophenotype, cytogenetic/molecular abnormalities, and early treatment response (including prednisone response on day 8 and bone marrow MRD levels on days 15 and 33). Although CD71 is not part of the minimal diagnostic panel for ALL, it was assessed at initial diagnosis using multiparameter flow cytometry as part of the routine “extended immunophenotyping panel” at participating academic centers. All CD71 expression data were retrieved from original, pre-existing diagnostic records; no retrospective staining or analysis of stored biological samples was performed for this research. This study was reported in accordance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines.²²

Characterization of CD71 expression and flow cytometry analysis

Bone marrow samples were collected at diagnosis, and the immunophenotype of leukemic blasts was analyzed using multiparameter flow cytometry. To minimize circadian variations and accommodate clinical fasting requirements for sedation, bone marrow aspirations were consistently performed in the morning (typically 08:00–11:00 AM) across all centers. The initial diagnostic panel mainly served for lineage assignment and preliminary immunophenotypic classification of acute leukemia, that is, to distinguish B-ALL, T-ALL, and myeloid leukemia. Specifically, the panel included B-lineage markers (CD19, CD22, CD20, CD79α, and CD10), T-lineage markers (CD3, cCD3, CD2, CD4, CD5, CD7, and CD8), and myeloid markers (MPO, CD13, CD33, CD14, CD11b, CD15, CD16, CD64, and CD71). Additional markers were incorporated for further immunophenotypic characterization, including precursor/immaturity-associated markers (CD34, CD117, and CD38) and others (CD45, HLA-DR, CD58, CD56, CD99, CD36, CD1a, CD41α, CD61, and CD42b, CD81). Detailed information regarding the complete antibody panel, including manufacturers, clones, and catalog numbers, is provided in Supplemental Table 2. Specifically, the CD71-FITC antibody (Clone L01.1, Cat. No. 665339; BD Biosciences, San Jose, CA, USA) was utilized for this assessment.

And the abnormal immunophenotypic features identified at diagnosis were subsequently used as reference for MRD assessment. For B-ALL, MRD was assessed by multiparameter flow cytometry based on leukemia-associated immunophenotypes identified at diagnosis. Briefly, CD19-positive B-lineage cells were first selected, after which abnormal immature B-cell populations were distinguished from normal mature B cells according to their aberrant antigen-expression patterns. In the representative clinical workflow, the abnormal blast population was characterized by a phenotype such as CD19-positive, CD45-dim/negative, CD34-dim, CD10-positive, CD38-bright, CD58-bright, CD22-positive, CD20-positive, CD81-positive, and CD15-negative, which was considered consistent with residual leukemic B lymphoblasts. The proportion of this aberrant population among total acquired nucleated cells was used to determine MRD status.

CD71 positivity was defined as the expression of CD71 on ⩾20% of blast cells, in accordance with the standard diagnostic criteria for ALL established in Principles and Practice of Pediatric Oncology.²³ Patients with <20% expression were classified as CD71-negative. Based on this cut-off value, patients were stratified into CD71-positive and CD71-negative groups for subsequent analysis of clinical characteristics and survival outcomes.

Outcomes and statistical analysis

The main outcome measures of this study were 5-year event-free survival (EFS) and CIR. EFS was calculated from the date of diagnosis to the first occurrence of an adverse event, defined as induction failure, relapse, death from any cause, or second malignancy. Patients without events were censored at the date of last follow-up. CIR was defined as the time from complete remission to relapse, with death in remission considered a competing risk. The data cutoff date was June 30, 2023.

All statistical analyses were performed using R software (version 4.5.2; R Foundation for Statistical Computing, Vienna, Austria). For baseline comparisons, continuous variables with non-normal distribution (e.g., age and initial WBC) were analyzed using the Wilcoxon rank sum test, while categorical variables (e.g., sex, risk group, genetic subtypes, and MRD status) were compared using Fisher’s exact test. Survival probabilities were estimated via the Kaplan–Meier method and compared using the log-rank test. For relapse analysis, CIR accounting for competing events was constructed by the method of Kalbfleisch and Prentice²⁴ and compared with Gray’s test.²⁵ To evaluate the independent prognostic value of CD71, a multivariate Cox proportional hazards model was utilized,²⁶ adjusting for covariates including age, initial WBC, risk group, and MRD status. All statistical tests were two-sided, and a p-value <0.05 was considered statistically significant.

