Leukaemia-associated immunophenotypes (LAIP) are observed in 90% of adult and childhood acute lymphoblastic leukaemia: detection in remission marrow predicts outcome
F. Griesinger et al., Leukaemia-associated immunophenotypes (LAIP) are observed in 90% of adult and childhood acute lymphoblastic leukaemia: detection in remission marrow predicts outcome, BR J HAEM, 105(1), 1999, pp. 241-255
Analysis of differentiation antigens on leukaemic blasts is routinely done
for diagnostic purposes, i.e. determination of stage of differentiation and
lineage assignment. Acute lymphoblastic leukaemias are also frequently cha
racterized by a leukaemia-associated immunophenotype (LAIP), either the coe
xpression of differentiation antigens physiologically restricted to other s
tages of differentiation (asynchronous LAIP) or cell lineages (aberrant LAI
P). We defined LAIP in 241 consecutive unselected B-lineage (n = 193) and T
-lineage (n = 48) ALL by three-colour now cytometry using directly conjugat
ed monoclonal antibodies. The incidence of LAIP was found to be 91.7%. In 6
3% of patients two to six leukaemia-associated expression patterns were det
ected, In order to study the specificity of LAIP in a therapy-relevant sett
ing, remission bone marrow samples from patients with B-lineage ALL were an
alysed for the expression of T-lineage-associated phenotypes on the normal
bone marrow cells and vice versa. The frequency of ail T-lineage LAIP(+) ce
lls and all aberrant B-lineage LAIP(+) cells was <1% in regenerating bone m
arrow samples at different timepoints. The incidence and clinical significa
nce of LAIP(+) cells was studied in 196 remission marrows of 70 ALL patient
s (55 remaining in CCR, 14 with bone marrow relapse, one with isolated CNS
relapse). The presence of >1% LAIP(+) at two consecutive timepoints predict
ed 5/8 bone marrow relapses in B-lineage ALL. The occurrence of LAIP(+) cel
ls >1% in T-lineage ALL after induction therapy predicted relapse in 7/7 ca
ses. In conclusion, flow cytometric detection of LAIP(+) cells appears to b
e a powerful tool for the prediction of outcome in ALL.