Efficacy and safety of avatrombopag in aplastic anemia: a comprehensive review of clinical evidence

Abstract

Aplastic anemia is a bone marrow failure disorder marked by cytopenias that impair oxygen delivery, immune defense, and hemostasis. Standard therapy traditionally combines immunosuppression with or without hematopoietic stem cell transplantation, and more recently incorporates thrombopoietin receptor agonists to stimulate residual hematopoiesis. Eltrombopag improved outcomes when added to immunosuppressive therapy, but its association with hepatotoxicity limits its suitability for patients with underlying liver disease or elevated baseline liver enzymes. Avatrombopag is a newer oral thrombopoietin receptor agonist that does not require dietary restrictions and does not undergo significant hepatic metabolism, which offers a potential therapeutic advantage in settings where liver function is compromised. This review evaluated ten clinical studies published from 2023 to October 2025 that investigated avatrombopag in acquired aplastic anemia across varied patient populations, including treatment-naive, relapsed or refractory cases, older adults, and patients with secondary aplastic anemia related to chemoradiation. Across these studies, overall response rates ranged from 55% to 85%, and complete response rates reached up to one-third of treated patients. Response onset typically occurred within 1–2 months, which aligns with clinical decision timelines for assessing therapeutic benefit. Avatrombopag supported reductions in transfusion requirements and sustained hematologic improvement in both severe and non-severe disease. Patients previously intolerant or non-responsive to eltrombopag also demonstrated clinical improvement, which suggests pharmacologic differences translate into meaningful therapeutic effects. Importantly, avatrombopag demonstrated a favorable safety profile in all reviewed settings. Reports did not identify clinically relevant hepatotoxicity, clonal evolution, or treatment-limiting adverse effects. Its tolerability in patients with liver dysfunction distinguishes it from earlier agents in this drug class. Ongoing trials will clarify optimal dosing strategies and define its future role within first-line therapy and salvage treatment pathways for aplastic anemia.

Plain language summary

A new treatment option for aplastic anemia: how avatrombopag may help patients with fewer side effects

Aplastic anemia is a rare but serious disease where the bone marrow stops making enough blood cells. People with this condition often feel tired, bruise easily, or get frequent infections. The standard treatment for those who can’t get a bone marrow transplant is a combination of medications that suppress the immune system. Another helpful medicine, eltrombopag, boosts blood cell production but can harm the liver, which limits its use in some patients. Avatrombopag is a newer medicine that also stimulates blood cell production but seems to be safer for the liver. It’s taken as a pill and does not interact with food or most other drugs. This review looked at eight studies from 2023 and 2024 to understand how well avatrombopag works and how safe it is. The studies included patients with different types and severities of aplastic anemia, including older adults and those with liver problems. Overall, more than half of the patients improved with avatrombopag, and some fully recovered. Most people began feeling better within one to two months. Avatrombopag was helpful even for people who didn’t respond to other treatments. It caused few side effects and did not damage the liver. Some patients even had improved liver function during treatment. These results suggest that avatrombopag is a promising new treatment for aplastic anemia. It may offer a better and safer option for patients, especially those who can’t tolerate other drugs. More research is underway to confirm these findings and explore its role in standard treatment plans.

Introduction

Aplastic anemia (AA) is a rare but life-threatening bone marrow failure syndrome characterized by pancytopenia and bone marrow hypoplasia. Immune-mediated destruction of hematopoietic stem cells underlies most acquired cases. For patients who are ineligible for hematopoietic stem cell transplantation, immunosuppressive therapy (IST) with antithymocyte globulin (ATG) and cyclosporine (CsA) remains the mainstay of treatment. However, only 50%–70% of patients respond to IST, and many experience relapse, incomplete hematologic recovery, or transfusion dependence. ¹

Thrombopoietin receptor agonists (TPO-RAs), particularly eltrombopag, have been shown to improve hematologic responses when added to IST, leading to higher complete response (CR) rates and faster platelet recovery. ² Despite these benefits, eltrombopag’s use is limited by its hepatotoxicity and drug–metal ion interactions. These challenges are particularly problematic in elderly patients and those with preexisting liver dysfunction, necessitating safer alternatives that retain or exceed the efficacy of eltrombopag.

