Three types of diagnostic findings fulfill the definition of acute hybrid/biphenotypic leukemias (AHL): coexpression of multiple myeloid/lymphoid markers on lymphoid/myeloid population, coexistence of two or more distinct populations from different lineages and lineage switch prior achievement of complete remission. Aberrant expression of myeloid/lymphoid antigens is frequent among ALL and AML, respectively. A scoring system (EGIL) was developed for arbitrary distinction of “scored AHL”. Published data contain cohorts of adults or combined cohorts of children and adults. There is lack of data specifically on childhood AHL as well as on consecutive cohorts analyzed from an immunophenotypic, genetic and clinical perspective. From 09/1996 to 08/2006, immunophenotype, DNA index and BCR/ABL and MLL/AF4 fusion gene status of 898 leukemic samples of de novo (730 children), relapsed (154 samples from 122 children), and secondary AL (14 children) were centrally evaluated. In all cases classified either as “scored AHL” or with lineage switch we analyzed clonality in Ig/TCR genes and other fusions of MLL gene. De novo AL samples consisted of: 572 ALL, 118 AML, 32 “scored AHL” (28 primary ALL/My+, 4 primary AML/Ly+), 2 cases of ALL switched to AML and 6 samples not appraisable by FC (morphology ALL). Relapsed AL samples consisted of 113 ALL, 32 AML and 9 “scored AHL” (6 primary ALL/My+, 3 primary AML/Ly+). Secondary AL consisted of 3 ALL, 10 AML and 1 “scored AHL” (primary ALL/My+). AHL primarily classified as ALL cases fell into the following immuno-genetic subsets: T ALL(6), 1 hyperdiploid(1), TEL/AML1(9), BCR/ABL(3), and MLL/AF4(3); one T ALL was BCR/ABL+. The most common genetic fusion was TEL/AML1 (n.s.). The MLL/AF4 and BCR/ABL were more frequent in AHL cases (p=0.0098 and p=0.044, respectively). The incidence of hyperdiploidy was lower among AHL cases (p=0.026). Prognosis of AHL in non mature B ALL is significantly worse in AHL BCP ALL only (in 3.5 yrs follow up 84%+−1.9% in non AHL versus 61%+−14% in AHL). Separate analysis in genetic subsets revealed significant differences in relapse free survival of TEL/AML1 (AHL 41%+−22% versus non AHL 91%+−3% in 3.5 yrs of fUp, p=0.01) and MLL/AF4 subset (AHL 0%, non-hybrid ALL 85+−13% in 1st year, p=0.0082). The incidence and maturity of clonal Ig/TCR rearrangements generally reflected the immunophenotype of primary lineage, with the exception of 4/6 T ALL/My+ lacking clonal rearrangements. All 4 cases with AML/Ly+ co-expressed T lymphoid markers and no clonal Ig/TCR rearrangement was found in any of them. Among 4 patients with AML/Ly+, 3 relapses occurred in the 1st year of follow up.

Conclusion: Principal lineage was determined by FC in all AHL cases. We found significantly worse prognosis of patients classified as AHL (primary BCP ALL). The prognosis of AHL (AML/Ly) should be confirmed on larger number of patients.

Disclosure: No relevant conflicts of interest to declare.

Author notes

Supported by MSM0021620813, MZdNR8269-3/2005, MZdNR/9531-3, GA UK 7393/2007, GACR301/06P162.