I. Baumann et al., IMAGE-ANALYSIS DETECTS LINEAGE-SPECIFIC MORPHOLOGIC MARKERS IN LEUKEMIC BLAST CELLS, American journal of clinical pathology, 105(1), 1996, pp. 23-30
This report outlines the morphologic classification of acute myeloid (
AML: French-American-British FAB classification: M1) and lymphoid (ALL
) leukemia by automatic image analysis and the correlation to immunolo
gic and cytochemical classification, The investigation was carried out
on Romanowsky-Giemsa stained bone marrow (n = 15) and blood smears (n
= 10) from 25 patients with primary acute leukemia. The cases had bee
n classified as of myeloid or lymphoid origin by three hematologic cen
ters using immunochemistry or cytochemistry, but the specimens were su
bmitted to the authors' laboratory without the diagnosis. The nuclear
and cytoplasmic pattern of the blast cells were analyzed by a high res
olution image analysis system and the measured and calculated cell fea
tures were sorted by means of a classifier program (CART), The image a
nalysis classification was then compared with the immunophenotypical a
nd cytochemical classification. Blood blast cells showed nuclear featu
res that were significantly correlated to a myeloid or lymphoid immuno
phenotype. In contrast, bone marrow blast cells displayed overlapping
and therefore nondiscriminating nuclear features, However, by generati
ng a learning data set using the immunophenotypes the classifier progr
am found specific cytoplasmic features that eventually permitted a dif
ferentiation into myeloid or lymphoid subtypes, In summary, the author
s suggest that high resolution image analysis of leukemic blast cells
detects nuclear and cytoplasmic features that are associated with the
immunophenotype and therefore with the lineage determination of the ce
ll. With this new objective and reproducible approach of morphologic c
ell analysis, it might not only be possible to classify blast cells wi
th minimal cellular differentiation, but furthermore to discover progn
ostic features because the remarkable difference in classification qua
lity between blood and bone marrow blast cells reported in this study,
might be of biologic relevance and requires further investigation.