Thrombocytosis in transient abnormal myelopoiesis: Association with megakaryocytic skewing and GATA1 mutation patterns

Abstract Background & Objectives: Transient abnormal myelopoiesis (TAM) is a unique neonatal hematological disorder associated with Down syndrome (DS), characterized by clonal proliferation of GATA1-mutated blast cells. While thrombocytopenia is a hallmark of TAM, cases with thrombocytosis have also been reported, suggesting a more heterogeneous hematopoietic phenotype. However, the frequency, clinical implications, and biological basis of thrombocytosis in TAM remain poorly understood. To elucidate the significance of thrombocytosis in TAM, we analyzed data from the JPLSG TAM-10 cohort, a nationwide prospective registry of 167 neonates with TAM. We hypothesized that thrombocytosis reflects a distinct biological subset and aimed to explore its associations with clinical features, CD surface marker expression, cytokine profiles, GATA1 mutation patterns. GATA1 mutations in TAM result in exclusive expression of a truncated isoform, GATA1s, which lacks the N-terminal transactivation domain. We previously reported that GATA1 mutations can be classified into GATA1s-high and GATA1s-low expression types (Kanezaki et al., Blood 2010). We also conducted functional validation using a transgenic mouse model to investigate the role of Gata1s expression levels in platelet production. Methods: Patients were stratified by platelet count at diagnosis into four groups: extremely high (≥1,000×10⁹/L), high (450–999), normal (150–449), and low (<150). Morphological review, flow cytometric immunophenotyping, cytokine analysis (27-plex assay), and GATA1 sequencing were performed centrally. GATA1 mutations were categorized based on predicted short-form (GATA1s) expression: “GATA1s-high” and “GATA1s-low”. Transgenic mouse models expressing different Gata1s levels were analyzed to validate findings. Results: Among 167 TAM patients, 7 (4%) had extreme thrombocytosis, 33 (20%) high, 52 (31%) normal, and 75 (45%) low platelet counts. Thrombocytosis was associated with elevated WBC counts, peripheral blasts, ALT, and hepatomegaly (p < 0.01), but not with demographic features or leukemia progression. No thrombotic events were observed in any group, including those with platelet counts ≥1,000×10⁹/L, and only one patient received low-dose aspirin. Morphologically, giant platelets were observed in thrombocytosis cases. Flow cytometric analysis demonstrated increased expression of megakaryocytic markers, including CD41 and CD61, as well as c-kit (CD117), alongside reduced expression of the myeloid marker CD33 in higher platelet groups. UMAP analysis based on surface marker expression showed clustering tendency of thrombocytosis cases. Cytokine profiling revealed that inflammatory mediators such as MIP-1b, TNF-α, and PDGF-bb were significantly elevated in thrombocytosis groups. Notably, GATA1s-high mutations were strongly enriched among patients with thrombocytosis (p < 0.0001), and these mutations correlated with higher platelet counts. In transgenic mice, higher Gata1s expression led to significantly elevated platelet counts, supporting a causal relationship. Conclusions: Thrombocytosis in TAM defines a distinct clinical and biological phenotype, characterized by megakaryocytic skewing, inflammatory cytokine elevation, and GATA1s-high type mutations. Despite markedly elevated platelet counts, no thrombotic complications were observed, suggesting that TAM-associated thrombocytosis may reflect a benign, possibly functionally impaired, megakaryocytic response. Platelet count at diagnosis in TAM may reflect underlying disease heterogeneity, but elevated counts should be interpreted with caution to avoid unnecessary treatment for thrombocytosis. Importantly, transgenic mouse models expressing different levels of Gata1s confirmed that higher Gata1s expression leads to increased platelet production, supporting a causal role of GATA1 mutation pattern in TAM-associated thrombocytosis.