Efficacy and safety of rabbit ATLG and ATG in allogeneic hematopoietic stem cell transplantation for children with acquired severe aplastic anemia

Abstract

Severe aplastic anemia (SAA) is a life-threatening bone marrow failure syndrome characterized by bone marrow hypoplasia and peripheral blood cytopenia. Without timely treatment, it frequently proves fatal. Rabbit anti-thymocyte globulin (ATG) and anti-human T lymphocyte globulin (ATLG) are widely used for graft-versus-host disease (GVHD) prophylaxis. However, their comparative efficacy in pediatric SAA remains undetermined. This study involved a single-center retrospective analysis of two ATG preparations in pediatric patients undergoing allo-HSCT. The primary endpoint was the incidence of GVHD and viral reactivation following HSCT. Secondary endpoints included overall survival (OS), GVHD-free and failure-free survival (GFFS), neutrophil engraftment, platelet engraftment, hemorrhagic cystitis (HC), tolerability, and toxicities within each group. A total of 124 pediatric SAA patients who underwent their first allo-HSCT between January 2019 and March 2024 were enrolled, with 35 receiving ATLG and 89 receiving ATG. OS, GFFS, GVHD, and HC incidence were comparable between the ATLG and ATG groups (OS: 95.2% vs. 92.9%, P = 0.617). ATLG significantly reduced the incidence of 180-day CMV (45.7% vs. 74.2%, P = 0.0062) and EBV reactivation (29.8% vs. 52.8%, P = 0.025). Additionally, ATLG was associated with fewer adverse events (AEs), including fever (P = 0.009) and rash (P = 0.018). ATLG demonstrated comparable efficacy to ATG in preventing GVHD and achieving OS in pediatric SAA patients undergoing allo-HSCT, while significantly reducing viral reactivation and AEs. These findings support ATLG as a safer alternative, warranting further prospective studies.

Introduction

Severe aplastic anemia (SAA) is a life-threatening bone marrow failure syndrome characterized by bone marrow hypoplasia and peripheral blood cytopenia. In the absence of timely treatment, it is often fatal [1]. Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is a curative approach widely utilized for both hematologic malignancies and non-malignant disorders [2]. In pediatric SAA patients, allo-HSCT remains one of the most effective treatment options, with a 2-year overall survival rate exceeding 90% [3].

Anti-thymocyte globulins (ATGs) are extensively used as immunomodulatory agents for the prevention and treatment of graft-versus-host disease (GVHD) in allo-HSCT and for the management of acute rejection in solid organ transplantation [4, 5]. Rabbit-derived anti-human T lymphocyte globulin (ATLG) and ATG are the most commonly used preparations. ATG is produced by immunizing rabbits with human thymocytes, whereas ATLG is derived from the Jurkat human T-lymphoblastic cell line [6]. Previous studies have investigated the effects of ATLG and ATG in adult patients with hematologic malignancies and SAA undergoing allo-HSCT [7–9]. However, data comparing their efficacy and safety in pediatric patients with SAA undergoing allo-HSCT remain scarce. This study aims to evaluate and compare the clinical outcomes and safety profiles of ATLG and ATG in this specific patient population.

Methods

Patients

Pediatric SAA cases undergoing their first allo-HSCT between January 2019 and March 2024 were enrolled. Inclusion criteria were as follows: (i) a diagnosis of SAA (absolute neutrophil count < 0.5 × 10⁹/L, platelets < 20 × 10⁹/L, reticulocytes < 20 × 10⁹/L, marrow cellularity < 25%) or very severe aplastic anemia (VSAA) (absolute neutrophil count < 0.2 × 10⁹/L), (ii) age below 18 years at the time of HSCT, and (iii) no prior history of HSCT. Exclusion criteria included: (i) inherited bone marrow conditions, such as Fanconi anemia, Diamond-Blackfan anemia, or dyskeratosis congenita, and (ii) missing data. The selection criteria for ATG vs. ATLG administration were strictly based on the recipient’s prior immunosuppressive therapy history. For pediatric SAA patients who had previously undergone immunosuppressive therapy with ATG combined with cyclosporine A (CSA), ATLG was systematically administered during the pre-transplant conditioning phase. Conversely, ATG remained the primary agent of choice for treatment-naïve patients without prior ATG-based immunosuppression.

