-
8/22/2019 M4 Eo_Articulo 1993
1/9
A c u te M y e l o id L e u k e m i a M 4 W i t h B o n e M a r
ro w E o s i n o p h i li a ( M 4 E o ) a ndinv( 16 ) (p1 3q22 ) E
xh ib i t s a Spec i f i c ImmunophenotypeW i t h C D 2 E x p r e s
s i o nBy Henk J . Adriaansen, Peter A.W. te Boekhorst, Anne M.
Hagemeijer, C. Ellen van der Schoot,
H. Ruud Delwel, and J acques J .M. van DongenExtensive
mmunologic marker analysis was performed tocharacterize the various
leukemic cell populations in eightpatients with inv(l6)(pl3q22)in
association with acutemyeloid leukemia with abnormal bone marrow
eosinophilia(AML-M4Eo). The eight AML cases consisted of
hetero-geneous cell populations; mainly due to the presence
ofmultiple subpopulations, which varied in size between
thepatients. However, the immunophenotype of these sub-populations
was comparable, independent of their relativesizes. Virtually all
AML-M~Eo ells were positive for thepan-myeloidmarker CDI3. In
addition, the AML were partlypositive for CD2, CDI 1b, CDI1 , CDI4,
CD33, CD34,CD36, CDw65, terminal deoxynucleotidyl transferase(TdT),
and HLA-DR. Double mmunofluorescence stainingsdemonstrated
coexpressionof the CD2 antigen and myeloidmarkers and allowed the
recognition of multiple AML sub-populations. The CD2 antigen was
expressed by immature
CUTE MYELOID LEUKEMIA (AML) is a phenotyp-A ically heterogeneous
disorder with a marked patient-to-patient variation and intraclonal
variation in mostpatients. The French-American-British (FAB) group
hasspecified different cytomorphological categories, such asAML-M4
with bone marrow eosinophilia (A M L- M ~E O ). ~~This FAB category
is highly correlated with abnormalities ofchromo some 16q22, eg,
del( 16)(q22), inv( 16)(pl3 q22), andt( 1 6 ; 1 6 ) ( ~ 1 3 ; q 2 2
) . ~ - ~he 16q22 abnormalities are found in6% of all AML
case^.'.^-'^ Larson et a1 suggested that thejuxtaposition of DNA
regions from 1 6 ~ 1 3 nd 16q22, asoccurs in inv( 16) an d t(
16;16), is probably required for theM4Eo phenotype? Only a few
patients with inv(l6) ort( 16;16) presented with another type of AM
L o r with myelo-d ysplas tic ~y n d r o m e . ~ - ~nitially it was
described that themetallothionein gene cluster was split by the
chromosome16 aberration in AM L-M4Eo.I6 However, recently this
genecomplex has been remapped to chromosome 16q13 and itwas not
found to be disrupted in AM L-M ~EO .In comparison to other types
of AML, patients with AML-M4Eo have a young median age,123L3ften
present with ahigh peripheral white blood cell (WBC ) count and
organo-megaly,4.8-LO213 nd generally have a high response rate to
in-duct ion ~ hemotherapy?~ ~*~- ~ n most studies, AM L- M ~E
opatients have a favorable p r o g n o ~ i s ~ ~ ~ ~ ~ ~ ~ ~ , ~
~lthough somereports do not support thi~.~3elapses in the c entral
nervoussystem (CNS) occur relatively frequently, generally
mani-festing as leptomeningeal disease and intracerebral myelo-
By use of cytomorphology, three different cell types canbe
recognized within each AM L-M ~E o,e, blast cells, mo no-cytic
cells, and eosinophil^.^^'^^^^ The eosinophils are oftendysplastic,
having abnormal eosinophilic granules, which areadmixed with
varying numbers of basophilic-staining gran-~les.4,5,8,9,1,13,21
They display aberrant positivity for chlo-roacetate esterase and
they often contain periodic acid-Schiff(PAS)-positivegranules. Wh
ether the eosinophilic cells and
blastomas.8, I , I3 ,I5 ,W
AML cells (CD34+, CD14-) and more mature monocyticAML cells
(CD34-, C DI4+),whereas TdT expression wasexclusively found in the
CD34+, CD14- cell population.The eight AML-M~Eo ases not only
expressed the CD2antigen, but also its ligand CD58 (leukocyte
function an-tigen-3). Culturing of AML-M~Eo ell samples showed
ahigh spontaneous proliferation in all three patients
tested.Addition of a mixture of CD2 antibodies against the T I1.I,T
I 1.2, and T I1.3epitopes diminished cell proliferation intwo
patients with high CD2 expression, but no inhibitoryeffects were
found inthe third patient with low frequencyand low density of CD2
expression. These results suggestthat high expressionof the CD2
molecule in AM L-M ~Eostimulates proliferation of the leukemic
cells, which mightexplain the high white blood cell count often
found in thistype of AML.0 993by The American Society of
Hematology.the monocytic cells in A M L-M ~E oepresent different
lineagesis not clear. Morphologically hybrid cells have been
describedwith nuclear folding and chromatin pattern
characteristicsof monocytes or promonocytes an d granules identical
to thosepresent in eosinophils. Immunologic m arker analysis,
es-pecially double immunofluorescence (IF) staining, is a pow-erful
tool to characterize subpopulations within phenotypi-cally
heterogeneous malignancies. Studies on markerexpression in A M L -M
~ Eo re scarce and only limited nu m-bers of markers have been
In this study, we performe d extensive imm unologic m
arkeranalysis to characterize the various subpop ulations in
eightcases of A M L- M ~E o ith inv( I6)(p 13q22). Various imm
a-ture and more mature subpopulations were detected in allcases.
Althoug h the proportion of each subpopulation variedfrom patient
to patient, the immun ophenotype of these sub-populations in all AM
L -M ~ E o ases was strikingly similar.A special finding in these
eight AML cases was the exp ressionof the T-cell marker CD 2 on a
portion of the leukemic cells.
