Thrombopoietin Receptor Agonists in Post-Hematopoietic Cell Transplantation Complicated by Prolonged Thrombocytopenia: A Comprehensive Review

Introduction

Hematopoietic cell transplantation (HCT) is a well-established, multi-step procedure involving the collection of hematopoietic stem cells, conditioning the patient with a regimen before infusing the stem cells, and ultimately reestablishing a new hematopoietic and immunological system.¹ It initially emerged in the early 1960s as a rescue treatment for cancer patients following intense chemotherapy and radiation treatments, as well as to address severe hematopoietic system deficits.² Since then, it has expanded into an adoptive immunological therapy for many conditions, including malignant and non-malignant hematological conditions and autoimmune diseases.³ HCT is classified as autologous or allogeneic based on the source of hematopoietic cells. In autologous HCT, after receiving a preparative regimen, the patient’s hematopoietic cells taken out before undergoing intense chemotherapy are reintroduced into the patient’s body. The sources of hematopoietic cells in autologous HCT can be from a related or nonrelated donor or umbilical blood cord units.⁴

Historically, the most feared outcomes of HCT are the relapse of the underlying condition and graft-versus-host disease (GVHD). Relapse of the underlying condition is frequently the main cause of mortality within the first four years following transplantation. Beyond this time, however, there is a considerable chance that patients will experience long-term survival provided they avoid recurrence.⁵ Despite the continuous evolution of HCT practice, GVHD is still a considerable threat to all patients. A trial testing the effects of using preemptive prednisolone therapy in patients with allogeneic HCT compared to placebo did not find a significant effect on the incidence or the severity of acute graft-versus-host disease (aGVHD), which is reported to contribute by as high as 50% to non-relapse-related mortality.⁶ Multiple studies report increased rates of GVHD incidence over the years. For instance, a study aimed to analyze the trends in the incidence and outcome of chronic graft-versus-host disease (cGVHD) over 12 years reported that the incidence has increased over time, even after adjusting for factors, such as donor type, graft type, and conditioning intensity.⁷ On the other hand, many studies reported improved mortality for HCT patients. The number of individuals who have undergone autologous and allogeneic HCT and survived is steadily rising. For patients who have maintained remission during the first two to five years after transplantation, it is estimated that around 80–90% of them will still be alive during the following ten years.⁵ These findings of increased survival and decreased rates of complications have been consistent among many studies that compared earlier versus recent eras of HCT practice.^8–11 Given the improved survival rates observed among post-HCT patients, a convergence with the general population’s lifespan underscores the urgency of addressing the associated long-term complications. A salient concern in this regard is post-transplant thrombocytopenia, a protracted condition that poses a substantial risk of morbidity for post-HCT patients. Despite the magnitude of this issue, a universally recognized treatment protocol for post-transplant thrombocytopenia has yet to be established. Thrombopoietin receptor (TpoR) agonists, including both first-generation and second-generation agents, have emerged as pivotal interventions due to their demonstrated efficacy in augmenting platelet production. The utilization of these agents is progressively expanding across diverse thrombocytopenic conditions. Consequently, we embarked on an exhaustive literature review to deliver an updated and comprehensive assessment of the use of TpoR agonists. Our aim is to explore the efficacy and safety of these agents specifically in the context of treating prolonged post-HCT thrombocytopenia.

Materials and Methods

Search Strategy and Selection Criteria

The review focus was centered on peer-reviewed clinical papers involving the use of first-generation TpoR agonists (recombinant human thrombopoietin [rhTpo]) and second-generation TpoR agonists (eltrombopag, romiplostim, avatrombopag, lusutrombopag, and herombopag) in patients who underwent HCT due to malignant and non-malignant hematologic conditions and had thrombocytopenia either due to poor graft function, delayed platelet engraftment, secondary failure of platelet engraftment or in few instances to test the efficacy of TpoR agonist to facilitate and speed the process of platelet engraftment. Studies were set as not eligible if it was in a foreign language, exceptions were made for a few studies about rhTpo that were reported in Chinese. Animal studies were not included.

A comprehensive search strategy was employed in the PubMed database, covering the period from 1996 to the first of November 2023, with the aim of thoroughly investigating the utilization of TpoR agonists in addressing thrombocytopenia following HCT. The predefined search strategy incorporated medical subject headings (MeSH) terms such as “Thrombopoietin”, “Hematopoietic Stem Cell Transplantation”, and “Thrombocytopenia”. The articles identified underwent assessment for eligibility by two independent reviewers, considering relevance. Subsequently, the selected articles were retrieved and subjected to full-text assessment by two reviewers independently, adhering to inclusion and exclusion criteria. Data were then extracted from the included articles and presented in tables.