Avatrombopag (AVA) is an oral TPO-RA with minimal hepatic metabolism and no dietary restrictions. It mimics endogenous thrombopoietin by binding to and activating the thrombopoietin receptor (c-Mpl) on megakaryocytes, triggering intracellular signaling cascades involving the JAK-STAT, MAPK, and PI3K-AKT pathways. AVA has demonstrated efficacy in phase II and III trials for treating thrombocytopenia in patients with chronic immune thrombocytopenia (ITP) ³ and in thrombocytopenic patients with chronic liver disease undergoing elective procedures.^4,5 More recently, it has been investigated in several clinical scenarios of AA.

Unlike eltrombopag, AVA does not chelate metal ions and is primarily metabolized via cytochrome P450 enzymes, thereby minimizing liver toxicity. Its consistent pharmacokinetics across ethnic groups and lack of significant drug-drug interactions enhance its suitability, particularly for patients with hepatic impairment or on polypharmacy. ⁶ In addition, AVA does not require dose adjustment in East Asian patients and may offer a more favorable hepatic safety profile. ⁷ Here, we present a comprehensive synthesis of available clinical data evaluating AVA across different AA subtypes and patient populations.

Patients and methods

This narrative review synthesizes clinical evidence on the use of avatrombopag (AVA) in patients with acquired aplastic anemia (AA), focusing on efficacy and safety outcomes across diverse clinical subgroups. A structured search of the literature was conducted using PubMed, Web of Science, and Embase databases from January 2020 to October 2025. Search terms included “avatrombopag,” “aplastic anemia,” “severe aplastic anemia,” “non-severe aplastic anemia,” “TPO receptor agonist,” and combinations thereof. Studies were included if they involved adult or elderly patients diagnosed with severe (SAA), very severe (VSAA), or non-severe AA (NSAA), treated with AVA either as monotherapy or in combination with immunosuppressive therapy (IST), and reported hematologic response and safety outcomes.

The included studies comprised both prospective and retrospective clinical trials, single-arm phase II trials, and observational cohort analyses. A total of ten studies were selected, each published in peer-reviewed journals and providing data on hematologic response rates, complete response (CR), partial response (PR), time to response, transfusion independence, and treatment-related adverse events. Populations represented treatment-naïve patients receiving first-line IST, patients with relapsed or refractory disease after prior therapy, including eltrombopag, elderly patients (⩾60 years), and those with hepatic comorbidities or baseline liver dysfunction.

Data extraction was performed independently by two reviewers. Information retrieved included study design, number and characteristics of patients, treatment regimen (AVA dosing, concurrent ATG and CsA), follow-up duration, definitions of response criteria, incidence of clonal evolution, and reported toxicities. Hematologic responses were defined according to standard AA response criteria: CR required normalization of peripheral counts (hemoglobin > 100 g/L, absolute neutrophil count > 1.5 × 10⁹/L, platelets > 100 × 10⁹/L); PR required transfusion independence and improvement in at least one lineage. Overall response rate (ORR) was calculated as the sum of CR and PR. Adverse events of grade ⩾2 and instances of treatment discontinuation were recorded.

Comparisons were made across studies for consistency in outcomes. Where available, subgroup analyses based on age, disease severity, and time from diagnosis to AVA initiation were included. The DIAAMOND trial, although ongoing, was included to provide a prospective design framework and endpoints. No formal meta-analysis was conducted due to heterogeneity in study design and outcome definitions.

Results

This review includes data from ten clinical studies comprising a total of 372 patients with acquired or secondary aplastic anemia treated with avatrombopag. The studies span a wide range of clinical contexts, including treatment-naïve, relapsed/refractory, elderly patients, those with liver dysfunction, and cases of chemoradiotherapy-induced bone marrow failure. The findings consistently demonstrate favorable hematologic responses, rapid onset of action, and a well-tolerated safety profile. Key outcomes from each study are summarized in Table 1.

Table 1.

Summary of avatrombopag studies in aplastic anemia.

