Biol Blood Marrow Transplant 25 (2019) 1825 1831 Biology of Blood and Marrow Transplantation journal homepage: www.bbmt.org Thrombopoietin Receptor Agonists for Severe Thrombocytopenia after Allogeneic Stem Cell Transplantation: Experience of the Spanish Group of Hematopoietic Stem Cell Transplant  María Bastida2, Irene García-Cadenas3, Estefania García-Torres4, Daniel Rivera2, Leyre Bento1,*, Jose Anna Bosch-Vilaseca3, Carlos De Miguel5, María Esther Martínez-Mun ~ oz5, Francesc Ferna ndez-Avile s6, Elisa Rolda n n7, Anabelle Chinea8, Lucrecia Ya ~ ez9, Teresa Zudaire10, Carlos Pinho Vaz11, 12 Ildefonso Espigado , Javier Lo pez , David Valca 8 rcel7, Rafael Duarte5, Rafael Cabrera5, Concepcio 4  Ramo  n Herrera , Jose  n Gonza lez-Porras2, Antonio Gutie rrez1, Carlos Solano13, 1 Antonia Sampol , on behalf of the Grupo Espan ~ ol de Trasplante Hematopoye tico (GETH) 1 n Sanitaria Islas Baleares, Palma de Mallorca, Spain Hematology Department, Hospital Universitario Son Espases, Instituto de Investigacio 2 Hematology Department, Hospital Universitario de Salamanca, Salamanca, Spain 3 Hematology Department, Hospital Sant Creu i Sant Pau, Barcelona, Spain 4 Hematology Department, Hospital Universitario Reina Sofía, Co rdoba, Spain 5 Hematology Department, Hospital Universitario Puerta de Hierro, Madrid, Spain 6 Hematology Department, Hospital Clínic de Barcelona, Barcelona, Spain 7 Hematology Department, Hospital Vall D’hebron, Barcelona, Spain 8 Hematology Department, Hospital Ramo n y Cajal, Madrid, Spain 9 Hematology Department, Hospital Universitario Marques de Valdecilla, Santander, Spain 10 Hematology Department, Complejo Hospitalario de Navarra, Pamplona, Spain 11 Hematology Department, Instituto Portugu^es de Oncologia, Oporto, Spain 12 Hematology Department, Hospital Virgen del Rocío, Sevilla, Spain 13 Hospital Clínico de Valencia, Valencia, Spain Article history: A B S T R A C T Received 22 March 2019 Persistent thrombocytopenia is a common complication after allogeneic hematopoietic stem cell transplantation Accepted 22 May 2019 (allo-SCT). Romiplostim and eltrombopag are the currently available thrombopoietin receptor agonists (TPO-RAs), and some studies with very small numbers of cases have reported their potential efficacy in the allo-SCT setting. Keywords: The present retrospective study evaluated the safety and efficacy of TPO-RAs in 86 patients with persistent throm- Thrombocytopenia bocytopenia after allo-HSCT. Sixteen patients (19%) had isolated thrombocytopenia (PT), and 71 (82%) had second- Allogeneic stem cell ary failure of platelet recovery (SFPR). TPO-RA therapy was started at a median of 127 days (range, 27 to 1177 transplantation days) after allo-SCT. The median initial and maximum administered doses were 50 mg/day (range, 25 to 150 mg/ Thrombopoietin receptor ago- day) and 75 mg/day (range, 25 to 150 mg/day), respectively, for eltrombopag and 1 mg/kg (range, 1 to 7 mg/kg) nists and 5 mg/kg (range, 1 to 10 mg/kg), respectively, for romiplostin. The median platelet count before initiation of TPO-RA therapy was 14,000/mL (range, 1000 to 57,000/mL). Platelet recovery to 50,000/mL without transfusion support was achieved in 72% of patients at a median time of 66 days (range, 2 to 247 days). Eighty-one percent of the patients had a decreased number of megakaryocytes before treatment, showing a slower response to therapy (P = .011). The median duration of treatment was 62 days (range, 7 to 700 days). Grade 3-4 adverse events (hepatic and asthenia) were observed in only 2% of the patients. At last follow-up, 81% of patients had discontinued TPO- RAs and maintained response, and 71% were alive. To our knowledge, this is the largest series analyzing the use of TPO-RAs after allo-SCT reported to date. Our results support the efficacy and safety in this new setting. Further prospective trials are needed to increase the level of evidence and to identify predictors of response. © 2019 American Society for Transplantation and Cellular Therapy. Published by Elsevier Inc. Financial disclosure: See Acknowledgments on page 1830. INTRODUCTION * Correspondence and reprint requests: Leyre Bento, MD, Hematology Persistent thrombocytopenia is a common complication after Department, Son Espases University Hospital, Ctra Valldemossa, 79, Planta +1, allogeneic