Introduction: Germline mutations in GATA2 lead to myelodysplastic syndromes (MDS) and bone marrow failure (BMF). Stem cell transplantation (SCT) remains the only curative therapy. In our previously described cohort of pediatric MDS/BMF patients with GATA2 mutation we noted a high transplant related mortality (TRM), a low relapse rate and frequent toxicities including thromboses. We hypothesized that analogous to other inherited syndromes the underlying GATA2 mutation may contribute to increased toxicity.

Methods: We performed a case control study of patients treated at our center in 2000-20014 comparing pediatric MDS/BMF patients with GATA2 mutations to MDS/BMF patients without the mutation (control A) and to patients with acute leukemia (ALL or AML) (control B). Controls were randomly selected and matched for age (0-10, 11-20), stem cell source (matched related donor (MRD), matched unrelated donor (MUD), mismatched unrelated donor (mmURD) or umbilical cord blood (UCB) and conditioning regimen.

We assessed if the proportion of patients who experienced a) high grade transplant related toxicity (TRT) including: infections, organ toxicities, secondary malignancies, thrombotic events; b) acute (grade II-IV) or extensive chronic GVHD (cGVHD); and, c) TRM within 100 and 365 days from transplant day, differed between the investigational and control cohorts. All pairwise comparisons of these factors were performed using a one-sided Fisher's Exact test.

We reported overall survival (OS), disease free survival (DFS) and event free-survival (EFS) (events: TRM, relapse, acute GVHD, cGVHD, secondary malignancy, thrombotic events using point estimates at 5 years post-transplant (± standard error).

Results: 80 patients were analyzed: 15 GATA2, 25 control A and 40 control B. The diagnoses in the GATA2 vs. the control A cohort were: RCC (n=8 vs 9), RAEB/ RAEB-T/AML-MRC (n=4 vs 11), AML (n=2 vs 0), BMF (n=0 vs 1), aplastic anemia (n=0 vs 4) and ALL (n=1 vs 0). Control B patients had AML (n=20) or ALL (n=20). Monosomy 7 was the most common cytogenetic abnormality in the GATA2 cohort [(n=9 (60%)] vs. 9 (36%) and 3 (8%) in Control A and B. The median follow-up for the study population was 3 years (range=0.14-12.1).

GATA2 patients were treated with the following conditioning regimens: cyclophosphamide (CY)/total body irradiation (TBI) +/-ATG (n=11), fludarabine/CY/TBI (n=2) and busulfan/CY (n=2) and transplanted using bone marrow from a MRD (n=3), MUD (n=5) and mmURD (n=3), or UCB (n=4). GVHD prophylaxis consisted of cyclosporine A (CSA)/MTX+/- prednisone (n=9) or CSA/ mycophenolate mofetil (n=3).

Thrombotic events were seen only in the GATA2 cohort (p< 0.001). For all other toxicities 10 (67%) occurred in the GATA2 group vs. 7 (28%) in control A (p=0.019), and 8 (20%) in control B (p=0.0018)]. Other comparisons were not statistically significant. Secondary malignancy was rare, but 1 (6.7%) was observed in the GATA2 cohort vs. 1 (4%) and 2 (5%) in control A or B. Severe lung disease requiring lung transplant was observed in 2 (13.3%) of the GATA2 patients and in 0 (0%) and 1 (2.5%) in control A and B, respectively. Acute and cGVHD were seen in 5 (33.3%) and 5 (33%) in the GATA2 group vs. 5 (20%) and 8 (32%) in control A and 10 (25%), 9 (22.5%) in control B. TRM within 100 and 365 days occurred in 2 (13%), 3 (20%) in the GATA2 cohort and 2 (8%), 4 (16%) in control A and 2 (5%), 7 (17.5%) in control B. TRM was the most common cause of death within 365 days in all groups. Relapse rate was not different between groups.

The 5-year OS in the GATA2 cohort was 65±13% compared to 58%±10% and 45%±8% for control A and B, respectively. The 5-year EFS were 7%±6%, 31%±11% and 33%±8% for cases, A, and B, respectively. The 5-year DFS was 51%±13% in the GATA2 group, 51% ±11% in control A, and 44%±8% in control B.

Conclusions: Pediatric MDS/BMF patients with GATA2 mutations had a significantly higher rate of TRT and thromboses compared to those without themutation or pre-treated leukemia patients. The 5-year EFS was lower in the GATA2 patients, but 5-year OS and DFS was comparable across the groups. Toxicities observed, especially frequent thromboses appear unique to the GATA2 cohort and warrant specific treatment considerations in the future.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.