Study	Population	Treatment	ORR (%)	CR (%)	Median Time to Response	Transfusion Independence	Safety
Zhang et al. (2024) 8	SAA; EPAG-refractory/intolerant (n = 20)	AVA 40–60 mg/day	85%	15%	1.7 months	50%	No grade ⩾2 AEs; well tolerated
Wan et al. (2023) 9	NSAA; relapsed/refractory/intolerant (n = 25)	AVA 20–60 mg/day	58%	25%	Not stated	Not stated	Well tolerated; no clonal evolution
Zhang et al. (2023) 10	R/R AA; AVA (n = 30) vs EPAG (n = 45)	AVA vs EPAG	55.2% (AVA) vs 51.4% (EPAG)	20.7% (AVA) vs 11.4% (EPAG)	1 mo (AVA) vs 3 mo (EPAG)	Not stated	Fewer AEs than EPAG (30% vs 53.3%)
Li et al. (2024) 11	SAA; newly diagnosed (n = 42 vs 84 controls)	AVA + ATG + CsA vs IST alone	71.4% (vs 51.2%)	31.0% (vs 14.3%)	Not stated	Not stated	No hepatotoxicity; well tolerated
McQuilten et al. (2024) 12	SAA; treatment-naïve and R/R (2 cohorts)	AVA + IST (FIRST) or AVA ± IST (NEXT)	Ongoing (target ORR 40%)	Ongoing (target CR 40%)	NA	Planned outcome	Ongoing; ACE monitored
McQuilten et al. 2025 (DIAAMOND-NEXT) 13	Relapsed/refractory SAA (n = 20)	AVA 60 mg/day + SOC	50	5	Not stated	Not stated	ACE 7%; Grade ⩾3 AEs 60%; 4 abnormal LFTs; 3 deaths
Zhang et al. (2024) 14	Elderly (⩾60 years) tD-NSAA (n = 26 vs. 26)	AVA + CsA vs CsA alone	76.9% (vs. 46.2%)	34.6% (vs. 7.7%)	1 mo (vs. 3 mo)	72.7% (vs. 31.6%) at 6 mo	Comparable AEs (~40%); well tolerated
Wang et al. (2023) 15	SAA with liver disease (n = 14)	AVA + IST (ATG + CsA)	57.1%	28.6%	1–9 months for CR	RBC: 9/14; Platelets: 8/14	Well tolerated; no hepatotoxicity; no clonal evolution
Chi et al. (2024) 16	Chemoradiotherapy-induced AA (n = 34)	AVA monotherapy	59% at 6 months	~33%	Not clearly specified	Improved with treatment duration	No clonal evolution or liver toxicity; well tolerated
Wang et al. 2025 17	Elderly (⩾60 years) newly diagnosed SAA ( 42 (ATG+CsA+AVA) vs 42 (CsA+AVA))	ATG+CsA+AVA vs CsA+AVA	65.9 vs 73.2 (p > 0.05)	19.5 vs 19.5 (p > 0.05)	2 vs 3 months (p = 0.522)	Planned outcome	CsA+AVA: fewer AEs (35.7% vs 64.3%, p = 0.009); fewer SAEs (9.5% vs 26.2%, p = 0.047)

In a multicenter prospective study by Zhang et al., 20 patients with SAA who were refractory or intolerant to eltrombopag received AVA 40–60 mg/day. After a median follow-up of 6 months, ORR was 85%, with CR in 15%, and PR in 50%. The median time to response was 1.7 months. Among eltrombopag-refractory patients, 75% responded to AVA. Transfusion independence was achieved in 50% of patients, and no grade ⩾2 adverse events were reported. ⁸

Wan et al. conducted a phase II trial in 25 patients with NSAA who were relapsed, refractory, or intolerant to standard therapy. AVA at 20–60 mg/day produced an ORR of 56% at 3 months and 58% at 6 months. CR was observed in 25% of patients. Multilineage hematologic responses were reported, and the response correlated significantly with earlier AVA initiation and relapsed/intolerant status rather than refractory disease. ⁹

In a retrospective study comparing AVA (n = 30) and eltrombopag (n = 45) in relapsed/refractory AA, Zhang et al. reported similar 6-month ORRs (55.2% vs 51.4%). AVA was associated with faster responses (median 1 vs 3 months, p = 0.012), higher platelet counts at month 2, and fewer adverse events (30% vs 53.3%, p = 0.046), particularly liver and renal toxicity. ¹⁰

Li et al. evaluated AVA combined with ATG and CsA in 42 newly diagnosed SAA patients compared to 84 matched controls treated with IST alone. At 6 months, CR and ORR were significantly higher in the AVA group (CR 31.0% vs 14.3%, p = 0.048; ORR 71.4% vs 51.2%, p = 0.048). Event-free survival favored the AVA arm (60.7% vs 49.6%), and no hepatotoxicity was observed even in those with pretreatment liver dysfunction. ¹¹

The DIAAMOND trial, as described by McQuilten et al., is an ongoing Bayesian Optimal Phase II study evaluating AVA in treatment naïve (FIRST cohort) and relapsed/refractory (NEXT cohort) SAA patients. AVA is administered at 60 mg/day for 6 months, with or without IST. The primary endpoints are 6-month CR (FIRST) and ORR (NEXT), with clonal evolution as a co-primary safety endpoint. ¹² Recent results from the DIAAMOND-AVA-NEXT trial in 20 evaluable patients with relapsed/refractory SAA showed a 6-month ORR of 50% (1 CR, 9 PR) and an acquired clonal evolution rate of 7%. ¹³