Included Articles

The summary of the electronic search strategy is presented using Figure 1. Our initial search identified 239 articles that underwent title and abstract screening for relevance. Of those, only 82 articles underwent full-text retrieval and assessment, and few studies were found ineligible for inclusion and were excluded, yielding a total of 64 studies to undergo data extraction and to be summarized in two tables. The first table includes ten studies on the use of rhTpo in patients with thrombocytopenia post-HCT, covering 777 patients. Meanwhile, the second table comprises 54 studies on the utilization of second-generation TpoR agonists.

Figure 1 Flowchart summarizing the findings of our search strategy. Abbreviations: rhTpo, recombinant human thrombopoietin; TpoR, thrombopoietin receptor.

Post-HCT Thrombocytopenia

Pathophysiology of Idiopathic Post-HCT Thrombocytopenia

Following myeloablative conditioning and autologous stem cell rescue, all patients experience profound thrombocytopenia, which is caused by bone marrow aplasia brought on by high-dose chemotherapy and/or total body irradiation (TBI). Following autologous hematopoietic progenitor cell transplantation (HPCT), the return to normal blood leukocyte and platelet counts typically occurs within the first month after transplant, with rising blood cell counts of lymphocytes, granulocytes, platelets, and eventually erythrocytes.¹² Post-transplant thrombocytopenia can be caused by a variety of reasons, including GVHD, infections (such as cytomegalovirus), immune-mediated factors, drug-related factors (such as ganciclovir and valganciclovir), disease recurrence, and thrombotic microangiopathy.^13,14 Historically, post-transplant thrombocytopenia was categorized as either secondary failure of platelet recovery (SFPR) or prolonged isolated thrombocytopenia (PIT).¹⁵ PIT is defined as the need for thrombocyte suspension transfusions for more than 60 days following HCT or sufficient engraftment of all peripheral blood lineages, except for platelets, being < 20x10⁹/L. On the other hand, SFPR is defined as losing independence on platelet transfusions for seven straight days after allogeneic HCT with a number of thrombocytes < 20x10⁹/L from over 50x10⁹/L. According to the reports, PIT and SFPR occur in 12–20% and 20–40% of patients, respectively.¹⁶ Post-HCT thrombocytopenia and lower initial posttransplant platelets are considered poor prognostic indicators as they reflect poor engraftment, can cause bleeding, and have been associated with higher mortality rates.^12,17

There are two basic theories explaining the origin of protracted isolated thrombocytopenia, while the precise etiology is unknown. Anti-platelet autoantibodies, splenic sequestration, or other conditions may cause normal platelets produced by the bone marrow to be prematurely destroyed in the peripheral circulation. Alternately, poor megakaryocyte differentiation may result in insufficient platelet production.¹⁸ One study measured glycocalicin index (GCI), a parameter that reflects platelet turn-over, serum thrombopoietin (Tpo) that is inversely proportional to megakaryocytes activity (ie, elevated in aplastic anemia) and circulating B-cells producing anti-GPIIb-IIIa in patients with or without thrombocytopenia post-HCT and patients with aplastic anemia, immune thrombocytopenia, and healthy individuals has found that CGI and Tpo levels in patients with post-HCT thrombocytopenia were similar to patients with aplastic anemia, suggesting a major role of impaired thrombocytosis as a cause of thrombocytopenia,¹⁴ with similar conclusions were reported in other study.¹⁹ In addition, antiplatelet antibodies were also detected in patients with thrombocytopenia proving the possible role of autoantibodies in the development of post-HCT thrombocytopenia and consistent with other findings in the literature as well.²⁰ These findings conclude a complex, multifactorial pathophysiology behind persistent post-HCT thrombocytopenia that remains not fully understood until now.

Treatment of Idiopathic Post-HCT Thrombocytopenia

Despite the advancements in the practice of HCT, definitive treatment regimens for post-HCT thrombocytopenia have not been established yet. Historically, the mainstay treatment of post-HCT thrombocytopenia was repeated platelets transfusion that has been used during the pancytopenia phase post-HCT, along with other blood products to maintain patients’ blood components.²¹ Potential morbidities associated with repetitive platelet transfusions include transfusion reactions, platelet alloimmunization, and increased costs.²² As a result, huge efforts were put to decrease the need for platelet transfusions and reduce bleeding complications.

