research paper

Bone marrow trephine findings in acute myeloid leukaemia

with multilineage dysplasia

Nyethane Ngo, Irvin A Lampert and Summary

Kikkeri N Naresh
Department of Histopathology, Hammersmith
Hospital & Imperial College, London, UK
Received 7 August 2007; accepted for
publication 20 August 2007
Correspondence: Prof. Kikkeri Naresh,
Department of Histopathology, Hammersmith
Hospital, Du Cane Road, London W12 0HS,
UK.
E-mail: k.naresh@imperial.ac.uk
Acute myeloid leukaemia (AML) with multilineage dysplasia (MD) is one of
the four main categories of AML in the World Health Organization (WHO)
classification. The role of bone marrow trephine biopsy (BMTB) histology
and immunohistochemistry in the diagnosis of AML-MD is currently
unclear. BMTBs were studied in 11 cases of AML-MD and two cases of
myelodysplasia that subsequently transformed to AML. Among them, six
cases showed trilineage dysplasia and seven showed bilineage dysplasia. With
respect to conforming to the WHO definition of AML, documentation of an
increased proportion of immature myeloid cells was possible on morphology
and counting of immature cells following immunostaining with
CD34, CD117 or HLA-DR antibodies. Recognition and quantification of
dysplastic features in the haemopoietic lineages was made easier by
immunohistochemistry with antibodies to ret40f (glycophorin C),
myeloperoxidase, CD61 and/or CD42b, CD34, CD117 and HLA-DR. Based
on this relatively small series of cases we show the utility of BMTB and
immunohistochemistry as an aid to the diagnosis of AML-MD. This has to be
seen not just in light of its utility at diagnosis, but also the role the diagnostic
BMTB would play for purposes of comparison when follow-up BMTBs are
submitted in this group of patients.
Keywords: acute leukaemia, AML, bone marrow pathology, MDS, immu-
nophenotype.

The classification of acute myeloid leukaemia (AML) was
originally based on morphological and cytochemical features
proposed by the French–American–British Co-operative
Group (FAB) in 1976 (Bennett et al, 1976). As techniques
such as immunophenotyping, cytogenetics and molecular
genetics developed, the World Health Organization (WHO)
proposed a new classification that embraced the newer
technologies, and incorporated AML with multilineage dys-
plasia (MD; AML-MD) as one of the four main categories
(Brunning et al, 2001).
The WHO classification defines AML-MD as an acute
leukaemia with ‡20% blasts in peripheral blood or bone
marrow, and dysplasia in two or more myeloid cell lines. The
dysplasia must be present in ‡50% of the cells of at least two cell-
lineages (Brunning et al, 2001). AML-MD may occur de novo or
evolve from a myelodysplastic syndrome (MDS). This entity
occurs more frequently in the elderly and is associated with
pancytopenia (Brunning et al, 2001; Goasguen et al, 1992).
Before the publication of the WHO classification the
frequency of trilineage dysplasia in de novo AML was reported
to be approximately 12% (Goasguen et al, 1992; Tamura et al,
1998). Using the WHO criteria of the presence of dysplasia in
at least two lineages, the frequency of AML-MD was reported
to be in the range of 24–38% (Arber et al, 2003; Bao et al,
2006). The identification of the presence of multilineage
dysplasia in AML cases is important because of its association
with a poor prognosis (Arber et al, 2003; Brunning et al, 2001;
Gahn et al, 1996; Tamura et al, 1998).
Islam et al (1985) found that bone marrow trephine
biopsy (BMTB) was superior to bone marrow aspiration
(BMA) for the evaluation of fat and marrow cellularity,
pattern and extent of blast cell infiltration, homogeneity of
the leukaemic infiltrate, frequency of mitoses and residual
haemopoietic activity. The present study documents how
BMTB morphology with the aid of immunohistochemistry
can be used as an adjunct to diagnose AML-MD (Islam