hematopoietic stem cell transplantation (allo-SCT) and dulo I, 07010 Palma, Illes Balears, Spain. Mo can lead to increased morbidity and mortality [1]. Underlying E-mail address: leyre.bento@ssib.es (L. Bento). https://doi.org/10.1016/j.bbmt.2019.05.023 1083-8791/© 2019 American Society for Transplantation and Cellular Therapy. Published by Elsevier Inc. 1826 L. Bento et al. / Biol Blood Marrow Transplant 25 (2019) 1825 1831 mechanisms are usually multifactorial but are poorly known [2]. transfusion after achievement of a platelet count 50,000/mL without trans- Prolonged isolated thrombocytopenia (PT), defined as persistent fusion for 7 days post-SCT [5]. Patients with severe hemorrhagic complica- tions were included as well. thrombocytopenia (<20,000/mL) with normal cell count in the Some patients had a platelet count >20 000/mL before the initiation of other hematopoietic lines or requirement of platelet transfusion TPO-RA therapy as a result of platelet transfusion. Patients with thrombocy- within the first 60 days after allo-SCT, has been described in 5% topenia related to primary disease recurrence, detected by immunopheno- to 20% of cases [3,4]. It may be caused by antibody-mediated typing and/or monitoring of chimerism, and those with thrombotic microangiopathy were excluded. This retrospective study was approved by platelet destruction, splenic sequestration, or delayed production the Ethics Committee of Son Espases University Hospital. due to impaired megakaryocytic differentiation. For this study, efficacy of treatment was defined as platelet recovery to The phenomenon of secondary failure of platelet recovery 50,000/mL without transfusion for 7 consecutive days after initiation of (SFPR) has been defined by the Seattle group as a decline in plate- TPO-RA treatment. Time to response, freedom from platelet transfusion, pre- let counts to <20,000/mL for 7 consecutive days or requirement dictors of response, and rate of successful taper of TPO-RAs without recur- rence of thrombocytopenia after successful platelet recovery were analyzed. for transfusion after achieving a platelet count 50,000/mL with- Adverse events were graded using the National Cancer Institute’s Common out transfusion for 7 days post-SCT. It occurs in an estimated Toxicity Criteria, version 4.0. Clot sections stained with hematoxylin and »20% of allo-SCT recipients [5]. Risk factors for the development eosin were used to assess the number of bone marrow megakaryocytes. The of SFPR include hematopoietic stem cell (HSC) dose, donor- number of megakaryocytes was considered normal at 1 per 1 to 3 low-power fields and decreased at 1 per 5 to 10 low-power fields. recipient HLA disparity, conditioning regimen intensity, immuno- suppression, infections, myelotoxic drugs, graft-versus-host Statistical Analysis disease (GVHD), and other immunologic processes. Treatment Descriptive statistics are used to present the patients’ general characteris- options are not well defined and based mostly on platelet trans- tics. Variables following a binomial distribution (ie, response rate, RR) are expressed as frequency and percentage. Comparisons between qualitative fusion. However, transfusion support is associated with several variables were done using Fisher’s exact test or the chi-square test. Compari- adverse events, including infusion reactions, platelet refractori- sons between quantitative and qualitative variables were performed using ness, acute lung injury, cardiac failure due to volume overload, the nonparametric Mann-Whitney Uor Kruskal-Wallis test. Time-to-event and viral transmission, which can impose a heavy financial bur- variables were measured from the date of therapy onset and estimated den. A boost of CD34 cells has also been proposed for patients with poor grant function, many of whom present with severe Table 1 thrombocytopenia, but this strategy is not always available and Patient and Allo-SCT Characteristics carries potential risks. Thus, it is important to identify new strate- Characteristic Value gies to manage this important post-SCT complication. Number of patients 86 Romiplostim and eltrombopag are currently available throm- Age, yr, median (range) 53 (8-74) bopoietin receptor agonists (TPO-RAs) that stimulate platelet Sex, male/female, n (%) 50 (58)/36 (42) production. Eltrombopag