Results

Expression pattern and prognostic significance of CD71 in ALL lineages

We evaluated CD71 expression and its lineage association in the initial exploratory cohort of 1348 newly diagnosed patients. Given the rarity of CD71 expression in B-ALL and its distinct prognostic implications, which were not observed in the T-ALL group, subsequent survival analyses focused exclusively on the B-ALL cohort. Using a cut-off of ⩾20% blast expression for positivity, CD71 expression exhibited significant lineage heterogeneity: the positivity rate was 15.6% (22/141) in T-ALL, compared to only 1.0% (12/1207) in B-ALL. Survival analysis revealed that the prognostic impact of CD71 was strictly lineage-specific. Among these CD71-positive B-ALL cases, the mean percentage of CD71 expression on blast cells was 32.75% (range: 21.02%–66.55%). In the T-ALL cohort, the 5-year EFS and CIR curves for CD71-positive and CD71-negative groups were superimposable (EFS: 76.5% vs 72.9%, p = 0.834; CIR: 19.0% vs 21.3%, p = 0.873; Figure 1(a)). In sharp contrast, CD71 positivity in the B-ALL cohort was associated with significantly inferior outcomes: the 5-year EFS was 60.6% (95% CI: 37.0%–100.0%) for the CD71-positive group versus 86.6% (95% CI: 84.6%–88.6%) for the negative group (p = 0.042). Correspondingly, the 5-year CIR was markedly higher in the CD71-positive group (39.4%; 95% CI: 15.6%–100.0%) compared to the negative group (10.1%; 95% CI: 8.3%–12.3%; p = 0.004; Figure 1(b)). Given the rarity of CD71 expression in B-ALL and its distinct prognostic implications, subsequent analyses focused exclusively on the B-ALL cohort.

Figure 1.

Prognostic impact of CD71 expression on survival outcomes in pediatric ALL lineages. (a) T-ALL Cohort: Kaplan–Meier curves for EFS (left) and CIR (right) stratified by CD71 status. No significant differences were observed (p > 0.05). (b) B-ALL cohort: Kaplan–Meier curves for EFS (left) and CIR (right). CD71 positivity is significantly associated with inferior EFS (p = 0.042) and higher relapse risk (p = 0.004).

Clinical characteristics of CD71-positive B-ALL

A total of 1207 B-ALL patients were stratified into CD71-positive (n = 12) and CD71-negative (n = 1195) groups. The clinical characteristics of these two groups are summarized in Table 1. Notably, the CD71-positive group presented with a significantly lower median initial WBC count compared to the CD71-negative group (4.0 × 10⁹/L (interquartile range (IQR): 2.0–8.0) vs 10.0 × 10⁹/L (IQR: 4.0–35.0); p = 0.015). Despite the lower initial tumor burden, a significantly higher proportion of CD71-positive patients were stratified into the clinical HR group compared to their negative counterparts (p = 0.044). In contrast, no statistically significant differences were observed between the two groups regarding age, sex distribution, or major fusion genes (e.g., BCR::ABL1, ETV6::RUNX1).

Table 1.

Clinical characteristics of B-ALL patients by CD71 expression.

