JMD 2000, Vol. 2, No. 4
Copyright © 2000 American Society for Investigative Pathology & Association for Molecular Pathology
Severe Chromosomal Aberrations in Pleural Mesotheliomas with Unusual Mesodermal Features
Comparative Genomic Hybridization Evidence for a Mesothelioma Subgroup
Michael Krismann,
Klaus-Michael Müller,
Malgorzata Jaworska and
Georg Johnen
From the Institute of Pathology, Professional Associations’ Clinic Bergmannsheil Bochum, University Clinic, Bochum, Germany
 |
Abstract
|
|---|
Malignant mesotheliomas are tumors known for their extensive
heterogeneity. Apart from the three classical patterns,
predominantly epithelioid, sarcomatoid, and
biphasic, some rare variants do exist. In some cases,
one can find uncommon additional mesodermal tumor components. These
tumors have previously been called "mesodermomas" and, like
regular mesotheliomas, are usually associated with a previous
asbestos exposure. We examined eight cases of mesodermomas by light
microscopy, immunohistochemistry and comparative genomic
hybridization (CGH). Besides biphasic and epithelioid areas,
unusual epithelial, chondroid, osseous, or even
angioblastic elements may be found to varying degrees.
Immunohistochemical analysis shows similar staining results as with
regular mesotheliomas. CGH reveals a high number of chromosomal
imbalances (16.5 per case; range, 11–27). In 10 classical
biphasic mesotheliomas that served as a control, defects of
comparable number and severity could not be detected (8 per case;
range, 2–16). The most frequent defects of mesodermomas
(losses on 1p, 4pq, 9p, 13q,
14q, and gains on 1q and 15q), however, could
also be found in mesotheliomas of the classical type. Thus, our
results support the classification of the so-called mesodermomas as a
separate tumor subgroup while maintaining the relationship to the
classical mesotheliomas. Therefore, we propose to use the term
mesodermoma for this subgroup.
|
Introduction
|
|---|
Every pathologist who is regularly confronted with mesotheliomas
knows the broad histomorphological spectrum of these tumors. In the
classical types of predominantly epithelioid, sarcomatoid, biphasic, or
poorly differentiated histological appearance, the diagnosis is not
very difficult to obtain if the panel of evaluated antibodies shows
unambiguous results. Nevertheless, a few unusual variants of
mesotheliomas do exist, like the desmoplastic, lymphohistiocytic, small
cell, or deciduoid peritoneal mesothelioma.1
However, in
some instances one may encounter histological structures that fit into
neither the patterns of classical mesotheliomas nor those of the
variants. These may be small cell solid areas, epithelial areas with
features of squamous epithelium, or uncommon mesenchymal components
like bone or cartilage
formation.In that context, the term mesodermoma was introduced in 1981 by Donna
and Betta for neoplasias that could be derived from undifferentiated
pluripotent mesoderm.2
It has never become popular; a
recent Medline search request returned only three articles citing the
term.2, 3, 4
Donna and Betta postulated that the
mesothelium, which lines the serous membranes, was of mesodermal
origin. The mesodermal epithelioid structure is supposed to persist
during the postembryonic phase. Structures that can be derived from the
mesoderm are connective tissue, cartilage, bone, muscles, cells of
blood and lymphatics, vessel walls, kidneys, gonads, adrenal cortex,
spleen, and serous membranes like the parietal and visceral pleura.
Tumors derived from the mesoderm can therefore exhibit a broad spectrum
of possible directions of differentiation. This could explain the
myoblastic, angioblastic, lymphoblastic, chondroblastic, osteoblastic,
fibroblastic, and uncommon epithelial differentiation of primary
neoplasias of the pleura. Thus, so-called mesodermomas seem to be a
variant type of mesotheliomas with additional characteristics of
differentiation. Among more than 500 cases of mesotheliomas observed
annually in the German Mesothelioma Registry, only one or two can be
classified as mesodermomas. Because mesodermomas are not commonly
known, they are rarely diagnosed. So far, no immunohistochemical or
genetic characterization is available.
