Immunophenotypic analysis of T-acute lymphoblastic leukemia. A CD5-based ETP-ALL perspective of non-ETP T-ALL

T-cell antigens [CD5,CD1a,CD8] define early T-cell precursor acute lymphoblastic leukemia (ETP-ALL). To understand immature T-ALL of which ETP-ALL is part, we used these antigens to subcategorize non-ETP T-ALL for examining expression of myeloid/stem cell antigens (M/S) and clinical features. Using CD5 (+/-) to start categorization, we studied 69 routinely immunophenotyped patients with T-ALL. CD5(-) was a homogenous (CD8,CD1a)(-) M/S(+) ETP-ALL group (n = 9). CD5(+) cases were (CD8,CD1a)(-) pre-T-ALL (n = 22) or (CD8,CD1a)(+) (n = 38) thymic/cortical T-ALL; M/S(+) 20/22 (90.91%) in former and 22/38 (57.89%) in latter (P = 0.007). ETP- and pre-T-ALL together (CD1a(-) ,CD5(-/+) immature T-ALL group) were nearly always M/S(+) (29/31; 93.55%). In multivariate analysis, only ETP-ALL predicted poor overall survival (P = 0.02). We conclude (i) CD5 negativity in T-ALL almost always means ETP-ALL. CD1a and CD8 negativity, as much as CD5, marks immaturity in T-ALL, and the CD5(+/-) /CD1a(-) /CD8(-) immature T-ALL group needs further study to understand the biology of the T-ALL-myeloid interface. (ii) ETP-ALL patients may be pre-T-ALL if CD2(+) ; CD2(+) , conversely, CD5(-) /CD1a(-) /CD8(-) pre-T ALL patients are ETP-ALL. (iii) Immunophenotypic workup of T-ALL must not omit CD1a, CD5, CD8 and CD2, and positivity of antigens should preferably be defined as recommended for ETP-ALL, so that this entity can be better evaluated in future studies of immature T-ALL, a group to which ETP-ALL belongs. (iv) ETP-ALL has poor prognosis.