Early Post-Transplant Serum Ferritin Levels as Predictive Biomarkers for Severe Acute Graft-Versus-Host Disease in Pediatric Umbilical Cord Blood Transplantation for Acute Leukemia

Introduction

Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is a valuable therapy option for patients with acute leukemia (AL). Umbilical cord blood, as a conveniently available donor source with low immunogenicity and less stringent human leukocyte antigen (HLA)-matching requirements, has emerged as an alternative donor source because of the scarcity of HLA-matched related donors [1,2]. Although the outcomes of umbilical cord blood transplantation (UCBT) have considerably improved, acute graft-versus-host disease (aGVHD) remains the most common and potentially life-threatening complication [3,4]. The risk factors for aGVHD and outcomes of aGVHD in adults using conventional donor sources are well documented. HLA matching, sex disparity between the donor and recipient, conditioning intensity, increased age, multiparous female donors, ineffective aGVHD prophylaxis, graft source, gut bacterial diversity, and total body irradiation have been implicated as risk factors for the onset of aGVHD [5–10]. However, little is known about risk factors for aGVHD after UCBT in children, and it is unclear whether the risk factors and outcomes are similar to those of conventional donor sources in adults.

Patients undergoing allo-HSCT are often transfused with multiple units of blood products, which can result in iron overload. Serum ferritin (SF) is an acute-phase reactant protein which can become elevated in response to iron overload. SF levels have been shown to correlate with long-term outcomes. According to a meta-analysis of 25 trials involving 4545 patients, a high pretransplant SF level was associated with worse overall survival (OS) [11]. Elevated post-transplant SF levels have also been linked to increased non-relapse mortality and lower OS [12–14]. Furthermore, Lin et al [15] demonstrated that increased pretransplant SF levels are linked to a higher incidence of grade II–IV aGVHD and grade III–IV aGVHD in patients with severe aplastic anemia (SAA). However, the relationship between post-transplant SF and aGVHD in children with AL after UCBT is yet to be clarified.

Therefore, we performed a retrospective study at our center to analyze the outcomes of aGVHD and possible risk factors in pediatric patients with AL after UCBT. The purpose was to identify risk factors for aGVHD to improve the quality of life after UCBT in children with AL.

Material and Methods

PATIENTS:

This study included children aged 1–18 years diagnosed with AL who underwent UCBT at Children’s Hospital of Soochow University, Suzhou, China, from 2017 to 2022. The patients’ parents and stem cell donors provided written informed consent before beginning the transplantation process. The patient selection criteria were as follows: (i) diagnosis of acute myeloid leukemia (AML), acute lymphoblastic leukemia (ALL), or mixed phenotype acute leukemia (MAL); (ii) aged 0–18 years at the time of transplantation; (iii) received a single-unit of cord blood; (iv) no prior allo-HSCT; and (v) neutrophils were successfully engrafted.

SUPPORTIVE CARE:

Calcineurin inhibitors (cyclosporine or tacrolimus) were used in combination with methotrexate or mycophenolate mofetil for GVHD prophylaxis. All patients were given routine supportive treatment, including antiviral (acyclovir) prophylaxis for herpes simplex virus-positive recipients, preemptive cytomegalovirus prophylaxis, and antifungal prophylaxis.

DEFINITIONS:

AGVHD was diagnosed and graded according to previously published consensus guidelines [16]. OS considered death from any cause as the event, and surviving patients were censored at the date of last contact. Neutrophil engraftment was defined as an absolute neutrophil count ≥0.5×109/L for ≥3 days. Platelet engraftment was defined as platelet count ≥20×109/L without transfusion support for ≥7 days.

STATISTICAL ANALYSIS:

To compare differences between groups, the chi-square test and Mann–Whitney U test were used to assess the statistical significance of differences in categorical and continuous variables, respectively, between groups. Univariate and multivariable analyses were performed using Gray’s test and the Fine and Gray proportional subdistribution hazards regression model. Death before day 180 without severe aGVHD was considered a competing event. Variables with P<0.10 were included in the multivariable model. OS was estimated using the Kaplan-Meier method, and differences were compared using log-rank tests. The statistical analyses were performed using RStudio software (version 4.3.1; RStudio Team, Boston, MA, USA; URL: http://www.rstudio.com/).

Results

OUTCOMES OF ACUTE GRAFT-VERSUS-HOST DISEASE:

We compared the OS and relapse-free survival (RFS) of the patients between groups of different grades of aGVHD. This study did not find a statistically significant difference in the OS and RFS of patients with grade I–IV aGVHD (Figure 1). However, patients with grade II–IV or grade III–IV aGVHD had worse OS (Figures 2 and 3, 69.4±10%, P=0.01 and 60.6± 11.6%, P=0.007, respectively).

CHARACTERISTICS OF PATIENTS WITH GRADE 0-II AND III–IV ACUTE GRAFT-VERSUS-HOST DISEASE:

The patients were divided into 2 groups: grade 0-II and grade III–IV aGVHD. The characteristics of patients in each group are shown in Table 1. The grade 0-II aGVHD group included 53 patients (27 boys and 26 girls) with a median age of 3.58 years (range: 0.67–12.57 years). There were 30 patients with AML and 23 patients with ALL. The median doses of mononuclear cells (MNC) and CD34+ cells were 6.20×107/kg (range: 1.30–11.33×107/kg) and 2.73×105/kg (range: 1.70–6.41×105/kg), respectively. The grade II–IV aGVHD group included 18 patients (8 boys and 10 girls) with a median age of 3.98 years (range: 1.42–8.83 years), 8 with AML, 8 with ALL, and 2 with MAL. The median doses of MNC and CD34+ cells were 5.90×107/kg (range: 1.37–11.79×107/kg) and 2.78×105/kg (range: 0.83–9.20×105/kg), respectively.