Transplantation procedure

The conditioning regimen for haploidentical donors (HID, reduced-intensity conditioning [RIC]) and matched unrelated donors (MUD, RIC) included fludarabine (45 mg/m²/day) on days − 7 to −4, busulfan (3.2 mg/kg/day) on days − 7 and − 6, cyclophosphamide (30 mg/kg/day) on days − 5 to −2, and either ATG (2.5 mg/kg/day) on days − 5 to −2 or ATLG (20 mg/kg total) on days − 4 to −2. For matched related donors (MRD, RIC), the conditioning regimen included fludarabine (40 mg/m²/day) on days − 6 to −3, cyclophosphamide (40 mg/kg/day) on days − 5 to −2, and either ATG (2.5 mg/kg/day) on days − 5 to −2 or ATLG (20 mg/kg total) on days − 4 to −2. For umbilical cord blood transplantation (UCB, myeloablative conditioning [MAC]), the regimen included fludarabine (45 mg/m²/day) on days − 5 to −2, busulfan (3.2 mg/kg/day) on days − 9 to −7, cyclophosphamide (40 mg/kg/day) on days − 6 to −4, and ATG (2.5 mg/kg/day) on days − 8 to −5. For pediatric patients demonstrating HLA antibody positivity, 375 mg/m² of rituximab was administered on day − 1 prior to HSCT to facilitate antibody clearance. UCB units sourced from cord blood banks in Shanghai, Shandong, or Guangdong Province, China, were infused 8 h prior to bone marrow (BM) infusion for recipients in the BM or BM + PB (peripheral blood) groups, and 8 h before PB transplantation in the PB group [10, 11].

Outcome assessment

The primary endpoint was the incidence of GVHD and viral reactivation post-HSCT. Secondary endpoints encompassed overall survival (OS), GVHD-free and failure-free survival (GFFS), neutrophil engraftment, platelet engraftment, hemorrhagic cystitis (HC), tolerability, and toxicities in each group. OS was defined as the time from allo-HSCT to death. GFFS was defined as the absence of grade III-IV acute GVHD, extensive chronic GVHD, graft failure, or death from any cause [12]. Neutrophil engraftment was defined as achieving an absolute neutrophil count ≥ 0.5 × 10⁹/L for 3 consecutive days. Platelet engraftment was defined as achieving a platelet count ≥ 20 × 10⁹/L for 7 consecutive days without transfusion support. HC was defined and graded according to previous criteria [13]. The diagnosis and grading of acute GVHD (aGVHD) and chronic GVHD (cGVHD) followed established criteria [14, 15]. CMV reactivation was confirmed when peripheral blood CMV DNA reached ≥ 500 copies/mL in two consecutive tests, and EBV reactivation was defined as EBV DNA ≥ 500 copies/mL in two consecutive tests.

Statistical analysis

The t-test was used for continuous variables with normal distributions, while the Mann–Whitney U test was applied for non-normally distributed variables. The chi-square test or Fisher’s exact test was employed to compare categorical variables. Survival analyses were conducted using the Kaplan-Meier method, with P values calculated by the log-rank test. Univariate and multivariate analyses were performed to compare hazard ratios obtained through Cox regression or competing risk models. Cumulative incidence curves were used in the competing risk setting to calculate the probabilities of outcomes in the presence of competing events. Variables with a P value ≤ 0.20 in univariate analysis were included in multivariate Cox proportional hazards regression and competing risk models. All statistical analyses were performed using IBM SPSS Statistics 25 (IBM, Armonk, NY, USA) and R version 4.3.3. A P value < 0.05 was considered statistically significant.

Results

Baseline characteristics

A total of 124 pediatric SAA patients undergoing their first allo-HSCT were included in this study. Among these patients, 35 received ATLG, while 89 received ATG for GVHD prophylaxis. The median age at HSCT was 8.3 years (range, 1.9–17.7 years), with no significant difference between the ATLG and ATG groups (8.3 [2.4–17.8] vs. 8.3 [1.9–16.2] years, P = 0.461). The cohort consisted of 69 SAA and 55 VSAA cases. No significant differences were observed between the two groups in the infusion doses of CD34 + cells or mononuclear cells. Furthermore, various other factors, including sex, disease severity, graft source, conditioning regimen, donor type, HLA matching, GVHD prophylaxis strategies, donor-patient sex matching, blood type, and the proportions of letermovir and rituximab use were comparable between the two groups. The baseline characteristics of the patients in both groups are summarized in Table 1.