From the Departments ofrm muno logy and Hematology ,
UniversityHospital Dijkzigt/Erasmus University, Rotterdam; the
Departmentof Cell Biology and Genetics, Erasm us University, Rotter
dam; theCentral Laboratory of the Netherlands R ed Cros s Blood Tra
nsfusionService and Laboratory or Experimental and Clinical
Immunology,University ofAm sterda m; and the Dr Daniel den Hoed
Cancer Center,Rotterdam, The Netherlands.Subm itted Septem ber 14,
199 2; accepted January 9, 1993.Address reprint requests to
H.J.Adriaansen, M D , PhD, Departmentof Immunology, Eras mus
University, PO B ox 1738, 3000 D R Ro t -terdam, The Netherlands.Th
e publication costs of th is article were defrayed in part by
pagecharge paym ent. This article must therefore be hereby
markedadvertisement in accordance with 18 U.S.C. section I73 4
solely toindicate this fact.0 993 by The American Society of
Hematology.0006-4971/93/81I 1 0025%3.00/0
3043lood, Vol81, No 11(Junel),993: pp3043-3051
-
8/22/2019 M4 Eo_Articulo 1993
2/9
3044 ADRIAANSEN ET AL
MATERIALS AND METHODSPatients and cytomorphology. Four children
(
-
8/22/2019 M4 Eo_Articulo 1993
3/9
AML-M4Eo WITH INV(16) EXHIBITS CD2+ PHENOTYPE 3045
analysis. We used the B-cell marker s CDlO (VIL-A1, Dr W. K napp
,Vienna, Aus tria), CD19 (L eu-12, Becton Dickinso n, San Jose,
CA),and CD20 (B l, Coulter Clone, Hialeah, FL); the T-cell markers
CD2(T 11 and TI 1 FITC, Coulter Clone; 6G4 [TI 1.1],4B2 [T 11.21,
andHIK27 [T11.3], Dr R. A. W. van Lier, Amsterdam, The
Netherlands),CD3 (Leu-4, Leu-4 FITC, and Leu-4 PE, Becton
Dickinson), CD4(Leu-3 PE, Becton D ickinson), CD7 (3A1, American
Type CultureCollection, Rockville, MD), and CD8 (Leu-2 PE, Becton D
ickinson);the myeloid monocytic markers CD13 (My7 and My7 PE,
CoulterClone), CD14 (My4, My4 FITC, and My4 PE, Cou lter Clone),
CD15(VIM-D5, Dr W. Knapp; Leu-MI FITC, Becton Dickinson), CD33(My9
, My9 FITC, and My9 PE, Coulter Clone), and CDw65 (VIM-2, Dr W.
Knapp; BMA-0210 FITC, Professor Dr W. Ax, BehringDiagnostica, M
arburg, Germany); the erythroid mark er glycophorinA (VIE-G4, Dr W.
Knapp); the platelet markers CD42a (FMC25,Dr H. Zola, Bedford Park,
Austra lia) and CD6 I (C17, Central Lab-oratory of the Blood
Transfusion Service, Amsterdam, The Neth-erlands); the
non-lineage-specific marke rsCD1 a(TB 133, Dr R.A.W.van Lier; CLB
LFA 1/2, Central Laboratory ofth e Blood Transfu sionService), CDl
lb (CLB mongran/l, Central Laboratory of the BloodTransfusio n
Service; Leu- 15 PE, Becton Dickinso n), CD 1 1c (Leu-M5, Becton
Dickinson), CD I8 (CLB54, Dr R.A.W. van Lier; CLBLFA 1/ , Central
Laboratory of the Blood Transfusion Service), CD34(BI-3C5, Seralab
, Crawley Down, U K; HPCA -2 FITC and HCPA-2PE, Becton Dickinson ;
CD34 FITC, Dr R. K urrle, Behring Diag-nostica), CD36 (OKM5, Ortho
Diagnostic Systems, Raritan, NJ),CD54 (BBL-4, British
Biotechnology, Oxford, UK), C D58 (T S2/9,Dr T. Schumacher,
Netherlands Cancer Institute, Amsterdam, TheNetherlands), and
HLA-DR (L243, L243 FITC, and L243 PE, BectonDickinso n). In case of
unconjugated MoAbs, we used an FITC -con-jugated goat anti-mouse
immunoglobulin (Ig) antiserum (CentralLaboratory of the Blood T
ransfusion Service) as a second-step reagent.Fluorescent labelings
of surface membrane antigens were measuredwith a FACScan flow
cytometer using FACScan-research software(Becton Dickinson). All
MNC as defined by forward and sidewardscatter patterns were gated;
only debris was excluded from analysis.For double -mem brane
stainings, only directly conjugated (FIT C orPE) MoAbs were used.
Expression of all m arkers tested in double IFstainingswas also
tested in single IF sta inings using direct or ind irec tFITC
labeling. As negative controls in all experim ents,we used
eitherunconjugated or isotype-matched unrelevant MoAbs, conjugated
withFITC or PE.The expression of terminal deoxynucleotidyl
ransferase (TdT) wasdetected as described by use of a rabbit a
nti-Td T antiserum and aHTC-conjugated goat anti-rab bit Ig
antiserum (Supertechs, Bethesda,MD).29Double IF staining for TdT
and several surface membraneantigens, ie, CD2 (T1 ), CD7 (3A1),
CDlO (VIL-A l), CD13 (My7),(BI-3C5), CDw65 (VIM -2), and HLA-DR
(L243), was performed asdescribed p r ev iou ~l y . ~~ ~ ' ~he
binding of the MoAbs on the surfacemem brane was demonstrated by
use of a tetrame thylrhodam ine iso-thiocyanate (TR1TC)-conjugated
goat anti-mouse Ig antiserum(Central Laboratory of the Blood
Transfusion Service). The T dT IFlabelings were analyzed on Zeiss
fluorescence microscopes (Zeiss,Oberk ochen, German y), equipped
with phase-contrast facilities.29MNC from three of the eight AML
patients(M.B., A.K., J.M .) were cultured. Before culturing, T-cell
depletionwas performed by use of CD3 (OKT3; Ortho Diagnostic
Systems)and magnetic cell separation (MACS system, Miltenyi Biotec,
Ber-gisch-Gladbach, Germ any) as described.'' For all cell culture
exper-iments, MNC were adjusted to a final concentration of 5 X IO5
ells/mL. Cells were cultured in a serum-free medium as
described."MoAbs added in optimal conc entra tions were CD2 (a
mixture of6G4 [T1 1.1], 4B2 [T11.2], and HIK27 [T11.3]), CD18
(CLB54),
CD I4 MY^), CD I5 (VIM-DS), CD 19(Leu- 2), CD33 MY^), CD34
Culture systems.
and/or CD58 (TS2/9). Control cultures did not contain MoAbs.