RISK FACTORS FOR GRADE III–IV ACUTE GRAFT-VERSUS-HOST DISEASE:

We assessed several other variables for a potential association with an increased risk of III–IV aGVHD using univariate analysis, including age, sex, disease, HLA matching, ABO matching, disease status at transplant, conditioning intensity, anti-thymocyte globulin (ATG), conditioning drug dose, neutrophil engraftment time, platelet engraftment time, MNC, CD34+, post-transplant SF within 2 weeks (all before aGVHD occurred), and pretransplant SF (Table 2). HLA matching (hazard ratio [HR]: 0.46, 95% CI: 0.18–1.14, P=0.093), conditioning intensity (HR: 0.32, 95% CI: 0.13–0.81, P=0.017), and SF level within 2 weeks post-transplant (HR: 3.98, 95% CI: 1.16–13.60, P=0.028) were associated with an elevated risk of developing III–IV aGVHD. Furthermore, conditioning intensity (HR: 0.34, 95% CI: 0.13–0.89, P=0.027) and SF level within 2 weeks post-transplant (HR: 3.61, 95% CI: 1.09–11.97, P=0.036) were independent risk factors for severe aGVHD in the multivariable analysis.

ACUTE GRAFT-VERSUS-HOST DISEASE AFTER UMBILICAL CORD BLOOD TRANSPLANTATION:

Seventy-one patients were included in the analysis, of whom 37 were in the aGVHD group (14 with grade I aGVHD, 5 with grade II aGVHD, 3 with grade III aGVHD, and 15 with grade IV aGVHD). The sites involved included the skin (31 patients), intestines (22 patients), and liver (10 patients). The patients were divided into 2 groups according to their SF level (Table 3).

INCIDENCE OF ACUTE GRAFT-VERSUS-HOST DISEASE:

The incidence of grade II–IV was 44.4% in patients with an SF level >1650 ng/mL, as opposed to 14.3% in patients with an SF level <1650 ng/mL, representing a statistically significant difference (P=0.011). Similarly, for grade III–IV aGVHD, the incidence was 35.4% in patients with an SF level >1650 ng/mL and 10.7% in patients with an SF level <1650 ng/mL (P=0.024) (Figure 4).

OUTCOMES ACCORDING TO SERUM FERRITIN LEVELS:

Finally, we compared the OS and RFS of patients in the SF level <1650 ng/mL and >1650 ng/mL groups. OS and RFS were not associated with the SF level (Figure 5, P=0.15 and P=0.058, respectively).

Discussion

This study assessed the prognosis of children with aGVHD after UCBT and identified risk factors for the development of grades III–IV aGVHD. In recent years, an increasing number of children with AL have undergone UCBT. Despite significant developments in UCBT, aGVDH remains a crucial factor restricting quality of life, especially in patients with severe aGVHD. Patients with grade II–IV and III–IV aGVHD had worse 5-year OS. Infection is the cause of death in most patients with aGVHD, which may have been exacerbated by more aggressive immunosuppressive strategies. Consequently, it is critical to understand the risk factors for aGVHD.

Several risk factors for aGVHD in patients undergoing UCBT, such as myeloablative conditioning, omission of ATG, and double umbilical cord blood, have been described previously [10,17–19]. In this study, we similarly found that conditioning intensity was an independent risk factor for III–IV aGVHD. We also identified the post-transplant SF level as a risk factor for severe aGVDH in the univariate and multivariable analysis.

Previous studies have found that the maximum post-transplant SF levels were significantly higher in patients during aGVHD after allo-HSCT [20]. However, it remains unclear whether post-transplant SF is associated with the occurrence of aGVHD after UCBT in children with AL. We found the post-transplant SF level within 2 weeks was a potential biomarker for predicting development of severe aGVHD. The SF level was also associated with the incidence of II–IV and III–IV aGVHD.

There are multiple potential causes of hyperferritinemia in patients undergoing UCBT. Brissot et al [21] reported that hyperferritinemia can be caused by inflammation, metabolic syndrome, hepatitis, and alcoholism. Iron overload can also induce hyperferritinemia after allo-HSCT, which may be related to transfusions, dyserythropoiesis, or associated genetic factors, and has been reported as a risk factor for aGVHD [22].

Ferritin can have an impact on effects of antigen presenting cells [23]. The activation of antigen-presenting cells is the main mechanism of aGVDH pathogenesis. Furthermore, ferritin enhances the susceptibility to aGVHD by organ toxicity due to reactive oxygen species [24]. Several in vivo studies have demonstrated that mitigating iron overload could potentially reduce the incidence of aGVHD [25,26].

In a retrospective analysis, Lin et al [15] found that pretransplant SF levels were a risk factor for aGVHD in patients with SAA after allo-HSCT. However, we found that the pretransplant SF level was not an independent predictor of aGVHD after UCBT in children with AL. This may be due to the absence of the need for repeated transfusions in patients with AL compared with patients with SAA before UCBT, which may result in no obvious elevation of pretransplant SF. Several studies have shown that post-transplant SF can be used to predict survival outcomes [12–14,27]. Nevertheless, the 5-year OS and RFS did not differ significantly according to the SF level in our analysis. This may be attributable to the limited sample size.

This study has some limitations, including the retrospective design and limited number of patients included in the analysis. Prospective multicenter trials are needed to confirm the risk factors for aGVHD after UCBT in children with AL.

Conclusions

In conclusion, the post-transplant SF level within 2 weeks was a potential biomarker for developing severe aGVHD. Higher levels of post-transplant SF are associated with a higher incidence of grade II–IV aGVHD and grade III–IV aGVHD. However, these results need to be confirmed in prospective studies.