is approved for the treatment of refrac- Disease, n (%) tory immune thrombocytopenia [6,7] and thrombocytopenia Myelodysplastic syndrome 6 (7) secondary to hepatitis C infection [8]. Recently, it has been asso- Acute myelogenous leukemia 33 (38) ciated with multilineage responses in some patients with refrac- Acute lymphoblastic leukemia 12 (14) tory severe aplastic anemia [9], supporting the idea that it may Non-Hodgkin lymphoma 16 (19) directly stimulate the few surviving HSCs [10]. Romiplostim is Hodgkin lymphoma 4 (5) also approved for treating refractory chronic immune thrombo- cytopenic purpura [11]. Both TPO-RAs have shown promising Aplastic anemia 5 (5) activity in the treatment of myelodysplastic syndrome-related Others 11 (13) thrombocytopenia [12,13]. Donor type, n (%) Given the success of TPO-RAs in several scenarios, the use HLA-identical sibling 21 (24) of these agents in the post-transplantation setting for persis- Unrelated 28 (33) tent thrombocytopenia has emerged. Some studies with small Haploidentical 32 (37) numbers of cases have reported encouraging results in allo- Cord blood/haploidentical cord 5 (6) SCT recipients. For this reason, the present study aimed to ana- Pre-SCT status, n (%) lyze the efficacy and safety of TPO-RAs in the largest series of Complete remission 55 (64) patients with severe and persistent thrombocytopenia after Partial remission 9 (10) allo-SCT reported to date. Active disease 22 (26) CD34+ cell dose, cells £ 106/kg, median (range) 5 (0.6-11.8) METHODS Stem cell source, n (%) Thirteen allo-SCT units from across Spain participated in this study. The Peripheral blood 74 (86) median number of patients included from each center was 5 (range, 1 to 19). Bone marrow 6 (7) Pediatric (age <18 years) and adult patients who underwent an allo-SCT between 2009 and 2017 and received TPO-RA therapy (eltrombopag or romi- Cord blood 6 (7) plostim) as compassionate use for severe thrombocytopenia were included. Conditioning regimen, n (%) These patients were retrospectively identified using both the transplantation Myeloablative 34 (39) database of each center and information from the Pharmacy Department, to avoid selection bias. Their clinical data were entered into a centralized database. Reduced intensity 52 (60) TPO-RA therapy was initiated according to the clinical criteria at each GVHD prophylaxis, n (%) center. Platelets were transfused in accordance with institutional protocols; Methotrexate + cyclosporine or tacrolimus 13 (15) clinically stable and afebrile patients received a transfusion when their plate- let count dropped to 10,000/mL, whereas patients with fever or hemorrhage Mycophenolate mofetil + cyclosporine 10 (12) and unstable patients received a transfusion when their platelet count or tacrolimus reached <20,000/mL. Tacrolimus + sirolimus 9 (10) PT was defined as the engraftment of all peripheral blood cell lines but Post-SCT cyclophosphamide 35 (41) with a platelet count <20,000/mL for 7 consecutive days or the need for trans- fusion within the first 60 days after allo-SCT [4]. SFPR was defined as a decline Antithymocyte globulin/alemtuzumab-based 19 (22) in platelet count to <20,000/mL for 7 consecutive days or the need for prophylaxis L. Bento et al. / Biol Blood Marrow Transplant 25 (2019) 1825 1831 1827 according to the Kaplan-Meier method. Comparisons between the variables myelotoxic drugs at therapeutic doses (19% with valganciclovir of interest were performed using the log-rank test. All reported Pvalues are or ganciclovir, 50% with valganciclovir or ganciclovir and sulfa- 2-sided, and statistical significance was defined at P < .05. methoxazole-trimethoprim, and 31% with other drugs). The median duration of follow-up after initiation of TPO-RA therapy RESULTS was 10 months (range, 1 to 59 months). Patient Characteristics Baseline patient and transplantation characteristics of the 86 Treatment Characteristics cases included in this study are presented in Table 1. The The median platelet count before initiation of TPO-RAs was median patient age was 53 years (range, 8 to 74 years). Indica- 14,000/mL (range, 1000 to 57,000 /mL), median neutrophil tions for allo-SCT were acute myelogenous leukemia in 33 count was 1740/mL (range, 0 to 13,900/mL), and median hemo- patients (38%), myelodysplastic syndrome (n = 