International Journal of Paleopathology 21 (2018) 56–63

Contents lists available at ScienceDirect

International Journal of Paleopathology

journal homepage: www.elsevier.com/locate/ijpp

Research article

Dry bone histology of bone tumours夽

H.H. (Hans) de Boer a,b,∗ , G.J.R. (George) Maat c
a
Dept. of Pathology, Academic Medical Center, Amsterdam, The Netherlands
b
Barge’s Anthropologica—Amsterdam, Dept. of Anatomy, Embryology and Physiology, Academic Medical Center, Amsterdam, The Netherlands
c
Barge’s Anthropologica—Leiden, Dept. of Anatomy and Embryology, Leiden University Medical Center, Leiden, The Netherlands

a r t i c l e i n f o a b s t r a c t

Article history: This article focuses on the application of dry bone histology in the diagnosis of a series of different bone
Received 28 June 2016 tumours. It provides a short introduction on bone tumour classification and how tumours may affect the
Received in revised form 9 November 2016 skeleton. To illustrate the value of dry bone histology in the diagnostic process we studied the ‘fresh’ and
Accepted 25 November 2016
‘dry’ bone histology of a series of well-documented, recent clinical cases of various benign and malignant
Available online 6 December 2016
bone tumours.
We show that histology is a valuable instrument to assess bone tissue architecture, which provides
Keywords:
information on the biological behaviour of a tumour. Though histology may reveal the specific ‘tumorous’
Paleopathology
Dry bone
bone deposition of high-grade conventional osteosarcomas, all other bone tumours display common,
Bone tumours unspecific features. This holds for the following tumours: osteochondroma, hyperostotic meningioma,
Diagnosis high-grade angiosarcoma, grade 2 chondrosarcoma, myoepithelial carcinoma, high-grade osteosarcoma
Histology and four carcinoma metastases.
Microscopy We conclude that histology is useful in cases where the biological behaviour of a tumour is to be defined,
and is particularly an aide in the diagnosis of high-grade conventional osteosarcomas. Nevertheless, the
differential diagnosis on the bone tumours in our series should primarily be based on a combination of
physical anthropological patient data (age, sex), gross anatomy (e.g. tumour morphology and location),
and radiography.
© 2016 Elsevier Inc. All rights reserved.

1. Introduction logical analysis is subject to discussion (De Boer et al., 2013a;

Schultz and Schmidt-Schultz, 2015; Schutkowski and Fernandez-
Paleopathologists study diseases in ancient remains. They Gil, 2010; Strouhal, 1991; Van der Merwe et al., 2010; Waldron,
reconstruct the medical history of deceased individuals, provide 2009; Weston, 2009), one of them being the unavoidable tissue
valuable insight in the living conditions of past populations, and destruction from sample excision.
utilise an unique opportunity to study the natural development This article focuses on the dry bone histology of a specific group
of diseases in time (Rühli et al., 2016). In most cases the studied of diseases, namely neoplasms. A short introduction of their clas-
material is in such an advanced state of decomposition that only sification and how they may affect the skeleton is followed by the
skeletonized or ‘dry bone’ tissue remains. This limits the range backbone of this paper, the difficulty of their diagnosis. To illus-
of diseases that can be analysed to those diseases that (eventu- trate the latter, we compare the ‘fresh’ and ‘dry bone’ histology of
ally) affect the skeleton. In addition, because of the missing of the specimens of a well-documented current series of clinical cases of
soft tissue component, it limits the reliability of a paleopathologi- neoplasms with bone involvement.
cal diagnosis (Waldron, 2009). In an effort to maximize diagnostic
reliability, paleopathologists tend to complement ‘standard’ gross
anatomical analysis with other diagnostic tools, such as histology.
1.1. The classification of bone tumours
However, due to various reasons, the diagnostic value of histo-
Neoplasms (ancient Greek: new formations) constitute a het-
erogeneous group of disorders with the common denominator:
monoclonality. This means that, although neoplasms generally con-
夽 The Special Issue on Paleo-Oncology was edited by Casey L. Kirkpatrick, Roselyn
tain more than one cell type, they originate from a single population
A. Campbell, Kathryn J. Hunt, Jennifer L. Willoughby.
∗ Corresponding author at: Dept. of Pathology, Academic Medical Center, of cells with a shared genetic or epigenetic anomaly. Although orig-
Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands. inally used for any type of tissue enlargement, the term tumour is
E-mail address: h.h.deboer@amc.uva.nl (H.H. de Boer). nowadays used as a synonym for neoplasm.

https://doi.org/10.1016/j.ijpp.2016.11.005
1879-9817/© 2016 Elsevier Inc. All rights reserved.
 H.H. de Boer, G.J.R. Maat / International Journal of Paleopathology 21 (2018) 56–63 57