Thrombopoietin and Thrombopoietin Receptors

TpoR also known previously as C-MPL or CD110 was first identified in the early 1990s.^23,24 The human TpoR is made up of various components. Its 466 amino acid extracellular portion has two cytokine receptor motifs (CRMs). In addition, there is a transmembrane section made up of 22 amino acids, followed by a cytoplasmic portion consisting of 122 amino acids.²⁵ Of the two extracellular CRMs, it seems that ligand binding occurs at the motif that is furthest from the membrane which also functions as a regulator when in the unbound state, preventing the cytokine receptor motif closest to the membrane from continuously signaling.²⁶

Tpo also known as C-MPL ligand, the glycoprotein composed of 332-amino acid, is the most potent cytokine that physiologically regulates platelet production, produced continuously in the liver and other organs, travels via the bloodstream to the bone marrow where it promotes the early emergence of numerous hematopoietic lineages.^27,28 Tpo levels are usually inversely proportional to thrombocytes in the blood and marrow megakaryocyte abundance.²⁹ Regulation of Tpo is thought to involve complex processes, one proposed mechanism is that Tpo is produced continuously and the abundance of platelets that have TpoRs binds, absorbs Tpo, and breaks it down, thus reducing Tpo levels when there is a high platelet count and when the platelet count is low, Tpo is not degraded and increases in the blood.^30,31 This model explains states of marrow failure such as aplastic anemia; however, it does not explain the unexpected decrease in Tpo levels in immune thrombocytopenia observed by some studies.^32,33 The theory was then adjusted after noticing elevated bone marrow megakaryocytes, which naturally express TpoRs explaining why blood Tpo is decreased in this disorder. Yet, the theory still does not explain the relationship between Tpo production and the abundance of platelets or megakaryocytes in other disorders such as reactive thrombocytosis associated with inflammation, or infection. Multiple factors that might induce increased Tpo levels have been described in the literature highlighting the complexity and multifactorial nature of Tpo regulation.²⁶

First Generation of Thrombopoietic Agents

Two molecules were developed in the early nineties that were found to have a stimulating activity on the thrombopoietic effects. The first was rhTpo, a glycosylated compound manufactured using Chinese hamster ovary (CHO) cells, encompassing the complete native human amino acid sequence. Administration of a sole rhTpo dose led to a rise in platelet count, commencing around the fifth day for the majority of patients and reaching its zenith at an average of day 12.³⁴ Pegylated human recombinant megakaryocyte growth and development factor (PEG-rHuMGDF), a truncated polypeptide Mpl-ligand derivatized with poly-(ethylene glycol), was produced in E-coli and had the first 163 amino acids of the human Tpo in its sequence,³⁵ an induced megakaryocyte endoreduplication and proliferation in vitro and in vivo. Changes in platelet production and function produced by PEG-rHuMGDF returned to baseline within two weeks after discontinuing treatment. Initial studies showed that PEG-rHuMGDF increases platelet production in a linear log-dose-dependent manner by stimulating megakaryocyte endoreduplication and new megakaryocyte formation from marrow hematopoietic progenitors in nonhuman primates.³⁶

Although PEG-rHuMGDF initially showed promise in early clinical trials by being safe and effective at increasing platelet levels, the development of neutralizing antibodies and severe thrombocytopenia in around 8% of exposed patients due to its homology with the native Tpo led to discontinuation of trails in the United States.^37,38 On the other side, similar adverse effects were not encountered with the use of rhTpo. Only a single study has reported that a patient tested positive for non-neutralizing transient antibodies to full length which was not associated with any clinical consequences.³⁴ As this finding was not reported elsewhere it did not hinder the clinical development of that potentially beneficial medication that has an array of indications, particularly in China where it was approved for the treatment of chemotherapy-induced thrombocytopenia (CIT).^39–41 The use of rhTpo for the treatment of post-HCT thrombocytopenia was investigated by many studies in the past with ongoing efforts to prove its benefits.⁴² Table 1 summarizes the available studies on the use of rhTpo for the treatment of post-HCT thrombocytopenia treatment. A cross all the reported nine studies that included more than 700 patients, rhTpo was reported to demonstrate a degree of efficacy for promoting platelet recovery after HCT and treatment of platelet engraftment failure.^42–51 A trial including 120 patients who underwent allogeneic HCT were randomized to receive rhTpo from day-1 post-transplant revealed a statistically significant cumulative incidence of platelet engraftment was significantly higher in the rhTpo group than in the control group on day-60 post-transplantation (91.7 ± 3.8% vs 74.5 ± 5.8%, P = 0.041) as well as a significantly lower rate of delayed platelet engraftment in the RhTpo group (8.3%, 5/60) than in the control group (23.3%, 14/60) (P = 0.024).⁴⁶ Another study included a cohort of 24 patients who underwent allogeneic HCT, 20 with DPE and four with SFPR have demonstrated a significant difference in overall response when compared to a historical group as 11 (45.8%) patients in the rhTpo group achieved platelet engraftment after 28 days of treatment versus the historical control group (12.2%, p < 0.001).⁴⁶

Table 1 Summary of ten Reported Studies on the Use of rhTpo for the Treatment of Post-HCT Thrombocytopenia