7; 7%), acute globin concentration was 9.7 g/dL (range, 6.4 to 13.5 g/dL). lymphoblastic leukemia in 12 patients (14%), non-Hodgkin lym- Eltrombopag was used in 51 patients (59%), and romiplostim phoma in 16 patients (19%), Hodgkin lymphoma in 4 patients was used in 35 patients (41%). The median starting and maxi- (5%), aplastic anemia in 5 patients (5%), and other hematologic mum doses for eltrombopag were 50 mg/day (range, 25 to 150 malignancies in 11 patients (13%). The median CD34+ cell dose mg/day) and 75 mg/day (range, 25 to 150 mg/day), respec- was 5 £ 106/kg (range, .6 to 11.8 £ 106/kg). Allo-SCT was per- tively and the median starting and maximum doses for romi- formed using a related donor in 21 patients (24%), an unrelated plostim were 1 mg/kg/week (range, 1 to 7 mg/kg/week) and 5 donor in 28 patients (33%), a haploidentical donor in 32 patients mg/kg/week (range, 1 to 10 mg/kg/week). TPO-RAs were started (37%), and cord blood or a haplo-cord donor in 5 patients (6%). at a median of 127 days (range, 27 to 1177 days) after allo-SCT. Fifty-five patients (64%) were in complete remission before allo- There were no differences between the 2 TPO-RAs at the time SCT, 9 (10%) were in partial remission, and 22 (26%) had active of initiation (P = .41), and no differences by type of thrombocy- disease. Stem cells were collected from peripheral blood in 74 topenia at the time of initiation, 78 days for PT and 155 days patients (86%), from bone marrow in 6 patients (7%) and from for SFPR (P = .37). The median time from a platelet count of cord blood in 6 patients (7%). Sixteen patients (19%) had PT and <20,000/m/L to the start of TPO-RAs was 32 days (range, 0 to 71 (82%) had SFPR. Before starting TPO-RA therapy, 7% of the 1016 days). Eighteen patients (21%) were previously treated patients had acute GVHD, 18% had cytomegalovirus infection, with cell infusion (67% with mesenchymal cells and 33% with 25% had other infections, and 25% of the patients had received CD34+cell boost). In addition, 78% of the patients with Figure 1. Overall survival of all series depending on TPO-RAs response. 1828 L. Bento et al. / Biol Blood Marrow Transplant 25 (2019) 1825 1831 Figure 2. Time to response with 20,000/mL platelet recovery depending on number of megacaryocytes before starting TPO-RAs. concomitant neutropenia had received granulocyte colony- 97 days (range, 7 to 700 days) (P = .001). At the last follow-up, stimulating factor before starting TPO-RA treatment, with 74% 63% of the patients had a neutrophil count 50,000/mL, and showing neutrophil responses, and 52% of the patients with 81% discontinued TPO-RAs, maintaining response. Sixty-one anemia had received erythropoietin (EPO) before starting TPO- patients (71%) were alive. Mortality was significantly lower in Ras, with 62% showing hemoglobin responses. responders to TPO-RAs compared with nonresponders (15% versus 53%; P < .001) (Figure 1). Causes of death were disease Efficay of TPO-RAs progression in 28%, infections in 48%, GVHD in 16%, and other The overall response rate (ORR) for platelet recovery causes in 8%. 50,000/mL was 72%, including 73% in the SFPR group and 67% in the PT group (P not significant). The response was achieved at a median time of 66 days (range, 2 to 247 days) after TPO-RA Predictors of Response initiation. The patients with PT required more time to response No differences in response rate were observed based on compared with those with SFPR (median, 93 days [range, 8 to type of thrombocytopenia (P not significant). Donor type and 217 days] versus 60 days [range, 2 to 247 days]). Also, the GVHD prophylaxis did not influence in response (P not signifi- median time from initiation of TPO-RAs to the last transfusion cant). Likewise, age and previous response to other growth fac- was 37 days (range, 0 to 298 days), including 33 days (range, 0 tors, such as EPO or granulocyte colony-stimulating factor, did to 298 days) for the SFPR group and 69 days (range, 14 to 188 not affect response. Seventy patients (81%) had a decreased days) for the PT group. On the other hand, 22 patients (25%) had number of megakaryocytes before treatment and showed a a neutrophil count <1000/mL before initiation of TPO-RAs, and slower response to therapy, with a median time to achieving a 17 (77%) achieved a count 1000/mL