Thalassemia is a prevalent global disease with a single-gene disorder (1). Thalassemia heterozygosity provides some immunity against malaria; however, a particularly high development of thalassemia in Central Asia, India, the Middle East, and Southern China is recognized. Meanwhile, β-thalassemia is widespread among people from these regions; those geographical areas are no longer limited as a result of migration to several places in the world (2, 3). Thalassemia syndromes include α and β gene disorders, resulting in α and β thalassemia, respectively, which are both inherited in an autosomal recessive manner (4). The β-thalassemia genotype is generally more severe than that of α-thalassemia. β-thalassemia comprises thalassemia minor due to single gene mutation and homozygous state in terms of double mutated genes and could be categorized by thalassemia major (β-TM) or intermedia (β-TI) according to clinical manifestations of patients. Furthermore, heterozygous genotypes, such as sickle cell β-thalassemia (S/β) and hemoglobin β-thalassemia (HbE/β), are observed. Currently, β-thalassemia is categorized into transfusion-dependent and transfusion-independent thalassemia (4).

Bone marrow and hematopoietic stem cell transplantation (SCT) are the rare medicinal ways for β-thalassemia nowadays due to the difficulty in identifying suitable donors and the relatively high cost. Moreover, gene therapy, an optimistic approach to cure thalassemia patients, could not be widely promoted and applied because its early-stage clinical trials have not been completed yet (5, 6). β-TM, severe β-TI, and severe HbE/β patients need a regular blood transfusion to maintain hemoglobin (Hb) and fetal hemoglobin (HbF) levels (6). However, significant risks such as acute life-threatening events (acute hemolytic reaction, bacterial infection, and anaphylaxis) are observed, and infections accompany chronic blood transfusions; meanwhile, iron remover may be required to avoid significant multi-system organ damage caused by iron overload (7).

Over the years, researchers have found that HbF inducer could reduce the anemia severity in β-thalassemia patients by elevating globin chain production and improving the excess chain imbalance. Several HbF inducers, such as erythropoietin, demethylating agents, 5-azacytidine, and short-chain fatty acids, have been studied in different combinations and individually to evaluate the clinical efficacy on β-thalassemia patients (8, 9). Hydroxyurea (HU), an impressively effective HbF inducer authorized by the U.S. Food and Drug Administration (FDA) in 1998, has been extensively used for sickle cell anemia patients. However, the use of HU in thalassemia was still restricted (10).

Many studies have shown that HU can significantly increase Hb and HbF levels in β-thalassemia patients, and no serious adverse events have been reported within a specific dose range (11, 12). Many meta-analyses described the HU treatment in lifelong transfusion-dependent (7, 13, 14), transfusion-independent (15, 16), and HbE/β (17) thalassemia patients. However, due to a single genotype of β-thalassemia, these studies lacked adequate power to identify the efficacy of HU for β-thalassemia patients with diverse genotypes or blood transfusion requirements. Furthermore, apart from one systematic review (18), which only included β-TM patients, no meta-analysis assessed the safety and efficacy of HU for β-thalassemia patients in the same article.

To estimate the safety and efficacy of HU on β-thalassemia patients from a relatively integrated viewpoint, we performed this meta-analysis, including clinical trials on the safety and efficacy of HU in β-thalassemia patients treatment with no limits to blood requirements or genotypes.

This study evaluated the safety and efficacy of HU in β-thalassemia patients for 1,352 patients who were blood transfusion-dependent and 396 patients who were blood transfusion-independent. The RR of HU in transfusion-dependent patients was 0.65 [95% CI, 0.53–0.76], and the good RR was 0.37 [95% CI, 0.26–0.49]; in transfusion-independent patients, the RR of HU was 0.53 [95% CI, 0.41–0.65] and the good RR was 0.20 [95% CI, 0.08–0.35]. In general, β-thalassemia patients could benefit from HU.