Group	Clinical parameters	Overall N = 1207	Negative N = 1195	Positive N = 12	p
Age, years	Median	4.5	4.5	4.8	0.9^a
	(IQR)	3.0–7.4	3.0–7.3	3.1–8.9
	<1	6 (0.5%)	6 (0.5%)	0 (0%)	>0.9
	1–9.9	1028 (85%)	1017 (85%)	11 (92%)	>0.9
	⩾10	173 (14%)	172 (14%)	1 (8.3%)	>0.9
Sex	Male	718 (59%)	709 (59%)	9 (75%)	0.4
	Female	489 (41%)	486 (41%)	3 (25%)
WBC, ×10⁹/L	Median	10	10	4	0.015 ^a
	(IQR)	4–34	4–35	2–8
	<50	980 (81%)	969 (81%)	11 (92%)	0.7
	⩾50	227 (19%)	226 (19%)	1 (8.3%)	0.7
Cytogenetics	Normal karyotype	713 (59%)	706 (59%)	7 (58%)	>0.9
	Hyperdiploidy	169 (14%)	169 (14%)	0 (0%)	0.4
	BCR::ABL1	70 (5.8%)	69 (5.8%)	1 (8.3%)	0.5
	KMT2A-rearranged (mll)	24 (2.0%)	24 (2.0%)	0 (0%)	>0.9
	ETV6::RUNX1 (tel/aml1)	133 (11%)	133 (11%)	0 (0%)	0.4
	IKZF1	112 (9.3%)	112 (9.4%)	0 (0%)	0.6
	Ph-like	46 (3.8%)	46 (3.8%)	0 (0%)	>0.9
	Others	147 (12%)	145 (12%)	2 (17%)	0.6
Risk group	Low	283 (23%)	279 (23%)	4 (33%)	0.5
	Intermediate	613 (51%)	605 (51%)	8 (67%)	0.4
	High	311 (26%)	311 (26%)	0 (0%)	0.044
Short-term outcomes	D15 MRD+	161 (13%)	161 (13%)	0 (0%)	0.4
	D33 MRD+	953 (79%)	945 (79%)	8 (67%)	0.3

Statistics: ^aCalculated using Wilcoxon rank-sum test. All other p values were calculated using Fisher’s exact test. Bold p values indicate statistical significance (p < 0.05).

B-ALL, B-cell acute lymphoblastic leukemia; CD71, transferrin receptor-1; IQR, interquartile range; MRD, minimal residual disease; WBC, white blood cell count.

Precise prognostic impact of CD71 in B-ALL subgroups

To determine whether the prognostic value of CD71 is confounded by other risk factors, we performed subgroup analyses (Table 2). The results demonstrated that the adverse prognostic impact of CD71 remained consistent across multiple key clinical subgroups. Notably, in patients with initial WBC <50 × 10⁹/L, the relapse rate was significantly elevated in CD71-positive patients (p = 0.039). Similarly, CD71 positivity indicated extremely poor outcomes within the normal karyotype subgroup (p < 0.001). Furthermore, among patients who achieved MRD negativity (<0.01%) on day 33 of induction therapy, the 5-year CIR remained significantly higher in CD71-positive patients (p = 0.002). This indicates potential chemoresistance or a tendency for early relapse in CD71-positive leukemia, necessitating vigilance even in the context of negative MRD.

Table 2.

Precise prognostic features of CD71 in B-ALL.

Parameter	Subgroup	n (CD71+/−)	5-Year EFS, % (95% CI)			5-Year CIR, % (95% CI)
Parameter	Subgroup	n (CD71+/−)	CD71+	CD71−	p	CD71+	CD71−	p
Age, years	<1	0/6	–	–	–	–	–	–
	1–9.9	11/1017	59.1 (30.6–100.0)	90.2 (87.8–92.8)	0.011	40.9 (0.0–84.8)	6.8 (4.5–9.1)	<0.001
	⩾10	1/172	0.0	76.5 (67.6–86.7)	0.681	0.0	18.2 (9.0–27.5)	0.735
Sex	Male	9/709	11.1	88.9 (85.8–92.0)	0.759	11.1	8.3 (5.5–11.2)	0.478
	Female	3/486	66.7	87.8 (83.6–92.1)	0.001	66.7	8.1 (4.3–11.9)	<0.001
WBC, ×10⁹/L	<50	11/969	67.5 (37.0–100.0)	91.1 (88.8–93.5)	0.154	32.5 (0.0–79.0)	6.4 (4.3–8.5)	0.039
	⩾50	1/226	100.0	76.9 (69.1–85.7)	0.014	100.0	16.0 (8.2–23.9)	<0.001
Risk group	Low	4/279	25.0	96.8 (94.6–99.0)	0.008	25.0	2.1 (0.2–4.0)	<0.001
	Intermediate/high	8/916	25.0	85.5 (82.2–88.8)	0.305	25.0	10.5 (7.4–13.6)	0.100
Cytogenetics	Normal karyotype	7/706	28.6	90.2 (87.3–93.3)	0.008	28.6	7.0 (4.2–9.8)	<0.001
	Hyperdiploidy	0/169	–	–	–	–	–	–
	BCR::ABL1	1/69	100.0	76.3 (62.4–93.3)	0.009	100.0	15.5 (1.3–29.8)	<0.001
	ETV6::RUNX1	0/133	–	–	–	–	–	–
	KMT2A-rearranged	0/24	–	–	–	–	–	–
	Others	2/145	50.0	83.9 (76.3–92.3)	0.161	50.0	9.3 (2.7–15.9)	0.031
D15 MRD	<0.01% (negative)	6/445	16.7	92.6 (89.3–96.0)	0.189	16.7	5.6 (2.5–8.7)	0.095
	0.01%–0.99% (low)	6/580	33.3	87.8 (84.0–91.7)	0.039	33.3	8.7 (5.1–12.2)	0.004
	⩾1% (high)	0/161	–	–	–	–	–	–
D33 MRD	<0.01% (negative)	4/250	25.0	96.7 (94.3–99.2)	0.004	25.0	2.9 (0.6–5.2)	0.002
	0.01%–0.99% (low)	8/629	25.0	88.6 (85.3–92.0)	0.150	25.0	8.1 (5.1–11.2)	0.036
	⩾1% (high)	0/316	–	–	–	–	–	–