after therapy. Regarding platelet count >20,000/mL of 43 days (range, 25 to 61 days) erythroid lineage, 10 patients (12%) had a hemoglobin concen- versus 28 days (range, 3 to 53 days) in patients with a normal tration <8 g/dL before TPO-RA therapy, and 5 (50%) achieved number of megakaryocytes (P = .019) (Figure 2). No difference 8 g/dL after therapy. in the median time to platelet recovery of 50,000/mL was The median duration of TPO-RA therapy was 62 days observed between patients with a normal number of megakar- (range, 7 to 700 days). No differences in duration of treatment yocytes and those with decreased megakaryocytes in the total by type of thrombocytopenia were observed. However, the series (P not significant); however, in the SFPR group, faster median duration was shorter in nonresponders than in responses were observed in patients with a normal number of responder-patients: 32 days (range, 7 to 299 days) versus megakaryocytes (P = .016) (Figure 3). L. Bento et al. / Biol Blood Marrow Transplant 25 (2019) 1825 1831 1829 Figure 3. Time to response with 50,000/mL platelet recovery depending on number of megacaryocytes in the SFPR group. Adverse Effects after TPO-RA Treatment received bone marrow instead of peripheral blood, whereas in The TPO-RAs were well tolerated, and no patients discontin- our study, only 7% of the patients received bone marrow. These ued treatment because of adverse events. Grade 3-4 liver abnor- differences might have had some effect on the differing malities and fatigue were observed in only 2% of patients. No response rate in the 2 studies. Although retrospective data are patients developed thrombosis or other grade 3-4 toxicities, at risk of bias, and we agree that the best design is a prospec- and no patients relapsed while receiving TPO-RA treatment. tive study, we included all consecutive patients from each cen- ter who had received TPO-RAs after allo-SCT in an attempt to DISCUSSION avoid any kind of selection bias. In this study, the largest series analyzing the use of TPO-RAs Fu et al [22] recently reported the experience of eltrombo- after allo-SCT reported to date, our results support the efficacy pag in 38 patients after haploidentical SCT. Eight patients had and safety in this setting with an overall response rate (ORR) of delayed platelet engraftment, 15 patients had SFPR, and 15 72% and few side effects. In addition, 81% of patients managed patients had poor graft function. Twenty-four patients to discontinue TPO-RA treatment. responded to eltrombopag treatment, and the cumulative inci- To date, 17 retrospective studies analyzed the potential dence of overall response was 63.2%, similar to our results. benefits of TPO-RAs in this setting, most of them with a small Similar results with a low number of cases (n = 13) were number of cases (Table 2) [14 30]. In addition, there are cur- recently published by Yuan et al [21], with an ORR of 62%. rently 7 registered ongoing clinical trials with similar scenarios However, Fu et al reported a shorter median time to response (www.clinicaltrials.gov). Data from clinical trials are not avail- of 17 days (range, 2 to 89 days), compared with 66 days (range, able, with the exception of a preliminary report from a phase II 2 to 247 days) in our cohort. Our data also confirm that PT placebo-controlled randomized trial [31]. In that study, which required more time to response compared with SFPR, 93 days enrolled 53 allo-SCT recipients and 7 autologous SCT recipients (range, 8 to 217 days) versus 60 days (range, 2 to 247 days). with persistent thrombocytopenia or neutropenia treated with During treatment, 5 patients (13.2%) developed liver injury, eltrombopag, the response rate in the experimental arm was with elevated transaminases >2.5 times normal values or bili- 36%, but results were statistically inconclusive in terms of rubin twice the normal level, but no patient discontinued superiority compared with the 28% response rate in the control eltrombopag because of adverse events or intolerability, the arm. That randomized study differs from ours in several areas. same as in our case series. First, it included 7 patients who received an autograft, whereas Hartranft et al [16] published the largest study on the use of ours included only allografts. In addition, 38% of their patients