Our aim was to investigate if there are any morphological,
immunohistochemical, or molecular clues to justify the classification
of mesotheliomas with uncommon mesodermal structures as a subgroup
within the mesotheliomas.
|
Materials and Methods
|
|---|
We have investigated eight cases of pleural mesotheliomas with
obvious uncommon mesenchymal or epithelial structures by
immunohistochemistry and comparative genomic hybridization (CGH). The
CGH is a well established method to screen tumor tissues for
chromosomal imbalances.5, 6
In comparison, we have analyzed
a series of 10 biphasic mesotheliomas. Patient data of both tumor
groups and the panel of antibodies are summarized in Tables 1
and 2
.
Immunohistochemical Staining
For the diagnosis of mesotheliomas we used an antibody panel that
has been proven to be reliable in more than a thousand mesotheliomas in
the German Mesothelioma Registry. This panel is consistent with
literature data.7, 8, 9
It consists of keratin AE1/AE3 (DAKO,
Hamburg, Germany, 1:800, protease pretreatment), calretinin (SWant,
Belliazoua, Switzerland, polyclonal, 1:3000, microwave
pretreatment, citrate buffer, pH 6.0), human epithelial antigen
(BerEP4, DAKO), vimentin (DAKO, 1:4000, microwave
pretreatment, citrate buffer, pH 8.0), and additionally Ki-67 (MIB-1,
Dianova, Hamburg, Germany, 1:800, microwave pretreatment, EDTA buffer,
pH 8.0). For differential diagnosis of predominantly sarcomatoid
mesotheliomas, antibodies used in soft tissue pathology have to be
applied. For this series, we have additionally tested smooth muscle
actin (SMA, Immunotech, monoclonal). In one case, thyreoglobulin (DAKO,
monoclonal, 1:50) had to be applied for differential diagnosis. All
immunohistochemical analyses were performed with a DAKO Techmate 500
immunostainer using the alkaline phosphatase-anti-alkaline phosphatase
method. Moreover, each tumor specimen was previously tested
histochemically for periodic acid-Schiff (PAS) with and without
diastase pretreatment.
CGH
In all cases, we used paraffin-embedded tissues. Twenty sections
of 10 µm each were cut. The selected tissue was microdissected by
using a scalpel. The genomic DNA was extracted and purified with the
help of a commercially available kit (QIAamp Tissue Kit, Qiagen,
Hilden, Germany). Similarly, reference DNA was isolated from normal
human peripheral blood with a QIAamp Blood Kit (Qiagen). A slightly
modified standard protocol for the hybridization was
used,10
similar to that published by Petersen et
al.11
Labeling of reference and tumor DNA was performed by
nick translation with digoxigenin-11-dUTP or biotin-16-dUTP (Roche,
Mannheim, Germany), respectively. The probe length, checked by agarose
gel electrophoresis, was between 300 and 3000 bp. Typically, 0.9 µg
labeled reference DNA and 0.9 µg labeled tumor DNA were hybridized to
normal human metaphase chromosomes (Vysis, Downers Grove, IL) for 3
days in the presence of 30 µg human Cot-1 DNA (Roche) and 10 µg
herring sperm DNA (Promega, Madison, WI). Staining of the hybridized
metaphase chromosomes was performed with 3.2 µg/ml
anti-digoxigenin-rhodamine (Roche) and 16 µg/ml fluorescein-avidin
(Vector Labs, Burlingame, CA). DAPI served as a counterstain for
karyotyping. Each tumor sample was labeled, hybridized, and stained at
least twice. The fluorescence images were recorded with a cooled 12-bit
charged coupled device camera (SensiCam; PCO, Kelheim, Germany) on an
Axiophot fluorescence microscope (Zeiss, Jena, Germany). Images
were processed and evaluated with the program Quips (Vysis) on a
Macintosh G3 PowerPC (Apple, Cupertino, CA). Depending on the quality
of a hybridization, 10 to 15 metaphases per case were analyzed and the
results averaged. A 99% confidence interval was calculated and is
depicted together with the average profile for each chromosome. The
threshold values of the fluorescence ratio fluorescein/rhodamine were
set to 0.80 and 1.20 for losses and gains, respectively. Losses are
shown as vertical red bars on the left side of a chromosome ideogram;
gains are shown as green bars on the right (Figure 2)
. As positive
control, we used a tumor cell line with known defects (MPE 600, Vysis).
