Introduction

Poor graft function (PrGF) is a serious complication of allogeneic hematopoietic cell transplantation (allo-HCT), and current PrGF therapies are only partially effective. PrGF diagnosis is based on the presence of two or more of the following criteria: hemoglobin (Hb) concentration < 100 g/L, neutrophil count < 1.0 × 10⁹/L, or platelet count < 30 × 10⁹/L, occurring ≥ 30 d post-transplant¹. Additional criteria include the need for red blood cell (RBC) and platelet transfusions, decreased bone marrow cellularity, complete donor chimerism, and absence of graft-versus-host disease (GvHD) or relapse.

PrGF is distinct from graft failure, which is due to retained recipient immune cells, and manifests as a loss of donor chimerism or late platelet recovery failure. PrGF incidence varies widely, with reports ranging from 5% to 27%, and its clinical course can range from spontaneous recovery to death due to cytopenia-related complications^1–3.

Treatment options for PrGF include RBC and platelet transfusions, intravenous immunoglobulin, hematopoietic growth factors (e.g., granulocyte-colony stimulating factor (G-CSF) and granulocyte-macrophage colony-stimulating factor (GM-CSF)), donor blood or bone marrow infusions, mesenchymal cell infusions, or a second transplant from the same or a different donor^4–8. However, these treatments are often partially effective. Eltrombopag, an oral thrombopoietin receptor agonist, has shown efficacy in the treatment of severe aplastic anemia, idiopathic thrombocytopenic purpura, thrombocytopenia associated with hepatitis C, myelodysplastic syndrome, and acute myeloid leukemia^9–15. Eltrombopag has also been reported to improve PrGF-induced cytopenia^16–24. This study aimed to enhance our understanding of eltrombopag use in the treatment of PrGF post-allo-HCT.

Materials and Methods

This retrospective, single-center study evaluated the safety and efficacy of eltrombopag in patients with PrGF post-allo-HCT. After obtaining approval from the Institutional Review Board of the King Hussein Cancer Center (18 KHCC 146) and exemption of consent, data were collected from electronic medical records. Patient-, disease-, and transplant-related variables are presented in Table 1. Complete response was defined as the normalization of blood counts, and partial response as transfusion independence²⁵. Statistical analyses were performed using IBM SPSS Statistics for Windows version 26, SPSS Inc. Here, we report 82 patients with PrGF, 23 of whom received eltrombopag from a single center in Jordan²⁵.

Continuous variables are presented as medians and ranges, and categorical variables are presented as frequencies (percentages). To compare groups, we used the chi-square or Fisher's exact test for categorical variables, and the Mann-Whitney U test for continuous variables. P values < 0.05 are considered statistically significant.

Results

Between January 2013 and December 2023, we documented that 82 of 1,000 consecutive allotransplant recipients (8.2%) who developed PrGF. Thirty-five patients received a second transplant, all for aplastic anemia or B-thalassemia, and 23 for eltrombopag therapy, all of which were female. The median patient age was 47 years (range: 26-68 years). Fourteen patients received reduced-intensity preparative conditioning and nine received conventional myeloablative conditioning. Fourteen donors were HLA-identical siblings and nine were HLA-mismatched related relatives. The median Hb concentration pre-eltrombopag administration was 8.5 g/dL (range: 6-14.2 g/dL), granulocyte count was 7.1 × 10⁹/L (range: 1.6-85 × 10⁹/L), and platelet count was 18 × 10⁹/L (range: 6-50 × 10⁹/L). Thirteen patients had platelet counts < 20 × 10⁹/L. Pre-eltrombopag administration, bone marrow aspirations and biopsies were normocellular with trilineage hematopoiesis in six patients and hypocellular with trilineage hypoplasia in 17 patients, and 15 patients had decreased or absent megakaryocytes. Chimerism was complete (> 95% donor cells) in all patients pre-eltrombopag treatment. The median CD34-positive cell dose was 6.10 × 10⁶/kg (range: 2.75-9.78 × 10⁶/kg). The median interval between transplantation and eltrombopag initiation was 105 d (range: 29-800 d). The median weekly dose of eltrombopag was 488 mg (range, 350-700 mg), and the median treatment duration was 75 d (range, 5-446 d). The transition of hemoglobin, platelet, and granulocyte levels are shown in Figure 1 (A, B, C).