ª 2007 The Authors First published online 31 October 2007

Journal Compilation ª 2007 Blackwell Publishing Ltd, British Journal of Haematology, 140, 279–286 doi:10.1111/j.1365-2141.2007.06882.x
N. Ngo et al
et al, 1985). We believe that diagnosing the entity can be
made easier when BMA morphology, flow cytometry and
cytogenetics are combined with the assessment of a well
prepared BMTB, which is adequately supported by immu-
nohistochemistry.
Materials and methods
The study involves thirteen cases of AML. The cases were
obtained from the archives of the Department of Histopa-
thology, Hammersmith Hospitals NHS Trust during the
period 2004–2007. In these cases, the BMTB report docu-
mented presence of dysplastic features in two or more
haemopoietic lineages and suggested that the proportion of
precursor cells warranted a diagnosis of AML. BMTBs had
been interpreted independent of peripheral blood (PB) or
BMA morphology, flow cytometry and cytogenetics. The
diagnosis of AML was later confirmed based on multiple
parameters that included clinical features, morphology (on
PB, BMA and BMTB), immunophenotyping, and cytogenet-
ics (where available). For the purposes of this study, cases
were subdivided into two groups: (i) AML with background
multilineage dysplasia (AML-MD; 11 cases), (ii) MDS that
subsequently transformed to AML (MDS fi AML; two
cases).
Bone marrow trephine biopsies had been transported and
fixed in acetic acid-zinc-formalin fixative, decalcified in 10%
formic acid–5% formaldehyde and processed to paraffin-wax
embedding. One-micron-thick sections had been cut in
accordance with our previously published protocol (Naresh
et al, 2006). At the time of initial diagnosis, sections had been
immunostained using antibodies for – ret40f (glycophorin C,
a marker for erythroid cells), myeloperoxidase, CD61 and/or
CD42b (markers for megakaryocytes), CD34, CD117, Tdt,
HLA-DR, CD99, CD68 (KP-1), CD68 (PGM-1) and CD10.
Immunohistochemistry was performed with appropriate
retrieval techniques, antibody dilutions and positive and
negative controls as per the previously published protocols
(Naresh et al, 2006). At the time of initial diagnosis and work-
up, BMTBs had been interpreted independent of BMA
morphology, and once BMTB reports had been authorised,
the cases had been jointly discussed with haematologists who
had interpreted the BMAs.
For the purposes of the current study, slides were reviewed
by one of the authors (NN) and the findings were reconfirmed
(KNN). The following histological parameters were docu-
mented.
1 Cellularity – assessed on haematoxylin–eosin-stained
(H&E) sections, as the proportion of haemopoietic marrow
when compared with fat spaces + haemopoietic marrow,
and expressed as a percentage.
2 Abnormal localization of immature precursors (ALIP) –
defined as the presence of at least three clusters of myeloid
precursors in the centre of the marrow space. Immunohis-
280 Journal Compilation ª 2007 Blackwell Publishing Ltd, British Journal of Haematology, 140, 279–286
tochemistry using CD117, CD34 and myeloperoxidase
aided the better recognition of ALIP (Naresh & Lampert,
2006).
3 Myeloid lineage cells were evaluated for maturation and for
the proportion of mature myeloid cells.
4 Dyserythropoiesis – identified by the presence of bi-/
multinucleation and/or irregular nuclear contours/lobation.
5 Megaloblastoid change in the erythroid lineage cells –
identified by the presence of erythroid cells with a size at
least twice that of normal erythroid lineage cells with
accompanying nuclear and chromatin abnormalities. Rec-
ognition of abnormal features in erythroid lineage cells was
easier on ret40f immunostained slides.
6 Megakaryocytes were evaluated in terms of the number of
megakaryocytes, size, nuclear lobation, and the presence of
obvious abnormal forms and micromegakaryocytes. Im-
munostaining for CD61 and/or CD42b was extremely
helpful.
7 Presence and prevalence of apoptotic cells – documented as
no, mild, moderate and marked increase in numbers.
8 The bone marrow trephines were also stained with Gomori
silver stain and the degree of reticulin fibrosis was assessed
(Manoharan et al, 1979).
9 The CD34+ cells were quantified at ·40 magnification,
excluding cortical and trabecular bone, periosteal connective
tissue, adipose tissue or areas of haemorrhage. The CD34+
cells were counted in five or more randomly selected fields
(equal to 0Æ98 mm2 of haemopoietic tissue) and the number
of CD34+ cells were expressed as a percentage of the total
number of bone marrow nucleated cells. The CD117+ cells
were quantified in the same manner as CD34+ cells.
Results
The study included nine men and four women (mean age:
61Æ5 years; range: 24–85 years). Among the 11 AML-MD
patients, seven were men and four were women (mean age:
59Æ7 years; range: 24–85 years). Both MDS fi AML patients
were men (63 and 80 years of age). At the time of analysis,
seven of 11 AML-MD patients were alive at a mean follow-up
interval of 19Æ6 months. However, both MDS fi AML patients
died of their disease.
The BMA blast count was ‡20% in all but two cases, where
the BMA quality was poor for the assessment of blasts. The
diagnosis in these two cases was based on a PB blast count of
‡20% and/or cytogenetics. Among the 11 cases with an
interpretable BMA, multilineage dysplasia was reported in all
eleven cases. Cytogenetic results were available for 10 of 11
AML-MD patients and for both MDS fi AML patients. Three
of the AML-MD patients and one MDS fi AML patient had
normal karyotypes. None of the cases had recurrent genetic
abnormalities associated with specific AML subtypes as defined
in the WHO classification (t(8;21), inv(16), t(16;16), t(15;17)
or variants or 11q23 abnormalities).
ª 2007 The Authors
 Bone Marrow Trephines in AML with Multilineage Dysplasia