In a study by Zhang et al. focusing on elderly patients (⩾60 years) with transfusion-dependent NSAA, AVA + CsA (n = 26) was compared to CsA alone (n = 26). ORR at 6 months was significantly higher in the AVA group (76.9% vs 46.2%, p = 0.023), and CR was 34.6% vs 7.7% (p = 0.017). Median time to response was shorter (1 vs 3 months), and platelet transfusion independence was significantly higher in the AVA group. Adverse events were comparable between groups (~40%) and manageable. ¹⁴

Wang et al. studied 14 patients with SAA and liver disease treated with AVA + IST. The ORR was 57.1%, with CR in 28.6%. All responders were under age 50, highlighting the importance of age as a response predictor. AVA was well tolerated, and liver function normalized in all but one patient. No clonal evolution or genetic progression was observed. ¹⁵

In a retrospective study of 34 patients with aplastic anemia secondary to chemoradiotherapy, avatrombopag demonstrated increasing efficacy over time, with overall response rates of 32% at 1 month, 56% at 3 months, and 59% at 6 months. Approximately one-third of patients achieved CR. Responses occurred regardless of prior eltrombopag exposure but were significantly associated with higher cumulative AVA doses. No clonal evolution, new liver dysfunction, or severe adverse events were observed. ¹⁶

A pivotal multicenter, prospective study by Wang et al. published in 2025 compared ATG + CsA + AVA (n = 42) with CsA + AVA (n = 42) in older adults (⩾60 years) with newly diagnosed SAA. The study found comparable efficacy between the two arms, with 6-month ORRs of 65.9% and 73.2%, respectively (p > 0.05). However, the CsA + AVA arm demonstrated a significantly better safety profile, with fewer adverse events (35.7% vs 64.3%, p = 0.009) and severe adverse events (9.5% vs 26.2%, p = 0.047). ¹⁷

Discussion

Thrombopoietin receptor agonists (TPO-RAs) have become an integral component of aplastic anemia therapy, particularly in enhancing hematologic recovery when combined with immunosuppressive regimens. Their expanding role in both frontline and relapsed settings underscores the importance of evaluating newer agents within this class. This review demonstrates the consistent efficacy of avatrombopag across diverse patient populations with AA. In both frontline and salvage settings, AVA yielded ORRs of 55%–85% and CR rates up to 34%, comparable to or exceeding those reported with eltrombopag. Notably, AVA responses occurred earlier (median ~1–2 months) and were associated with higher platelet recovery and fewer adverse events.

A recent meta-analysis by Zhang et al. provided strong evidence that adding eltrombopag to standard IST improves early hematologic responses in SAA. Among 2148 patients across 16 studies, the addition of EPAG significantly increased the ORR and CRR at 3 and 6 months. ¹⁸ Importantly, the combination improved overall survival without significantly increasing clonal evolution or adverse events. These findings confirm the therapeutic value of TPO-RAs in frontline SAA treatment and provide a benchmark against which emerging agents like avatrombopag may be compared.

Although direct comparisons are limited, retrospective data show that avatrombopag may offer faster responses and better tolerability than eltrombopag, particularly with respect to liver and kidney toxicity. ¹⁰ Unlike eltrombopag, AVA does not require dietary restrictions or dose adjustments for hepatic metabolism, making it suitable for patients with liver disease. The study by Wang et al. confirmed its safety and efficacy in patients with significant hepatic dysfunction, a population often excluded from TPO-RA trials. ¹⁵ A 2025 pharmacovigilance study based on the FDA Adverse Event Reporting System further supports this, reporting lower hepatotoxicity signals for avatrombopag compared to eltrombopag. ¹⁹

The benefit of AVA when used upfront with IST, as demonstrated in the Li et al. study, marks a significant shift in treatment strategy. The combination was associated with improved response depth and quality, faster hematologic recovery, and greater transfusion independence compared to IST alone. Importantly, this enhanced efficacy was achieved without compromising safety, even in patients with preexisting liver dysfunction. ¹¹