Eltrombopag was initiated at a dose of 50 mg/day and increased in 25 mg increments every 2 weeks if no response was achieved. Fifteen allotransplant recipients (65%) responded to the treatment. The median interval to response was 30 d (range: 7-122 d). Eltrombopag was gradually tapered off after achieving a stable platelet count of 25 mg every 1-2 weeks. One patient lost response after eltrombopag discontinuation because of primary disease relapse and dense marrow fibrosis. Adequate bone marrow megakaryocytes were predictive of response (p=0.019). There was no statistically significant difference in the objective response between patients with primary and secondary graft failure (p=0.074).

At 2 years, graft function was better preserved in the responders than in the non-responders (67% vs 12.5%; p=0.0001). The median follow-up period for responders was 30 months (range, 7-122 months). Twelve of the 15 responders survived with normal blood cell concentrations. The 2-year survival rate was 48% (range: 27%-70%) (Figure 2), which was significantly improved in responders compared to that in non-responders (67.0% vs 12.5%; p=0.0001; Figure 3). Three responding patients and seven non-responders died (none owing to relapse). The causes of death, GvHD (n=3), infection (n=3), and pulmonary hemorrhage (n=2). One patient responded, but died due to disease progression. No major adverse events were reported.

Conclusion

Twenty-three patients with PrGF following allo-HCT in whom prior therapies failed responded to eltrombopag treatment, as it improved their peripheral blood cell concentration. Thus, patient response to eltrombopag treatment is predictable, and eltrombopag has significant impact on patient survival outcomes.

Discussion

Our study reports the use of eltrombopag to resolve PrGF in patients post-allo-HCT. Fifteen patients (65%) achieved an objective response and became transfusion independent after a median treatment duration of 75 d. Twelve patients were alive with normal blood cell concentrations, and the 2-year survival rate was 48% (range, 27%-70%). Therefore, eltrombopag administration is safe, with no reported adverse events at the highest daily dose of 100 mg.

The efficacy of thrombopoietic agonists in severe aplastic anemia, in which the bone marrow is hypocellular, led to their use in the management of post-allo-HCT cytopenias²⁶. Giammarco et al. reported the largest cohort of patients with PrGF-induced post-transplant cytopenia (n=48) who were treated with eltrombopag (50-150 mg) for a median of 120 d. They reported an overall response rate of 75% (n=36), with complete normalization of blood counts in 24 patients (50%). This multicenter study included prolonged idiopathic thrombocytopenia as a criterion for PrGF²⁷. In our cohort, eltrombopag was used to treat PrGF-induced post-transplant cytopenia (uni-, bi-, or tri-lineage cytopenia). We observed an increase in platelet, hemoglobin, and granulocyte counts. Most of our cohort was on supportive G-CSF, and a few were on erythropoietin agonists.

Kırcalı et al. documented 39 patients with post-transplant cytopenia (8 with isolated thrombocytopenia, 19 with bi-lineage cytopenia, and 12 with pancytopenia), including nine patients after autologous bone marrow transplants. In their cohort, 84.6% (n=33) of patients responded to eltrombopag after a median of 82 d. Furthermore, they observed a trend toward improved 1-year overall survival rates for responders (75% vs 66.7%; p=0.3), with no major safety concerns. This is consistent with our results, which showed that survival at 2 years was significantly better in responders than in non-responders (67.0% vs 12.5%; p=0.0001). Additionally, 61.5% of patients in their study discontinued eltrombopag successfully after a median of 82 d, which is comparable to that in our cohort (65%) at a median of 75 d.

This study has many limitations, including its retrospective single-center design and small sample size. Moreover, the variable definitions of PrGF across different studies made comparing results challenging. Larger real-world data, registry reports, and randomized studies are warranted to draw stronger conclusions.

Acknowledgments

We acknowledge healthcare providers, patients and their families

Author Contributions

KH contributed to the conception, design of the study, and interpretation of results, wrote the first draft and final manuscript. MA, IM, RA, WD, and RA did the data acquisition. MM, HA, and MS, designed the study. AT analyzed results. All authors approved the final manuscript.

Conflicts of Interest

The authors declare no conflict of interest. Disclosure forms provided by the authors are available on the website.

Acknowledgments

We acknowledge healthcare providers, patients and their families

Data Sharing Statement

All data generated or analyzed during this study are included in this article.