Five samples from healthy tissues and a case of pleuritis of a patient
without any known malignancies served as negative controls. In
addition, we used reverse labeling of reference and tumor DNA (with
fluorescein and rhodamine, respectively) to confirm the CGH results of
the mesodermomas.
View larger version (25K):
[in this window]
[in a new window]
|
Figure 2. CGH results. Fluorescence ratio profile of case M3a, 66-year-old woman
without known history of asbestos exposure. High number of genetic
imbalances. Losses are represented as red bars on the left side of each
chromosome ideogram, gains are represented as green bars on the right
side. The reference DNA was from a male donor. Therefore, the X
chromosome shows a gain and the Y chromosome a loss.
|
|
|
Results
|
|---|
Immunohistochemistry
The histomorphological heterogeneity of the studied malignant
mesotheliomas with uncommon mesodermal histoarchitectural features
(Figure 1A
and 1B)
is reflected in varying immunohistochemical results (Table 2)
. Keratin shows a positive reaction with the solid pseudoglandular
and partially with the sarcomatoid areas. Vimentin is positive in
fusiform cells as well as in a varying number of epithelioid cells.
Calretinin is moderately to strongly positive in the epithelioid areas,
but in some cases the pattern of expression is restricted to a few
areas (Figure 1C)
. The spindle cell areas show only focally a positive
reaction with calretinin, as it is found in common mesotheliomas. HEA
is completely negative in all cases. Thyreoglobulin shows no positive
immunoreactivity in case M1, which exhibits thyroid-like epithelial
features (Figure 1A)
. Interestingly, one of the so-called mesodermomas
shows a positive reaction with the SMA antibody not only in the
sarcomatoid areas, but also in the epithelioid parts (Figure 1D)
. A
positive actin stain in spindle cell areas can be observed in eight of
the ten biphasic mesotheliomas of the control group, but in none of
their epithelioid areas. The proliferative activity, as measured by
expression of the Ki-67 antigen (MIB-1), is somewhat higher than in
common types of mesotheliomas.
View larger version (133K):
[in this window]
[in a new window]
|
Figure 1. Morphological findings with mesodermomas. A: Pleural
mesodermoma, 60-year-old man with a known history of asbestos exposure
(case M1). Epithelial
follicular tumor area resembling thyroid gland tissue. Thyreoglobulin
was not detectable immunohistochemically
(H&E). B: Same
case as presented in A. Chondroid area of the tumor that was
included in the CGH analysis
(H&E).
Immunohistochemical results. C: Calretinin staining.
Heterogeneous staining pattern with positive reaction in the large
pleomorphic cells, but negative result in the small cell areas
(case M8). D:
Smooth muscle actin staining with positive reaction in the small cell
portion of the tumor, the stromal cells being negative
(case M8).
|
|
CGH
Detailed results of the CGH analyses are depicted in Tables
3and 4, a comparison of both tumor groups is summarized in Table 5
, and an overview is given in the two "superkaryograms"
(superposition of all CGH profiles of each tumor group) in Figure 3
.
Reverse labeling confirmed all major and most of the other defects.
Each of the eight cases of mesodermomas shows a remarkably high genetic
instability. An example of a single CGH profile is given in Figure 2
. Inverse labeling shows reproducible results. We have found 16.5
defects per case on average, with a range between 11 and 27 defects per
case. Losses were more common than gains (11.3 on average; range,
8–15). Gains of chromosomal material could be detected in 5.2
locations per case, with a range between zero and 13. Recurrent losses
were found in all cases in regions of chromosome arms 1p, 4q, and 9p
(100%). Further frequent losses were localized on sections of
chromosomes 14q, 4p, 13q (75%), 15q (63%), 6q, and 18q (50%), as
well as 3p, 3q, 10q, 21q, and 22q (38%). In all, we have found 14
different recurrent (ie, occurring in >30% of the tumors) losses of
chromosomal material in the tumors selected for this study (Table 5)
.
Recurrent gains of DNA material were also detected. Chromosome 15q was
affected in 88% of the cases, followed by chromosomes 1q (63%), 7q,
and 11q (38%). In all, four recurrent gains were found. Thus, an
overall number of 18 different recurrent defects could be observed.