Formeasurement of ['HI-thymidine incorpora tion, MNC were
culturedin 96-well flat-bottom tissue culture plates ( o5
cells/well; Cos tar,Cam bridge, MA). Incubation was performed for 3
days at 37"C, with100% elative humidity and a P co2 of 5% .MNC were
cultured for 3 days.['HI-thymidine (specific activity, 6.7 C
i/mmol; Amersham Interna-tional , Amersham , UK ) pulsing was for 6
hours using 0.5 gCi/well.After the 6-hou r pulse, the cells were ha
rvested using an autom aticcell harvester (Skatro n, Lier, Norway).
['HI-thym idine inco rpora tionwas measured with a Betaplate Liquid
Scintillation Counter (LKBWallac, Turk u, Finland) . Each determ
inatio n was performed in trip-licate.
[3H]-thymidine ncorporation.
RESULTSCytomorphology and cytogenetics. The eight leukemias
were classified as AML-M~Eo.Cytogenetics at initial diag-nosis
showed an inv(l6)(p13q22) in 60% to 100% of themetapha ses in the
eight cases studied. In patien t M.V., a tri-somy 22 was seen in
18% of the metaphases in addit ion toinv( 16).
The results of the im-munologic marker analysis are summarized
in Tables 1through 3 an d in Figs 1 an d 2. With the exception of
pat ientE .K., the MN C samples con ta ined m ore than 80% of
leu-kemic blasts. Th e PB M N C s am p le of patient E.K.
enclosedapproximately 10% of C D 3 + T cells, approximately 25%
ofCD19+ B cells, and approximately 65% of leukemic blasts.T h e M N
C s am p le of patient M.V. contained approximately10 % of C D 3 +
T cells, whereas the o ther s ix AM L samplescontained less tha n
5% of C D 3 + T cells (Table 1).Virtually all AML cells were
positive for the pan-myeloidmarker C D 13. Double IF s taining al
lowed th e recognit ionof immatu re myeloid cells (C D I3+,CD34')
and mo re maturemonocytic cells (CDI3 +, CD14') within the CD 13+
cell pop-ulation (Fig 1, Table I ) . In addit ion, small subpopulat
ionsof CD14+ , CD34' and probably also CD14-, CD34- cel lswere
present (Fig 2). Only in patient E.E. and M.B. weregreater than
10%C D l4 +, CD 34+ cel ls detected (Table 1). Ingeneral, the C D
14' cells had th e strongest CD13 fluorescenceintensity (Fig 1). In
each AML , the CD 33+ cel l populat ionwas less prom inent th an
the CD13' cell popula tion (Table1). The difference between these
two pan-myeloid markerswas caused by the fact that within the imm
ature C D3 4+ sub-populat ion C D33 expression was lower than C D
13 expression(Table I), whereas the percentages CD13+,CD I4+ cells
an dCD33+ , CD14+ were com parable (Figs 1 and 2 , Tab le
1).Although a considerable part of the leukemic cells was HLA-D R +
, a relatively large fraction of the C D3 4+ cells was HLA-DR-
(Table I) . The myeloid markers CD l lb , C DI IC, CD15 ,CD36, and
CDw65 were especial ly expressed by the moremat ure leukemic cel ls
.In al l eight M NC samples, the percentages of CD 2+ cel
lsexceeded the CD3 and CD7 percentages (Tables 1 and 2).Table 2
summarizes the resul ts of double IF s tainings forC D 2 and
several other differentiation markers. In additionto a small CD3+,
CD 2+ T-cell po pulat ion, the majori ty ofthe CD 2+ cells
expressed CD13, C D33, an d HLA -DR (Table2, Fig 1). Despite some
patient-to-patient variation, CD2expression was foun d in both imm
ature (CD34') and m ore
Immunologic marker analysis.
-
8/22/2019 M4 Eo_Articulo 1993
4/9
3046 ADRIAANSEN ET AL
.,n0
CD14(My4)
CD, [ T l l )mature monocytic (CD14+) eukemic subpopulations
(Table2, Fig 2).
Four AML samples were tested for CD4 and CD8 expres-sion, which
showed weak CD4 reactivity on a portion of theleukemic cells. The
leukemic cells were positive for the leu-kocyte function antigen
(LFA) CD 1a/CD 18 (LFA-1) andits ligand CD54 (ICAM-I), as well as
CD58 (LFA-3) (Ta-ble I ).
Fig 1. Dot-plot analysis ofthree double IF labelings in alleight
AM L -M ~E opatients.Green (FITC) fluorescence isshown on the
X-axis. Red (PE)fluorescence is shown on theY-axis. (Top panel)
CD14 (My4FITC). CD13 (My7 PE); (middlepanel) CD2 ( T l1 FITC),
CD13(My7 PE); (bottom panel) CD2(T11FITC), CD3 (Leu-4 PE).
TdT+ cells were detected in all eight MNC samples.
Thepercentages of TdT+ cells varied from 0.1% (patient E.K.) to40%
(patient E.E.) (Table I ) . To establish the precise
im-munophenotypeof the TdT+cells and to prove their leukemicorigin,
we performed extensive double IF stainings for TdTand a series of
membrane-bound differentiation marker (Ta-ble 3). The precursor
B-cell markers CDlO and CD19 weredetected on only a few TdT+ cells.
