MIC2 is characteristically expressed in lymphoblastic lesions and Ewing's/p
rimitive neuroectodermal tumor sarcomas. Although MIC2 has recently been re
ported in chloroma and rare terminal deoxynucleotidyl transferase-positive
acute myelogenous leukemia (AML), the incidence and the significance of MIC
2 (CD99) immunoreactivity in myeloid lesions is not clear. In this study, w
e evaluated MIC2 positivity in a variety of myeloid diseases and normal mar
row to determine its incidence and distribution in myeloid diseases; its co
rrelation with dow cytometric and cytogenetic data in AML; and its associat
ion with leukemic transformation, relapse, and chloroma formation Paraffin
sections of 11 chloromas and 94 bone marrow core biopsies from 66 patients
were stained with CD99 monoclonal antibody 12E7. Of 94 bone marrow core bio
psies, there were 30 AML, (fragment antigen binding M0 to M6), 23 remission
s, 5 relapses, 12 myeloproliferative disorders, 13 myelodysplastic syndrome
s, and 11 normal marrows from patients who did not have leukemia CD99 immun
oreactivity was evaluated with light microscopy. MIC2 expression was seen i
n leukemic blasts in 6 of 11 chloromas (55%) and 13 of 30 AML (43%) but rar
ely in myeloproliferative disorders, myelodysplastic syndromes, remission,
and normal marrow CD99 tended to be positive in M1-, M3-, and HLA-Dr-negati
ve AML and negative in AML with relapse. MICE expression did not correlate
with the karyotype independent of French-American-British Cooperative Group
classification and the disease remission or occurrence of chloroma in AML.
We concluded that MIC2 is commonly expressed in leukemic blasts of AML and
is not predictive of leukemic transformation from myeloproliferative disor
ders and myelodysplastic syndromes or chloroma formation. Caution should be
taken when using MIC2 as a marker for Ewing's sarcoma/ primitive neuroecto
dermal tumor or lymphoblastic lymphoma on paraffin sections of either soft
tissue or bone marrow specimens.