romiplostim, comprising 13 patients with SFPR and PT, until 1830 L. Bento et al. / Biol Blood Marrow Transplant 25 (2019) 1825 1831 Table 2 Experience with TPO-RAs for Persistent Thrombocytopenia in the Allo-SCT Setting Thrombocytopenia Type N TPO-RAs Transfusion Independence Response Rate, Reference Platelets 50 £ 109/L, n/N (%) SFPR 1 Romiplostim Yes 1/1 (100) Beck et al, 2010 [14] SFPR 7 Romiplostim Yes 7/7 (100) Calmettes et al, 2011 [15] PT 1 Eltrombopag Yes NR Reid et al, 2012 [23] SFPR 1 Romiplostim Yes 1/1 (100) Bollag et al, 2012 [24] PT, SFPR 3 Romiplostim Yes 3/3 (100) Poon et a, 2013 [25] SFPR 1 Romiplostim No NR DeRemer et al, 2013 [26] SFPR 1 Romiplostim Yes 1/1 (100) Buchbinder et al, 2015 [27] PT 1 Eltrombopag Yes 1/1 (100) Fujimi et al, 2015 [28] SFPR 7 Romiplostim Yes (n = 6) 6/7 (86) Maximova et al, 2015 [29] SFPR 3 Romiplostim Yes 3/3 (100) Battipaglia et al, 2015 [30] PT, SFPR 13 Romiplostim Yes (n = 7) 7/13 (54) Hartranft et al, 2015 [16] PT, SFPR 12 Eltrombopag Yes (n = 9) 9/12 (75) Tanaka et al, 2016 [17] PT, SFPR 20 Eltrombopag Yes (n = 12) 12/20 (60) Bosch-Vilaseca et al, 2018 [18] Romiplostim PT, SFPR 14 Eltrombopag Yes 8/14 (57) Rivera et al, 2108 [19] PT, SFPR 13 Eltrombopag Yes 7/13 (54) Marotta et al, 2018 [20] PT, SFPR 13 Eltrombopag Yes (n = 8) 8/13 (62) Yuan et al, 2019 [21] PT, SFPR 38 Eltrombopag Yes (n = 24) 24/38 (63) Fu et al, 2019 [22] now. In that case series, 54% of patients achieved the primary Death rate was significantly lower in responders to TPO-RAs, endpoint of a platelet count 50,000/mL at a median of 35 days 15% versus 53% in nonresponders (P < .001). This confirms that (range, 14 to 56 days) after initiation of TPO-RAs. The time to refractory thrombocytopenia is an adverse prognostic factor for response was shorter than in our series, although the response survival in allo-SCT setting and likely a surrogate marker of rate was lower. defective hematopoiesis or an altered immune system. In our study, 81% of the patients had a decreased number of Nonetheless, the present study has several limitations. First, megakaryocytes before treatment and showed a slower this was a retrospective study, which may have influenced the response to therapy (median time to 20,000/mL platelets, uncertainty about effect size—response rate in this case. Sec- 43 days versus 28 days; P= .019). Tanaka et al [17] reported a ond, although this study had the largest number of patients of series of 12 patients treated with eltrombopag, with a faster any study reported to date, the cohort size is still small. Ulti- and higher rate of platelet recovery in patients with a normal mately, there was no untreated control population against number of megakaryocytes before starting treatment com- which to better measure the impact of TPO-RAs. pared with patients with decreased megakaryocytes. These In conclusion, to our knowledge this is the largest series results suggest that the number of megakaryocytes in bone analyzing the use of TPO-RAs after allo-SCT reported to date. marrow may better predict the response to TPO-RAs than type Our results support the efficacy and safety in this new setting of thrombocytopenia after allo-SCT. with an ORR of 72% and few side effects. Further prospective Eltrombopag induces differentiation of CD34+ hemato- trials are needed to increase the level of evidence and to iden- poietic precursor cells into committed CD41+ megakaryo- tify predictors of response. cyte progenitor cells and stimulates the proliferation of megakaryocyte progenitor cells [32]. On the other hand, ACKNOWLEDGMENTS this synthetic small-molecule thrombopoietin agonist stim- Financial disclosure: The authors have nothing to disclose. ulates c-MPL receptors and can improve hematopoiesis at Conflict of interest statement: L.B. has received honoraria for the level of primitive cells (platelet, erythroid and neutro- serving as a speaker or on advisory boards for Janssen, Amgen, phil lineages) [9,33,34]. In contrast to eltrombopag, the use and Roche. F.F.A. has served as a consultant for Celgene. I.E. has of which has been shown to induce a multilineage response received honoraria for serving as a speaker or on advisory boards in refractory aplastic anemia [9,35], there are little pub- for Astellas, Pfizer, MSD, Janssen, and Jazz Pharmaceuticals. D.V. lished