For subgroups with limited sample size (n < 10), crude rates are reported without confidence intervals to avoid statistical instability. Statistics: EFS, log-rank test; CIR, Gray’s test. Bold p values denote comparisons reaching statistical significance.

B-ALL, B-cell acute lymphoblastic leukemia; CD71, transferrin receptor-1; CI, confidence interval; CIR, cumulative incidence of relapse; EFS, event-free survival; MRD, minimal residual disease; WBC, white blood cell count.

Univariate and multivariate analysis of prognostic factors for relapse

In univariate analysis for treatment outcomes, CD71 positivity was identified as a significant risk factor for both adverse events (hazard ratio (HR) = 3.105, p = 0.042) and relapse (HR = 4.970, p = 0.004). In addition to CD71, age ⩾ 10 years, initial WBC ⩾ 50 × 10⁹/L, HR group assignment, and positive D15/D33 MRD were confirmed as indicators of poor prognosis. Notably, although infants (age <1 year) are typically considered a HR group, this variable did not reach statistical significance in the univariate analysis of this cohort (p > 0.05), likely due to the limited sample size (Supplemental Table 3). To rigorously evaluate the independent prognostic value of CD71, we incorporated CD71 status into a multivariate Cox proportional hazards model alongside established prognostic factors, including age (<1 or ⩾10 years), initial WBC, clinical risk group, and MRD status. The results demonstrated that even after adjusting for these key clinical confounders, CD71 positivity remained an independent risk factor for relapse (HR = 5.940, 95% CI: 1.837–19.204, p = 0.003) and adverse events (HR = 4.868, 95% CI: 1.512–15.671, p = 0.008; Table 3). These findings confirm that CD71 expression provides additional prognostic information independent of the current risk stratification system.

Table 3.

Multivariate analysis of treatment outcomes in B-ALL.

Variable	Number of patients (no/yes)	Number of events (no/yes)	HR	95% CI	P
Event-free survival
CD71 positive	1195/12	94/3	4.868	1.512–15.671	0.008
High-risk group	896/311	52/45	2.528	1.593–4.014	<0.001
Age <1 or ⩾10 years	1028/179	71/26	1.988	1.250–3.162	0.004
WBC ⩾50 ×10⁹/L	980/227	65/32	1.492	0.936–2.379	0.093
BCR::ABL1 or KMT2Ar	1113/94	85/12	0.810	0.422–1.555	0.526
D15 MRD ⩾0.01%	460/747	25/72	1.087	0.632–1.870	0.763
D33 MRD ⩾0.01%	254/953	8/89	1.653	0.689–3.962	0.260
Cumulative incidence of relapse
CD71 positive	1195/12	58/3	5.940	1.837–19.204	0.003
Male gender	489/718	22/39	1.122	0.664–1.895	0.667
Age <1 or ⩾10 years	1028/179	44/17	2.370	1.321–4.255	0.004
WBC ⩾50 × 10^9/L	980/227	42/19	1.815	1.001–3.291	0.050
BCR::ABL1 or KMT2Ar	1113/94	53/8	1.230	0.552–2.742	0.613
D15 MRD ⩾0.01%	460/747	17/44	1.422	0.712–2.842	0.318
D33 MRD ⩾0.01%	254/953	7/54	1.147	0.415–3.172	0.792