Oprelvekin, a recombinant human interleukin 11 analogs that received the United States food and drug administration (FDA) approval for the prevention of CIT in patients with non-myeloid malignancies after demonstrating modest efficacy as per multiple studies in the literature.^41,52 For instance, a meta-analysis demonstrated a reduction in the time needed for platelet counts to reach 50 × 10⁹/L (weighted mean difference [WMD] = −4.19 days; 95% CI: −5.01, −3.37), the time to reach 100 × 10⁹/L (WMD = −4.45 days; 95% CI: −4.85, −4.06), platelet transfusion volume (WMD = −6.14 units; 95% CI: −9.20, −3.09) in patients with acute leukemia patients with CIT.⁵³ Despite the reported benefits of oprelvekin in patients with CIT, the FDA did not recommend oprelvekin following myeloablative chemotherapy due to increased rates of adverse effects such as edema, conjunctival bleeding, hypotension, and tachycardia, in addition to other fatal adverse effects reported in the post-marketing period of the drug. Few studies were conducted in China on the use of oprelvekin for the treatment of patients post-HCT have demonstrated faster platelet recovery time as compared to placebo and have also reported adverse effects associated with the use of oprelvekin.^54–56 These adverse effects, along with the high cost of rhIL-11 led the manufacturer to discontinue producing the drug in the United States.⁵⁷

Second Generation of TpoR Agonists

Second-generation TpoR agonists are short peptide sequences that serve as analogs that activate TpoRs, unlike the first-generation molecules, they have no sequence homology with Tpo, therefore eliminating the dilemma of autoantibodies formation to endogenous Tpo.⁵⁸ Late in the nineties of the last century, efforts finally discovered a 14-amino acid peptide was among the earliest identified molecules, which was able to bind to and activate the TpoR, with even higher affinity by thousands after it was dimerized.^59,60 With covalently linking the tandem peptide dimers with two disulfide-bonded human IgG1 kappa light chains, the stability, and half-life of that molecule were extended, yielding the novel peptide TpoR agonist Romiplostim.⁵⁸ Romiplostim was approved by the FDA for the treatment of immune thrombocytopenia (ITP) in adults and children aged one and above. Later, another non-peptide TpoR agonists were developed that showed efficacy for the treatment of many thrombocytopenia-related conditions such as eltrombopag that was FDA approved for many conditions such as ITP and refractory aplastic anemia, avatrombopag and lusutrombopag that are FDA approved and hetrombopag that is being under investigation currently. Efforts have been focusing on the promising benefits of these novel drugs for patients with post-HCT platelet engraftment failure and here we highlight the latest evidence for the benefits and adverse effects of utilizing these medications from this aspect. Table 2 provides a summary of all the reported studies on the utility of second-generation TpoR agonist agents.^61–114

Table 2 Summary of 54 Studies Done to Assess the Efficacy and Safety of Second-Generation TpoR Agonists for the Treatment of Thrombocytopenia

Romiplostim (AMG 531, Nplate^®, Romiplate^®)

Since 1997, the primitive structure of the first molecule that had no sequence homology with Tpo started developing as an anticipated substitute for first-generation TpoR agonists.⁵⁹ Romiplostim, a peptide TpoR agonist that’s structure was achieved by linking four of the 14-amino-acid peptides to the C-terminus of an IgG1 Fc fragment creating a peptibody. Using polyglycine linkers, two of the four 14-amino-acid peptides were attached to the Fc-gamma chain. The peptibody has functional constant domains (CH2 and CH3) that can bind to the FcRn receptors which internalize it and then release it back into the circulation by exocytosis thus explaining its prolonged half-life.¹¹⁵

Although it is believed that romiplostim has lower affinity than native Tpo, it produces an immediate activation cascade starting with phosphorylation of the receptor and initiation of the JAK2 and STAT5 pathways stimulating CFU-Mk growth and increased megakaryocyte ploidy in a dose-dependent fashion.¹¹⁶ It was approved by the FDA in 2008 for ITP and administered weekly as a subcutaneous injection at a dose of 1 to 10 mcg/kg and also approved for patients with radiation injury-related thrombocytopenia. In addition romiplostim has been tested for patients with chemotherapy-induced thrombocytopenia,^117–120 myelodysplastic syndrome with thrombocytopenia,^121–123 perioperative thrombocytopenia,^124,125 in approved in Japan for thrombocytopenia in aplastic anemia after studies have shown benefit for its use there.^126,127