Age emerged as a critical predictor of response, with younger patients consistently demonstrating better outcomes. In elderly patients, AVA combined with CsA still offered superior response rates compared to monotherapy, suggesting that age should inform but not exclude TPO-RA use. ¹⁴ The 2025 study by Wang et al. provides compelling prospective evidence that for older adults with SAA, an ATG-free regimen of CsA and avatrombopag offers comparable efficacy to the standard triple therapy but with a significantly improved safety profile, making it a highly attractive frontline option for this population. ¹⁷ Multilineage responses with AVA were frequently observed, especially in patients with early initiation or higher baseline reticulocyte counts, suggesting a robust stimulation of hematopoietic progenitors across all lineages.^9,11

Avatrombopag shows efficacy in patients with secondary AA following chemoradiotherapy and in those previously unresponsive to eltrombopag, and its efficacy appeared dose dependent. Importantly, no clonal evolution or liver toxicity was observed. ¹⁶

Beyond hematopoiesis, TPO-RAs may also exert immunomodulatory effects that could contribute to their therapeutic efficacy in marrow failure syndromes, as illustrated and proposed by Schifferli and Kühne through the mechanism of immune tolerance induction via increased platelet mass, TGF-β release, and the immunosuppressive activity of platelet-derived microparticles. ²⁰ Zhu et al. further demonstrated that TPO-RAs enhance myeloid-derived suppressor cells, which suppress cytotoxic T cells and promote regulatory T cell expansion. ²¹ These findings, primarily observed in ITP, but may be relevant in SAA, which requires more evidence work, where autoimmune destruction of hematopoietic stem cells is central to pathogenesis. The potential immunoregulatory role of TPO-RAs warrants further exploration in the context of AA.

Concerns persist regarding eltrombopag’s potential to promote clonal evolution and relapse in AA, particularly in relapsed settings. Although frontline data suggest comparable risks to standard IST, long-term outcomes remain uncertain. A possible thrombotic risk in patients with clonal hematopoiesis has been reported, underscoring the need for caution and further study.^2,22 Within the available clinical evidence, no report shows AVA has treatment-related clonal evolution, marrow dysplasia, or significant hepatotoxicity. Even in patients with preexisting chromosomal abnormalities or PNH clones, AVA was not associated with disease progression.^10,11,15 The DIAAMOND-AVA-NEXT trial reported a 7% rate of acquired clonal evolution at 6 months in a relapsed/refractory population, a rate that requires continued monitoring in larger, long-term studies but appears comparable to other TPO-RAs in this setting. ¹³

While many of the initial studies on avatrombopag in AA originated from China, the growing body of international evidence, including the Australian DIAAMOND trial and the recent multicenter study by Wang et al., enhances the generalizability of these findings. It is worth noting that the incidence of AA is two to three times higher in Asia than in Western countries, a difference potentially linked to genetic predispositions, such as polymorphisms in HLA types and detoxification enzyme genes (e.g., GSTM1 and GSTT1), rather than purely environmental factors. ¹⁹ The prevalence of hepatitis-associated AA also appears slightly higher in East Asia (4%–10%) compared to the West. ²³ However, pharmacokinetic studies have shown no clinically significant differences in avatrombopag metabolism between Japanese, Chinese, and White subjects, suggesting that dose adjustments based on ethnicity are not required.^7,24 This contrasts with eltrombopag, where higher plasma levels have been observed in Asian populations, sometimes necessitating dose modifications. ²⁵

Limitations

This review has several limitations that should be acknowledged. First, the available evidence on avatrombopag in AA, while growing, is still limited, with a scarcity of large-scale, randomized controlled trials directly comparing it to other TPO-RAs. Many of the included studies were retrospective or single-arm prospective trials with relatively small sample sizes. Second, there was significant heterogeneity across studies in terms of patient populations (treatment-naïve vs relapsed/refractory, SAA vs NSAA), avatrombopag dosing schedules, and definitions of response criteria, which complicates direct comparisons. Finally, while the updated literature search includes recent prospective data, long-term follow-up on outcomes such as clonal evolution, relapse-free survival, and overall survival is still maturing. The early closure of the DIAAMOND-AVA-NEXT trial due to slow enrollment also limits the robustness of its conclusions.

Conclusion

Avatrombopag is an effective and well-tolerated TPO-RA for the treatment of AA. It demonstrates consistent efficacy across diverse clinical scenarios and patient populations, including those with liver comorbidities and the elderly. Its favorable safety profile, particularly the lack of significant hepatotoxicity, positions it as a promising therapeutic option in the management of AA. Ongoing and future studies will further clarify its optimal role in the treatment landscape of this challenging disease.