Immune aplastic anemia (SAA) disproportionally affects children and young adults. Immunosuppression (IST) remains the treatment of choice for patients less than 40 years of age without a fully human leukocyte antigen (HLA) matched sibling. In the pediatric population (aged <18 years), from NIH historic studies, horse antithymocyte globulin (h-ATG) and cyclosporine(CSA) regimes have produced good results, with an overall response rate (ORR) of 74% at 6 months, a relapse rate of 33% at 10 years, and an overall survival (OS) of 80% at 10 years (Scheinberg P, J Pediatrics 2008), generally better outcomes than in adults. Eltrombopag (EPAG) is an oral thrombopoietin-receptor agonist that has been shown to significantly improve response rates when combined with h-ATG and CSA for treatment-naïve patients (Townsley D, NEJM 2017). We report here a pediatric subgroup analysis of our clinical trial (NCT01623167) to assess whether the addition of EPAG to IST improved ORR in children with SAA.


Between 2012 and 2018 a total of 39 patients <18 years of age with SAA were treated with front line EPAG in combination with h-ATG and CSA. We assessed ORR at 6 months, relapse rate, rate of clonal evolution, and OS. Additionally, we monitored for the development of clonal evolution and for the emergence of somatic clones. Germline testing for inherited marrow failure was performed in all patients. Response was defined by blood counts no longer satisfying criteria for SAA. Outcomes were compared to two treatment-naive cohorts: adult patients aged 18 years and older who received EPAG in combination with immunosuppression (adult EPAG group), and a historical cohort (from 1989 to 2010) of 87 pediatric patients aged <18 years who were treated with immunosuppression (pediatric IST group). The pediatric IST group included all patients treated with a h-ATG based regimen including h-ATG with CSA, and h-ATG with CSA and additional immunosuppression (either mycophenolate mofetil or sirolimus). A historical adult group (Adult IST group) of 322 patients treated from 1989 to 2010 with the same IST regimens as the pediatric IST group was also used as a comparator. Analyses were based on intention-to-treat.


Median age in the pediatric EPAG group was 15 years and in the pediatric IST group 11 years. Thirty-two patients (of an initial 39) were evaluable for response at 6 months. Seven patients (18%) were taken off study prior to 6 months to undertake bone marrow transplant, due to either failure to respond by 3 months or clinical urgency in count recovery. Overall, 28 (72%) of all treated patients achieved a response at 6 months, by comparison, responses were observed in 74% of the pediatric IST group (p=0.836) and in 83% in the adult EPAG group (p=0.143). In contrast, patients in the adult EPAG group had a significant difference in 6 month ORR compared with the adult IST group (83% versus 58%, p= <0.0001). The PK analysis of EPAG suggested comparable exposure in pediatric (6-17 years) and adult population. Of the 28 patients who achieved a response at 6 months, 43% relapsed, compared to 28% in the pediatric IST group (p=0.252). Median time to relapse in the pediatric EPAG group was 565 days from IST. There was no significant difference in survival between the pediatric EPAG group and either the pediatric IST group or the adult EPAG group. Clonal evolution occurred in 3 (8%) of patients (monosomy 7, del 5q, and translocation 5:12); 7% of patients evolved in the pediatric IST group. Somatic mutation testing was performed in 36 patients at baseline and was negative in all. At the 6 month timepoint, 5 patients had developed new somatic mutations (2 ASXL1, 2 ATRX, 1 BCOR), three clones appeared in responders and two in non-responders. Germline testing did not reveal any pathogenic variants.


Eltrombopag added to immunosuppression did not improve ORR at 6 months in pediatric patients with SAA compared to our historical cohort of pediatric IST patients. In contrast, adults had much improved responses with EPAG. Some patients were taken off study prior to 6 month evaluation (18% compared to 6% in the pediatric IST group) to pursue BMT, likely reflecting recent improvements in outcomes and donor availability. In our study the addition of eltrombopag to IST did not provide any obvious therapeutic benefit to pediatric patients with severe aplastic anemia.

This research was supported in part by the Intramural Research Program of the NIH and NHLBI.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract

Sign in via your Institution