The utility of romiplostim use in patients with thrombocytopenia post-HCT has been questioned in multiple case reports and retrospective case series reported in Table 2.^{108,110,113,114} The only available prospective trial in this context reported that 18 patients achieved a sustained platelet (>50x 10⁹/L) of 24 patients with delayed platelet engraftment (n = 10) and secondary thrombocytopenia (n = 14) due to either GVHD or infection. In addition, a hemoglobin response was also observed in 21 of 22 patients and a response in neutrophil counts in four patients who had <1000x 10⁹/L before treatment initiation.⁸⁹ However, spontaneous recovery could not be excluded as the cause of these effects because the study lacked a control group. Another retrospective study included 13 pediatric patients with SFPR (n = 9) and engraftment failure (n = 4) reported platelet transfusion independence after one week of treatment with romiplostim despite a platelet count of <10x10⁹/L before treatment initiation.¹⁰² As with other TpoR agonists, the response to romiplostim seems to correlate inversely with low bone marrow functionality reflected by low megakaryocyte counts on bone marrow biopsies, which is thought to explain the cause of resistance to the treatment of romiplostim in some of the reported cases.^19,108

Eltrombopag (SB497115, Promacta^®, Revolade^®)

Eltrombopag is the first non-peptide Tpo agonist identified by using screening strategies of chemical libraries.¹²⁸ Interestingly, many compounds were found to display bioactivity on TpoRs, Hydrazone compounds were described as thrombopoietin receptor mimetics as they were found to be activating STAT proteins in a Tpo responsive cell line.¹²⁹ As a member of the bioarylhydrazone class of chemical compounds, eltrombopag has a metal chelate group in the center with an acidic (COOH) group at one end and lipophilic (CH3) groups at the other end.¹¹⁵ In vitro eltrombopag is a highly potent activator of the STAT and MAPK signaling pathways and induces the differentiation of bone marrow precursor cells and proliferation of Tpo-dependent cell lines. In cells expressing other factors other than TpoRs, eltrombopag did not show activity on these cascades highlighting that the activation of JAK/STAT by eltrombopag is due to the specific activation of the Tpo.¹⁰⁸ Similar to all nonpeptide TpoR agonists eltrombopag binds to the transmembrane region of the TpoR at the histidine 499 amino acid, therefore it does not compete with Tpo for the binding site and its effects seem to be additive to Tpo in various in vitro experiments.¹³⁰

In 2008, eltrombopag was initially approved as the first oral agent to increase platelet count and to be used for rare blood disorders, mainly ITP, and later its utility has been expanding after many clinical trials proved its efficacy.¹³¹ These approvals were based on evidence from trials on conditions such as aplastic anemia,^132–134 and thrombocytopenia due to Hepatitis C-related cirrhosis.^135,136 Among the TpoR agonists, eltrombopag is the most studied agent in the context of post-HCT thrombocytopenia with 24 reports of case series and retrospective studies and three published clinical trials including 687 patients (Table 2), as well as few case reports. Collectively, eltrombopag has demonstrated efficacy in raising platelet counts for patients with SFPR or PIT post HCT and decreasing platelet transfusion dependence. A recent trial including 60 patients randomized to receive eltrombopag or placebo demonstrated a statistically significant difference in the rates of achieving platelet count 50,000/µL (the study’s secondary endpoint) despite having a nonsignificant difference between the eltrombopag group and the placebo group at a threshold of 30,000/µL (the study’s primary endpoint) and therefore highlighting the efficacy of eltrombopag in raising platelet counts aside from the spontaneous recovery of platelets.⁷⁶

On the other side, a prospective single-armed trial analyzed the efficacy of eltrombopag treatment from day one after transplantation until platelet count exceeded 50,000/µL for 14 consecutive days in 12 patients undergoing allogeneic transplantation compared to a historical cohort that concluded no statistically significant between the two groups in regards to partial and complete platelet engraftment.⁷⁷ Therefore questioning the need for routine administration of anti-TpoR agents for patients undergoing HCT. Eltrombopag was shown to be effective for the treatment of SFPR and DPR, a study on a total of 86 adult patients post HCT, 16 with PT and 71 with SFPR reported overall response for platelet recovery was 72%, including 73% in the SFPR group and 67% in the PT group without a statistical significance between the groups and a longer time to response for patients with PT compared with those with SFPR with a median of 93 days and 60 days respectively.⁹⁰ Another retrospective study on 46 patients who underwent HCT reported that 23 patients responded to treatment and the cumulative incidence of successful platelet recovery was 48%. The cumulative incidences of platelet recovery were 38% among patients with DPR and 50% among those with SFPR for allogeneic transplantation, respectively.⁶⁸

