Peripheral Blood Blast Erythrophagocytosis After Chemotherapy in a Patient with Acute Monoblastic Leukemia

: Hemophagocytosis is a property of histiocytic lineage cells; it is rarely performed by leukemic blasts. This phenomenon happens in approximately 1% of acute leukemias, particularly acute myeloid leukemia (AML) and especially those of monoblastic or monocytic lineage. Associations have been found with some chromosomal abnormalities, especially t(8;16). The presence of blast hemophagocytosis may speed-up the diagnosis of some of these cytogenetic abnormalities. We aim to report a case of peripheral blood blast erythrophagocytosis after chemotherapy for the treatment of acute monoblastic leukemia. It is about 29 years old male treated for acute monoblastic leukemia without chromosomal abnormalities at the onset, with two relapses for the first and second induction cures. Peripheral blood smear performed during the follow up showed 73% of blasts, 3.6% of them with erythrophagocytosis. Several genetic abnormalities are known to be associated to leukemic cell hemophagocytosis, involving particularly the chromosome 8 and explaining a part of the leukemogenesis. These abnormalities are rare and associated to a poor prognosis. The mechanism behind this blast behavior is still unclear and authors suggest some hypothesis to explain this phenomenon: the presence of binding receptors involved in phagocytosis on blasts, karyotype instability and evolution during the course of the disease and the involvement of treatment toxicity.


Introduction
Hemophagocytosis is a property of histiocytic lineage cells; it is rarely performed by leukemic blasts. Blast hemophagocytosis must not be confounded with Malignancy related Hemophagocytic Lymphohistiocytosis (M-HLH) occurring generally after chemotherapy and requiring specific diagnosis and treatment.
Blast hemophagocytosis is found in about 1% of acute leukemia, particularly acute myeloid leukemia (AML) and especially those of a monoblastic or monocytic lineage. It may be associated or not to M-HLH [1]. This phenomenon is more rarely observed in lymphoblastic leukemia [2].
Our aim is to report a case of peripheral blood blast erythrophagocytosis after chemotherapy for the treatment of FAB M5a acute monoblastic leukemia without chromosomal abnormalities at the onset and to report the mechanisms trying to explain this phenomenon.

Case Report
It is about a previously healthy 29 years old male without medical antecedents except a chronic smoking, admitted to onco-hematology department for abdominal pain and vomiting since a month with fever and alteration of general condition (Table 1). Clinical examination showed a fever at 39°C, mucocutaneous pallor, conjunctival bleeding and tumoral syndrome made of splenomegaly, hepatomegaly and bilateral cervical, axial and inguinal lymphadenopathies. The first peripheral blood examination showed hyperleucocytosis at 90 G/l with 38% of blasts and normochromic normocytic anemia (Hemoglobin: 6,3 g/dl MCV: 100 fl MCHC: 37 pg) and thrombocytopenia with platelet count at 33 G/l. Bone marrow aspiration showed an infiltration with monoblastic cells HLA-DR+, CD34+, CD33+, CD11c+, CD 36+ and CD 14 +. B and T lymphoid markers were negative except an aberrant expression of CD5 and CD7. The karyotype was normal. The patient was diagnosed as acute monoblastic leukemia without chromosomal abnormalities (M5a) and he was conducted under chemotherapy. The patient relapsed to the first and second induction cures. Following bone marrow aspiration showed blasts infiltration with rare granulations and weak myeloperoxidase and same immunophenotyping found at the time of presentation. Peripheral blood smear showed the presence of 73% of blasts, 3.6% of them presented erythrophagocytic ( Figure 1). The blasts contained both single and double cytoplasmic vacuoles occuping a large part of the cytoplasm and mimiking the aspect of red blood cells. The second karyotype weren't performed since the patient lost to follow-up.