Variables for analysis: CD71 positivity, high-risk group, male gender, age (<1 or ⩾10 years), diagnostic WBC (⩾50 × 10⁹/L), genetics (BCR::ABL1 or KMT2Ar), MRD ⩾0.01% at D15 or D33. Statistics: Cox proportional hazards model. Bold p values denote comparisons reaching statistical significance.

B-ALL, B-cell acute lymphoblastic leukemia; CD71, transferrin receptor-1; CI, confidence interval; HR, hazard ratio; MRD, minimal residual disease; WBC, white blood cell count.

Discussion

This study, utilizing a large-scale multicenter cohort, evaluates the prognostic significance of CD71 (transferrin receptor 1) in pediatric B-ALL. Although the CD71-positive phenotype is infrequent in B-ALL, it identifies a specific patient subgroup associated with poor outcomes. Multivariate analysis indicates that CD71 serves as a prognostic predictor for relapse and adverse events, independent of conventional risk factors such as MRD, age, and cytogenetic abnormalities.

The prevalence of CD71 positivity in B-ALL observed in our study (1.0%) is significantly lower than that reported in previous small-scale, single-center studies (approx. 20%–40%).¹⁸ This discrepancy is likely attributable to differences in sample size and methodological heterogeneity. Previous studies were limited by small sample sizes (e.g., n = 75), rendering them susceptible to selection bias. In contrast, our study enrolled 1207 consecutive unselected cases, thereby providing a more accurate epidemiological estimation of this phenotype in the general pediatric B-ALL population. Regarding detection methodology, variations in flow cytometry platforms, antibody clones, and fluorochromes are inevitable between studies and within a multicenter consortium.²⁷ As noted by Leung et al. in a similar large-scale leukemia study, such inter-laboratory variability tends to filter out cases with marginal or weak antigen expression. Consequently, the 1% subset identified in our study likely represents a cell population characterized by robust overexpression, effectively excluding cases with low-level background expression. This stringent biological selection may explain why the prognostic association of CD71 in our cohort (HR = 5.940) is markedly stronger than previously reported.

Although high CD71 expression is typically linked to cellular proliferation, this poor-prognosis subgroup presented with a lower tumor burden. Rodriguez et al.²⁸ highlighted that cancer cells upregulate CD71 to increase iron uptake, leading to an expanded intracellular labile iron pool, and that this heightened iron dependency renders cancer cells metabolically vulnerable to oxidative stress. This vulnerability appears particularly relevant in the context of B-ALL. Lalonde et al.²⁹ demonstrated via genome-wide CRISPR screening that B-ALL cells typically exhibit low expression of the ferroptosis suppressor FSP1 (AIFM2). Consequently, they rely heavily on the GSH-GPX4 axis for antioxidant defense and display inherent sensitivity to ferroptosis, an iron-dependent form of cell death. This might lead to lower tumor burden and suggests that inducing ferroptosis, for instance through GPX4 inhibition, could represent a rational therapeutic strategy for this refractory subgroup, warranting further preclinical investigation.

Clinical characterization revealed an inverse association between CD71 expression and initial peripheral WBC counts. One possible explanation is that CD71-positive tumors may restrict iron availability to surrounding immune cells, thereby suppressing the proliferation of immune cells. In line with this, iron deficiency has been shown to induce a hypoproliferative state in T cells.^30,31

The aberrant overexpression of CD71 in this HR subgroup represents a distinct metabolic deviation from the typical B-ALL phenotype, aligning instead with the dysregulated iron homeostasis characteristic of aggressive malignancies.³² A recent study by Yanagiya et al.³³ demonstrated that CD71 expression in leukemic cells is dynamically regulated by the microenvironmental nutrient status (e.g., amino acid supply) and is preferentially maintained within specific niches such as the bone marrow. Considering the link between CD71 and leukemia stem cell characteristics described by Cosialls et al.,³⁴ CD71-positive blasts may preferentially retain within the bone marrow niche rather than mobilizing into the peripheral circulation.