In previous studies, Algiraigri et al., published a series of meta-analyses about different genotypes of β-thalassemia patients treated with HU, including lifetime transfusion-dependent thalassemia (7), transfusion-independent thalassemia (15) and HbE/β thalassemia (17), furthermore using an analysis of the overall effect of value extracting the figures of single arms. In line with their findings, we also endorse the utilization of HU. In addition, two published meta-analyses, including RCTs, were found in the Cochrane Database of Systematic Reviews. One of them just reported transfusion-dependent β-thalassemia patients (14) and did not enroll any literature; another one, dependent on transfusion-independent β-thalassemia patients (16), enrolled one literature (25), which was included in our study.

Blood transfusion is a recommended therapy for β-thalassemia patients, a hereditary disease that has long afflicted the world (50); however, numerous problems like iron excesses accompany this method. Subsequently, iron agents are required after blood transfusion. Other treatments, such as SCT and gene therapy, have been approved to be significantly effective; nevertheless, they were difficult to implement (6) due to the economic and operational constraints of these therapies make them unsuitable for patients from ordinary families or developing countries, leading to a need for more studies on the most convenient and cost-effective treatment (51). Currently, HbF inducer is an increasingly powerful drug that elevates HbF levels by enhancing the severity of β-thalassemia patients and reducing the imbalance of α/β chain. One of them is HU, a globin synthesis inducer, which leads to stress erythrocyte production through its cytotoxicity, resulting in increased HbF levels (8).

The FDA authorized HU as a key drug for sickle cell disease treatment; however, the HU use in thalassemia patients has not been confirmed. Numerous studies have shown that HU can effectively lower the blood transfusion frequency in thalassemia patients (23, 39, 40, 43), raise the level of Hb (11, 28, 29, 31), improve the HbF level (32, 48, 52), and improve their quality of life (53, 54). Meanwhile, The data collected were RR and good RR in β-thalassemia patients who received HU. The combined RRs were above 50% for both transfusion-dependent and transfusion-independent β-thalassemia patients, which showed significant curative impacts of HU on β-thalassemia patients.

Consistent with the description in the introduction, most of the people included in this study came from Southest Asia and West Asia. Among them, 6 studies of Iranian (total 19 studies) reported transfusion-dependent patients, while 4 studies of Indian (total 11 studies) reports transfusion-independent patients. Many types of β-thalassemia patients were included in our study. Herein, HU significantly reduced or ceased blood transfusion of transfusion-dependent β-TM patients and significantly improved the Hb level of transfusion-independent β-TI patients. Moreover, in comparison to HU dose above 15 mg/kg/d, the efficacy was better when using HU less than 15 mg/kg/d in transfusion-dependent β-thalassemia patients. However, no statistically significant variation was identified. While HU’s effectiveness in treating β-thalassemia differs by country and genotype (p < 0.05), it can be applied to other populations or settings with regular blood monitoring due to its efficacy and safety. A 13-year experience in Iran suggested that HU at a dose of 8–15 mg/kg/day was effective in decreasing or effecting cessation of the need for regular blood transfusion, as well as in increasing Hb levels in β-thalassemia patients, without any major side effects (40). Another study found that the response of HU treatment is best in patients having both XmnI polymorphism and α-deletion (24). Several HbF inducers have also been successfully tried in thalassemia syndromes besides HU. These include 5-Azacytidine, Butyrates, Erythropoietin etc. (4) All these agents have shown promising results in thalassemia, but to a lesser extent in thalassemia major. Additionally, combining hydroxyurea with recombinant human erythropoietin (rHuEPO) shows promise by significantly boosting hemoglobin levels and reducing blood transfusion needs in β-thalassemia patients, while maintaining good safety (30). Based on what was above, long-term HU efficacy, optimal dosing for different genotypes, and combined therapies could be crucial points for researchers to dig out and clear off in future.

