1. Introduction
In the treatment of hemoglobinopathies, such as sickle cell disease (SCD) and thalassemia, red blood cell (RBC) transfusions are essential for the management of symptomatic anaemia [
1,
2]. Repeated transfusions lead to accumulation of iron in soft tissues, resulting in inflammation, susceptibility to infections, end-organ damage and death. These include cardiomyopathy, liver cirrhosis, multiple endocrine abnormalities such as hypothyroidism, growth retardation, hypogonadism, delayed puberty, and diabetes, with an increased risk of developing hepatic malignancies [
3]. Although iron chelation therapy is available, an optimal and durable response is not guaranteed [
4,
5,
6]. This may be due to several factors that include non-compliance to therapy, inadequate dosage, drug resistance due to changes in the iron chelator’s target binding sites, variable tissue distribution, or genetic variations that may affect the metabolism, absorption, or excretion of iron and chelating agents [
4]. Erythrocytapheresis, an automated method of RBC exchange, has been reported to reduce iron overload in chronically transfused patients. Although a significant number of publications on the use of erythrocytapheresis in disease control in both children and adults diagnosed with SCD have been published, only a limited number of reports utilising erythrocytapheresis in patients with SCD and co-commutant iron overload are available [
7,
8,
9,
10]. In patients with SCD, erythrocytapheresis has been applied not only as a chronic transfusion modality in treating SCD complications such as stroke, acute chest syndrome, priapism, dactylitis, and splenic sequestration but to optimise the clinical condition in patients towards haemopoietic stem cell transplantation and the acute indications where an end organ is threatened [
11]. These reports, however, focus on the disease pathophysiology rather than regulating iron overload.
Patients with thalassemia are in need of chronic RBC transition therapy and are also at risk of developing iron overload. Erythrocytapheresis has no role in disease regulation in patients with thalassemia, but as in patients with SCD, even fewer reports of erythrocytapheresis for the treatment of iron overload in patients with thalassemia have been published [
12]. These reports are limited to adults [
12].
The gold standard for evaluating tissue iron overload is a biopsy, but because of its invasive nature, magnetic resonance imaging (MRI) has become the primary outcome measure for clinical trials of iron chelation therapy [
4,
13]. In children, both methods are logistically problematic and resource-intensive. Therefore, the most efficient and cost-effective method for quantifying iron overload is to determine serum ferritin levels, especially before clinical signs of iron overload appear. It is also an effective method for monitoring iron chelation therapy, even in the presence of factors that affect serum ferritin levels [
4].
At the Antwerp University Hospital, a tertiary healthcare facility, erythrocytapheresis has been implemented as a standard of care during the management of iron overload in patients with SCD, who received chronic blood transfusions, and with β-thalassemia major, who are transfusion dependent, to decrease ferritin levels as well as prevent permanent complications of iron overload in these patients. The aim of this study was to evaluate erythrocytapheresis, either as a standalone therapy or in combination with iron chelation therapy, in children and young adults with hemoglobinopathies in whom current iron chelation therapy is not sufficient in decreasing the iron overload during management. Two secondary objectives were (1) to evaluate if erythrocytapheresis has similar benefits in regulating iron overload in both patients with SCD and β-thalassemia major and (2) to evaluate the cost-benefit ratio involved when implementing erythrocytapheresis during management compared to iron chelation therapy alone.
2. Materials and Methods
This retrospective study included patients diagnosed with SCD or thalassemia, in need of chronic blood transfusions between 2015 and 2022, who developed iron overload based on serum ferritin levels where iron chelation therapy did not provide effective treatment in decreasing serum ferritin levels. Patients in the study population were of Central African, South East Asian, and Middle Eastern origin. Antwerp serves a large immigrant community, who are in need of significant social support. Newly immigrated patients also present with untreated iron overload from their native countries.
Poor adherence to oral iron-chelation therapy was defined as the patient or caregiver reporting poor adherence, irrespective of the reason, or documented unclaimed medication from pharmacy records.
Patients were divided into two groups: (1) a case cohort who received erythrocytapheresis (irrespective of the duration) alone or in combination with iron chelation therapy and (2) a control cohort who received oral iron chelation therapy alone. Where a sub-analysis was possible, the case cohort was divided into two subgroups: (1) erythrocytapheresis alone and (2) iron chelation therapy in combination with erythrocytapheresis.
All patients commenced iron chelation therapy when ferritin levels reached 1000 μg/L or higher and continued as ferritin levels did not decline sufficiently. In our study, all patients received Deferasirox (DFX), which is an oral iron chelator. Institutional indications for starting erythrocytapheresis are (1) rising ferritin levels despite adequate iron chelation therapy, (2) medical reasons for low compliance to iron chelation, and (3) social factors limiting compliance to iron chelation therapy.
The collected epidemiological data included sex/gender, age, and diagnosis. Laboratory test results were also collected from the patient’s file. These blood examinations were performed during the standard follow-up consultation for patients treated with iron chelation therapy every 1 to 3 months for patients treated with erythrocytapheresis, biweekly to monthly, depending on the treatment frequency. Blood examinations were performed using the following devices: for biochemical evaluations, the Attelica Solution Immunoassay and Clinical Chemistry Analyzer ® by Siemens® (Munchen, Germany) and for the full blood counts, the XN 9100 by Sysmex® (Kobe, Hyogo, Japan).