Bone marrow trephine findings Table I. Immunophenotype of the precursor cells in AML-MD and
AML fi MDS.
Quantifying precursor cells on trephine sections. In nine of
AML-MD patients, leukaemic blasts expressed CD34. In four Antibody Cases positive (%)
of nine CD34-positive cases, the CD34+ cells amounted to
CD34 11/13 (85)
‡20% of all marrow cells. In one case, CD34+ cells amounted CD117 10/11 (91)
to only 7%. However, in this case, HLA-DR positive cells HLA-DR 4/5 (80)
amounted to over 30% and >50% of the cells in the BMTB had Tdt 2/7 (29)
morphological features of ‘blasts’. In two of the nine CD34- CD99 1/1 (100)
positive cases, the percentage of CD34+ cells was 16% and 14% Myeloperoxidase 11/13 (85)
respectively. In these two cases, the blast count on the aspirate CD68 (KP1) 4/4 (100)
was 23% and 36% respectively. In another two cases, the Ret40f 2/13 (15)
percentage of CD34-positive cells was 3% and 9%. However, in
Blasts were negative for megakaryocytic markers (CD42b and/or CD61;
both these latter cases the ratio of myeloid to erythroid all cases), CD10 (six cases), CD68 (PGM-1; four cases) and lymphoid
precursors (M:E) was approximately 0Æ1. Hence, the markers (four cases).
percentage of CD34-positive cells among the myeloid
component would amount to >20%. In two AML-MD cases Figs 2–4). Megakaryocytic markers (CD42b and/or CD61; all
where the blasts were CD34-negative, the blasts were CD117+ cases), CD10 (six cases), CD68 (PGM-1; four cases) and
and the CD117+ cells amounted to 25% and 33% of all marrow lymphoid markers (four cases) were negative.
cells (Fig 1). In the two MDS fi AML cases, the blasts were
CD34+ and they amounted to 83% and 100% of the marrow Recognition and quantification of ‘dysplasia’ on trephine
cells. sections. Dysplasia in the erythroid lineage was always
associated with megaloblastoid features. This feature, which
Immunophenotype of the neoplastic precursor cells on trephine could be easily recognized on the H&E and Giemsa stained
sections. CD34 expression was noted in the blastic cells from sections, was confirmed on sections immunostained with
nine of eleven AML-MD cases. Among them, HLA-DR was ret40f. Other features of dyserythropoiesis, such as bi-/
positive in three of three cases and CD99 in one of one case. All multinucleation, abnormal lobation, nuclear budding and
four cases tested for Tdt were negative. Among the CD34- nuclear atypia were often observed. Seven of 11 AML-MD
positive cases, eight of nine expressed myeloperoxidase, six of cases showed evidence of dyserythropoiesis. The erythroid
seven were CD117-positive and two of two were CD68 (KP1)- population was too scanty to determine dyserythropoiesis
positive. In two of the CD34 positive cases, cells with blastic amongst the MDS fi AML cases.
morphology also strongly expressed ret40f. The proportion of Abnormalities in myeloid maturation manifested as matu-
CD34 positive cells and ret40f positive cells suggested co- ration arrest of myeloid cells (lack of mature myeloid cells),
expression in a good proportion of cells. Overall, these two increased layers of early myeloid cells surrounding the
cases had features compatible with acute erythroleukaemia. trabeculae and ALIP. Foci of ALIP, a feature that can be seen
Among the two CD34-negative CD117-positive AML-MD on H&E and Giemsa stained sections, was highlighted on
cases, blastic cells expressed HLA-DR and myeloperoxidase in immunostains for CD34, CD117 and myeloperoxidase. Nine of
one case and CD99, Tdt, CD68 (KP1) and myeloperoxidase in 11 AML-MD cases showed foci of ALIP of varying degrees of
the other case. prominence.
In both MDS fi AML cases, blasts were CD34+. One case Dysmegakaryopoiesis manifested as the variation in size of
also expressed CD68 (KP1), CD117 and Tdt (50% cells). The megakaryocytes and in nuclear lobation. Seven of 14 (50%)
other case expressed CD117 and myeloperoxidase (Table I; cases displayed hypolobated nuclei and seven of 14 (50%) cases

