Acute myeloid leukemia (AML) induces profound impairment of healthy hematopoiesis. The production deficit in the bone marrow (BM) leads to development of peripheral anemia, thrombocytopenia and neutropenia, which is a major cause of AML-associated morbidity and mortality. Despite much progress in understanding of AML biology, the mechanisms by which AML blasts interact with elements of normal hematopoiesis to cause cytopenia are unclear.

Conventional wisdom has it that blasts infiltrate the marrow and displace normal hematopoiesis. If this concept were to be true, there should be a strong correlation between BM blast count and peripheral cytopenia. Surprisingly, analysis of 223 patients with newly diagnosed AML at a tertiary referral center revealed lack of correlation between initial BM blast count [% of cellularity] and hemoglobin level (ρ=-0.11, P=0.12), platelet count (ρ=-0.00, P=0.53) and absolute neutrophil count (ρ=0.13, P=0.06). This indicates that mechanisms other than displacement of normal hematopoiesis dictate the severity of cytopenia in AML patients.

Hematopoiesis is tightly regulated by cytokines. Among them, thrombopoietin (TPO) acts through its receptor c-Mpl as the master regulator of megakaryopoiesis, but also exerts upstream effects on hematopoietic stem and progenitor cells (HSPC). TPO levels are controlled by receptor-mediated scavenging by cells carrying c-Mpl on the surface, with platelets representing the lion's share in a healthy organism. This negative feedback loop results in strong negative correlation between serum TPO concentration and platelet count in the steady state. When we examined this relationship in our AML cohort, TPO levels did not follow the expected negative correlation with platelet counts (ρ=-0.10, P=0.59). Comparison with historic controls with thrombocytopenia induced by chemotherapy for non-hematopoietic malignancy revealed that the lack of correlation was driven by AML cases with severe thrombocytopenia that had lower than expected levels of TPO in the serum.

As HSPC are known to express c-Mpl, we hypothesized that HSPC-derived AML blasts may also express the receptor and cause insufficiency of hematopoiesis by means of receptor-mediated TPO scavenging. To test this hypothesis, we compared c-Mpl expression on blasts in AML cases with severe thrombocytopenia and low TPO concentration (potential scavenger cases) to cases with TPO levels adequate for the degree of cytopenia. Both surface flow cytometry and qPCR demonstrated higher c-Mpl expression in potential scavenger cases (3.1-fold, P=0.02).

To determine whether this difference in expression translates into increased serum TPO clearance, we incubated AML blasts with high (c-Mpl+) and low (c-Mpl-) receptor expression in serum containing recombinant human TPO at a concentration of 100 pg/mL. After 2h, TPO clearance reached 45 pg per 106 cells in wells with c-Mpl+ blasts, compared to only 4 pg per 106 cells in wells with c-Mpl- blasts (P=0.02). This confirms the hypothesis that AML blasts can lower TPO levels by virtue of their c-Mpl expression.

Validation studies in an independent, multi-center Dutch-Belgian-Swiss cohort of 437 AML cases confirmed lack of correlation between initial BM blast count and cytopenia. Ranked gene list correlation analysis of whole genome microarray data proved significant enrichment of the MPL transcript in patients with severe thrombocytopenia when compared to patients with average platelet counts (rank 27/20'589, FDR<10-6). MPL enrichment could also be observed in patients with severe neutropenia (P<0.01), but there was no correlation between MPL transcript level and degree of anemia.

Lastly, we asked if MPL expression was related to cytogenetic or molecular AML subtype: indeed, microarray analysis showed higher MPL expression in cases of AML with t(8;21) than in any other subtype (P<10-4). Concurrently, these patients displayed significantly lower platelet count (40 vs 83 x 109/L, P=0.02) when compared to all other AML cases.

In summary, our study demonstrates that cytopenia in AML is independent of BM blast count, but strongly correlated with c-Mpl expression on blasts. We show that c-Mpl+ blasts clear TPO, causing insufficient TPO levels and contributing to development of thrombocytopenia and neutropenia. The work may have important ramifications for treatment of AML-induced cytopenia, especially in the relapsed or refractory setting.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.