Table 1.

Clinical characteristics of 124 pediatric SAA cases

Variable	ATLG (n = 35)	ATG (n = 89)	P value
Age, median (range), years	8.3(2.4–17.8)	8.3(1.9–16.2)	0.461
Sex, [n (%)]			0.515
Male	17(48.6%)	49(55.1%)
Female	18(51.4%)	40(44.9%)
Disease, [n (%)]			0.163
SAA	16(45.7%)	53(59.6%)
VSAA	19(54.3%)	36(40.4%)
Graft source, [n (%)]			0.185
BM + PB + CB	9(25.7%)	13(14.6%)
PB	14(35.0%)	40(44.9%)
BM + PB	9(25.7%)	30(33.7%)
BM + CB	1(2.9%)	0(0%)
PB + CB	2(5.7%)	2(2.3%)
CB	0(0%)	4(4.5%)
Conditioning regimen, [n (%)]			0.486
MAC	1(2.9%)	1(1.1%)
RIC	34(97.1%)	88(98.9%)
CMV serostatus, [n (%)]			1.000
Positive	34(97.1%)	87(97.7%)
Negative	1(2.9%)	2(2.3%)
EBV serostatus, [n (%)]			0.701
Positive	33(94.3%)	84(94.4%)
Negative	2(5.7%)	5(5.6%)
Donor type, [n (%)]			0.111
HID	24(68.6%)	48(53.9%)
MRD	2(5.7%)	18(20.2%)
MUD	9(25.7%)	19(21.4%)
UCB	0(0%)	4(4.5%)
HLA matched, [n (%)]			0.302
5/10	10(28.6%)	21(23.6%)
6/10–8/10	13(37.1%)	24(27.0%)
9/10–10/10	12(34.3%)	44(49.4%)
GVHD prophylaxis, [n (%)]			0.274
CsA + MMF + others	19(54.3%)	62(70.0%)
FK506 + MMF + others	16(45.7%)	27(30.0%)
MNC, median (range), ×10 8 /kg	7.41 (1.81–11.45)	6.87 (0.274–15.67)	0.766
CD34+, median (range), ×10 6 /kg	6.08 (0.57–15.00)	6.85 (0.107–17.47)	0.266
Donor-patient sex matched, [n (%)]			0.253
Female–male	8(22.9%)	24(27.0%)
Female–female	10(28.5%)	12(13.5%)
Male–female	8(22.9%)	28(31.5%)
Male–male	9(25.7%)	25(28.0%)
Blood type, [n (%)]			0.922
ABO comparability	20(57.1%)	51(57.3%)
Major ABO incomparability	6(17.1%)	12(13.5%)
Minor ABO incomparability	8(22.9%)	21(23.6%)
Major–minor incomparability	1(2.9%)	5(5.6%)
Letermovir, [n (%)]			0.176
Yes	6(17.1%)	6(6.7%)
No	29(82.9%)	83(93.3%)
RTX, [n (%)]			0.809
Yes	22(62.9%)	58(65.2%)
No	13(37.1%)	31(34.8%)

Abbreviations: ATLG Rabbit anti-human T lymphocyte globulin, ATG, anti-thymocyte globulin, SAA severe aplastic anemia, VSAA very severe aplastic anemia, BM bone marrow, PB peripheral blood, CB cord blood, MAC myeloablative conditioning, RIC reduced-intensity conditioning, HID haploidentical donor, MRD matched related donor, MUD matched unrelated donor, UCB umbilical cord blood, HLA human leukocyte antigen, GVHD graft-versus-host disease, MNC mononuclear cells, RTX rituximab

Long time survival

As of July 1, 2024, the median follow-up duration was 33.0 months (range, 4–56 months). A total of six patients died: five (83.3%) due to severe infections and one (16.7%) due to severe GVHD. The 3-year OS rates post-HSCT were 95.2% for the ATLG group and 92.9% for the ATG group, with no significant difference observed (P = 0.617, Fig. 1a). Similarly, the 3-year GFFS rates were comparable between the two groups, at 79.9% for the ATLG group and 80.7% for the ATG group (P = 0.780, Fig. 1b).