Proportion of precursor cells

AML-MD (11 cases) AML->MD (2 cases)

CD34+ Precursor cells

83% & 100%

Precursor cells >20%

CD34+ precursor cells CD34+ precursor cells
(7 cases)
16% & 14% in two cases amounted to
CD34+ - 4 cases
- aspirate blast count was 3% & 9% in two cases; in
CD117+ - 2 cases
23% & 36% these two cases M:E<0·1·
HLADR - 1 case

Fig 1. Quantifying blasts in BMTB samples. M:E, ratio of myeloid to erythroid precursors.

ª 2007 The Authors

Journal Compilation ª 2007 Blackwell Publishing Ltd, British Journal of Haematology, 140, 279–286 281
N. Ngo et al
H&E X200
CD42bX200
CD34X400
Fig 2. Marrow shows an increased proportion of blastic cells expressing myeloperoxidase, CD34 and CD117. The dysplastic features among
megakaryocytic cells and ALIP are easily recognized.
showed micromegakaryocytes. Micromegakaryocytes were bet-
ter recognized with CD42b or CD61 immunostains. All 11
AML-MD cases showed features of dysmegakaryopoiesis.
Dysmegakaryopoiesis was also apparent in both MDS fi AML
cases.
Among AML-MD cases, five cases showed trilineage
dysplasia and six showed bilineage dysplasia (dyserythropoi-
esis/abnormal myeloid maturation and dysmegakaryopoiesis).
In the MDS fi AML cases, one showed trilineage dysplasia
and one showed bilineage dysplasia (Figs 2–4).
Other features. Among AML-MD cases, the bone marrow
cellularity varied from 25% to 100% (69Æ5 ± 14Æ3%), and
the M:E ratio varied from 0Æ1 to 20 (5Æ7 ± 4Æ1). The
cellularity noted among MDS fi AML cases was 75% and
85% and the M:E ratio was >20 in both cases. Most of the
AML-MD cases (70%) showed mild to moderate reticulin
fibrosis, while both MDS fi AML cases showed marked
reticulin fibrosis. Apoptotic cells were seen in increased
numbers in 10 of 11 AML-MD cases and in both
MDS fi AML cases. Among the AML-MD cases, the
282 Journal Compilation ª 2007 Blackwell Publishing Ltd, British Journal of Haematology, 140, 279–286
H&E X60
Myeloperoxidase X400
CD117 X400
increase was mild in two, moderate in seven and marked
in one. The increase in aopototic cells among the two
MDS fi AML cases was mild.
Summary of bone marrow trephine features of previous biopsies
performed on patients with MDS fi AML. One patient had
trilineage dysplasia and the other patient had bilineage
dysplasia (myeloid and megakaryocytic) in BMTBs prior to
developing AML. The percentage of CD34-positive cells was
10% and 14% while the BMA blast count was 2% in both
cases. ALIP was observed in both cases. The M:E ratio (1:2) in
contrast to the higher M:E ratio (>20) seen in the BMTB with
AML. There was a marked increase in the numbers of
apoptotic cells in both cases (Fig 4).
Discussion
Acute myeloid leukaemia-MD and MDS fi AML have an
increased frequency of complex cytogenetic abnormalities,
deletions of chromosomes 5 or 7, trisomies, abnormalities of
3q21 and poor prognosis. The diagnosis of AML with
ª 2007 The Authors