Our study has several limitations. First, its retrospective design and the small absolute number of CD71-positive B-ALL cases resulted in wide confidence intervals and limited statistical power for detailed subgroup analyses. Second, the median follow-up time was 28.7 months because patient recruitment continued in recent years. As a result, the 5-year estimates were less precise than those at earlier time points. Nevertheless, 17.3% of the cohort (n = 209) had follow-up of at least 5 years, which enabled the identification of late relapse events in the CD71-positive group occurring around 4 years after diagnosis. These findings suggest that reporting 5-year endpoints remains clinically informative, although the long-term estimates should still be interpreted with caution.

Nevertheless, given the rarity of this phenotype (~1%) in clinical practice, the present analysis provides essential data characterizing this specific subpopulation. Future prospective studies are warranted to validate the clinical utility of incorporating CD71 into diagnostic screening. Furthermore, elucidating the iron-dependent vulnerabilities of these cells could support the development of targeted strategies, such as ferroptosis induction, potentially offering alternative therapeutic options for this specific refractory subgroup.

Conclusion

Supplemental Material

sj-doc-4-tam-10.1177_17588359261445060 – Supplemental material for CD71 as an independent prognostic factor in pediatric B-ALL: a regional multicenter retrospective study in Southern China (SCCLG-ALL-2016)

Supplemental material, sj-doc-4-tam-10.1177_17588359261445060 for CD71 as an independent prognostic factor in pediatric B-ALL: a regional multicenter retrospective study in Southern China (SCCLG-ALL-2016) by Xing-Fei Chen, Wen-Guang Jia, Ying Liu, Li-Li Liu, Xing-Jiang Long, Li-Hua Yu, Wen-Yong Kuang, Yun Cai, Liu-Hua Liao, Bei-Yan Wu, Qiao-Ru Li, Ying Fu, Hao-Yan Ren, Guo-Hua Chen, Li-Na Wang, Jian Wang, Nan Zhong, Zi-Jun Zhen, Hui-Rong Mai, Xian-Ling Kong, Jia-Jie Chen, Ya-Yun Ling, Chuan Wen, Yun-Yan He and Yan-Lai Tang in Therapeutic Advances in Medical Oncology

Supplemental Material

sj-docx-1-tam-10.1177_17588359261445060 – Supplemental material for CD71 as an independent prognostic factor in pediatric B-ALL: a regional multicenter retrospective study in Southern China (SCCLG-ALL-2016)

Supplemental material, sj-docx-1-tam-10.1177_17588359261445060 for CD71 as an independent prognostic factor in pediatric B-ALL: a regional multicenter retrospective study in Southern China (SCCLG-ALL-2016) by Xing-Fei Chen, Wen-Guang Jia, Ying Liu, Li-Li Liu, Xing-Jiang Long, Li-Hua Yu, Wen-Yong Kuang, Yun Cai, Liu-Hua Liao, Bei-Yan Wu, Qiao-Ru Li, Ying Fu, Hao-Yan Ren, Guo-Hua Chen, Li-Na Wang, Jian Wang, Nan Zhong, Zi-Jun Zhen, Hui-Rong Mai, Xian-Ling Kong, Jia-Jie Chen, Ya-Yun Ling, Chuan Wen, Yun-Yan He and Yan-Lai Tang in Therapeutic Advances in Medical Oncology

Supplemental Material

sj-docx-2-tam-10.1177_17588359261445060 – Supplemental material for CD71 as an independent prognostic factor in pediatric B-ALL: a regional multicenter retrospective study in Southern China (SCCLG-ALL-2016)

Supplemental material, sj-docx-2-tam-10.1177_17588359261445060 for CD71 as an independent prognostic factor in pediatric B-ALL: a regional multicenter retrospective study in Southern China (SCCLG-ALL-2016) by Xing-Fei Chen, Wen-Guang Jia, Ying Liu, Li-Li Liu, Xing-Jiang Long, Li-Hua Yu, Wen-Yong Kuang, Yun Cai, Liu-Hua Liao, Bei-Yan Wu, Qiao-Ru Li, Ying Fu, Hao-Yan Ren, Guo-Hua Chen, Li-Na Wang, Jian Wang, Nan Zhong, Zi-Jun Zhen, Hui-Rong Mai, Xian-Ling Kong, Jia-Jie Chen, Ya-Yun Ling, Chuan Wen, Yun-Yan He and Yan-Lai Tang in Therapeutic Advances in Medical Oncology