Fig. 1.

Kaplan-Meier curves of (a) OS between ATLG and ATG; (b) GFFS between ATLG and ATG

Engraftment and GVHD

The median time to neutrophil engraftment was 11 days in both groups (P = 0.792, Fig. 2a), while the median time to platelet engraftment was 12 days in the ATLG group and 11 days in the ATG group (P = 0.624, Fig. 2b). In total, 59 patients developed grade I-IV aGVHD at a median of 18 days (range, 10–81 days) after HSCT. Meanwhile, 20 patients developed grade II-IV aGVHD at a median of 17 days (range, 11–56 days). The cumulative incidence of 100-day grade I-IV aGVHD was 52.0% in the ATLG group and 46.1% in the ATG group, with no significant difference between the two groups (P = 0.479, Fig. 3a). The cumulative incidence of grade II-IV aGVHD also did not show a significant difference between the groups, being 20.1% for ATLG and 14.6% for ATG (P = 0.441, Fig. 3b). Furthermore, the 3-year cumulative incidence of cGVHD was comparable between the ATLG group and the ATG group (35.2% vs. 43.4%; P = 0.270; Fig. 4a).

Fig. 2.

Cumulative incidence of (a) neutrophil engraftment between ATLG and ATG; (b) platelet engraftment between ATLG and ATG

Fig. 3.

Cumulative incidence of (a) 100-day I-IV aGVHD between ATLG and ATG; (b) 100-day II-IV aGVHD between ATLG and ATG

Fig. 4.

Cumulative incidence of (a) 3-year cGVHD between ATLG and ATG; (b) 3-year HC between ATLG and ATG

HC and viral infection

The 3-year cumulative incidence of HC was comparable between the two groups, standing at 34.5% in the ATLG group and 29.3% in the ATG group (P = 0.421, Fig. 4b). In contrast, significant differences were observed in viral reactivation rates. The cumulative incidence of 180-day CMV reactivation was significantly lower in the ATLG group compared to the ATG group (45.7% vs. 74.2%, P = 0.0062; Fig. 5a). Similarly, patients in the ATLG group exhibited a significantly lower cumulative incidence of 180-day EBV reactivation compared to the ATG group (29.8% vs. 52.8%, P = 0.025; Fig. 5b). To further explore the factors influencing CMV and EBV reactivation, univariable and multivariable competing risk model analyses were performed. Univariable analysis identified that the use of ATLG and letermovir were associated with a reduced incidence of CMV reactivation (Fig. 6a). In multivariable analysis, these two factors remained independent protective factors for CMV reactivation (P = 0.007 and P = 0.0029, respectively; Table 2). Additionally, an HLA-matched donor was also identified as an independent protective factor for CMV reactivation (Table 2). Regarding EBV reactivation, univariable and multivariable analyses demonstrated that the use of ATLG and rituximab (administered at a dose of 375 mg/m² on day − 1 before HSCT) was associated with a lower incidence of EBV reactivation (Fig. 6b). Moreover, the analyses indicated that the use of a matched donor was a risk factor for EBV reactivation (Table 3).

Fig. 5.

Cumulative incidence of (a) 180-day CMV reactivation between ATLG and ATG; (b) 180-day EBV reactivation between ATLG and ATG

Fig. 6.

Cumulative incidence of (a) 180-day CMV reactivation between letermovir and control; (b) 180-day EBV reactivation between rituximab and control

Table 2.

Univariable and multivariable completing risk model analyses of 180-day CMV reactivation

Factor	Univariable analysis		Multivariable analysis
	HR (95% CI)	P value	HR (95% CI)	P value
Male vs. female	0.923(0.599–1.420)	0.710
Age at HSCT	0.999 (0.994–1.001)	0.670
CsA vs. FK506	0.997 (0.635–1.570)	0.990
PB vs. others	0.749 (0.482–1.160)	0.200
ABO matched vs. mismatched	0.943 (0.612–1.450)	0.790
HLA matched vs. mismatched	0.665 (0.424–1.040)	0.075	0.537 (0.335–0.862)	0.01*
HID vs. others	1.270 (0.822–1.970)	0.280
MNC ≥ 7.055 × 108/kg	1.060 (0.631–1.780)	0.830
CD34 + ≥ 6.615 × 106/kg	0.917 (0.597–1.410)	0.690
Letermovir, yes vs. no	0.236 (0.0852–0.655)	0.0056	0.237 (0.092–0.611)	0.0029*
RTX, yes vs. no	0.767 (0.494–1.190)	0.240
ATLG vs. ATG	0.472 (0.268–0.834)	0.0057	0.438 (0.241–0.789)	0.007*

※Divided bt the median. †Use at 1 day before HSCT

Table 3.