The far majority of the
-
8/22/2019 M4 Eo_Articulo 1993
5/9
AML-M4Eo WITH INV(16) EXHIBITS CD2+PHENOTYPE 3047
CD2(Tll) CD2(Tll)
J M I E , i MVCD14(My4)
. MVJ M I F I . ..CD14(My4)
Fig 2. Dot-plot analysis of four double IF abelings n patients
J.M. and M.V. Green (FITC) fluorescence is shown on the X-axis. Red
(PE)fluorescence is shown on th e Y-axis. (Top left ) CD2 (T11
FITC), CD34 (HPCA-2 PE); (top right) CD2 (T11 FITC), CD14 (My 4
PE); (bottomleft) CD14 (MY4 FITC), CD34 (HPCA-2 PE); (bottom right)
CD14 (My4 FITC), CD33 (M Y9 PE).
TdT' cells were positive for HLA-DR and the myeloidmarkers CD13,
CD33, and CDw65, and to a lesser extentfor the CD15 antigen.
Virtually all TdT+ cells were CD34+,while only a few CD14+, TdT'
cells could be detected, in-dicating that TdT was particularly
expressed by the immatureAML cells (Table 3). These results
indicate that the majorityof the TdT+ cells represented leukemic
cells, independent ofthe relative size of the TdT+
subpopulation.
In three patients (M.B., A.K., and M.V.)both BM and PBsamples
were tested at diagnosis. The relative size and theimmunophenotype
of the various subpopulations in the cor-responding cell samples of
these three patients were com-parable.
MNC from three of the eight AMLsamples were cultured with or
without MoAbs against LFA(CD2, CD18, and CD58). Because of shortage
of cells, the
Culture ofAML cells.
culture experiments were performed only once, but each cul-ture
was performed in triplate. Before culture, the AML sam-ples were
depleted of T cells with CD3 MoAb using magneticcell separation.
After T e l l depletion, the percentage of CD3+T cells in these
samples was less than 1%. The results of theculture experiments are
shown in Fig 3. High spontaneousproliferation was observed in all
three AML cases. Additionof CD2 MoAb to the culture medium
inhibited the prolif-erative response in patients M.B. and J.M.,
but in patientA.K. the CD2 MoAb did not influence the
proliferation,which was probably caused by the relatively low
frequencyand low density of CD2 expression in this patient (Fig
1,Table 1). Addition of CD18 or CD58 MoAb did not resultin
significantly higher proliferative responses in patient M.B.and
J.M. However, in patient A.K., addition of CD18 andCD58 MoAb or a
mixture ofCD2, CD18 and CD58 MoAb
Table 2. Immunologic Marker Analysis of CD2+ CellsPat
ientImmunologic
Markers E.K. E.E. M.B. N.S. A.K . J.M. M . V . D.W.CD3 (Leu-4)
18 8 12 7 13 6 20 4CD13 (My71 78 94 84 93 83 92 74 95CD33 (My9) 62
64 35 94 95CD34 (HPCA-2) 4 83 68 43 78 13 29 33CD14 (My4) 52 25 32
18 20 58 57 56
HLA-DR (L243) 62 78 92 89Data represent percentage positivity
for the various markers per CD2+cells as determinedbydouble IF
staining.
-
8/22/2019 M4 Eo_Articulo 1993
6/9
3048 ADRIAANSEN ET ALTable 3. Immunolo gic Marker Analysis of
TdT+ Cells
PatientImmunologicMarkers E.K. E.E. M.B. N.S. A.K. J.M. M.V.
D.W.CD2 (T l l )CD7 (3A1)CD13 (My7)CD14 (My4)CD19 (Leu-12)CD33
(My9)
CDlO (VIL-A1)
CD15 (VIM-D5)
CD34 (61-3C5)CDw65 (VIM-2)HLA-DR (L243)
31 7810
92 920 0
3186 81
100 998294
2110
660901
6718
5836
922
6191979896
2700
62010
209712
3650
6501
368480
3130
69001
45814516
4800
890
56868423
Data represent percentage posit ivity or surface-membrane marker
expression per TdT' cells as determined by double IF staining.
resulted in an increased proliferative response, probably
re-lated to stimulation by the CDI 8 and CD58 MoAb and ab-sence of
inhibition by the CD2 MoAb (Fig 3).
DISCUSSIONIn this study, we could show that AML-M4Eo'with
inv( 16)(p13q22) s associated with a specific immunopheno-type
with CD2 expression. The eight leukemias consistedofheterogeneous
cell populations mainly caused by the presenceof multiple
subpopulations, which varied in size between thepatients. However,
the immunophenotype of these subpopulations was comparable,
independent of their relative size.Not only the presence or absence
of various immunologicmarkers was comparable, but also the
fluorescence intensity(Fig I ) . Whereas a close association
between a specific chro-mosome aberration and a particular
(immuno)phenotype iswell known in acute lymphoblastic leukemia,
this has notfrequently been observed in AML, except for the typical
phe-notype ofAM LM 3 with t( 15;17)(q22;q12),CD19 expressionin AML
with t(8;2 )(q22;q22), and a few associations in otherAML types.334
This may be explained by the heterogeneouscomposition of most AML,
which can only be characterizedproperly, if multiparameter analysis
is pe rf~rmed ."~. ~*n ad-dition, small leukemic subpopulations may
be missed, if rigidcut-off values of 15% to 25% positivity are
used, which isoften the case in routine immunologic marker
analysis.Therefore, we performed detailed immunophenotyping byuse
of multiple double IF stainings in our series of AML-M4Eo
patients.
Virtually all AML-M~Eo ells were positive for the pan-myeloid
marker CD13. In addition, the AMLs were partlypositive for CD2, CDI
Ib, CDI IC , CD14, CD33, CD34,CD36, CDw65, TdT, and HLA-DR. The
double IF stainingsdemonstrated coexpressionof the CD2 antigen and
myeloidmarkers and allowed us to recognize multiple AML subp
opulations (Figs I and 2). Within the CD13+ cell
population,immature cells (CD34+,CD14-) and more mature
monocyticcells (CD34-, CD14+) could be identified. The CD2
antigenwas expressed by immature cells, as well as more
maturemonocytic cells, whereas TdT expression was exclusivelyfound
in the CD34+, CD14- subpopulation. Although therelative size of the
TdT+ subpopulation varied from 0.1% to
40% among patients, our extensive double IF stainings
dem-onstrated the homogeneous leukemic immunophenotype ofthis
subpopulation in all eight patients (Table 3). Based onthe double
IF staining results obtained in our AML-M~Eocases, a hypothetical
diagram of the subpopulations in thistype of AML is shown in Fig
4.
