Between September 2021 and October 2022, 30 patients who underwent allo-HSCT in the Department of Pediatrics, Nanfang Hospital, Southern Medical University were enrolled. Inclusion criteria were defined for those patients who had PGF, SFPR, or PIT with complete engraftment of donor cells. Exclusion criteria were defined for patients with leukemia relapse or abnormality of hepatic function (more than 1.5 times the upper limit of normal serum glutamate pyruvate transaminase and bilirubin). A total of 32 thrombocytopenia events occurred and were treated with avatrombopag in these patients. The institutional review board at Nanfang Hospital hospital approved the retrospective study, and all consent forms approved by the institution were signed.

The donor selection and the transplant protocol were reported in previous studies (12–14). In general, busulfan, cyclophosphamide, and fludarabine with or without thiotepa were administrated as a myeloablative conditioning regimen for both HLA-matched and HLA-mismatched patients. Posttransplantation cyclophosphamide (PTCY) with or without fludarabine at day +3 and day +4 was utilized as a haploidentical transplant regimen.

PGF was defined as persistent neutropenia (absolute neutrophil count <0.5 × 10⁹/L), thrombocytopenia (platelets <20 × 10⁹/L), and/or hemoglobin <70 g/L for at least 3 consecutive days by day 28 post-HSCT with transfusion requirement in the presence of complete donor chimerism without disease relapse (15, 16). SFPR was defined as platelets <20 × 10⁹/L for 7 consecutive days or requirement of transfusion following reaching platelets ≥50 × 10⁹/L without transfusion for 7 days post-HSCT (17). Meanwhile, the promotion of platelet engraftment was proposed to facilitate platelets ≥20 × 10⁹/L without transfusion for 7 days post-HSCT. The number of megakaryocytes in bone marrow (BM) smear less than 7 in 1.5 cm × 3.5 cm area was considered as inadequate megakaryocyte status (18, 19).

Since this was a retrospective pilot study, randomization, blinding, and power analysis were not applicable. Avatrombopag was administrated for the patients with PGF or SFPR. It was indicated for engraftment-promotion patients who had a history of immune cytopenia, immune autoantibodies detected positive post-HSCT (direct Coomb's test or platelet antibody test), or encountered hemorrhage complications post-HSCT. The initial dosage was 10 mg once a day for a weight of less than 30 kg patients while 20 mg for ≥30 kg patients. Subsequently, the elevated dosage by an interval of 10 mg every two weeks according to the response with a maximum dosage of up to 40 mg/day (Supplementary Table S1). Avatrobopag was tapered or ceased if the platelet count achieved ≥50 × 10⁹/L or 100 × 10⁹/L, respectively, without transfusion. The overall response (OR) was defined as the platelet ≥20 × 10⁹/L for at least 7 consecutive days without transfusion. Meanwhile, the complete response (CR) was defined as the platelet ≥50 × 10⁹/L for at least 7 consecutive days without transfusion (18). Adverse events and severe adverse events were evaluated according to the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI-CTCAE) version 5.0.

Two-tailed t-test or Kruskal–Wallis test were performed for continuous variables between groups. Categorical variables were compared using the χ² test or Fisher's exact test. The univariate and multivariate logistic regression analyses were conducted to identify risk factors of treatment of avatrombopag. The factors in univariate analysis with a p-value < 0.05 were input into multivariate analysis. The cumulative incidence of OR and CR was calculated by a competing risk model. The overall survival probability (OS) was determined by the Kaplan–Meier analysis and compared with the log-rank test. It was considered statistically significant if the p-value <0.05. Data analyses were completely conducted with RStudio (version 2022.02.0 + 443).