data on the effect of romiplostim in multilineage has served as a consultant or advisor for Celgene, Amgen, Glax- responses [36]. Scant data are available to explain such dif- oSmithKline, Novartis, Takeda, and Pfizer and has served on ferences, since both agonists activate the same molecular speakers’ bureaus for Celgene, Novartis, Amgen, GlaxoSmithK- pathways. In our study, 22 of 86 patients (25%) had <1000/ line, Astellas, and Pfizer. A.G. has recived consultancy and mL neutrophils before TPO-RA therapy, and 17 of 22 (77%) research funding from Roche and Janssen. The other authors achieved 1000/mL after therapy. Regarding erythroid line- have no conflicts of interest to disclosure. age, 10 patients (12%) had hemoglobin <8 g/dL before TPO- Authorship statement: L.B. performed the research. L.B., J.M. RA therapy and 5 (50%) achieved 8 g/dL after therapy. B., I.G.C., E.G.T., D.R., A.B.V., C.D.M., M.E.M.M., F.F.A., E.R., A.C., L. However, in all these patients, we could not rule out the Y., T.Z., C.P.V., and I.E. contributed clinical data. L.B., J.M.B., I.G. possibility that the hematologic response was due at least in C., and A.G. contributed to data analysis and interpretation. A. part to some improvement or resolution of the possible G. served as the project statistician. All authors contributed to causes underlying the development of cytopenia as infec- the review, provided their comments on this manuscript, and tions or GVHD. approval the final version. L. Bento et al. / Biol Blood Marrow Transplant 25 (2019) 1825 1831 1831 REFERENCES 20. Marotta S, Marano L, Ricci P, et al. Eltrombopag for post-transplant cyto- 1. Kuzmina Z, Eder S, Bo € hm A, et al. Significantly worse survival of patients penias due to poor graft function. Bone Marrow Transplant. 2019;54: with NIH-defined chronic graft-versus-host disease and thrombocytope- 1346–1353. nia or progressive onset type: results of a prospective study. Leukemia. 21. Yuan C, Boyd AM, Nelson J, et al. Eltrombopag for the treatment of throm- 2012;26:746–756. bocytopenia post-allogeneic stem cell transplantation. Biol Blood Marrow 2. Akahoshi Y, Kanda J, Gomyo A, et al. Risk factors and impact of secondary Transplant. 2019;25:1320–1324. failure of platelet recovery after allogeneic stem cell transplantation. Biol 22. Fu H, Zhang X, Han T, et al. Eltrombopag is an effective and safe therapy Blood Marrow Transplant. 2016;22:1678–1683. for refractory thrombocytopenia after haploidentical hematopoietic stem 3. Bielski M, Yomtovian R, Lazarus HM, Rosenthal N. Prolonged isolated cell transplantation. Bone Marrow Transplant. 2019;54:1310–1318. thrombocytopenia after hematopoietic stem cell transplantation: mor- 23. Reid R, Bennett JM, Becker M, et al. Use of eltrombopag, a thrombopoietin phologic correlation. Bone Marrow Transplant. 1998;22:1071–1076. receptor agonist, in post-transplantation thrombocytopenia. Am J Hema- 4. First LR, Smith BR, Lipton J, Nathan DG, Parkman R, Rappeport JM. Isolated tol. 2012;87:743–745. thrombocytopenia after allogeneic bone marrow transplantation: exis- 24. Bollag RJ, Sterett M, Reding MT, Key NS, Cohn CS, Ustun C. Response of tence of transient and chronic thrombocytopenic syndromes. Blood. complex immune-mediated thrombocytopenia to romiplostim in the set- 1985;65:368–374. ting of allogeneic stem cell transplantation for chronic myelogenous leu- 5. Bruno B, Gooley T, Sullivan KM, et al. Secondary failure of platelet recovery kemia. Eur J Haematol. 2012;89:361–364. after hematopoietic stem cell transplantation. Biol Blood Marrow Trans- 25. Poon LM, Di Stasi A, Popat U, Champlin RE, Ciurea SO. Romiplostim plant. 2001;7:154–162. for delayed platelet recovery and secondary thrombocytopenia fol- 6. Bussel JB, Cheng G, Saleh MN, et al. Eltrombopag for the treatment of chronic lowing allogeneic stem cell transplantation. Am J Blood Res. 2013;3: idiopathic thrombocytopenic purpura. N Engl J Med. 2007;357:2237–2247. 260–264. 7. Gonzalez-Porras JR, Bastida JM. Eltrombopag in immune thrombocytope- 26. DeRemer DL, Katsanevas K, Bradley A, Awan FT. Romiplostim resistance nia: efficacy review and update on drug safety. Ther Adv Drug Saf. in secondary failure of platelet recovery. J Oncol Pharm Pract. 2013; 2018;9:263–285. 19:369–372. 