To evaluate the status of iron overload and the maintenance of stable RBC parameters, the following laboratory results were analysed: haemoglobin (Hb), haematocrit (Hct), reticulocyte count (Ret), serum ferritin (Fe), and lactate dehydrogenase (LDH) as potential indicators. To evaluate end-organ damage, aspartate aminotransferase (AST), alanine aminotransferase (ALT), gamma-glutamyl transferase (GGT), alkaline phosphatase (ALP), total serum bilirubin (TBil), serum conjugated bilirubin (cBil), and serum unconjugated bilirubin (uBil) were selected as indicators.
Data to evaluate erythrocytapheresis therapy long-term outcomes were collected at five different time points (at the start of either erythrocytapheresis or DFX alone therapy, after 1 month, after 3 months, 6 months, 9 months, and 1 year) from the start of erythrocytapheresis therapy until one year post commencing erythrocytapheresis therapy. Ferritin and Hb were collected at five standardised time points to prevent confounding by short-term outcomes of acute events that could affect acute phase reactants. Similarly, Hb values were evaluated over the long-term trend, thereby preventing confounding by fluctuations due to intermittent red blood cell transfusions.
In the absence of clinical symptoms or indications, no echocardiograms and advanced imaging, using MRI or Ferriscan to detect cardiac or liver iron overload, were performed during the follow-up period of one year.
2.1. The Erythrocytapheresis Protocol
Long-term venous access is established in all patients utilising one of two methods: (1) two high-pressure intravenous 14-gauge catheters (Port-a-Cath®, Smiths Medical, Minnesota, USA), one placed in the subclavian vein, and one placed in the inguinal vein or (2) a surgical arteriovenous fistula is created in the cubital fossa.
Erythrocytapheresis was performed with the Spectra Optia® system (Caridian BCT, Inc., Colorado, USA). The erythrocytapheresis volume was calculated using the patient’s height, weight, and sex. Before the start of erythrocytapheresis and at the end, a blood examination is performed.
The total procedure takes approximately three to four hours while the patient is monitored for cardiovascular and respiratory parameters. Patients receive erythrocytapheresis every 2 to 4 weeks based on the ferritin levels. No international guidelines exist to guide this frequency of treatments.
2.2. Statistical Analysis
Data were analysed using IBM SPSS version 28 (IBM Corporation, USA). A p-value of <0.05 was considered statistically significant.
Data were expressed as median with minimum and maximum and quartiles (25% and 75% percentiles) due to the small study sample and distinctly non-normal distribution of quantitative variables (tested with Kolmogorov–Smirnov tests). Associations between two categorical variables were assessed using Pearson chi-square analysis or Fisher’s exact two-sided tests as appropriate. One-way ANOVA was used to analyse the trends of association between categorical and continuous normally distributed variables. The Mann–Whitney U test was used to compare distributions of abnormally distributed variables between two categories. The Kruskal–Wallis test and Friedman test were used to compare medians of more than two categories between and within subjects, respectively.
Some cohorts were too small (less than five) for meaningful calculations and were reported as trends. The single sickle-thalassemia patient was included in all analyses where inclusion criteria were met but was never analysed as a single patient cohort as it does not constitute a representative cohort.
4. Discussion
Erythrocytapheresis, in combination with iron chelating therapy, provides a significant reduction in serum ferritin levels in patients with hemoglobinopathies and maintains the haemoglobin levels at a more stable level. Transfusion-related iron overload is a global challenge in patients with hemoglobinopathies receiving chronic transfusions, and they have potential long-term complications due to iron storage in tissues [
14]. Although symptoms or complications of iron overload occur over several years of excessive iron intake or conditions that cause increased iron absorption, timely and efficient intervention remains important in the prevention of long-term effects [
15]. Oral iron chelation therapy has shown good results in the treatment of iron accumulation; however, it has not achieved optimal results due to factors such as poor tolerance and treatment compliance [
4].
A limited number of studies evaluating erythrocytapheresis for the treatment of iron overload in chronically transfused patients with SCD concluded that erythrocytapheresis is a safe and efficient technique [
8,
9,
10,
11]. Therapy with erythrocytapheresis achieved a significant decrease or stabilization of serum ferritin with or without Deferoxamine iron chelation therapy [
7,
8,
9,
10]. In our study, we could not prove that erythrocytapheresis alone could significantly reduce serum ferritin levels compared to iron chelation therapy alone. However, combination therapy was better than single-modality therapy. Hilliard et al. and Singer et al. showed stabilization of serum ferritin in three and four non-chelated patients [
9,
10], while Kim et al. and Adams et al. showed a significant decrease in serum ferritin in two and three non-chelated patients [
7,
8] (see
Table 2). Similarly, we demonstrated that erythrocytapheresis significantly decreased serum ferritin levels in patients who received concomitant Deferasirox. In this study, three patients showed a decrease in serum ferritin levels when receiving erythrocytapheresis alone, although not statistically significant.