Supplemental Material

sj-docx-3-tam-10.1177_17588359261445060 – Supplemental material for CD71 as an independent prognostic factor in pediatric B-ALL: a regional multicenter retrospective study in Southern China (SCCLG-ALL-2016)

Supplemental material, sj-docx-3-tam-10.1177_17588359261445060 for CD71 as an independent prognostic factor in pediatric B-ALL: a regional multicenter retrospective study in Southern China (SCCLG-ALL-2016) by Xing-Fei Chen, Wen-Guang Jia, Ying Liu, Li-Li Liu, Xing-Jiang Long, Li-Hua Yu, Wen-Yong Kuang, Yun Cai, Liu-Hua Liao, Bei-Yan Wu, Qiao-Ru Li, Ying Fu, Hao-Yan Ren, Guo-Hua Chen, Li-Na Wang, Jian Wang, Nan Zhong, Zi-Jun Zhen, Hui-Rong Mai, Xian-Ling Kong, Jia-Jie Chen, Ya-Yun Ling, Chuan Wen, Yun-Yan He and Yan-Lai Tang in Therapeutic Advances in Medical Oncology

Footnotes

Acknowledgements

We express our gratitude to all the patients and their families for their participation in this study. We are also grateful to the clinical and research staff at the participating centers for their assistance with patient management and data collection.

Declarations

ORCID iDs

Xing-Fei Chen

Yan-Lai Tang

Supplemental material

Supplemental material for this article is available online.

References

Evans

Melton

Liu

, et al. Do surveillance tests for relapse improve survival after chemotherapy for paediatric acute leukaemia? A systematic review. Pediatr Blood Cancer 2025; 72: e31973.

Inaba

Mullighan

CG.

Pediatric acute lymphoblastic leukemia. Haematologica 2020; 105: 2524–2539.

Kato

Manabe

Treatment and biology of pediatric acute lymphoblastic leukemia. Pediatr Int 2018; 60: 4–12.

Anagnostou

Riaz

Hashmi

, et al. Anti-CD19 chimeric antigen receptor T-cell therapy in acute lymphocytic leukaemia: a systematic review and meta-analysis. Lancet Haematol 2020; 7: e816–e826.

Bhojwani

Pui

C-H.

Relapsed childhood acute lymphoblastic leukaemia. Lancet Oncol 2013; 14: e205–e217.

L-H

Geng

Liao

, et al. Prognostic significance of CNSL at diagnosis of childhood B-cell acute lymphoblastic leukemia: a report from the South China Children’s Leukemia Group. Front Oncol 2022; 12: 943761.

Teachey

Hunger

SP.

Predicting relapse risk in childhood acute lymphoblastic leukaemia. Br J Haematol 2013; 162: 606–620.

Hunger

Mullighan

CG.

Acute lymphoblastic leukemia in children. N Engl J Med 2015; 373: 1541–1552.

Nguyen

Devidas

Cheng

S-C

, et al. Factors influencing survival after relapse from acute lymphoblastic leukemia: a Children’s Oncology Group study. Leukemia 2008; 22: 2142–2150.

10.

Daniels

Delgado

Rodriguez

, et al. The transferrin receptor part I: Biology and targeting with cytotoxic antibodies for the treatment of cancer. Clin Immunol 2006; 121: 144–158.

11.

Wei

Y-Y

Zhang

X-Z

Zhang

, et al. [Expression of CD71 on cell proliferation in hematologic malignancy and its correlation with Ki-67]. Zhongguo Shi Yan Xue Ye Xue Za Zhi 2015; 23: 234–240.

12.

Crépin

Goenaga

A-L

Jullienne

, et al. Development of human single-chain antibodies to the transferrin receptor that effectively antagonize the growth of leukemias and lymphomas. Cancer Res 2010; 70: 5497–5506.

13.

Basuli

Tesfay

Deng

, et al. Iron addiction: a novel therapeutic target in ovarian cancer. Oncogene 2017; 36: 4089–4099.

