Comparison of oral iron chelators in the management of transfusion-dependent β-thalassemia major based on serum ferritin and liver enzymes

Background Excess iron deriving from a chronic transfusion and dietary intake increases the risk for cardiac complications in β-thalassemia major patients. Deferiprone and deferasirox are commonly prescribed to thalassemic patients who are at risk of iron overload. This study aimed to compare the performance and toxicity of deferiprone and deferasirox in β-thalassemia major patients. Methods A cross-sectional observation was performed on 102 patients with β-thalassemia major. Serum ferritin along with total, indirect, and direct bilirubin levels were measured. Levels of liver enzymes, transaminase (ALT), and aspartate transaminase (AST), were also determined. Ferritin correlations with serum ALT, AST, and total bilirubin were constructed based on Spearman’s rank correlation. Statistical differences based on the serum parameters were analyzed between deferiprone and deferasirox groups. The differences of iron chelators’ effects between those receiving short-term (≤7 years) and long-term (>7 years) blood transfusion were also analyzed. Results The averaged levels of bilirubin, ALT, AST, and ferritin were found to be high. Ferritin was positively correlated with ALT (r=0.508 and p<0.001) and AST ((r=0.569; p<0.001). There was no statistical difference in ferritin levels between the deferiprone and deferasirox groups ( p=0.776). However, higher total bilirubin and ALT were observed in the deferasirox group than in the deferiprone group ( p=0.001 and 0.022, respectively). Total ( p<0.001), indirect ( p<0.001), and direct bilirubin levels ( p=0.015) were significantly higher in patients with long-term transfusions than those receiving short-term transfusions. Higher ferritin was found with a statistical significance of p=0.008 in the long-term transfusions group. Conclusions Ferritin is high in people with transfusion-dependent β-thalassemia major and positively correlated with ALT and AST. Deferasirox might pose a higher risk of developing hepatic injury as compared with deferiprone. Yet, no significant change of deferasirox efficacy (based on ferritin level) was found between those receiving short-term and long-term transfusions.


Introduction
Though the disease was known to be centered around the Southeast Asian and Middle Eastern regions, a recent review has highlighted the increasing prevalence of β-thalassemia in North America and Western Europe. 1 Approximately 1.5% of the world's population are carriers of β-thalassemia, where the number varied across regions, including in specific countries. 2For example, among Southeast Asian countries the number of β-thalassemia carriers ranged between 0.5% and 12.8% as reported in Myanmar and Malaysia, respectively. 3In Indonesia, it is estimated that 3.0-10.0% of the population have the β-thalassemia trait depending on the population. 4In 2019, data from Indonesian Pediatric Association indicated at least there were 10,500 registered thalassemia patients but this figure is far below the estimation. 4ople with β-thalassemia require a lifelong blood transfusion due to their inability to produce sufficient normal hemoglobin A. Ineffective erythropoiesis in β-thalassemia patients occurs because of the inherited genetic mutations causing impaired β-globin protein and consequently the imbalance in αand β-globin chains production. 5The condition leads to the inevitable hyperabsorption of irons deriving from either dietary irons or those from blood transfusion. 6xcessive iron and its deposition could cause cardiac complications, which currently affects 25% of total 26,893 β-thalassemia major patients according to a meta-analysis. 7To prevent the toxicity of iron overload, the patients should receive iron chelating therapy.The life expectancy of people with β-thalassemia major could be improved, even to be similar to those with β-thalassemia intermedia if managed with an iron chelating agent. 8sed on a systematic literature review, a person with transfusion-dependent thalassemia should pay $128,062 for blood transfusion and iron chelating therapy. 9In Indonesia itself, the access to iron chelation therapy is still a challenge, especially in rural areas. 10It is worth mentioning that Indonesia lies in a thalassemia belt area owing to the high prevalence and carrier rate. 11Therefore, it is of importance to focus the research on the Indonesian population to improve the clinical management and monitoring of iron overload among patients with β-thalassemia major.There are two oral iron chelating drugs usually used in Indonesia deferiprone and deferasirox.Deferiprone has two ligands which could coordinate with Fe forming a complex with three chelator molecules. 12Meanwhile, deferasirox has three ligands that could coordinately bind one Fe atom to form a complex with two chelator molecules. 12Understanding the performance of these two drugs in reducing the iron level could provide an insight into more effective and cost-efficient management.Although these drugs are the main oral iron chelating drugs in Indonesia, 4 studies comparing their effects on liver function are limited and focused in Java. 13,14In addition, the iron storage protein, ferritin, has been suggested as an indicator of iron overload and has been used to monitor iron chelation therapy efficacy. 15,16Its correlation with liver enzymes could be an indicator for liver damage monitoring. 17,18e objective of the study was to compare the performance and adverse effects of two oral iron chelating drugs (i.e., deferiprone (Ferriprox) and deferasirox (Exjade)) on the liver in β-thalassemia major patients with chronic transfusion.The effect on the liver were assessed by measuring the levels of serum ferritin, transaminase (ALT), aspartate transaminase (AST), total bilirubin, indirect bilirubin and direct bilirubin among patients with β-thalassemia major and chronic transfusion.In addition, we performed a correlation test between serum ferritin with ALT, AST, and total bilirubin.