Although eltrombopag displayed consistent safety and tolerability in the majority of the studies, few studies reported adverse events that required discontinuation of the drug. Only a single study reported that one of 39 adult patients who received treatment for PGF developed two episodes of grade II deep vein thrombosis.⁶³ Eltrombopag was also associated with transaminitis, a case series on eight adult patients with PGF reported three events of grade two liver enzyme elevations were reported but resolved with conservative treatment.⁷⁴ Another study on 43 pediatric patients with PIT and SFPR reported liver injury occurred in six children (14%) where transaminases were 2.5 times higher than the normal value or the bilirubin level was twice the normal value, however, treatment was continued.⁸⁴ Similarly another study reported that five patients out of 38 had similar events.⁹² Hyperbilirubinemia (grade III) was also reported in six out of 24 patients and was significantly higher in the treatment group of a recent trial.⁷⁶ However, a study reported that metabolites of eltrombopag interact with the laboratory measurements of bilirubin and therefore cause inaccurate bilirubin measurements.¹³⁷

Avatrombopag (E5501, AKR-50, YM-477, Doptelet^®)

Avatrombopag is a newer non-peptide Tpo memetic and three times more potent than eltrombopag in raising platelet counts.¹³⁸ Like other non-peptide agents, it has similar pharmacological properties such as being orally available and its binding site at the transmembrane domain that does not compete with endogenous Tpo at its distal domain binding site. The combination of avatrombopag with Tpo increased the number of differentiations of CD34+ to megakaryocytes to about 200% of that generated with Tpo only reflecting the additive properties with endogenous Tpo and its effect to induce differentiation of human stem cells.¹³⁹ A phase-I clinical trial revealed that avatrombopag caused platelet count to increase after 3–5 days post-administration, with maximum changes of >370 × 10⁹/L over baseline with 20 mg daily after 13–16 days in healthy individuals demonstrating that its effect on platelet counts depended on dose, concentration and treatment duration.¹³¹ Avatrombopag induces proliferation of human TpoR-expressing murine Ba/F3 cells in a concentration-dependent manner in in vitro experiments with a maximum activity similar to that of rhTpo activating the downstream signaling cascade through tyrosine phosphorylation of STAT3 and STAT5, and threonine phosphorylation of MAPK (ERK) activating PI3K/AKT and MAPK signaling pathways.¹⁴⁰ Similar to eltrombopag, without TpoR no response was noted.¹⁴¹

Avatrombopag was the first drug to be approved for the treatment of thrombocytopenia in cirrhosis patients undergoing elective procedures by the FDA in 2018. Approval was based on large phase II and III clinical trials assessing its clinical utility in this patient population that showed both efficacy and safety.^142,143 Further studies have demonstrated similar efficacy of avatrombopag. A phase-II clinical trial on Japanese patients with CLD and thrombocytopenia have illustrated a significant improve in platelet counts as compared to placebo with no marked adverse effects.¹⁴⁴ In 2019, the usage of avatrombopag has been expanded to include patients with thrombocytopenia due to ITP that failed to improve with other treatments as clinical trials provided evidence of the overall benefits of avatrombopag which was tolerated and effective for the treatment of chronic ITP.^145,146 Recent post-hoc analysis on Phase III trial data by Jurczak et al, 2018 has demonstrated consistent conclusions and highlighted the high durability of response to treatment.¹⁴⁷ Moreover, a study on patients with ITP treated with romiplostim or eltrombopag showed that patients achieved very high response rates even in those without initial response following switching to avatrombopag which illustrated the value of switching between TpoR agonists when prior TpoR agonists do not provide adequate effectiveness, convenience, or tolerability and potency of avatrombopag.¹⁴⁸ In addition, avatrombopag, when compared to other TpoR agents, has practical oral dosing with a single pill strength as compared to rhTpo, has lower hepatoxicity rates, and does not require dietary restrictions in opposition to eltrombopag that requires to be only taken one to two hours before food for good absorption of the drug and to obtain a baseline liver function tests for starting the medication.¹⁴⁹

So far, the utility of avatrombopag for post-HCT thrombocytopenia has been tested in three recent studies including 104 patients only where significant findings emerged. The first study, conducted on 61 patients, highlighted avatrombopag effectiveness, revealing an overall response rate of 68.9% and a complete response rate of 39.3% in patients with SFPR and DPE. Notably, the presence of adequate megakaryocytes prior to treatment was associated with a significantly higher likelihood of achieving both overall and complete responses.⁶⁵ The second study involving 16 patients focused on avatrombopag combined with mesenchymal stem cells (MSCs). This combination exhibited promising outcomes with 81.3% achieving a complete response, suggesting that this regimen could expedite platelet recovery after transplantation.⁷⁵ The third study, conducted on 30 pediatric patients with SFPR, PGF, and for promoting engraftment of platelets post-HCT, revealed a notably higher overall response rate (91%) and complete response rate (78%). This study identified different patient groups, distinguishing their response rates based on graft function. Specifically, avatrombopag was significantly more effective in the engraftment-promotion group compared to the PGF/SFPR recovery group. Risk factors such as severe GVHD and inadequate megakaryocytes were associated with reduced complete response rates.⁶⁴ Importantly, all three studies reported avatrombopag as a well tolerated and potentially effective treatment option for post-HCT thrombocytopenia in adults and children, although the combination with MSCs in one study showed promise but also a severe adverse event leading to mortality most likely related to cytomegalovirus infection.