Univariable and multivariable completing risk model analyses of 180-day EBV reactivation

Factor	Univariable analysis		Multivariable analysis
	HR (95% CI)	P value	HR (95% CI)	P value
Male vs. female	1.25 (0.738–2.110)	0.410
Age at HSCT	0.997 (0.991–1.001)	0.340
CsA vs. FK506	1.450 (0.818–2.570)	0.200
PB vs. others	1.59 (0.950–2.670)	0.077	0.809 (0.384–1.707)	0.580
ABO matched vs. mismatched	1.470 (0.856–2.540)	0.160	1.439 (0.822–2.519)	0.200
HLA matched vs. mismatched	1.730 (1.030–2.920)	0.038	2.296 (1.119–4.711)	0.023*
HID vs. others	0.759 (0.450–1.280)	0.300
MNC ≥ 7.055 × 108/kg	1.160 (0.751–1.780)	0.510
CD34 + ≥ 6.615 × 106/kg	1.140 (0.676–1.910)	0.630
Letermovir, yes vs. no	0.474 (0.169–1.330)	0.100	0.662 (0.214–2.049)	0.470
RTX, yes vs. no	0.163 (0.094–0.280)	<0.001	0.130 (0.074–0.229)	<0.001*
ATLG vs. ATG	0.461 (0.228–0.935)	0.032	0.327 (0.150–0.714)	0.005*

Adverse events

Both treatment groups exhibited tolerable toxicities. Commonly observed adverse events (AEs) included fever ≥ 38.0 °C (111/124, 89.5%), rash (45/124, 36.3%), and capillary leak syndrome (1/124, 0.8%). The incidence of fever was significantly lower in the ATLG group compared to the ATG group (77.1% vs. 94.4%, P = 0.009). Similarly, the incidence of rash was significantly reduced in the ATLG group (20.0% vs. 42.7%, P = 0.018). Capillary leak syndrome occurred in only one case within the ATG group, with no significant difference between the two groups (0% vs. 1.1%, P = 1.000; Fig. 7).

Fig. 7.

Comparion of adverse events between ATLG and ATG

Immune reconstitution

The immune reconstitution profiles of lymphocyte subsets in the ATLG and ATG groups at 30 days after HSCT are presented in Fig. 8. No statistically significant intergroup differences were observed in absolute counts of total lymphocytes (median: 527.00 vs. 419.00 cells/µL, P = 0.391), T lymphocytes (386.13 vs. 241.31 cells/µL, P = 0.266), CD4 + T-cell subsets (67.87 vs. 50.82 cells/µL, P = 0.375), CD3-CD19 + B cells (0.29 vs. 0.28 cells/µL, P = 0.092), or CD3-(CD16 + CD56+) natural killer cells (137.53 vs. 94.56 cells/µL, P = 0.821). Notably, quantitative analysis demonstrated a consistent trend toward higher median values across all lymphocyte subpopulations in the ATLG cohort compared with the ATG group, suggesting a potential acceleration of early immune reconstitution in the ATLG group.

Fig. 8.

Comparative evaluation of immune recovery between ATLG and ATG at 30 days post-transplant

Discussion

Owing to disparities in manufacturing processes, rabbit ATLG and ATG manifest distinct pharmacokinetic and pharmacodynamic profiles. Both agents primarily exert immunosuppressive effects through depleting T lymphocytes; however, ATLG exhibits higher specificity towards activated T lymphocytes [16]. While most studies have focused on their effects in adult HSCT recipients, investigations within pediatric populations remain relatively scarce. The influence of different ATG preparations on survival outcomes and complications in HSCT recipients continues to be debated.