Reports on immunologic marker analysisof AML-M~Eowith a
chromosome 16 aberration are scarce and only a min-imal number of
markers have been used.*'-26f tested, CDI 3was found to be
positive.*'.*' In addition, expression of CD14,CD33, and HLA-DR has
been reported, confirming themonocytic ph en ~t yp e. ~~ .~ '.
~~aietta et al described one patientwith an AML-M~Eowith inv( 16)
in whom the cells wereCDw65+ and TdT', but CD14-.22 It is not clear
whether thisleukemia indeed differs from our eight cases, or
whether thedifference can be explained by the presence of a
relativelysmall (
-
8/22/2019 M4 Eo_Articulo 1993
7/9
AML-M4Eo WITH INV(16) EXHIBITS CD2+ PHENOTYPE
I CD13 I3049
/ CD33 IHLA-DR
I CD34 \CD14
CD 2
TdT4 - -- bimm ature cells mon ocytic cellsFig 4. Hypothetical
diagram of immunologic marker expression
by the various leukemic cell populations in AML-M~Eo
amples.Virtually all AML-M~Eoells express CD13. Most cells are
positivefor either CD34 (immatur e subpopulation) or CD14 (more
maturemonocytic subpopulation), whereas CD14+, CD34+ cells
andCD14-, CD34- cells are scarce. CD33, HLA-DR,and CD2 are
ex-pressedon a part of the cells in both subpopulations. TdT
expressionis restricted to the CD34+ subpopulation. This diagram is
based onthe results of double IF staining experiments given in
Tables 1through 3.
The consistent expression of the CD 2 antigen in ou r eightAM L
cases was detectable by M oAbs against the three
CD2(Tll.I,T11.2,andTI1.3)epitopes.noneAML (D.W.) , theexpression of
the T I 1. I epitope was confirmed by the abilityto form rosettes
with sheep red blood cells (data not shown).44These results are in
line with the report by Ball et al,26wh odemonstrated CD2 mRNA in a
case of AML-M~Eo.Thenegativity for other T-cell markers, such as
CD3 and CD7,as well as the coexpression of CD2 and myeloid
markers,argues against T-cell lineage c0mmitment.4~According tothe
literature, CD 2 expression can be found in 6%to 2 1% ofAML
cases.26,4649So far, an association between CD2expression and a
specific type of AM L has only been reportedfor a subgroup of
AML-M3 with a so-called 5' PML-RARfusion region.40The prognostic
value of CD2 expression inAM L is controversial, because CD2
expression has been as-sociated with both favorable an d poor In
arecent study in childhood A ML, expression of the CD2 an-tigen was
not found to be prognostica lly ~ ign if i ca n t .~ ~hi
scontroversy may be explained by the fact tha t CD2' AM Lform a
mixture of AML w ith a relatively good progno sis (eg,AMLM4Eo) and
AML with poor prognosis, such as im-mature types of AML an d acute
undifferentiated leukemias.%The C D2 antigen interacts with its
ligand the CD 58 (LFA-3) molecule, a cell-surface glycoprotein with
broa d tissue dis-tribution, including expression on erythrocytes,
epithelialcells, endothelial cells, fibroblasts, and most cells of
hema-topoietic rigi in.^',^' Studies in T cells have dem onstrated
thatCD2-CD58 interactions can induce activation in both CD 2+cells
and CD58+ cells, leading to proliferation an d expansionof the
activated T cells.53-55 he A M L- M ~E o ases in ou r
study were not only positive for the C D2 (LFA-2) antigen,but
they also expressed the CD58 (LFA-3) antigen. We cul-tured three A
M LM 4E o samples, enabling cell-to-cell contact,and observed a
high spontaneous proliferation in all threesamples. This may
explain the reported high success rate ofthe detection of
chromosome 16 aberrations, if cytogeneticanalysis is performed on
cultured AMLM4Eo cells insteadof freshly obtained cells.56 n our
culture experiments, ad-dition of CD2 MoA b inhibited cell
proliferation in tw o pa-tients, suggesting that C D2(LFA
-2)-CDSs(LFA-3) interactionsupports the high spontaneous
proliferation of the AML-M4Eo cells. However, in the third patient
(A.K.), additionof CD2 MoAb did not have inhibitory effects on cell
prolif-eration, w hich might be explained by the low frequency
andlow density expression of the CD 2 antigen in this patient (F
ig1). MoA bs against LFA molecules, such as CD2, CD18, a ndCD58,
may not only abrogate cell-cell interactions, b ut theycan also
function as agonist for signaling through these mo
l-ecules.44,54,55,57-59 Ou r data suggest that, in a t least one
patient(A.K.), CD18 and CD58 M oAbs might indeed induce
someadditional cell proliferation (Fig 3 ) . In such cases, it is
notknown whether binding of CD18 an d CD58 M oAbs directlyinduces
cell proliferation or whether this is indirectlycaused by other
mechanisms such as the production ofinterleukin- I .55Finally, it
is intriguing to speculate about some uniqueclinical and biologic
characteristics of A M L- M ~E o,whichmight be related to the
expression of the CD2 and CD58antigens. Th e proliferation-inducing
effect of the CD 2-CD 58mediated cell-cell contact may contribute
to the high WBCcou nt i n AM L- M ~Eo .~ ' .~ 'n addition, based on
the distinctexpression of the C D58 a ntigen on e ndothelial cells,
Plunk ettet a16' speculated that CD2-CD58 interaction may
supportextravasation of activated T lymphocytes at sites of immu
nereaction. Therefore, it is intriguing that patients with AM
L-M4Eo frequently have enlarged lymph nodes, hepatom egaly,a n d /
o r ~ p l e n o m e g a l y . ~ ~ ~ ~ ' ' ~ ' ~urthermore, in AM
L-M ~Eo ,relatively high incidence of CNS leukem ia has been
observed,manifesting as leptomeningeal disease and
intracerebralmyeloblastoma^.*^"^'^^'^*'^ Also, in most of our
patients, ahigh WBC count, hepatosplenomegaly, and/or CN S
leukemiawere observed (Table 1). Whether expression of the CD2
andCD58 antigens induces high WBC co unts an d facilitates
dis-semina tion of leukem ic cells to lymphoid tissues and the
CNSneeds further investigation.