Lusutrombopag (S-888711, Mulpleta^®)

Lusutrombopag was initially identified by Shionogi & Company, Limited, as a non-peptide second-generation TpoR agonist that shares similar pharmacological properties as other non-peptide TpoR agonists such as being orally bioavailable and acting on the transmembrane domain of human TpoRs found in megakaryocytes.^150,151 A Trial of multiple studies was conducted to evaluate the impact of food and calcium carbonate on the pharmacokinetics of lusutrombopag in healthy subjects. The results from 48 patients indicated that lusutrombopag exposure remained largely unchanged, with only a minor decrease when taken with food, and there was no significant effect when coadministered with calcium carbonate, suggesting that lusutrombopag administration does not require specific restrictions regarding meals or mineral supplements, unlike eltrombopag.¹⁵² Initial and recent trials have demonstrated the efficacy of lusutrombopag for the treatment of cirrhosis Child-Pugh class A and B patients with thrombocytopenia who are planned to undergo surgical or dental procedures.¹⁵³ Newer trials have addressed cirrhosis patients with Child-Pugh Class C and thrombocytopenia. The analysis of data from multiple studies demonstrated that lusutrombopag effectively increased platelet counts in these patients and was found to be safe and well tolerated, with no treatment-related serious adverse events.^154,155 Lusutrombopag was initially approved by the FDA for the treatment of cirrhosis patients with thrombocytopenia who are planned to undergo surgical or dental procedures as well as for patients with ITP. Given the novelty of the drug, its benefits and safety for patients with post-HCT thrombocytopenia or compared to other TpoR agonists have not been examined so far in the literature, although theoretically, it might be a promising option in the future.

Herombopag (Hengqu^®)

Herombopag is a second-generation nonpeptide TpoR agonist that was developed first by Jiangsu Hengrui pharmaceutical by structural modification of eltrombopag to improve potency.¹⁵⁶ Herombopag has the same mechanism of action as eltrombopag but with greater potency as evidenced by both in vitro and in vivo experiments.^157,158 According to the national medical products administration, herombopag has received it is first approval in China as a treatment for chronic ITP that is refractory to immunotherapy and as a conditional approval for the treatment of aplastic anemia.

As a promising treatment option for post-HCT thrombocytopenia, only a single pilot study has tested herombopag for its efficacy in 17 patients who had platelet engraftment failure after HCT as part of acute myelogenous leukemia (AML) population, compared to a matched historical cohort of patients.⁶¹ Results from this study showed a statistically significant higher incidence of partial platelet engraftment (PPE – A platelet count exceeding 20,000/μL for seven consecutive days without transfusion) and complete platelet engraftment (CPE – A platelet count exceeding 50,000/μL for seven consecutive days without transfusion) in the herombopag group as compared to placebo group with a median time to PPE and CPE of 13 days (range 8–24), 20 days (range 14–45), respectively. In addition to the demonstrated efficacy, herombopag was well tolerated among the patients with no grade IV adverse effects reported.

Safety and Adverse Events

The safety profile of recombinant rhTpo in treating post-HSCT thrombocytopenia has been questioned across multiple studies. In Nash et al’s 2000 study involving 37 patients, no significant adverse effects were observed, establishing an initial safety baseline for rhTpo. A similar finding was stated by Wolff et al, who reported no serious adverse events or neutralizing antibodies in 33 breast cancer patients treated with rhTpo following high-dose chemotherapy and autologous bone marrow transplantation (ABMT). Subsequent studies, including Lui et al and Han et al, echoed these results, with no significant adverse effects reported in 19 and 120 patients, respectively. Additionally, Wang et al and Song et al both noted the absence of severe adverse events, further reinforcing rhTpo’s safety. Minor adverse events were occasionally reported, such as grade I pain at the injection site (Sun et al, 2019) and localized pain leading to treatment discontinuation in two patients (Tang et al, 2020). Overall, the collective data suggest that rhTpo is generally well tolerated with minimal safety concerns in the context of post-HSCT thrombocytopenia.