A total of 30 patients with 32 thrombocytopenia events were included in this pilot study (Figure 1). The details of the transplant were summarized in Table 1. The median age of the patients was 6 years old. Thalassemia major (TM) accounted for 59% of the disease composition. All patients received myeloablative conditioning. In addition, all patients were treated with avatrombopag as indicated, including 15 (47%) patients with PGF, 5 (16%) patients with SFPR, and 12 (37%) patients with the aim of engraftment promotion. No one PIT patient was documented during the study period. In general, 56% of patients underwent HLA-mismatched transplants with 31% cord blood engraftment. PTCY was the main transplant regimen which accounted for 88%. The median days of neutrophil and platelet recovery were 21 days and 18 days, respectively. Eleven (34%) patients and three patients suffered from III–IV acute GVHD (aGHVD) and extensive chronic GVHD (cGVHD), respectively. Sixteen (50%) patients experienced CMV infection. Autoimmune hemolytic anemia (AIHA) and immune thrombocytopenia (ITP) occurred in 19 (59%) and 4 (12%) patients, respectively. Five (16%) patients had veno-occlusive disease (VOD). Two (6%) patients underwent transplant-associated thrombotic microangiopathy (TA-TMA). Hemorrhagic cystitis was the most common hemorrhage event accounting for 12 (38%) patients. Moreover, the overall survival probability (OS) was 93.2% ± 4.7% [95% CI (84.5%–100%)] (Figure 2A).

Based on the indication of avatrombopag, patients were divided into two groups (PGF/SFPR and engraftment-promotion) for further analysis (Table 2). The median day of initial avatrombopag treatment post-HSCT was at day +25. The median accumulative dosage of avatrombopag was 325 mg in PGF/SFPR group which was markedly higher than the engraftment-promotion group (p = 0.007). Nine (28%) patients received recombinant human thrombopoietin (rhTPO), and two patients were exposed to umbilical cord blood-derived mesenchymal stem cell (MSC).

The OR rate (ORR) and CR rate (CRR) in the entire population were 91% and 78%, respectively. Additionally, 17 (85%) patients and 13 (65%) patients achieved OR and CR, respectively in the PGF/SFPR group, while both ORR and CRR were 100% in the engraftment-promotion group (ORR p = 0.300 and CRR p = 0.029). Similarly, the cumulative ORR and CRR were significantly lower in PGF/SFPR group compared to the engraftment-promotion group (86.7% vs. 100%, p = 0.002 and 65.0% vs. 100%, p < 0.001, respectively) (Figure 3). The median time of OR achievement in the PGF/SFPR group was significantly longer than in the engraftment-promotion group 16 days vs. 7 days (p = 0.003). It required a predominantly longer time to achieve CR in PGF/SFPR group compared to the engraftment-promotion group (30 days vs. 11 days, p = 0.002). Surprisingly, the median dose of platelet and red blood cell (RBC) transfusion were 2 units and 1.5 units, respectively in the whole population.

As result, the OS of the non-CR thrombocytopenia population dramatically dropped to 68.6% ± 18.6% [95% CI (4.03%–100%)] compared to the CR population which was 100% (p = 0.005) (Figure 2B). Interestingly, hemoglobin level was significantly elevated by avatrombopag in both PGF/SFPR and engraftment-promotion groups (p = 0.001 and p = 0.001, respectively), whereas neutrophil level was increased only in engraftment-promotion groups (p = 0.020) (Figure 4).

Grade II–III alanine aminotransferase increase occurred in three (9%) patients. All of these three cases were considered as not likely related to avatrombopag. No other adverse effect was found. Furthermore, no patients were intermitted or suspended from the avatrombopag treatment. In the end, the causes of the two dead patients were severe respiratory syncytial virus pneumonia and the progress of EB virus relative-hemophagocytic syndrome.

Firstly, a statistically higher incidence of NR was found in both aGVHD (p = 0.023) and inadequate megakaryocyte status (p = 0.004) subgroups (Table 1). Secondly, in the univariate analysis (Table 3), grade III–IV aGVHD and inadequate megakaryocyte status were identified as risk factors (p = 0.03 and p = 0.01, respectively). Of note, late platelet recovery (p = 0.05), hemorrhagic cystitis (p = 0.05), and CMV infection (p = 0.06) were close to risk factors despite no statistical difference. Furthermore, both aGVHD and megakaryocyte status were further analyzed in multivariate analysis. Although inadequate megakaryocyte status potentially indicated non-CR achievement (p = 0.06), there was no significant difference in these two factors (Table 3).