ACKNOWLEDGMENTWe gratefully acknowledge Professor Dr R. Benner
and Dr H.
Hooijkaas for their continuous support; Drs K. Hihien,
A.C.J.M.Holdrinet, P. Sonneveld and M.B. van 't Veer for sending
patientmaterial and providing clinical information; Drs W. Ax, W.
Knapp,R. Kurrle, R.A.W. van Lier, and T. Schumacher for kind ly
providingmon oclon al antibodies; P.W.C . Adriaansen-Soeting,J.G.te
M arvelde,and A.F. Wierenga-Wolf for excellent techn ical
assistance; T.W . vanOs for preparation o f the figures;and
A.D.Korpershoek for secretarialsupport.
REFERENCES1 . McCulloch EA , Kelleher CA, M iyauchi J, Wang C,
ChengGYN,
Minden MD , Curtis JE: H eterogeneity in acu te myeloblastic
leukemia.Leukemia 2:38S, 1988 (suppl)
-
8/22/2019 M4 Eo_Articulo 1993
8/9
3050 ADRIAANSEN ET A L
2. Bennett JM, Catovsky D, Daniel M-T, Flandrin G, GaltonDAG, G
ralnick H R, Sultan C: Proposals for the classification of theacute
leukaemias. French-American-British (FAB ) co-operative grou p.Br J
Haematol 33:451, 19763. Bennett JM, Catovsky D, Daniel MT, Flandrin
G, Galton DAG ,Gralnick HR, Sultan C: Proposed revised criter ia
for the classificationof acute myeloid leukemia. A report of the
French-A merican-Britishcooperative group. Ann Intern Med 103:626,
19854. Arthur DC, Bloomfield CD: Partial deletion of the long arm
ofchromosome 16 and bone marrow eosinophilia in acute
nonlym-phocytic leukemia: A new association. Blood 61:994, 1983
5 . Le Beau MM, Larson R A, Bitter MA, Vardiman JW , Golom bHM,
Rowley JD. Association of an inversion of chromosome 16with
abnormal marrow eosinophils in acute myelomonocytic leu-kemia. A
unique cytogenetic-clinicopathological ssociation. N EnglJ Med
309:630, 19836. De la Chapelle A, Lahtinen R: Chromosome 16 and
bone-marrow eosinophilia. N Engl J Med 309:1394, 19837. Testa JR,
Hogge DE, Misawa S, Zandparsa N Chromosome16 rearrangements n acute
myelomonocytic eukemia with abnormaleosinophils. N Engl J M ed
310:468, 19848. Holmes R, K eating MJ, Cork A, Broach Y, Trujillo
J, DaltonWT, McCredie KB, Freireich EJ:A unique p attern of central
nervoussystem leukemia in acute m yelomonocytic leukemia associated
withinv(l6)(pl3q22). Blood 65:1071, 19859. Larson RA , Williams SF,
Le Beau MM, Bitter MA, VardimanJW, Rowley JD: Acute myelomonocytic
leukemia with abnormaleosinophils and inv( 16) or t( 16; 16) has a
favorable prognosis. Blood68:1242, 198610. Bloomfield CD, De la Ch
apelle A: Chromosome abno rmalitiesin ac ute non lymp hocy tic
leukem ia: Clinical and b iologic significance.Semin Oncol 14:372,
19871 I . Second MIC cooperative study group: Morphologic,
immu-nologic and cytogenetic (MIC) working classification of the
acutemyeloid leukaemias. Br J Haematol 68:487, 198812. Schiffer CA,
Lee EJ , Tomiyasu T, Wiernik PH, Testa J RPrognostic impact of
cytogenetic abnormalities in patients with denovo acute nonlymph
ocytic leukemia. Blood 73:263, I98913. Bemard P, Dachary D,
Reiffers J, Marit G ,We n Z , JonveauxP, David B, Lacombe F,
Broustet A: Acute nonlymphocytic eukemiawith marrow eosinophilia
and chromosome 16 abnormality: A reportof 18 cases. Leuk emia
3:740, 198914. Raim ondi SC , Kalwinsky DK , Hayashi Y, Behm FG,
MirroJ, Williams DL: Cytogenetics of childhood acute
nonlymphocyticleukemia. Cancer Ge net C ytogenet 40: 13, 198915.
Kalwinsky DK, Raimondi SC, Schell MJ, Mirro J, SantanaVM, Behm F,
Dah1 GV , Williams D: Prog nostic importance of cy-togenetic
subgroups in de novo pediatric acute nonlymph ocytic leu-kemia. J
Clin Oncol 8:75, 199016. Le Beau MM , Diaz MO, K arin M, Rowley JD:
Metallothioneingene cluster is split by chrom osome 16
rearrangements in m yelo-monocytic leukaemia. Natu re 3 13:709,
198517. Su therland GR , Baker E, Callen DF, Garson O M, West
AK:The hum an metallothionein gene cluster is not disrupted in
myelo-monocytic leukemia. G enomics 6: 144, 199018. Arthur DC,
Berger R, Golomb HM , Swansburry GJ, ReevesBR, Alimena G, Van den B
erghe H, Bloomfield CD, De la ChapelleA, Dewald GW, Garson OM,
Hagemeijer A, Kaneko Y, MitelmanF, Pierre RV, Ruutu T, Sakurai M,
Lawler SD, Rowley JD: Theclinical significance of karyotype in
acute myelogenous leukemia.Cancer Genet Cytogenet 40:203, 198919. O
hyashiki K, Ohyashiki JH, Iw abuchi A, Ito H, Toyama KCentral
nervous system involvement in acute nonlymphocytic leu-
kemia w ith inv( 16)(p13q22). Leukemia 2:398, 1988
20. Bitter MA, Le Beau MM, Larson RA, Rosner MC, GolombHM,
Rowley JD, Vardiman JW: A morphologic and cytochemicalstudy of
acute myelomonocytic leukemia with abnormal marroweosinophils
associated with inv( 16)(pl3q 22). Am J Clin Pathol 81:733, 198421.