H&E X100
H&E X400
CD117 X400
Fig 3. Marrow with trilineage dysplasia and a prominent erythroid component. Dysplastic features in erythroid cells and megakaryocytic cells are
highlighted by ret40f and CD42b stains. Blastic cells express CD117.
multilineage dysplasia in the WHO classification is based
mainly on the PB and BMA findings (Brunning et al, 2001). To
date, the utility of BMTB as a diagnostic tool in acute
leukaemias has been unclear. Islam et al (1985) observed that
the BMTB features could be useful adjuncts to BMA in the
diagnosis of AML. A later study by Horny et al (1990)
suggested that blasts in AML had a heterogeneous immuno-
phenotype. However, this was before the introduction of the
robust antibodies that are currently used on BMTBs. The
utility of BMTB cases with hypoplastic AML or where the
BMA is either inadequate or unsatisfactory (e.g. because of
increased reticulin fibrosis) is well recognized (Orazi, 2007).
The question of whether BMTB combined with immunohis-
tochemistry would be a useful adjunct to BMA combined with
flow cytometry needs to be addressed. The obvious advantage
with BMTB is the ability to review archived samples on later
occasions. With the availability of newer antibodies, samples
can be re-evaluated, especially at the time of disease relapse or
suspected relapse.
This study describes for the first time, the BMTB appearance
and immunohistochemistry results in a series of AML-MD and
ª 2007 The Authors
Journal Compilation ª 2007 Blackwell Publishing Ltd, British Journal of Haematology, 140, 279–286
Bone Marrow Trephines in AML with Multilineage Dysplasia
H&E X400
ret40f X400
CD42b X400
MDS fi AML cases. Identification of dysplastic features
among erythroid and megakaryocytic lineages is relatively easy
on BMTB, especially with the use of immunostains. Megalo-
blastoid change, bi-/multinucleation, irregular nuclear con-
tours/lobation are easily identifiable features especially with
ret40f immunostain. Similarly, recognition of the abnormal/
dysplastic megakaryocytic cells is far easier with immunostains
(CD42b or CD61) than with routine H&E or Giemsa stains.
On the other hand, classical features of dysmyelopoiesis, such
as small size, nuclear hypolobation (pseudo Pelger-Huet) and
hypersegmentation, hypogranularity and pseudo Chediak-
Higashi granules are difficult to recognize on sections.
However, there are other ‘markers’ for dysmyelopoiesis on
BMTB; these include maturation arrest in myeloid cells,
increased layers of early myeloid cells surrounding the
trabeculae and ALIP. Among the 13 cases in this study, six
showed trilineage dysplasia and seven showed bilineage
dysplasia. Dysmegakaryopoiesis was the constant feature.
Presence of an increased proportion of precursor cells and
ALIP can be identified both on routine H&E or Giemsa
stains, and also on immunostains – CD34, CD117 and
283
N. Ngo et al

H&E X200 H&E X400

H&E X400 CD42b X400

CD34 X200 CD34 X400

MDS AML
Fig 4. The left panels are from BMTB that showed features of MDS. There was trilineage dysplasia, but the CD34% was much <20%. The right panels
show the subsequent BMTB with AML – sheets of CD34+ blasts and persistent dysplastic megakaryocytes that are highlighted by CD42b are seen.

myeloperoxidase. ALIP was initially defined as a collection of
immature myeloid precursors in the central, intertrabecular
region of the BMTB section not in the vicinity of blood vessels
or endosteal surfaces (Tricot et al, 1984). While the utility of
CD34 and CD117 immunostains in identifying ALIP has been
previously reported, myeloperoxidase immunostain can also
be useful. Myeloperoxidase expression is often stronger in early
myeloid cells, for instance promyelocytes. Presence of clusters
of myeloperoxidase positive larger cells with fine chromatin or
nucleoli is a helpful feature to spot ALIP in BMTBs.
Prominence of ALIP is a well-recognised feature among
high-risk MDS and this prominence among cases of MDS
may help to identify cases that are in transformation to AML
(Orazi, 2007).
The main perceived advantage of the BMA is the
provision to perform accurate blast counts, the criterion
by which AML is defined. We have, however, demonstrated
that with the help of immunostains, the quantification of
leukaemic precursor cells as a fraction of all marrow cells or
of myeloid cells (when M:E ratio is low) is possible using
the BMTBs in most cases. This is best evaluated with CD34
and, in cases where leukaemic blasts are CD34 negative, with
CD117 or HLA-DR. Quantification in this manner helps us
to conform to the criteria required for diagnosing AML as
per the WHO classification. The importance of CD34 has
been previously explored in the classification and diagnosis
of MDS (Baur et al, 2000; Soligo et al, 1994; Torlakovic
et al, 2002). In comparison with flow cytometry, BMTB
represents marrow in its entirety, whereas aspirated marrow
often under-represents the paratrabecular areas. Further-
more, aspirated marrow is diluted with peripheral blood to
varying extents. Hence, immunohistochemical evaluation of
BMTB should provide a more accurate estimate of the
proportions of cells expressing particular antigens despite the
fact that the total numbers of cells scanned in flow
cytometry is far higher.