8. McHutchison JG, Dusheiko G, Shiffman ML, et al. Eltrombopag for throm- 27. Buchbinder D, Hsieh L, Mahajerin A, Puthenveetil G, Soni A, Nugent D. bocytopenia in patients with cirrhosis associated with hepatitis C. N Engl J Successful treatment of secondary graft failure following unrelated cord Med. 2007;357:2227–2236. blood transplant with hematopoietic growth factors in a pediatric patient 9. Olnes MJ, Scheinberg P, Calvo KR, et al. Eltrombopag and improved hema- with Fanconi anemia. Pediatr Transplant. 2015;19:E181–E184. topoiesis in refractory aplastic anemia. N Engl J Med. 2012;367:11–19. 28. Fujimi A, Kamihara Y, Hashimoto A, et al. Identification of anti-thrombo- 10. Marsh JC, Mufti GJ. Eltrombopag: a stem cell cookie? Blood. poietin receptor antibody in prolonged thrombocytopenia after allogeneic 2014;123:1774–1775. hematopoietic stem cell transplantation treated successfully with eltrom- 11. Kuter DJ, Bussel JB, Lyons RM, et al. Efficacy of romiplostim in patients bopag. Int J Hematol. 2015;102:471–476. with chronic immune thrombocytopenic purpura: a double-blind rando- 29. Maximova N, Zanon D, Rovere F, Maestro A, Schillani G, Paparazzo R. mised controlled trial. Lancet. 2008;371:395–403. Romiplostim for secondary thrombocytopenia following allogeneic 12. Kantarjian HM, Giles FJ, Greenberg PL, et al. Phase 2 study of romiplostim stem cell transplantation in children. Int J Hematol. 2015;102:626– in patients with low- or intermediate-risk myelodysplastic syndrome 632. receiving azacitidine therapy. Blood. 2010;116:3163–3170. 30. Battipaglia G, Ruggeri A, Brissot E, et al. Safety and feasibility of romi- 13. Mittelman M, Platzbecker U, Afanasyev B, et al. Eltrombopag for advanced plostim treatment for patients with persistent thrombocytopenia after myelodysplastic syndromes or acute myeloid leukaemia and severe allogeneic stem cell transplantation. Bone Marrow Transplant. 2015; thrombocytopenia (ASPIRE): a randomised, placebo-controlled, phase 2 50:1574–1577. trial. Lancet Haematol. 2018;5:e34–e43. 31. Popat UR, Ray G, Bassett Jr RL, et al. Eltrombopag for post-transplant 14. Beck JC, Burke MJ, Tolar J. Response of refractory immune thrombocytope- thrombocytopenia: results of phase ii randomized double blind placebo nia after bone marrow transplantation to romiplostim. Pediatr Blood Can- controlled trial. Blood. 2015;126:738. cer. 2010;54:490–491. 32. Erickson-Miller CL, Delorme E, Tian SS, et al. Preclinical activity of eltrom- 15. Calmettes C, Vigouroux S, Tabrizi R, Milpied N. Romiplostim (AMG531, bopag (SB-497115), an oral, nonpeptide thrombopoietin receptor agonist. Nplate) for secondary failure of platelet recovery after allo-SCT. Bone Mar- Stem Cells. 2009;27:424–430. row Transplant. 2011;46:1587–1589. 33. Alexander WS, Roberts AW, Nicola NA, Li R, Metcalf D. Deficiencies in pro- 16. Hartranft ME, Clemmons AB, DeRemer DL, Kota V. Evaluation of romiplostim genitor cells of multiple hematopoietic lineages and defective megakaryo- for the treatment of secondary failure of platelet recovery among allogeneic cytopoiesis in mice lacking the thrombopoietic receptor c-Mpl. Blood. hematopoietic stem cell transplant patients. J Oncol Pharm Pract. 2015;0:1–8. 1996;87:2162–2170. 17. Tanaka T, Inamoto Y, Yamashita T, et al. Eltrombopag for treatment of 34. Alvarado LJ, Huntsman HD, Cheng H, et al. Eltrombopag maintains human thrombocytopenia after allogeneic hematopoietic cell transplantation. hematopoietic stem and progenitor cells under inflammatory conditions Biol Blood Marrow Transplant. 2016;22:919–924. mediated by IFNg. Blood. 2019;133:2043–2055. 18. Bosch-Vilaseca A, García-Cadenas I, Rolda n E, et al. Usefulness of thrombo- 35. Townsley DM, Scheinberg P, Winkler T, et al. Eltrombopag added to stan- poietin receptor agonists for persistent clinically relevant thrombocytopenia dard immunosuppression for aplastic anemia. N Engl J Med. 2017; after allogeneic stem cell transplantation. Eur J Haematol. 2018;101:407–414. 376:1540–1550. 19. Rivera D, Bastida JM, Lopez-Corral L, et al. Usefulness of eltrombopag for 36. Zhao LP, Sicre De Fontbrune F, Contejean A, et al. Nationwide survey in treating thrombocytopenia after allogeneic stem cell transplantation. France on the use of romiplostim in patients with refractory severe aplas- Bone Marrow Transplant. 2019;54:757–761. tic anemia. Bone Marrow Transplant. 2019;54:1161–1163.