Tantra vahi R, Schwe nn M, H enkle C, Ne11 M, Leav itt PR,Griffin
JD, Weinstein HJ: A pericentric inversion of chromosome16 is
associated with dysplastic marrow eo sinophils in acute
myelo-monocytic leukemia. Blood 63:800, 198422. Paietta E,
Papenhausen P, Azar C, Wiemik PH, Spielvogel A:Inv(l6) occurring in
a case of acute biphenotypic leukemia lackingmonocytic markers:
multiple but short remissions. Cancer GenetCytogenet 25:367,
198723. Maseki N, Kaneko Y, Sakurai M: Nonrandom
additionalchromosome changes in acute nonlymphocytic leukemia
withinv( 16)(pl3q22). Cancer Genet Cytogenet 26:309, 198724.
Borowitz MJ, Gockerman JP, Moore JO, Civin CI, Page SO,Robertson J,
Bigner SH: Clinicopathologic and cytogen ic features ofCD34 (MY
10)-positiveacute nonlymphocytic leukemia. Am J ClinPathol 91:265,
198925. Geller RB, Zahurak M, Hurwitz CA, Burke PJ, Karp JE
,Piantadosi S, Civin CI: Prognostic importance of
immunophenotyp-ing in adults w ith acute myelocytic leukaemia: The
significance ofthe stem-cell glycoprotein CD34 (MylO). Br J Haem
atol76 :340, 199026. Ball ED, Davis RB, Griffin JD, Mayer RJ, Davey
FR , ArthurDC, urster-Hill D, No11 W, Elghetany MT, Allen SL, Rai
K, LeeEJ, Schiffer CA, Bloomfield CD: Prognostic value of
lymphocytesurface markers in acute myeloid leukemia. Blood 77:2242,
199127. Hagemeijer A, Smit EME, Bootsma D: Improved
identificationof chromosomes of leukemic cells in
methotrexate-treated cultures.Cytogenet Cell Genet 23:208, 197928.
Hamden DG, linger HP, eds. ISCN (1985): An InternationalSystem for
Human Cytogenetic Nomenclature. Published in collab-oration with
Cytogenet Cell Genet. Basel, Switzerland, Karger, 198529. Van
Dongen JJM, Adriaansen HJ, Hooijkaas H: Immuno-logical marker
analysis of cells in the various hematop oietic differ-entiation
stages and their m alignant counterparts, in Ru iter DJ,Fleuren GJ,
Wam aar SO(eds): Application of Monoclonal A ntibodiesin T um our
Pathology. Dordrecht, The Netherlands, Martinus Nijhoff,1987, p
8730. Adriaansen HJ, Van Dongen JJM, Kappers-Klunne MC,Hahlen K ,
Van 't V eer MB, Wijdenes-de Bresser JHF M, H oldrinetACJM,
Harthoom-Lasthuizen EJ, Abels J, Hooijkaas H:
Terminaldeoxynucleotidyl transferase positive subpopulations occur
in themajority of ANL L Implications for the detection of minimal
disease.Leukemia 4:404, 19903 . Miltenyi S, Muller W, Weichel W,
Radbruch A High gradientmagnetic cell separation with MAC S.
Cytometry 11:231, 199032. Salem M, Delwel R, Touw I, Mahmoud L,
Lowenberg B:Human AML colony growth in serum-free culture. Leuk Res
12:157, 198833. Adriaansen HJ, Van Dongen JJM, Hooijkaas H, Hahlen
K ,Van 't Veer MB, Lowenberg B, Hagem eijer A: Translocation
(6;9)may be associated with a specific TdT-positive immu nological
phe-notype in A NLL. Leukem ia 2: 136, 198834. De Rossi G, Avvisati
G, Coluzzi S, Fenu S, Lo Coco F, LopezM, Nanni M, Pasqualetti D,
Mandelli F Immunological definitionof acute promyelocytic leukemia
(FAB M3): A study of 39 cases. EurJ H aematol 45: 168, 199035. Sun
G, Sparkes RS, Worm sley S, Naeim F,Champlin R, GaleR P Are some
cases of acute leukemia with t(8;21) hybrid leukemias?Cancer Genet
Cytogenet 49: 177, 199036. Touw I, Donath J, Pouwels K, Van
Buitenen C, Schipper P,Santini V, Hagemeijer A, Lowenberg B, Delwel
R: Acute myeloid
-
8/22/2019 M4 Eo_Articulo 1993
9/9
AML-M4Eo WITH INV(16) EXHIBITS CD2+ PHENOTYPE 305
leukemias with chromosomal abnormalities involving the 2 lq2 2
re-gion identified by their in vitro responsiveness to
interleukin-5. Leu-kemia 5:687, 199137. Lo Coco F, Avvisati G,
Diverio D, Biondi A, Pandolfi PP,Alcalay M, De R ossi G, Petti MC ,
Cantd-Ra jnoldi A, Pasqualetti D,Nanni M, Fenu S, Frontani M, M
andelli F: Rearrangements of theRAR-(Ugene in acute promyelocytic
leukaemia: Correlations withmorphology and immunoph enotype. Br J
Haem atol78:49 4, I99 I
38. Cuneo A, Michaux J-L, Ferrant A, Van Hove L, Bosh A,Stul M,
Dal Cin P, V andenberghe E, Cassiman J-J, Negrini M, PivaN,
Castoldi G, Van den Berghe H: Correlation of cytogenetic
patternsand clinicobiological features in adu lt acute m yeloid
leukemia ex-pressing lymphoid markers. Blood 79:720, 199239 . Kita
K, Nakase K, Miwa H, M asuya M, Nishii K, Morita N,Takakura N,
Otsuji A, Shirakawa S, Ueda T, Nasu K, Kyo T, D ohyH, Kamada N:
Phenotypical characteristics of acute myelocytic leu-kemia
associated with the t(8;2 l)(q22;q 22) chrom osomal abnor-mality:
Frequent expression of immature Bcell antigen CD 19 togetherwith
stem cell antigen CD34. Blood 80:470, 19 9240. Claxton DF, Reading
CL, Nagarajan L, Tsujimoto Y, An -
dersson BS, Estey E, Cork A, Huh YO , Trujillo J, Deisseroth
AB:Correlation of CD2 expression with PM L gene breakpoints in
patientswith acute promyelocytic leukemia. Blood 80582, 199241.