Parts of the articles retrieved showed that HU was effective for β-thalassemia patients (55–57); however, the full texts were unavailable, resulting in an exclusion for incomplete data. In addition, after the search deadline, we found a high-quality placebo-controlled RCT (58) of 10-20 mg/kg/d oral hydroxyurea for transfusion-dependent β-thalassaemia, which made a comparison of amount of blood volume, the percentage of fetal hemoglobin percentage and erythropoietic stress as measured by soluble transferrin receptor concentration between two groups, indicating that HU was effective in increasing HbF levels and decreasing blood transfusions in β-thalassemia patients, but ultimately excluded it due to inconsistent inclusion criteria with our study.

To evaluate the safety of β-thalassemia patients who received HU, we focused on the incidence of bone marrow suppression and hematological, neurological, dermatologic, and gastrointestinal side effects. Myelosuppression is a relatively common bone marrow toxic effect of HU, while previous studies of HU used for thalassemia reported little myelosuppression in some studies (49, 59). Among the articles reporting myelosuppression in this study, myelosuppression in patients was reversible (41); one patient recovered after reducing HU dose (30), and some patients with myelosuppression entirely recovered after cessation of drug therapy (23, 34). Leucopenia was the most frequent adverse event in hematology effects, followed by neutropenia (Table 8). Neurologic and dermatologic effects adverse effects were less reported, and gastrointestinal effects (mainly nausea, vomiting, and diarrhea) were more reported. Kattami et al. (60) reported that dermatologic effects like pigmentation, alopecia, and macular papules were most common, and gastrointestinal side effects such as tarry stools and diarrhea were second most common in included patients. Simultaneously, the product specification (61) for HU lists common adverse reactions affecting the hematologic, gastrointestinal symptoms, nervous system, and skin. Additionally, in a study of 299 sickle cell anemia patients (62), the most frequent side effects were hematologic, including neutropenia and reduced reticulocyte and platelet levels. This might be due to HU’s inhibition of DNA synthesis and cell growth, impacting bone marrow’s hematopoietic function, directly stimulating gastrointestinal epithelial cells, and triggering oxidative stress and inflammation, leading to adverse reactions. Although start dose of HU was 10 mg/kg per day and increased by 5 mg/kg per day every 4–6 weeks until toxicity or according to clinical response, the study reported no significant correlation between the HU dosage and the occurrence of adverse effects (49). It is worth mentioning that our study demonstrated that β-thalassemia patients who received HU had a low incidence of adverse events, which were mild and recoverable. If it was possible, we could compare response rates between dependent and independent β-thalassemia patients, or both with individuals without β-thalassemia in same dose of hydroxyurea in future studies.

The limitations of our study includes that firstly, most of the articles about HU in the β-thalassemia treatment were single-armed and retrospective observational studies, which present clear limitations in research design, such as the absence of parallel control groups and limited comparability. Secondly, the use of external controls may introduce confounding variables that could compromise the integrity of the evaluation. Thirdly, the subgroups were categorized by HU dose, country, and β-thalassemia genotype, with complete data on these characteristics. Sensitivity analysis and bias assessment support the study’s primary outcomes. However, the unavailability of some primary outcomes in the included studies may impact the accuracy and reliability of the analysis. Multicenter RCTs (compared with placebo or other available treatments, such as iron chelation and blood transfusion) with large sample size are needed to evaluate the safety and efficacy of β-thalassemia patients treated with HU (14).

Although the outcome indicators did not clearly and directly verify the purpose of our study, and the quality of included studies was flawed, this study collected more studies as comprehensively as possible, included a wider range of population, and used subgroup analysis to make comparison between groups, explaining the drug use characteristics of different groups of population, and providing some reference for future research. Existing studies have shown that HU works well; moreover, it can be a key drug in the β-thalassemia treatment, which requires more economic-effect evaluations to evaluate the cost savings of HU in transfusion, management of iron-chelation-toxic-related diseases and equipment (18).

β-thalassemia patients can benefit from HU therapy because it may raise Hb levels in transfusion-independent patients and lower the blood transfusion needs in transfusion-dependent patients. Meanwhile, neutropenia and leucopenia were the prevalent adverse events in β-thalassemia patients treated with HU; the incidence rates of other side effects were relatively lower, according to our study. It can be concluded that HU is safe and well-tolerated in β-thalassemia patients.