Thrombopoietin is a major factor that regulates megakaryocyte production. The main mechanism of action of thrombopoietin receptor agonists (TPO-RAs) is to stimulate bone marrow megakaryocytes [24,25]. Eltrombopag and romiplostim are TPO-RA drugs approved by the U.S. Food and Drug Administration for use in children. Both drugs showed efficacy and safety compared with placebo in children, but no clinical trials comparing the two drugs have been conducted. The ASH 2019 guidelines recommend the use of TPO-RA rather than rituximab in children with ITP who do not respond to first-line treatment [26].

Eltrombopag is a once-daily, oral TPO-RA requiring some dietary restrictions. Peak levels are obtained 2 to 6 hours after oral ingestion. After a single oral dose of 75 mg, the bioavailability was > 52%. When taking eltrombopag with a high-fat, high-calcium diet versus on an empty stomach, the area under the curve decreased by 59%, the maximum serum concentration (C_max) decreased by 65%, and the time to reach C_max was delayed by 1 hour [27]. Additionally, eltrombopag is metabolized in the liver, and its half-life ranges from 21 to 32 hours [28]. The dose of eltrombopag depends on the patient’s age, ethnicity, and hepatic function. In addition, the dose should be adjusted according to platelet response (maximum dose, 75 mg/day).

The efficacy and safety of eltrombopag in children were verified in two multicenter randomized trials, PETIT and PETIT2 [29,30]. Eltrombopag showed response rates of 62% in PETIT and 75% in PETIT2. It also showed a reduction in bleeding events and the requirement for additional treatments. The median time to response was 12 to 20 days depending on the patient’s age.

In a recent retrospective multicenter study conducted in 17 centers affiliated with the Italian Association of Pediatric Hematology and Oncology, 68% and 44% of patients achieved platelet counts of 30×10⁹/L and 100×10⁹/L, respectively [31].

In a phase 1/2 trial with children, romiplostim showed an improvement in platelet counts compared with that of placebo [32]. In that study, 88% of patients in the romiplostim group had platelet counts ≥50×10⁹/L and 20×10⁹/L above baseline for 2 consecutive weeks. In contrast, none of the patients in the placebo group showed improvement. In the open label extension study after the phase 1/2 trial, the median average weekly dose was 5.4 μg/kg [33]. In a phase 3 trial, romiplostim showed a high rate of platelet response and reduced rate of bleeding events in pediatric patients with chronic ITP [34]. A durable platelet response was achieved in 52% of patients in the romiplostim group compared with 10% in the placebo group.

The starting dose is 1 µg/kg/week and is increased by 1 µg/kg/week according to the platelet count to a maximum of 10 µg/kg/week. The goal is to determine the minimum dose that maintains a platelet count of at least 50×10⁹/L. If there is no response after 4 weeks of maximum weekly dosing, other treatments should be considered.

Unlike eltrombopag, romiplostim is not affected by diet. It is known that the serum concentration of the drug is the same in adults and in children over 1 year of age. Platelet production peaks on days 8 to 15 after romiplostim injection and returns to baseline on days 22 to 28 [35]. In an adult study, 32% of patients maintained a platelet count of >50×10⁹/L after discontinuing romiplostim following 12 months of treatment [36]. Romiplostim was also effective in patients with an ITP duration of ≤1 year who failed first-line treatment [37]. In the pediatric study, there were no treatment-related serious adverse events. Headache and epistaxis are the most common side effects in both pediatric and adult patients [32-34]. Bone marrow examination was not routinely performed in pediatric studies despite the controversial issue of bone marrow fibrosis in adults. However, no bone marrow reticulin or fibrosis was observed in five bone marrow biopsies performed in the pediatric extension study [33].

Rituximab is a monoclonal antibody against CD20+ B cells that produce autoantibodies to platelets [38]. Rituximab is infused at 375 mg/m² once weekly over 4 weeks [39-42]. Rituximab produced a complete response in 22% to 79% of pediatric patients with ITP. The 58% of patients who responded to rituximab maintained platelet counts of >50×10⁹/L for at least 1 year after rituximab treatment, and 25% to 30% maintained a response for more than 5 years. Most patient relapse occurred within 2 years [43,44]. Rituximab, when administered in combination with dexamethasone in adults, showed superior effects compared to dexamethasone alone [45-47]. In a study in which rituximab and dexamethasone were combined in children, 30% of patients showed prolonged remission after treatment with a standard dose of rituximab and 4 days of dexamethasone [48]. Reported toxicities of rituximab, including neutropenia, infection, hypogammaglobulinemia, and infusion-related reactions, were minimized with steroid premedication.

Dapsone is an inexpensive and effective therapeutic option for the treatment of chronic ITP. Since it was first known to be effective against chronic ITP in 1988, dapsone has been one of the oldest and safest agents in the management of chronic ITP [49]. The mechanism of action of dapsone is not well understood. Hemolysis caused by dapsone is known to interfere with platelet destruction in the reticuloendothelial system and suppress antiplatelet antibodies.

Dapsone has a response rate of 50% to 72%, a complete response of 20% to 48%, and a partial response of 17% to 48% [50-52]. The duration to response is approximately 1 to 3 months, and the relapse rate is approximately 10% to 20%. As well-known hematologic adverse effects, methemoglobinemia and hemolysis occur in 10% to 20% of cases. Rarely, agranulocytosis and aplastic anemia may also occur, and characteristic hypersensitivity reactions such as fever, eosinophilia, and skin rash may occur with dapsone [50-52].

There are few reports on the use of dapsone in children. However, because it is a drug whose safety has been verified for a long time, it is worth considering as an alternative if the disease is refractory to other treatments.

Splenectomy is an effective therapeutic option for children with chronic ITP. Recently, TPO-RAs and rituximab have been preferred, and splenectomy is performed less frequently today. Owing to the development of new treatments, approximately 90% of patients now show complete remission, and less than 5% of patients show refractoriness.

The splenectomy registry of the Intercontinental Cooperative ITP Study Group collected the splenectomy data of pediatric patients with ITP. The overall response rate was 93%, and 81% of patients showed complete response. Of the latter patients, 76% showed a sustained complete response and 24% showed fluctuation of platelet counts to <100×10⁹/L [53].

Although there is a low risk of perioperative adverse events, the long-term outcomes and quality of life should be considered. Fatal sepsis and lifelong susceptibility to bacterial infection are of concern. The incidence of sepsis and venous thromboembolism in adults with ITP is approximately 10%, and age and comorbidities are important factors [54]. Prognostic factors that can be expected for remission after splenectomy are older patient age and good response to intravenous immunoglobulin and steroids in children [53,55-59].