Study design
This research was conducted prospectively on children (1 to 18 years old) with β-thalassemia major who were hospitalized at the Children's Thalassemia Center at Dr. Zainoel Abidin Hospital in Banda Aceh within the February-March 2022 period.The research aimed to seek the correlation between the administration of the iron chelating agent deferiprone with aspartate aminotransferase (AST), alanine aminotransferase (ALT), ferritin, and albumin.Ethical clearance was provided by the Ethical Committee of Dr. Zainoel Abidin Hospital, Banda Aceh (013/EA/FK-RSUDZA/2022, KEPPKN No 1171012P).Written informed consent for all the patients' participation was obtained from the patients' guardians.

REVISED Amendments from Version 1
In this new version, we have re-stressed the aim of the study at the end of the Introduction.The data of Thalassemia from Indonesia, the dosage of the iron chelation used in the study has been added in the revised article.We have added the approach to exclude the possibility of inflammatory state among the patients.The dosage of the iron chelation used in the study has been added.In addition, the height data have been changed to meter unit.More discussion has been added in particular the difference between the two types of oral chelating drugs (deferiprone and deferasirox).

Patient recruitment
Diagnosis of β-thalassemia major was established based on hemoglobin (Hb) electrophoresis.The patient was included if having serum ferritin >1500 ng/mL, treated with oral Ferriprox ® (containing deferiprone with dose of 75 mg/kg body weight each day) or Exjade ® containing deferasirox with dose of 20 mg/kg body weight each day in single dose), and they were not prescribed with other iron chelation therapy (i.e., injectable deferoxamine) within the last three months.Patients having histories of hepatitis virus (HPV) B and HPV C were excluded.In addition, since the level of ferritin could be influenced by some conditions, all patients that had infection (assessed through physically and laboratory tests including routine lab, C-reactive protein level and procalcitonin) and autoimmune diseases were excluded.
Consecutive sampling was employed to select the study subjects with sampling size calculated based on Equation 1.

Serum parameters analysis
Venous blood (10 mL) was drawn from each subject and analyzed for serum bilirubin, ferritin, AST, and ALT.Serum bilirubin was measured using direct diazo reaction.Serum ferritin level was obtained from an analysis using the immunoassay technique.Serum AST and ALT determinations were based on non-pyridoxal phosphate Inverse Fast Fourier Transform (IFFT).

Statistical analysis
Continuous data were presented as meanAEstandard deviation (SD).Normality of the data distribution was judged based on a Kolmogorov-Smirnov test.Depending on the normality test, the significant difference was statistically analyzed using independent t-test or Mann-Whitney U test.All statistical analyses were performed on SPSS version 22 (SPSS Inc., Chicago, IL, USA) (SPSS, RRID: SCR_019096).

Characteristics of patients
The patients included in this study were a balanced number of males (50%) and females (50%) with <10 years old as the most predominant age, followed by 10-15 years old and >15 years old, respectively (Table 1).The average body weight and height of the patients were 25.56AE9.75kg and 1.25AE0.18m.As many as 45.1% of the total patients were in the healthy weight category, while another 45% were categorized as underweight.Almost all patients required 1 bag of blood each month.The average ferritin level of 3531.80AE2322.44ng/mL exceeded the normal range (7-140 ng/mL).The averaged total, indirect, and direct bilirubin levels were 1.86AE1.00,1.25AE0.88,and 0.62AE0.42mg/dL, respectively.The averaged ALT and AST levels were also found to be higher than the upper limit for children (51.57AE32.77and 49.19AE48.32,respectively).High SDs in ALT and AST indicate that there were some patients having these values in the normal range (10-40 UI/mL) but some were abnormal.