ª 2007 The Authors

284 Journal Compilation ª 2007 Blackwell Publishing Ltd, British Journal of Haematology, 140, 279–286

Care is essential in interpreting CD117 stain as
immunostaining of the myeloid precursors by CD117 is
often weak and as cells other than myeloid precursors in the
marrow are CD117 positive. Hence, quantification can be less
precise than with CD34. Apart from mast cells (which express
CD117 strongly), moderate intensity staining may also be
seen in a proportion of early erythroid cells. Cytological
assessment of the CD117+ population and correlation with
ret40f immunostained sections is essential to make these
distinctions. Other antibodies that can be used for quanti-
fication of myeloid precursors include CD99 (mic2) and Tdt
(Kang & Dunphy, 2006). It should be noted that, among the
blastic population, there can be heterogeneity in the intensity
of expression of some of these antigens like CD34 in some
cases. This was noted in three CD34-positive cases in this
series where the percentage of CD34-positive cells was 7%,
14% and 16%. Clearly the blast percentage was higher in
these cases; in one case estimation of the correct blast count
was aided by determining HLA-DR positive cells and in the
other two, only the aspirate blast count resolved the issue.
The current 20% criterion for the blast percentage is based
on BMA and PB evaluation. Well organized multi-institu-
tional studies can lead to defining ‘precursor’ cell percentage
criteria specifically for BMTBs. This would also be relevant to
cases with suboptimal BMA samples and with a lower
percentage of PB-blast.
Acute myeloid leukaemia is a heterogeneous entity, as are
the blasts in AML with respect to antigen expression. The
European Group for the Immunological Classification of
Leukaemias (EGIL) proposed an immunological classification
of acute leukaemias with the intention to clearly distinguish
AML from ALL on immunophenotypic grounds (Bene et al,
1995; Legrand et al, 2000). In this classification, AML was
defined immunologically by the expression of two or more of
the following myeloid markers: myeloperoxidase, CD13,
CD33, CDw65 and CD117. In another study, CD33, myelop-
eroxidase and CD13 were the most frequently expressed
myeloid antigens, followed by CD65s and CD15 (Thalham-
mer-Scherrer et al, 2002). In the current, relatively small series,
most cases of AML-MD and MDS fi AML had precursor cells
that were positive for CD34 (11 of 13), myeloperoxidase (11 of
13) and CD117 (10 of 11) by immunohistochemistry.
This study further highlights the overlap between AML-MD
and acute erythroleukaemia. However, as M:E ratio and CD34
percentage can be quantified on immunohistochemistry,
reaching a diagnosis of AML and recognition of dysplastic
features in more than one lineage is easier with BMTB than
BMA. In BMA, megakaryocytic abnormalities are often
underrepresented. Earlier studies have shown that cases
diagnosed as erythroleukaemia, including FAB-M6 type, were
often associated with myelodysplasia, with dysplastic features
seen in at least two lineages in as many as 86% of cases (Arber,
2001; Davey et al, 1995).
The increased numbers of apoptotic cells in BMTBs from
MDS patients was first described in 1990 (Clark & Lampert,
ª 2007 The Authors
Journal Compilation ª 2007 Blackwell Publishing Ltd, British Journal of Haematology, 140, 279–286
Bone Marrow Trephines in AML with Multilineage Dysplasia
1990). Later, studies using in-situ end labelling confirmed an
increased proportion of apoptotic cells in MDS marrow
samples (Mundle et al, 1994; Raza et al, 1995). In the current
study, a marked increase in the numbers of apoptotic cells was
documented in two MDS samples (those prior to development
of AML). Furthermore, increased numbers of apoptotic cells
were seen in most AML-MD (nine of 11) and both
MDS fi AML cases, though to a degree lesser than MDS.
Our findings are contrary to those of Thiele et al (1996) who
did not report an increase in the proportion of apoptotic cells
among any of the FAB-AML categories. It is possible that the
increased numbers of apoptotic cells is a feature particular to
AML-MD and MDS fi AML cases, signifying the ‘dysplastic’