Terstappen LWMM, Konemann S, Safford M, Loken MR,Zurlutter K,
Buchner Th, Hiddemann W, Wormann B: Flow cyto-metric
characterizationof acute m yeloid leukemia. Part I. Significanceof
light scattering properties. Leukemia 5 :315, 199142. T erstappen
LWMM , Safford M, Konema nn S, Loken MR,Zurlutter K, Buchner T,
Hiddemann W, Wormann B: Flow cyto-metric characterizationof acute
myeloid leukemia. Part 11. Phenotypicheterogeneity at diagnosis.
Leukemia 5:757, 199143. Hogge DE, Misawa S, Parsa NZ, Pollak A,
Testa JR: Abnor-malities of chromosome 16 in association with acute
myelomonocyticleukemia and dysplastic bone marrow eosinophils. J
Clin Oncol 2:550, 198444. Meuer SC, Hussey RE, Fabbi M, Fox D,
Acuto 0, FitzgeraldKA , Hodgdon JC, Protentis JP, Schlossman SF,
Reinherz E L Analternative pathway of T-cell activation: A
functional role for the 50kd T11 sheep erythrocyte receptor
protein. Cell 36997, 198445. Van Dongen JJM, Hooijkaas H,
Comans-Bitter M, HahlenH, De Klein A, Van Zanen GE, Van 't Veer MB,
Abels J, BennerR: Hu ma n bone marrow cells positive for terminal
deox ynucleotidyltransferase (TdT), HLA -DR, an d a T cell marker
may represent pro-thymocytes. J Immunol 135:3144, 198546. Mirro J,
Antoun G R, Zipf TF, Melvin S, StassS: Th e E rosette-associated
antigen of T cells can be identified on blasts from patientswith
acute my eloblastic leukemia. Blood 65:363, 198547. Cross AH,
Goorha RM, N u s R, Behm FG, Mu rphy SB, Kal-winsky DK, R aimondi
S, Kitchingman GR, Mirro J: Acute myeloidleukemia with T-lymph oid
features: A distinct biologic and clinicalentity. Blood 725 79,
1988
48. Del Vecchio LD, Schiavone EM, Ferrara F, Pace E, Lo PardoC,
Pacetti M, Russo M, Cirillo D, Vacca C: Immunodiagnosis ofacute
leukemia displaying ectopic antigens: Proposal for a
classifi-cation of promiscuous phenotypes. Am J Hematol 31:173,
198949. Smith FO, Lamp kin BC, Versteeg C, Flowers DA, DinndorfPA,
Buckley JD, Woods WG, H amm ond GD , Bernstein ID: Expres-sion of
lymphoid-associated cell surface antigens by childhood acutemyeloid
leukemia cell lacks prognostic significance. Blood 79 :24
15,I992
50. Pui C-H, Behm FG, Kalwinsky K, Murphy SB, Butler DL,Dah1 GV,
M irro J: Clinical significance of low levels of myeloper-oxidase
positivity in childhood acute nonlymphoblastic leukemia.Blood 7051,
19875 I . Dustin ML, Sanders ME, Shaw S, Springer TA: Purified
lym-phocyte function-associated antigen 3 binds to CD 2 and
mediates Tlymphocyte adhesion. J Exp Med 165:677, 198752. Wallner
BP, Frey AZ, Tizard R, Mattaliano RJ, Hession C,Sanders ME, Dustin
ML, Springer TA: Primary structure of lym-phocyte
function-associated antigen 3 (LFA -3). The ligand of the
Tlymphocyte CD 2 glycoprotein. J Exp Med 166:923, 198753. Vollger
LW, Tuck D T, Springer TA, Haynes BF, Singer KH:Thymocyte binding
to hu man thymic epithelial cells is inhibited bymonoclonal
antibodies to CD-2 and LFA-3 antigens. J Immunol138:358, 198754.
Shaw S, Ginth er Luce GE, Quinones R, Cress RE, SpringerTA, Sanders
ME: Two antigen-independent adhesion pathways usedby human
cytotoxic T-cell clones. Nature 323:262, 198655. Le P, Denning S,
Springer T, Haynes B, Singer K: Anti-LFA-3 monoclonal antibody
induces interleukin 1 (IL 1) release by thym icepithelial (TE)
cells and monocytes. Fed Proc 4 6:447, 198756. Harth H, Pees H,
Zankl H: Acute myelomono cytic leukemia(M4) with eosinophilia:
Problems concerning chromosome 16 ab-normality. Can cer Genet
Cytogenet 23: 127, 198657. Dustin ML, Springer TA: Role of
lymphocyte adhesion re-ceptors in transient interactions and cell
locomotion. Ann Rev Im-munol 9:27, 199
58 . Hunig T, Tiefenthaler G, Meyer zum Biischenfelde K-H,Meuer
SC: Alternative pathway activation of T cells by binding ofCD2 to
its cell-surface ligand. Nature 326:298, 198759. Van Noesel C,
Miedema F, Brouwer M, De Rie MA, A ardenLA, Van Lier RAW:
Regulatory properties of LFA-I alpha a nd betachains in human
T-lymphocyte activation. Nature 333:850, 198860. Nara N, McCulloch
EA: Membranes replace irradiated blastcells as growth requirement
for leukemic blast progenitors in sus-pension culture. J Exp Med
162:1435, 19856 1. Reilly IAG, Kozlowski R, Russell NH: T he role
ofcell contactand autostimulatory soluble factors in the pro
liferation of blast cellsin acu te myeloblastic leukemia. Leukemia
3: 145, 198962. Plunkett ML, Sanders ME, Selvaraj P, Dustin ML,
SpringerTA: Rosetting of activated human T lymphocytes with
autologouserythrocytes. Definition of the receptor and ligand
molecules as CD 2and lymphocyte function-associated antigen 3
(LFA-3). J Exp Med165:664, 1987