Second-generation TpoR agonists, including Eltrombopag, Romiplostim, and Avatrombopag, exhibit varied safety profiles in the management of post-HSCT thrombocytopenia. Eltrombopag has been the most extensively studied with consistent findings supporting its safety and efficacy. Pasvolsky et al conducted a phase 2a trial involving 12 patients, demonstrating a median time to platelet engraftment of 66 days without significant toxicity or dose reductions, even at maximal doses. Yaman et al observed no treatment discontinuations due to adverse events among 18 pediatric patients, highlighting its well-tolerated nature in younger populations. Similarly, Giammarco et al and Gupta et al reported favorable safety profiles with Eltrombopag, noting minimal adverse effects and effective platelet recovery in their respective studies. Although most of the studies reported no significant concerns regarding the safety of eltrombopag, few studies reported adverse effects that in most of the times did not results in treatment discontinuation such as transaminitis,^{73,74,84,87,90,92,97} hyperbilirubinemia,^76,92 thromboembolism, and bone marrow fibrosis.⁸⁵ Kırcalı et al, 2023 for instance, reported that one patient discontinued eltrombopag treatment because of two consecutive venous thromboembolisms and reported that mild bone marrow fibrosis developed in five patients that was however clinically insignificant. Another study reported that the longer the duration of treatment response, the higher the grade of bone marrow fibrosis was noticed.⁷¹ Other nonspecific symptoms such as nausea and vomiting have been reported in few studies.⁶⁷ Given the current data, these agents generally appear safe and effective in clinical use, supporting their role in managing post-HSCT thrombocytopenia.

Limitations

Despite the promising safety profiles and potential efficacy observed with recombinant thrombopoietin agonists (rhTpo) such as romiplostim, as well as second-generation TpoRas like eltrombopag in treating post-HSCT thrombocytopenia, the existing studies have several notable limitations. Studies investigating rhTpo and second-generation agents frequently suffer from small sample sizes and a lack of control groups, which restricts the applicability of their findings across broader patient populations. Similarly, research on second-generation TpoRas, as highlighted in studies by Nampoothiri et al and Qiu et al, often relies on retrospective analyses involving diverse patient cohorts and varying dosing regimens and till now very few randomized controlled trials have been published. This heterogeneity in study designs, patient demographics and the absence of reported effect sizes complicates direct comparisons and meta-analyses, potentially hiding nuanced safety and efficacy outcomes.

In addition, given the complicated clinical course and the comorbid conditions of most of the patients who undergo HSCT, even the reported adverse effects can be hardly attributed to the use of TPO-Ras. Moreover, the scarcity of long-term safety data for both rhTpo and second-generation TpoRas raises concerns about the emergence of unforeseen adverse events, such as bone marrow fibrosis. These collective limitations emphasize the critical necessity for large-scale, well-designed prospective trials with standardized protocols. Such studies are essential to definitively assess the efficacy, safety profiles, and optimal clinical use of rhTpo and even the newer second-generation TpoRas in the management of post-HSCT thrombocytopenia, which is currently being rarely reported on few occasions that are mostly case reports and case series.

Conclusion

Clinical studies of second-generation TpoR agonists, including romiplostim, eltrombopag, avatrombopag, and herombopag demonstrated promising effectiveness in promoting platelet recovery post-HCT, albeit with variations in response rates. Second-generation TpoR agonists seem to be more convenient than subcutaneous rhTpo as they are orally administered with a reasonable platelet recovery as. Figure 2 shows a simplified flowchart for the suggested use of second-generation TpoR agonists in post-HCT thrombocytopenia. Nevertheless, the overall efficacy and safety of TpoR agonist agents post-HCT necessitate further exploration through larger, well-structured clinical trials to optimize their use and ensure their place in standard clinical practice.

Figure 2 Flowchart illustrating our suggestions for second-generation TpoR agonists use in post-HCT thrombocytopenia. A SFPR definition: losing Independence on platelet transfusions for seven straight days after allogeneic HCT with a number of thrombocytes < 20x109/L from over 50x109/L. PIT definition: the need for thrombocyte suspension transfusions for more than 60 days following HCT or sufficient engraftment of all peripheral blood lineages, except for platelets, being < 20x109/L. BThere is currently no evidence supporting the early use of TpoR agonists to accelerate the recovery of platelet counts. * Regular individualized follow-up including screening, prevention, and counseling for all patients based on the recommendations from the guidelines. ** Investigations are based on clinical suspicion and may include laboratory work-up and histologic confirmation. + Suggested based on the level of evidence from the literature, taking into consideration the availability, costs and pharmacological properties of different agents. #Does not require dietary restriction (versus eltrombopag which is only taken one to two hours before food for good absorption). Abbreviation: HCT, hematopoietic cell transplantation; TpoR, thrombopoietin receptor; GVHD, graft versus host disease; SFPR, secondary failure of platelet recovery; PIT, prolonged isolated thrombocytopenia; LFT, liver function tests; BMB, bone marrow transplant; SC, subcutaneous.