relationship to MDS. Among the two cases where sequential
samples were available, BMTB representing MDS had a much
higher apoptotic rate when compared with subsequent AML
samples. Our study also suggests that increased marrow
reticulin is common among AML-MD and this is likely to
result in a poor quality aspirate, highlighting the importance of
performing BMTBs and adequately work-up of the trephines.
In conclusion, we have demonstrated the utility of BMTB
and immunohistochemistry as an additional aid in the
diagnosis of AML-MD and AML fi MDS. Quantifying the
proportion of precursor cells as a fraction of all marrow cells
and of myeloid cells, identifying the phenotype of the
precursor cells and identifying dysplastic features in all three
haematopoietic lineage is possible on BMTB. While the
advantage of BMA is the determination of an accurate blast
count and recognition of dysmyelopoeisis, BMTB scores better
in identifying dysplastic features among erythroid precursors
and megakaryocytic cells, and accurately quantifying precursor
cells defined by a phenotype such as CD34 or CD117.
References
Arber, D.A. (2001) Realistic pathologic classification of acute myeloid
leukemias. American Journal of Clinical Pathology, 115, 552–560.
Arber, D.A., Stein, A.S., Carter, N.H., Ikle, D., Forman, S.J. & Slovak,
M.L. (2003) Prognostic impact of acute myeloid leukemia classifi-
cation. Importance of detection of recurring cytogenetic abnor-
malities and multilineage dysplasia on survival. American Journal of
Clinical Pathology, 119, 672–680.
Bao, L., Wang, X., Ryder, J., Ji, M., Chen, Y., Chen, H., Sun, H., Yang,
Y., Du, X., Kerzic, P., Gross, S.A., Yao, L., Lv, L., Fu, H., Lin, G. &
Irons, R.D. (2006) Prospective study of 174 de novo acute mye-
logenous leukemias according to the WHO classification: subtypes,
cytogenetic features and FLT3 mutations. European Journal of
Haematology, 77, 35–45.
Baur, A.S., Meuge-Moraw, C., Schmidt, P.M., Parlier, V., Jotterand, M.
& Delacretaz, F. (2000) CD34/QBEND10 immunostaining in bone
marrow biopsies: an additional parameter for the diagnosis and
classification of myelodysplastic syndromes. European Journal of
Haematology, 64, 71–79.
Bene, M.C., Castoldi, G., Knapp, W., Ludwig, W.D., Matutes, E.,
Orfao, A. & van’t Veer, M.B. (1995) Proposals for the immuno-
logical classification of acute leukemias. european group for the
285
N. Ngo et al
immunological characterization of leukemias (EGIL). Leukemia, 9,
1783–1786.
Bennett, J.M., Catovsky, D., Daniel, M.T., Flandrin, G., Galton, D.A.,
Gralnick, H.R. & Sultan, C. (1976) Proposals for the classification of
the acute leukaemias. French–American–British (FAB) Co-operative
Group. British Journal of Haematology, 33, 451–458.
Brunning, R.D., Matutes, E., Harris, N.L., Flandrin, G., Vardiman, J.,
Bennett, J. & Head, D. (2001) Acute myeloid leukaemia with mul-
tilineage dysplasia. In: World Health Organization Classification of
Tumours. Pathology & Genetics. Tumours of Haematopoietic and
Lymphoid Tissues (ed. by E.S. Jaffe, N.L. Harris, H. Stein & J.W.
Vardiman), pp. 88–89. IARC Press, Lyon.
Clark, D.M. & Lampert, I.A. (1990) Apoptosis is a common histo-
pathological finding in myelodysplasia: the correlate of ineffective
haematopoiesis. Leukemia and Lymphoma, 2, 415–418.
Davey, F.R., Abraham, N., Jr, Brunetto, V.L., MacCallum, J.M., Nelson,
D.A., Ball, E.D., Griffin, J.D., Baer, M.R., Wurster-Hill, D. & Mayer,
R.J. (1995) Morphologic characteristics of erythroleukemia (acute
myeloid leukemia; FAB-M6): a CALGB study. American Journal of
Hematology, 49, 29–38.
Gahn, B., Haase, D., Unterhalt, M., Drescher, M., Schoch, C., Fonatsch,
C., Terstappen, L.W., Hiddemann, W., Buchner, T., Bennett, J.M. &
Wormann, B. (1996) De novo AML with dysplastic hematopoiesis:
cytogenetic and prognostic significance. Leukemia: Official Journal of
the Leukemia Society of America, Leukemia Research Fund, United
Kingdom, 10, 946–951.
Goasguen, J.E., Matsuo, T., Cox, C. & Bennett, J.M. (1992) Evaluation
of the dysmyelopoiesis in 336 patients with de novo acute myeloid
leukemia: major importance of dysgranulopoiesis for remission and