Previous studies all performed erythrocytapheresis in patients with SCD. Our study reports novel findings showing a significant reduction in serum ferritin in paediatric patients with β-thalassemia major treated with erythrocytapheresis. A single study from Wall et al. evaluated erythrocytapheresis in adult patients with β-thalassemia major without benefit [
12]. In this study, erythrocytapheresis did not have benefits in treating patients with established iron overload [
12]. We postulate that our study did reduce serum ferritin levels, indicating the potential benefits for the treatment and prevention of iron overload in children. This study analysed different parameters concerning iron metabolism. Recent studies reported novel markers of iron metabolism, such as soluble hemojuvelin and hepcidin, that are more sensitive and specific than ferritin [
16,
17]. As this was a retrospective study, these parameters should be implemented in new studies evaluating the efficacy of erythrocytapheresis in the management of iron overload due to chronic transfusions [
16,
17].
Importantly, previous studies mentioned several complications in vascular access because children have small vessels. Our study showed optimal vascular access using a Port-a-Cath
® system, placing one in the shoulder and one in the groin. Moreover, no complications were reported with the prudent use of this system. Therefore, the additional risks associated with the surgical placement of the catheters to perform erythrocytapheresis as standard therapy for transfusion iron overload are less than the complications resulting from the sequelae of iron overload over time. Previous studies have shown that venous access is a barrier to standardization of erythrocytapheresis [
7,
8,
9,
10] that may be eliminated with the Port-a-Cath
® system.
There is no international consensus for starting erythrocytapheresis, which explains our varied study population. According to the literature, the inclusion criteria for erythrocytapheresis are high iron loads with/or without end-organ damage, poor chelation use, good venous access, and a history of stroke [
7,
8,
9,
10]. Standardizing international criteria for the indication to start erythrocytapheresis should, therefore, be prioritised. The data regarding the social impact of both therapies was not robust enough for evaluation in our study. However, reports on long-term iron chelation have shown variable results, especially with regard to compliance. Therefore, more comparative studies evaluating the quality of life during therapy with erythrocytapheresis and iron chelation therapy are needed. As the follow-up period in this study was only one year, the period to analyse long-term complications of iron overload was insufficient as complications developed over years [
16].
A more in-depth cost-benefit analysis is indicated, but as this is a new technique, it is important to evaluate the erythrocytapheresis and iron chelation therapy implications of both singular and combined therapies. Excluding the costs for gaining venous access, the monthly costs for the single therapies are comparable, except if erythrocytapheresis is performed more than once a month. This may be necessary in the initial phase of starting erythrocytapheresis therapy but may become less over time. However, iron chelation therapy doses may need to be escalated when the initial response is inadequate or when the age-related doses increase. These costs become wasted if compliance is poor. In this study, only a single Port-a-Cath® was removed due to infections, but it increases the cost during erythrocytapheresis management. When combination therapy is needed, the costs are naturally higher. The question remains whether either therapy can be discontinued when good control has been achieved. In the case of non-compliance, continuation of erythrocytapheresis seems to be more beneficial. If these two therapies are used to achieve stem cell transplantation, the costs are for a limited period of time.
Considering the added benefit of erythrocytapheresis to lower Hb-S levels in patients diagnosed with SCD to achieve disease control increases the cost-effectiveness of this treatment modality. In patients receiving iron chelation therapy alone, hydroxyurea for disease control also needs to be included in the cost analysis. This increases the cost of oral therapies, increases the number of medications to be taken daily, and increases the burden on adherence.
In contrast to the short-term costs, the costs related to end organ damage are potentially much greater than the monthly costs of disease control. This may also include higher morbidity, mortality, and burden of disease [
3]. A further consideration is the economic impact of progressive disease and complications related to iron overload [
3,
14].
The question remains whether either therapy can be discontinued once good control has been achieved. In the case of non-compliance, continuation of erythrocytapheresis would be potentially more beneficial. If the two therapies are used to achieve stem cell transplantation, the cost is for a limited period of time.
Although this study was conducted in a high-income setting, the risk-benefit ratios, particularly in socioeconomic terms, change when the impact is assessed in low-and middle-income countries, such as Africa and South-East Asia, which have the highest burden of SCD and β-thalassemia major [
18,
19].
This study was limited because of its retrospective nature and incomplete and limited data, as it was conducted in a single institution. The lack of international standards for the initiation of erythrocytapheresis has added to the diversity in the groups. Outcomes and benefits were defined by trends in haemoglobin and ferritin levels during a 1-year follow-up period; no analysis of long-term complications of iron overload, such as cardiac and liver MRI, was performed. In general, an increased risk of morbidity and mortality is associated with ferritin levels above 2500 ng/mL [
4], which partly explains the absence of complications in this study, as chelation therapy was started once ferritin levels reached 1000 ng/mL. International recommendations recommend an initial MRI scan after at least 10 transfusions of 15 mL/kg per transfusion or before reaching 10 transfusions if clinical symptoms are present [
4].
We acknowledge a possible inherent bias in the study because of no international standards for the indications of erythrocytapheresis. This study used the objective failure to decrease serum ferritin levels to standardise the indication.