AGVHD constitutes a major complication subsequent to allo-HSCT, contributing to transplant-associated morbidity and mortality [15]. Reported incidence rates of II-IV aGVHD vary from 21.74 to 43% in the ATLG group and 20.46–42% in the ATG group [7, 16, 17]. In the present study, the incidence of grade II-IV aGVHD was 20.1% in the ATLG group and 14.6% in the ATG group, both lower than those previously reported. This discrepancy may be attributed to differences in hematologic diseases as well as the higher proportion of matched donors in our cohort. Prior studies have suggested that neither ATG preparation significantly affects aGVHD incidence in hematologic malignancies following HSCT [8, 18]. In both pediatric and adult SAA populations, Liu et al. reported that ATG was associated with a higher risk of III-IV aGVHD compared to ATLG [7]. Nevertheless, our findings demonstrated no significant differences in the incidence rates of I-IV or II-IV aGVHD between the ATLG and ATG groups, potentially due to differences in underlying diseases and ATG dosing regimens. The reported incidence of cGVHD ranged from 11.8 to 66.7% [19]. Previous eviendence have demonstrated that haplo-HSCT are associated with an elevated risk of GVHD [2]. The relatively higher proportion of haplo-HSCT (72/124, 58.1%) and lower proportion of MRD-HSCT (20/124, 16.1%) in our cohort may contribute to the observed increased incidence of cGVHD. Several studies have reported that ATG reduces the incidence of extensive cGVHD compared with ATLG in patients undergoing unrelated donor HSCT for hematologic malignancies [16–18]. In this study, the incidence of cGVHD was also lower in the ATLG group than in the ATG group (35.2% vs. 43.4%).

It is commonly suggested that ATLG requires a 2–3-fold higher dose than ATG to achieve comparable immunosuppressive efficacy and clinical equivalence. Huang et al. compared the clinical outcomes of 10 mg/kg ATG and 20 mg/kg ATLG in patients with hematological malignancies undergoing unrelated PBSCT and observed significant lower incidences of cGVHD and fever/chills with ATLG [17]. In another two comparative studies of patients with hematological malignancies receiving either unrelated or haplo-HSCT, Wang et al. [18] and Zhang et al. [19] demonstrated that a 20 mg/kg dose of ATLG exhibited comparable aGVHD, cGVHD and survival with 10 mg/kg ATG, but fewer CMV infections. Despite differences in disease types between our enrolled patient cohort and those in previous studies, the outcomes related to GVHD and safety profiles demonstrated similarities. Liu et al. [7] reported that 10 mg/kg ATG was associated with a significantly lower incidence of grade III-IV aGVHD and faster platelet recovery compared to 20 mg/kg ATLG. However, no statistically significant differences were observed in 5-year OS or GRFS between the two treatment groups. The discrepancies between our findings and those reported may be attributed to variations in age of patient, donor types and stem cell sources across the study cohorts. To the best of our knowledge, our study represents the first comparison of 10 mg/kg ATG and 20 mg/kg ATLG in pediatric SAA/VSAA patients.

Notably, Zhou et al. observed that ATLG was associated with a lower incidence of EBV infections in patients with hematologic malignancies undergoing haplo-HSCT [8]. Another study indicated that ATLG reduced CMV titers in patients undergoing unrelated donor HSCT [18]. Our results corroborate these findings, revealing lower incidence rates of CMV and EBV reactivation in the ATLG group. This effect could be related to faster immune reconstitution associated with ATLG use [20]. Our results indicated that the immune reconstitution at 30 days post-HSCT was faster in the ATLG group compared to the ATG group. Moreover, ATLG is linked to reduced thymic output, while ATG is associated with increased T cell activation (CD38 + DR + Ki67+) and senescence (CD8 + CD57 + CD28-) [21], which may explain the observed differences in viral reactivation rates between the groups.