Correlations between serum ferritin and liver enzymes
Correlation tests were performed to see if the ferritin level was correlated with ALT, AST, or total bilirubin, where the results have been presented in Table 2. Based on Spearman's rank correlation analysis, serum ferritin was found to be positively correlated with serum ATL based on r=0.508 and p<0.001.Similarly, serum ferritin was also positively correlated with serum AST with a statistical significance (r=0.569;p<0.001).These correlations suggest the increase in serum ferritin level contributes to the increase in serum ALT and AST levels.No correlation was found between serum ferritin and total bilirubin.

Comparisons between deferiprone and deferasirox
Comparisons of serum ferritin, bilirubin, ALT, and AST among patients receiving either deferiprone or deferasirox have been presented in Table 3.In this analysis, only 90 patients were included while 12 others were excluded because information about blood types was missing for some patients, or they had been treated with drug combinations.Levels of serum total, indirect, and direct bilirubin were found to be significantly different between the two groups (p<0.05).Higher levels of serum ALT and AST were found in the deferasirox group as compared with the deferiprone group, yet statistical significance was only achieved in terms of serum ALT level.
We further analyzed the differences in serum ferritin, bilirubin, ALT, and AST levels between patients receiving longterm and short-term transfusions.The data of this analysis have been presented in Table 4. Significantly higher serum ferritin level was observed in the group that had undergone long-term transfusion (p<0.05)among the deferiprone recipients.As in patients receiving deferasirox, long-term blood transfusion did not give a significant effect on ferritin levels.However, among those receiving deferasirox, long-term blood transfusion yielded significantly higher serum total, indirect, and direct bilirubin levels.Neither deferiprone nor deferasirox prescription yielded statistically different effects on serum ALT and AST levels between those undergoing long-term and short-term transfusions.