View larger version (22K):
[in this window]
[in a new window]
|
Figure 3. Summary of CGH results. Defects are represented in superkaryograms by
superposition of all CGH profiles of each tumor group. A:
Chromosomal imbalances of eight mesodermomas. B: Chromosomal
imbalances of 10 biphasic mesotheliomas. Losses are shown as vertical
lines on the left side of a chromosome, gains are shown on the right
side.
|
|
On the contrary, in the group of biphasic mesotheliomas (Table 4)
that
served as a control, we only found five defects that occurred in more
than 30% of the cases. The most frequent defects were losses on 22q
(70%) and 9p (60%) and different gains on chromosome 7 (40 to 50%).
The average number of total defects per case was 8, with 5.4 losses and
2.6 gains per case. The locations of the chromosomal defects were
similar in both groups (Table 5
, Figure 3
).
For our CGH analysis, we have excluded possible artifact regions,
especially for gains, on chromosomes 1p33-pter, 9q34, 16, 17, 19, 20,
and 22. (For discussion of artifact regions, see Larramendy et
al.12
)
|
Discussion
|
|---|
The immunohistochemical spectrum of the eight tested tumors with
uncommon mesodermal histoarchitectural features (so-called
mesodermomas) supports their histogenetic relationship to the
mesotheliomas. A positive calretinin stain is typical for mesotheliomas
as well as mesodermomas. At the moment, a positive calretinin stain in
combination with a negative histochemical PAS staining after diastase
pretreatment, a staining positive for vimentin and negative for HEA is,
based on the experiences at the German Mesothelioma Registry, the best
differential diagnostic panel to distinguish between mesotheliomas and
metastases of adenocarcinomas. Actin is commonly found in both biphasic
mesotheliomas and mesodermomas. Antibodies to keratin and Ki-67 also
showed similar results in both tumor groups. Thus, the two groups
cannot be distinguished by immunohistochemical analysis alone.
In contrast, our CGH results support, based on the markedly high number
of chromosomal aberrations (16.5 defects per case), the unique position
of the mesodermomas within primary malignant neoplasias of the serous
membranes. None of the mesodermomas had fewer than 11 defects, whereas
only two of the biphasic mesotheliomas had 11 or more defects (average,
8 defects per case). On the other hand, the classification of the
mesodermomas as part of the mesotheliomas is also documented by the CGH
analysis, because several corresponding defects were demonstrated. The
recurrent defects that we found both in mesotheliomas and mesodermomas
were located on chromosomes 1p, 4p, 4q, 6q, 9p, 13q, 14q, and 22q
(losses) as well as 1q, 7q, and 15q (gains). On the contrary, losses on
chromosome 15q could be detected only in the mesodermoma group. Whereas
most defects showed a higher frequency in mesodermomas, losses on 22q
and gains on chromosome 7 were more prominent in the biphasic
mesotheliomas. In general, defects in mesodermomas appeared more
pronounced and tended to affect whole chromosome arms. The consensus
regions of the defects in the two tumor groups were similar, but not
always identical. This might in part also reflect the lower number of
total defects derived from the investigated biphasic mesotheliomas and
could change somewhat with an increasing numbers of cases. We also
compared our results to published CGH studies of
mesotheliomas.6, 13, 14, 15, 16, 17
The known defects that have been
found in more than one CGH study on malignant mesotheliomas are also
found in our series of mesodermomas, like losses on consensus
chromosome regions 1p21–22, 4cen-p15.3, 4cen-q24, 4q33-qter, 6q16–22,
9p21–22, 10q23, 13q21–22, 14q21–24, and 22q, as well as gains on
chromosome arms 1q and 15q (compare Table 5
). A well defined loss on
the upper parts of chromosome arm 15q (around 15q13) that only had been
described once for cultured mesothelioma cell
lines13
can be seen in five of the eight presented cases.