Discussion
To our knowledge, no case has been reported of blast hemophagocytosis in peripheral blood smear. In the bone morrow, this phenomenon may interest all types of hematopoietic cells; erythrocytes, platelets, leucocytes and even blasts [11].
Blast hemophagocytosis is frequently reported during monoblastic/monocytic acute leukemia, rarely in other types of acute myeloid leukemia as M0, M1, M2 and M7 and it is unusual in acute lymphoblastic leukemia [1-3, 9, 11, 12]. This phenomenon is not considered to distinguish between acute monoblastic leukemia and other type of leukemia. It may also be present in non-hematopoietic malignancies [13]. Blast hemophagocytosis is mostly reported at the time of the onset and is associated to abnormal karyotype [1,[3][4][5][6][7][8][9]11], it rarely happens after chemotherapy, as in our case, and seems to be associated to karyotype evolution in patients presenting with normal karyotype at the onset [1].
It remains important to us to understand the mechanism related to this phenomenon, its implication in cytogenetic diagnosis and prognosis.
Hemophagocytosis by leukemic cells or blasts must not be confounded with M-HLH, even if both phenomenon may be present in the same patient [1].
The pathogenic mechanism related to leukemic cells with phagocytic activity remains unclear. Phagocytosis is defined by the ingestion of particles following two steps; binding and ingestion. But receptors involved in binding, like Fc receptor, are not normally present in myeloblasts; Park et al suppose that more complex factors are involved in blast hemophagocytosis other than the presence or not of Fc receptors and Yeh et al postulate this may be related to the presence of C3 receptor, expressed in leukemic blasts of AML, which binds blast cells to erythrocytes known by expressing the receptor for C3b and leading to erythrophagocytosis [2,6].
Blast hemophagocytosis seems to be associated to some specific genetic abnormalities diagnosed at the onset or during the course of the disease ( Table 2) [3]. Translocation t(8;16) (p11;p13) is frequently reported in acute monoblastic leukemia with hemophagocytosis by leukemic cells and more rarely the translocations t(3;8;17) and t(3;8) (q27;q22) [1,4,9]. In fact, the chromosome 8 rearrangement in 8p11 region is frequently implicated in monocytic AML with hemophagocytosis and seems to play also a role in leukemogenesis [3]; abnormalities of MYST 3 gene situated in this region may improve myelomonocytic proliferation by increasing related growth factors as M-CSF and G-CSF, hemophagocytosis may be a consequence of cytokines and growth factors dysregulation as INFγ, M-CSF, GM-CSF [9,10].
Imashuku S and al suggest that the presence of blast hemophagocytosis may speed up the diagnosis of t(16;21) (p11;q22). They report 12 cases of AML with t(16;21) all with blast hemophagocytosis in the bone morrow [3]. Blieden et al suppose that the presence of monocytoid blast morphology with eryhtrophagocytosis suggested the presence of the t(8;16) in this AML [5]. Close association between this morphologic feature and cytogenetic abnormalities may be useful.
Complex karyotype has been also reported in acute myeloid leukemia with monocytoid blasts and erythrophagocytosis [5].
Very few cases of lymphoblastic hemophagocytosis have been reported [2,15], one case with the t(12;21) (p13;q22); TEL-AML1 (ETV6-RUNX1) [2]. This phenomenon may be explained by the presence of the Fc receptor in blasts of lymphoid lineage [2]. All these genetic abnormalities seem to be associated to poor response to therapy and poor prognosis [1,3,[10][11][12]. Hemophagocytosis by leukemic blast may be found at the onset of the disease or during the treatment like in our case, the apparition of cytogenetic abnormalities in a patient with a normal karyotype at the initial presentation may be related to high genetic instability and chemotherapy toxicity [1,16]. In fact, evolution to more complex karyotype has been shown in 53% of myeloid leukemia [16].
Many authors suggest the implication of the chemotherapy in the appearance of blast hemophagocytosis [1,16]; Farias et al suggest this phenomenon to be explained by the cytogenetic aberrations, sometimes complex, acquired after chemotherapy [1], For Foadi et al, prolonged chemotherapy expose cells to a variety of drugs and alterations and may contribute to the appearance of more aggressive clone of leukemic cells with phagocytic features in cases of lymphoid blast hemophagocytosis [17].
Unlike blast hemophagocytosis, M-HLH is generally reported after chemotherapy.
In the M-HLH, hemophagocytosis is performed by cells of histiocytic lineage and is largely explained by the immune dysregulation of T lymphocytes and natural killer or immune suppression caused in part by the treatment and resulting in the proinflammatory cytokines release and excessive activation of hemophagocytic monocytes, associated to fatal infections especially in patients who have already achieved remission [18,19]. It is a serious and life-threatening acute complication due to the treatment of different malignancies [19,20].
The HLH Study Group of the Histiocyte Society proposed diagnostic guidelines for HLH; five of this eight criteria must be fulfilled: fever, splenomegaly, cytopenia, hypertriglyceridemia and/or hypofibrinogenemia, hemophagocytosis, low or absent Narural killer cell activity, hyperferritinemia, and high soluble interleukin 2 receptor (s-IL2-receptor) [21]. Early diagnosis and immediate initiation of adequate treatment are mandatory to overcome this severe condition [20].
Unlike M-HLH, no specific treatment is administered to patients with hemophagocytic blasts in all reported cases, considering, indirectly, this phenomenon as a "normal" behavior of leukemic cells without specific treatment implications. Further explorations related to HLH are also not indicated for those patients.

Conclusion and Recommendations
Phagocytic activity in leukemic blasts is a very rare phenomenon, it may be closely linked to some cytogenetic abnormalities and explain a part of the leukemogenesis. Morphology examination is still relevant and may provide a useful orientation to this type of abnormalities. The mechanism behind this phenomenon is still unclear and few hypotheses have been proposed in the literature. To date, no specific explorations or treatment are dedicated to this morphologic feature.
Further investigations are necessary to determine the precise mechanism behind this behavior and its implications on the prognostic and treatment.