survival. Leukemia, 6, 520–525.
Horny, H.P., Campbell, M., Steinke, B. & Kaiserling, E. (1990) Acute
myeloid leukemia: immunohistologic findings in paraffin-embedded
bone marrow biopsy specimens. Human Pathology, 21, 648–655.
Islam, A., Catovsky, D., Goldman, J.M. & Galton, D.A. (1985) Bone
marrow biopsy changes in acute myeloid leukaemia. I: observations
before chemotherapy. Histopathology, 9, 939–957.
Kang, L.C. & Dunphy, C.H. (2006) Immunoreactivity of MIC2 (CD99)
and terminal deoxynucleotidyl transferase in bone marrow clot and
core specimens of acute myeloid leukemias and myelodysplastic
syndromes. Archives of Pathology & Laboratory Medicine, 130,
153–157.
Legrand, O., Perrot, J.Y., Baudard, M., Cordier, A., Lautier, R., Si-
monin, G., Zittoun, R., Casadevall, N. & Marie, J.P. (2000) The
immunophenotype of 177 adults with acute myeloid leukemia:
proposal of a prognostic score. Blood, 96, 870–877.
Manoharan, A., Horsley, R. & Pitney, W.R. (1979) The reticulin
content of bone marrow in acute leukaemia in adults. British Journal
of Haematology, 43, 185–190.
286 Journal Compilation ª 2007 Blackwell Publishing Ltd, British Journal of Haematology, 140, 279–286
Mundle, S., Iftikhar, A., Shetty, V., Dameron, S., Wright-Quinones, V.,
Marcus, B., Loew, J., Gregory, S. & Raza, A. (1994) Novel in situ
double labeling for simultaneous detection of proliferation and
apoptosis. The Journal of Histochemistry and Cytochemistry, 42,
1533–1537.
Naresh, K.N. & Lampert, I.A. (2006) CD117 expression as an aid to
identify immature myeloid cells and foci of ALIP in bone marrow
trephines. American Journal of Hematology, 81, 79.
Naresh, K.N., Lampert, I., Hasserjian, R., Lykidis, D., Elderfield, K.,
Horncastle, D., Smith, N., Murray-Brown, W. & Stamp, G.W.
(2006) Optimal processing of bone marrow trephine biopsy: the
hammersmith protocol. Journal of Clinical Pathology, 59, 903–911.
Orazi, A. (2007) Histopathology in the diagnosis and classification of
acute myeloid leukemia, myelodysplastic syndromes, and myelo-
dysplastic/myeloproliferative diseases. Pathobiology: Journal of
Immunopathology, Molecular and Cellular Biology, 74, 97–114.
Raza, A., Gezer, S., Mundle, S., Gao, X.Z., Alvi, S., Borok, R., Rifkin, S.,
Iftikhar, A., Shetty, V. & Parcharidou, A. (1995) Apoptosis in bone
marrow biopsy samples involving stromal and hematopoietic cells in
50 patients with myelodysplastic syndromes. Blood, 86, 268–276.
Soligo, D.A., Oriani, A., Annaloro, C., Cortelezzi, A., Calori, R., Po-
zzoli, E., Nosella, D., Orazi, A. & Deliliers, G.L. (1994) CD34
immunohistochemistry of bone marrow biopsies: prognostic sig-
nificance in primary myelodysplastic syndromes. American Journal
of Hematology, 46, 9–17.
Tamura, S., Takemoto, Y., Wada, H., Itoh, T., Mori, A., Saheki, K.,
Okada, M., Takatsuka, H., Fujimori, Y., Okamoto, T. & Kakishita, E.
(1998) Significance of trilineage myelodysplasia in de novo acute
myeloid leukaemia during remission rather than at diagnosis. British
Journal of Haematology, 101, 743–748.
Thalhammer-Scherrer, R., Mitterbauer, G., Simonitsch, I., Jaeger, U.,
Lechner, K., Schneider, B., Fonatsch, C. & Schwarzinger, I. (2002)
The immunophenotype of 325 adult acute leukemias: relationship to
morphologic and molecular classification and proposal for a mini-
mal screening program highly predictive for lineage discrimination.
American Journal of Clinical Pathology, 117, 380–389.
Thiele, J., Lorenzen, J., Zirbes, T.K., Kvasnicka, H.M., Schilling, C.,
Schwonzen, M., Diehl, V. & Fischer, R. (1996) Apoptosis in acute
myeloblastic leukemia: follow-up study on trephine biopsies of the
bone marrow. Leukemia and Lymphoma, 22, 77–82.
Torlakovic, G., Langholm, R. & Torlakovic, E. (2002) CD34/QBEND10
immunostaining in the bone marrow trephine biopsy: a study of
CD34-positive mononuclear cells and megakaryocytes. Archives of
Pathology and Laboratory Medicine, 126, 823–828.
Tricot, G., De Wolf-Peeters, C., Vlietinck, R. & Verwilghen, R.L. (1984)
Bone marrow histology in myelodysplastic syndromes. II. Prognostic
value of abnormal localization of immature precursors in MDS.
British Journal of Haematology, 58, 217–225.
ª 2007 The Authors