Interestingly, we found a frequent combination of defects in the
mesodermomas on 15q: the loss at 15q13 and an additional gain of
material on the lower part of 15q (15q21–26) was found in four of the
eight cases (50%). This gain on 15q was in all four cases a high-level
amplification (ratio = 1.5), one of these with a ratio of 2.0
(case M2). Besides frequent losses on 15q13 and 18q22-qter, cultured
mesothelioma cell lines also shared a higher total number of genetic
changes with the mesodermomas. However, we were not able to detect
recurrent gains on 5p and losses on 8p that were found by two
independent groups in the mesothelioma cell lines.13, 16
We also examined different areas of the same tumors. For example, in a
heterogenic portion of mesodermoma case 3 (M3a, Table 3
) the gains were
located on different chromosomal regions than in a more epithelioid
area (M3b) or a desmoplastic area of low cellularity (M3c), whereas
most losses were similar. In another case (M7), hybridizations of
different tumor areas of similar histology (M7a and M7b) showed a few
differences in the localization of the losses but identical gains. This
reflects the intratumor heterogeneity and reminds not to overestimate
hybridization results obtained from small areas of large tumors in view
of chromosomal alterations that may be involved in the molecular
pathogenesis of tumors. However, the chromosomal position of the most
frequent defects can serve as a guide for further genomic localization
of candidate genes. In total, we found a relatively high degree of
similarity of genetic defects within the collection of mesodermomas.
The remarkable genetic patterns were also mirrored in the histological
appearance.
In conclusion, nearly 20 years after Donna and Betta’s more
general definition of mesodermomas, we propose to use this term for the
described mesothelioma subclass. We suggest using this term for four
reasons. First, we have documented that the discussed tumors are
sufficiently distinct to be regarded as a special subgroup of
mesotheliomas. Second, the term mesodermoma reflects the embryological
background and makes sense in view of the pathogenesis of these highly
heterogenic tumors. Third, the rather bulky and impractical term
"mesothelioma with uncommon mesodermal histoarchitectural features"
should be avoided. Fourth, the term "mesoblastoma," though it would
also be appropriate for description, could be confused with the
mesoblastic nephroma. This renal tumor was first reported in the field
of pediatric pathology in 1967.18
In summary, we propose to diagnose a tumor as mesodermoma if (i) the
histological features of a primary tumor of the serous membranes
include uncommon patterns of mesodermal differentiation like
myoblastic, chondroblastic, osteoblastic, or uncommon epithelial
patterns; (ii) immunohistochemical analysis shows a positive result for
keratins, vimentin, calretinin, and negative results for CEA and HEA
(Ber-EP4) as well as a negative histochemical result for the PAS
staining after diastase pretreatment; and (iii) if a markedly increased
number of genetic imbalances by the use of molecular genetic methods
like CGH can be demonstrated.
The presented analytical tools should improve diagnosis of these
mesodermomas and help to further our knowledge about the development of
malignant tumors of the serous membranes.
|
Footnotes
|
|---|
Address reprint requests to Dr. Michael Krismann, Institute of Pathology, Professional Associations‘ Clinic Bergmannsheil Bochum, University Clinic, Bürkle-de-la-Camp-Platz 1, D-44789 Bochum, Germany. E-mail: patho-bhl{at}ruhr-uni-bochum.de
Accepted for publication July 28, 2000.