14.

Zhongqing

Uddin

Wence

Elevated expression of transferrin receptor-1 in pancreatic cancer: clinical implications and prognostic significance. Clin Exp Med 2025; 25: 307.

15.

Szulc

Woźniak

Targeting ferroptosis as the Achilles’ heel of breast cancer: mechanisms and therapeutic opportunities from a comprehensive systematic review. Int J Mol Sci 2025; 26: 9902.

16.

Pande

Dorwal

Jain

, et al. Expression of CD71 by flow cytometry in acute leukemias: more often seen in acute myeloid leukemia. Indian J Pathol Microbiol 2016; 59: 310–313.

17.

Liu

Wang

, et al. Significance of CD71 expression by flow cytometry in diagnosis of acute leukemia. Leuk Lymphoma 2014; 55: 892–898.

18.

Hagag

Badraia

Abdelmageed

, et al. Prognostic value of transferrin receptor-1 (CD71) expression in acute lymphoblastic leukemia. Endocr Metab Immune Disord Drug Targets 2018; 18: 610–617.

19.

Płoszyńska

Ruckemann-Dziurdzińska

Jóźwik

, et al. Cytometric evaluation of transferrin receptor 1 (CD71) in childhood acute lymphoblastic leukemia. Folia Histochem Cytobiol 2012; 50: 304–311.

20.

Dai

J-T

Y-Y

Zhang

Y-T

, et al. A multicenter study on the prognostic value of prednisone response in pediatric acute lymphoblastic leukemia. Ann Hematol 2025; 104: 5855–5866.

21.

Qiu

Luo

, et al. Prognostic value and outcome for ETV6/RUNX1-positive pediatric acute lymphoblastic leukemia: a report from the South China Children’s Leukemia Group. Front Oncol 2021; 11: 797194.

22.

Von Elm

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 2007; 370: 1453–1457.

23.

Poplack

Steuber

Rabin

, et al. Acute lymphoblastic leukemia. In: Pizzo

Poplack

(eds) Principles and practice of pediatric oncology. Philadelphia: Lippincott Williams & Wilkins, 2011, pp. 518–565.

24.

Kalbfleisch

Prentice

RL.

The statistical analysis of failure time data. 2nd ed. Hoboken, NJ: J. Wiley, 2002.

25.

Gray

RJ.

A class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann Statist 1988; 16: 1141–1154.

26.

Cox

DR.

Regression models and life-tables. J R Stat Soc Ser B Stat Methodol 1972; 34: 187–202.

27.

Leung

Cai

Liu

, et al. Prognostic implications of CD9 in childhood acute lymphoblastic leukemia: insights from a nationwide multicenter study in China. Leukemia 2024; 38: 250–257.

28.

Rodriguez

Schreiber

Conrad

Persister cancer cells: Iron addiction and vulnerability to ferroptosis. Mol Cell 2022; 82: 728–740.

29.

Lalonde

M-E

Sasseville

Gélinas

A-M

, et al. Genome-wide CRISPR screens identify ferroptosis as a novel therapeutic vulnerability in acute lymphoblastic leukemia. Haematologica 2023; 108: 382–393.

30.

Sun

J-L

Zhang

N-P

R-C

, et al. Tumor cell-imposed iron restriction drives immunosuppressive polarization of tumor-associated macrophages. J Transl Med 2021; 19: 347.

31.

Voss

Sewell

Krystofiak

, et al. Elevated transferrin receptor impairs T cell metabolism and function in systemic lupus erythematosus. Sci Immunol 2023; 8: eabq0178.

32.

Torti

FM.

Iron and cancer: more ore to be mined. Nat Rev Cancer 2013; 13: 342–355.

33.

Yanagiya

Miyatake

Watanabe

, et al. Amino acid influx via LAT1 regulates iron demand and sensitivity to PPMX-T003 of aggressive natural killer cell leukemia. Leukemia 2024; 38: 1731–1741.

34.

Cosialls

El Hage

Dos Santos

, et al. Ferroptosis: cancer stem cells rely on iron until “to die for” it. Cells 2021; 10: 2981.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.02 MB

0.07 MB

0.00 MB

0.01 MB