Discussion
Increased ferritin levels in the sera of thalassemic patients are expected since they have ineffective erythropoiesis conditions, allowing hyperabsorption of dietary irons in the intestines.Multiple reports have suggested the effect of iron overload-associated liver abnormalities, indicated by the raised level of liver enzymes.Egyptian patients with β-thalassemia, aged 1.1 to 17 years old, had a mean serum ferritin level of 1367.52 AE 856.22 ng/mL, which exceeded the normal range in children (7-140 ng/mL). 19Following the blood transfusion, irons are retained in the macrophages due to the increased hepcidin, hence elevation in serum ferritin. 20Hepcidin production is suppressed due to erythropoietic activity which allows increased absorption of dietary iron in addition to the iron obtained from blood transfusion.Elevated ferritin as an iron storage marker was reported to have a strong correlation with iron overload condition. 21mong 291 thalassemic children in Cipto Mangunkusumo Hospital, Indonesia, there was a predominance of mild, moderate, and severe degrees of iron overload (28.2%, 34.7%, and 27.8%, respectively). 22eraccumulation of iron in the liver could damage the organ following reactive oxygen species (ROS) formation through Haber-Weiss and Fenton reactions. 23Ferric iron could be reduced to ferrous iron resulting in the formation of hydroxyl radicals which promotes DNA and protein damages, oxidation of amino acid side chains, and peroxidation of phospholipid. 24In this present study, we found the abnormally high averaged values of serum AST and ALT (51.57AE32.77and 49.19AE48.32,respectively), suggesting the presence of hepatic injury.The correlation tests revealed that serum AST and ALT levels are strongly influenced by ferritin, hence hepatoxicity of iron overload.Similarly, a study on 100 patients with β-thalassemia major reported the strong positive correlation between serum ferritin and AST or ALT. 19Based on the study with a slightly higher number of transfusion-dependent thalassemic patients (n=138), ferritin was also found to be positively correlated with liver enzymes (AST and ALT). 17Among patients with acute dengue fever, similar relationships among serum ferritin, AST, and ALT were also reported, where serum ferritin was proposed as an indicator for liver damage. 18In the light of this iron toxicity, it is of importance for thalassemic patients undergoing blood transfusion to receive iron chelating therapies.
Herein, we observed differences of total, indirect, and direct bilirubin and ALT levels between deferiprone and deferasirox groups.All the aforementioned biomarkers were statistically higher in the latter group.In the correlation test, we did not find a correlation between ferritin and bilirubin, indicating that increased bilirubin was not associated with the efficacy of the iron chelator.6][27] In drug-induced hepatic injury, the drugs induce the detrimental immune modulations after attaching to the enzyme molecules. 28The neoantigens eventually initiate the immune response cascade which could result in cytotoxicity and apoptosis. 28Nonetheless, the reactions are delayed and required multiple exposures.Another proposed mechanism of the deferasirox-induced hepatic injury is the blockage of salt transport proteins (cholestatic injury), 25,26 which is supported by the findings of higher bilirubin levels in this present study.However, a case report suggested mitochondrial injury as the most explaining mechanism of deferasirox-induced liver injury which is consistent with elevations of transaminases. 27r data indicated that the levels of serum total, indirect bilirubin, and direct bilirubin were significantly higher in patients treated with deferasirox than deferiprone.Deferasirox has three ligands that could bind one Fe atom to form a complex with two chelator molecules while deferiprone only has two ligands. 12The hepatoxicity effect of deferasirox was found to be more potent in patients with long-term transfusion as compared to those with short-term transfusion.However, interestingly, deferasirox could maintain its iron chelating efficacy even though the patients have undergone blood transfusion for more than 7 years.In the case of deferiprone, serum ferritin was produced at a higher level in patients with long-term blood transfusion.This is an indication that deferiprone lost its efficacy as an iron chelating agent as the patient received more blood transfusion.Our findings are in agreement with a study that continually monitored the longterm efficacy of deferasirox, which appeared to be sustained for 3.5 years. 29Deferasirox uniquely has a longer half-life and bioavailability in the plasma (lasting for 24 hours) than other iron chelating therapies. 30It is worth noting that, among patients receiving short-term blood transfusion herein, lower ferritin was observed in the deferiprone group than in the deferasirox group.In previous studies, deferasirox had a lower rate of effectiveness. 16,31r study had some limitations.For example, we did not simultaneously assess the effects of deferasirox or deferiprone in the thalassemic patients.The number of patients recruited in the study was too few, limiting the results in terms of being clinically meaningful.Moreover, we did not measure iron overload in specific organs, which could give an important insight since the drugs' efficacies are dependent on the targeted organ.Diagnosis using magnetic resonance imaging (MRI) or other definitive tools should be carried out to conclude the hepatic injury in the patients.

Conclusions
Patients with β-thalassemia major have a higher level of serum ferritin, which is strongly correlated with the elevations of serum ALT and AST levels, respectively.Deferasirox poses a higher risk of developing liver dysfunction than deferiprone, as suggested by increased levels of bilirubin and ALT.While deferiprone has reduced efficacy among patients managed with long-term blood transfusion, deferasirox could yield a similar efficacy regardless of the length of the chronic transfusion.These findings could be used as guidance for medical practitioners in determining the oral chelating therapies for iron-overload conditions among β-thalassemia major patients, by considering the net clinical benefits of each drug.Although a higher number of participants is required to obtain meaningful clinical conclusions, medical practitioners should be aware of the net clinical benefit of prescribing the oral chelators to transfusion-dependent β-thalassemia major patients.
dependent β-thalassemia major based on serum ferritin and liver enzymes", the work is a reasonable study to compare between 2 oral chelating drugs (Ferriprox (deferiprone) and Exjade (Deferasirox)) and their action in decreasing the level of serum ferritin and their effects on serum transaminases and serum bilirubin and the correlation between serum ferritin and the mentioned parameters.In my opinion the work was well done, and only few notes are required to be revised as follows: Method: The doses of the 2 drugs were not mentioned which are very important as different protocols of treatment use different daily drugs doses.

1.
Under the title "serum parameters analysis" methods for the estimation of serum; ferritin, AST, and ALT were mentioned while method of estimation of serum bilirubin was not mentioned.