Our study also verified the protective roles of letermovir and rituximab in preventing CMV and EBV reactivation. Letermovir, an antiviral agent, inhibits CMV replication by binding to components of the terminase complex [22]. Previous studies have demonstrated that letermovir is both effective and safe in reducing the risk of CMV infection in HSCT recipients [23, 24]. Rituximab, a CD20 antibody, was the first FDA-approved antibody for cancer treatment in 1997 [25]. Patel et al. reported that administering rituximab before HSCT significantly reduces the incidence of EBV reactivation and post-transplant lymphoproliferative disorder (PTLD) in HSCT recipients [26]. Similarly, Marjańska et al. showed that adding rituximab on day + 5 post-HSCT decreases the risk of EBV infection and EBV-PTLD in pediatric SAA HSCT recipients [27]. Our findings suggest that administering rituximab one day before HSCT can significantly reduce EBV reactivation in pediatric SAA patients undergoing allo-HSCT, offering a novel preventive strategy against EBV infection.

Interestingly, in our multivariable analysis, HLA-matched donors were identified as an independent risk factor for EBV reactivation, which contrasts with previous findings [28]. This observed discrepancy could potentially be attributed to the heterogeneous distribution of pre-transplant EBV serostatus (IgG) among HLA-matched and HLA-mismatched donor-recipient pairs, particularly regarding donor-recipient serostatus constellations (e.g., D+/R- vs. D+/R+), although the absence of detailed serological data precluded further validation of this hypothesis through subgroup stratification analysis.

The prognostic roles of ATLG and ATG remain inconsistently reported in the literature [7–9, 29]. In patients with hematological malignancies undergoing HSCT, ATLG has been shown to be associated with higher OS and GRFS compared to ATG [9, 29]. However, in patients with severe aplastic anemia, neither ATLG nor ATG preparations were linked to significant differences in OS or GRFS [7]. This discrepancy suggests that the effects of ATLG may vary depending on the underlying disease. Given the relatively small sample sizes in these studies, more high-quality randomized controlled trials (RCTs) are needed to better elucidate the prognostic roles of ATLG and ATG.

Regarding AEs associated with ATG treatment, previous studies have documented a lower incidence of AEs in the ATLG group compared to the ATG group [30, 31]. Our study also indicates that ATLG is associated with a lower incidence of fever and rash, which might translate to improved quality of life for patients undergoing HSCT.

Several limitations of this study warrant emphasized. Firstly, this was a retrospective study with a relatively small sample size. Secondly, the viral antibody status of both donors and recipients was not available, thereby restricting our ability to further investigate their potential impact on complications and survival in this cohort. Third, owing to the limited sample size, we were unable to pursue further investigation of CMV or EBV infections with clinical significance. Larger-scale, high-quality prospective studies are necessary to further explore the outcomes of these two ATG preparations in pediatric SAA patients undergoing allo-HSCT.

In conclusion, for pediatric SAA allo-HSCT recipients, ATLG demonstrates similar effects on OS and GVHD prophylaxis as ATG. Notably, ATLG reduces the incidence of viral reactivation after allo-HSCT and is associated with a lower risk of acute ATG-related adverse reactions. Larger-scale RCTs are warranted to further confirm these conclusions.

Author contributions

Senlin Zhang, Qi Ji, Li Gao, Qingwei Wang, and Kai Cui: Data and manuscript draft. Minyuan Liu, Bohan Li, Yixin Hu, and Yongping Zhang: Conceptualization and methodology; Yuanyuan Tian and Shengqin Cheng: Statistic analysis and project administration; Jun Lu and Shaoyan Hu: Supervise manuscipt and project administration.

Funding

The National Key Research and Development Program of China (no.2022YFC2502700), the National Natural Science Foundation of China (NSFC 82170218, 82470221) to Shaoyan Hu, NSFC 82100229 to Yuanyuan Tian, NSFC 82200177 to Li Gao, NSFC 82470127 to Yixin Hu, NSFC 82300244 to Bohan Li, NSFC 82400264 to Yongping Zhang, Suzhou Projects (DZXYJ202305, GSWS2023048, 2020ZKPB02) to Shaoyan Hu, 2023QN07 to Shengqin Cheng, and the Suzhou Municipal Key Laboratory (SZS201615, SKY2022012, SZS2023014)to Shaoyan Hu. Soochow University of Medical School, ML13101223 to Shaoyan Hu.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval

The study protocol was approved by the ethics committee of the Children’s Hospital of Soochow University and conducted according to the Declaration of Helsinki. Written informed consent was obtained from all patients.

Consent to publish

Not applicable.

Competing interests

The authors declare no competing interests.