|
References
|
|---|
-
Battifora H, McCaughey WTE: Tumors of the Serosal Membranes. Edited by Rosai J and Sobin LH. Washington, DC, Armed Forces Institute of Pathology, 1995, pp 41–52
-
Donna A, Betta PG: Mesodermomas: a new embryological approach to primary tumours of coelomic surfaces. Histopathology 1982, 5:31-44
-
Donna A, Betta PG, Bianchi V, Ribotta M, Bellingeri D, Robutti F, Marchesini A: A new insight into the histogenesis of "mesodermomas: " malignant mesotheliomas. Histopathology 1991, 19:239-244[Medline]
-
Donna A, Betta PG: Differentiation towards cartilage and bone in a primary tumour of pleura: further evidence in support of the concept of mesodermoma. Histopathology 1986, 10:101-108[Medline]
-
Hermsen MAJA, Meijer GA, Baak JPA, Joenje H, Waldbloomers JJM: Comparative genomic hybridization: a new tool in cancer pathology. Hum Pathol 1996, 27:342-349[Medline]
-
Knuutila S, Aalto Y, Autio K, Björkqvist AM, El-Rifai W, Hemmer S, Huhta T, Kettunen E, Kiuru-Kuhlefelt S, Larramendy ML, Lushnikova T, Monni O, Pere H, Tapper J, Tarkkanen M, Varis A, Wasenius VM, Wolf M, Zhu Y: DNA copy number losses in human neoplasms. Am J Pathol 1999, 155:683-694[Abstract/Free Full Text]
-
Ordonez NG: Role of immunohistochemistry in differentiating epithelial mesothelioma from adenocarcinoma. Am J Clin Pathol 1999, 112:75-89[Medline]
-
Riera JR, Astengo-Osuna C, Longmate JA, Battifora H: The immunohistochemical diagnostic panel for epithelial mesothelioma: a reevaluation after heat-induced epitope retrieval. Am J Surg Pathol 1997, 21:1409-1419[Medline]
-
Wiethege T, Philips S, Müller KM: Calretinin als Marker zur Differentialdiagnose von Mesotheliomen und Adenokarzinomen. Verh Dtsch Ges Pathol 1997, 81:417 (abstr)
-
Krismann M, Adams H, Jaworska M, Müller KM, Johnen G: Patterns of chromosomal imbalances in benign solitary fibrous tumours of the pleura. Virchows Arch 2000, 437:248-255[Medline]
-
Ried T, Petersen I, Holtgreve-Grez H, Speicher MR, Schroeck E, du Manoir S, Cremer T: Mapping of multiple DNA gains and losses in primary small cell lung carcinomas by comparative genomic hybridization. Cancer Res 1998, 54:1801-1806[Abstract/Free Full Text]
-
Larramendy ML, El-Rifai W, Knuutila S: Comparison of fluorescein isothiocyanate- and Texas red-conjugated nucleotides for direct labeling in comparative genomic hybridization. Cytometry 1998, 31:174-179[Medline]
-
Balsara BR, Bell DW, Sonoda G, de Rieuzo A, du Manoir S, Jhanwar SC, Testa JR: Comparative genomic hybridization and loss of heterozygosity analyses identify a common region of deletion at 15q11.1-15 in human malignant mesothelioma. Cancer Res 1999, 59:450-454[Abstract/Free Full Text]
-
Björkqvist AM, Tammilehto L, Nordling S, Nurminen M, Anttila S, Mattson K, Knuutila S: Comparison of DNA copy number changes in malignant mesothelioma, adenocarcinoma and large-cell anaplastic carcinoma of the lung. Br J Cancer 1998, 77:260-269[Medline]
-
Björkqvist AM, Tammilehto L, Anttila S, Mattson K, Knuutila S: Recurrent DNA copy number changes in 1q, 4q, 6q, 9p, 13q, 14q and 22q detected by comparative genomic hybridization in malignant mesothelioma. Br J Cancer 1997, 75:523-527[Medline]
-
Kivipensas P, Björkqvist AM, Karhu R, Pelin K, Linnainmaa K, Tammilehto L, Mattson K, Kallioniemi QP, Knuutila S: Gains and losses of DNA sequences in malignant mesothelioma by comparative genomic hybridization. Cancer Genet Cytogenet 1996, 89:7-13[Medline]
-
Knuutila S, Armengol G, Björkqvist AM, El-Rifai W, Larramendy ML, Monni O, Szymanska J: Comparative genomic hybridization study on pooled DNAs from tumors of one clinical-pathological entity. Cancer Genet Cytogenet 1998, 100:25-30[Medline]
-
Bolande R, Brough A, Izant RJ: Congenital mesoblastic nephroma of infancy: a report of eight cases and the relationship to Wilms’ tumor. Pediatrics 1967, 40:272-278[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
I. Opitz, A. Soltermann, M. Abaecherli, M. Hinterberger, N. Probst-Hensch, R. Stahel, H. Moch, and W. Weder
PTEN expression is a strong predictor of survival in mesothelioma patients
Eur. J. Cardiothorac. Surg.,
March 1, 2008;
33(3):
502 - 506.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A Scattone, A Pennella, M Gentile, M Musti, P Nazzaro, A L Buonadonna, A Marzullo, D Cavone, L Pollice, and G Serio
Comparative genomic hybridisation in malignant deciduoid mesothelioma
J. Clin. Pathol.,
July 1, 2006;
59(7):
764 - 769.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