2.
Results: The mean length of the patients that was measured in meters was written as "125.96±18.06".This is in cm and if a meter unit is required then it is 1.2596±0.1806.The same fault was repeated in table 1.

3.
Discussion: "Following the blood transfusion, irons are retained in the macrophages due to the increased hepcidin, hence elevation in serum ferritin.17".Hepcidin production is suppressed due to erythropoietic activity which allows increased absorption of dietary iron in addition to the iron obtained from blood transfusion.

4.
The authors used the scientific names of the drugs which are general names of the main constituent of the drug.I think they have to use the trade names of the drugs and in 2 brackets the constituents of the drugs as follows: Ferriprox (deferiprone) and Exjade (deferasirox).

5.
Aim of the study was mentioned in the "background" only and have to be repeated in more details at the end of the "Introduction".

6.
In addition, in aim of the study mentioned in the "Background" I suggest adding (on the liver) and substituting the word (toxicity) by (adverse effects) so it become (This study aimed to compare the performance and adverse effects of deferiprone and Deferasirox on the liver in β-thalassemia major patients).
In my opinion the authors went many times out of the scope of the aim of the study which leads to a relatively long article and can be summarized to fulfill the aim of the study. 1.

Is the work clearly and accurately presented and does it cite the current literature? Yes
Is the study design appropriate and is the work technically sound?Yes

If applicable, is the statistical analysis and its interpretation appropriate? Yes
Are all the source data underlying the results available to ensure full reproducibility?Yes

Are the conclusions drawn adequately supported by the results? Yes
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Clinical Biochemistry with Special interest in Thalassemia I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

Introduction
Since similar studies with greater number of subjects have been established.Please provide the novelty of this study. 1.
Please provide data from Indonesia 2.

Methods
Please provide data about the dosage of iron chelation used in this study.1.
Considering most of the patients with transfusion dependent thalassemia had organomegaly, please re-assessed the nutritional status based on the mid-upper arm circumference.

2.
Since ferritin is easily affected by inflammatory status, please provide the technique to exclude the possibility of inflammatory state 3.

Results
Please provide the analysis data between dosage used and the dependent variables 1.
Please revised the height data in "metre" unit 2.

Discussion
Please provide more discussion about the results findings, especially the difference between the two types of drugs. 1.

Competing Interests: I do not have conflict of interest
The benefits of publishing with F1000Research: Your article is published within days, with no editorial bias • You can publish traditional articles, null/negative results, case reports, data notes and more • The peer review process is transparent and collaborative • Your article is indexed in PubMed after passing peer review • Dedicated customer support at every stage • For pre-submission enquiries, contact research@f1000.com

Results 6 .
presented and does it cite the current literature?YesIs the study design appropriate and is the work technically sound?PartlyAre sufficient details of methods and analysis provided to allow replication by others?PartlyIf applicable, is the statistical analysis and its interpretation appropriate?YesAre all the source data underlying the results available to ensure full reproducibility?YesAre the conclusions drawn adequately supported by the results?exclude the possibility of inflammatory state.Response: Thank you for your question.In all patients, the level of ferritin was measured only the patients had no infection (assessed through physically and laboratory tests including routine lab, C-reactive protein level and procalcitonin) and had no autoimmune diseases.Please provide the analysis data between dosage used and the dependent variables.Response: Thank you for your suggestion.Unfortunately, all the patients in our study were treated with the same dose of iron chelation drugs (deferiprone with dose 75 mg/kg body weight each day and deferasirox with dose of 20 mg/kg body weight each day for single dose).7.Please revised the height data in "metre" unit Response: The unit of the height has been changed to unit metre.Discussion 8. Please provide more discussion about the results findings, especially the difference between the two types of drugs.Response: More discussion has been added in particular the difference between the two types of oral chelating drugs (deferiprone and deferasirox) used in this study.
*Statistically significant at p<0.01 based on Spearman's rank correlation.

Table 4 .
Levels of serum ferritin, bilirubin, ALT, and AST in patients receiving long-term and short-term transfusions.
b Transfusion category was divided into long-term (> 7 years) and short-term (≤ 7 years) based on the median.a Otherwise stated, p-value was obtained from independent t-test.b Obtained from Mann-Whitney U test.*Statistically significant at p<0.05.
*Statistically significant at p<0.05 using Mann-Whitney U test.