Neoplasms are usually classified by their biological behaviour bone tissue by excreting erosive substances. The resulting resorp-
and their tissue type. Benign neoplasms are non-invasive and their tion bays, Howship’s lacunae, remain visible in dry bone histology.
name is customarily named an aggregation of their dominating tis- According to the present state of science, osteoclasts do not undergo
sue type and the suffix −oma. Benign neoplasms of bone tissue malign transformation. Their bone tissue resorption activity can be
are for instance called osteomas whereas benign cartilaginous neo- regarded as a secondary, reactive process.
plasms are chondromas. Malign neoplasms, or cancers, are defined
by their potential to spread (metastasise) to other parts of the body.
1.3. The diagnosis of bone tumours
Since almost every cell in the body has the potential to undergo
malign transformation, the number of different types of cancers is
In current medical practice, the diagnosis of bone tumours
enormous. The most relevant ones from a paleopathological per-
requires a combination of clinical, radiographic, histological and
spective are derivatives of epithelial or mesenchymal tissue, which
presently in an increasing number of cases molecular analysis
are respectively called carcinomas and sarcomas.
(Czerniak, 2016; Waldron, 2009). Especially primary bone tumours
Alternatively, bone tumours (tumours situated in or near the
have a tendency to occur in specific age groups, in specific skele-
skeleton), can be classified as being primary or secondary. Primary
tal elements, and at preferential positions within a long bone (its
bone tumours are those neoplasms that originate from skeletal tis-
diaphyseal, metaphyseal or epiphyseal part) (Fletcher et al., 2003).
sue itself (e.g. bone, cartilage) and can either be benign or malign.
The location and size of the tumour, as well as its potency to pro-
The current incidence of benign primary bone tumours is unknown
duce extracellular matrix (e.g. cartilaginous or osseous tissue) is
as most of them grow (very) slowly or cause no clinical symptoms.
best assessed by radiography. Radiographic analysis also provides
Malign primary bone tumours are very rare, having an incidence of
information on the biological behaviour of the tumour by showing
0.8–2 cases per 100,000 individuals per year (Fletcher et al., 2003).
its pattern of bone destruction and its related periosteal reaction
Secondary bone tumours originate elsewhere in the body and have
(Fletcher et al., 2003; Miller, 2008; Priolo and Cerase, 1998). His-
spread to the skeleton. As such they are malign by definition. Sec-
tological analysis of tissue samples, sometimes combined with
ondary bone tumours are much more common than primary bone
molecular analysis, is the final step of the clinical diagnostic process
tumours. In the USA an estimated 350,000 people per year die hav-
(Mangham and Athanasou, 2011).
ing metastases in their bones (Vigorita et al., 2016).
Strategically, an analogous diagnostic approach is aspired in
paleopathology (Rothschild and Rothschild, 1995; Waldron, 2009).
1.2. The effect of bone tumours on the skeleton
Though frequently missing are selective data for diagnoses such as
the age and sex of the diseased and the presence of soft tissues.
Only neoplasms that leave traces in the skeleton can be studied
In a clinical context the histological diagnosis is primarily based on
in dry bone tissue. Essentially these traces are limited to the for-
soft-tissue features such as soft tissue architecture and cytonuclear
mation of mineralised tissue, the resorption or destruction of bone
morphology. A lack of soft tissue thus greatly hinders the applica-
tissue or a combination of the two.
tion of histology as a diagnostic tool. Due to the focus on soft tissue
Mineralised tissue is formed either by ossification or calcifica-
features, pathology textbooks hardly mention the histology of the
tion (Vigorita et al., 2016). Ossification is the formation of new
mineralised, ‘dry bone’ part of the various bone tumours.
bone and requires the deposition of osteoid and its subsequent
There is a limited number of paleopathology case reports of
mineralisation by osteoblasts. If bone tissue is formed in a slow
bone tumours in which histology was used (Alt and Adler, 1992;
and organised manner it displays a distinct lamellar architecture
Anderson et al., 1992; Campillo, 1991; et al., 1984Campillo and
and is referred to as lamellar bone. Accelerated bone deposition
Marcí-Balcells, 1984; De La Rúa et al., 1995; Grupe, 1988; Merczi
results in bone tissue with an irregular, disordered architecture and
et al., 2014; Molnár et al., 2009; Plenk Jr., 1999; Schultz, 1993;
is referred to as woven bone. Subsequent remodelling may con-
Schultz et al., 2007; Šefčáková et al., 2001; Strouhal, 1991, 1993;
vert woven bone into lamellar bone. Given a certain time lapse, the
Strouhal et al., 1996, 1997; Suzuki, 1987; Tkocz and Bierring, 1984;
degree of progress of the remodelling process is indicative for the
Vyhnánek et al., 1999; Wakely et al., 1995, 1998). These reports
growth speed of a neoplasm. A high growth speed is usually indica-
indicate that histology is especially valuable to exclude pseu-
tive of poor differentiation or immaturity and of (consequently)
dopathology, which is best recognized by an irregular distribution
high malignity, although exceptions exist.
that does not correspond with the normal distribution of disease.
By definition, osteogenic bone tumours produce tumorous bone
However, the value of histology to differentiate between different
tissue by means of their tumorous osteoblasts. Benign osteogenic
tumour types varies strongly (De Boer and Maat, 2012; De Boer
tumours produce bone tissue resembling the texture of nor-
et al., 2013a). As a result, the potential of histology in the diagnoses
mal bone, whilst the bone texture in malign osteogenic tumours
of bone tumours in dry bone tissue remains subject to debate.
(osteosarcomas) is markedly disordered and immature (less dif-
This study therefore compares the ‘fresh’ and ‘dry bone’ histol-
ferentiated). Prompted by the growth of any bone tumour, locally
ogy of specimens of a well-documented current series of clinical
also reactive, secondary bone tissue may be formed by normal,
cases of bone tumours. Such an approach has two important bene-
non-tumorous osteoblasts, the so-called ‘wild type’ osteoblasts.
fits. Firstly, all cases are diagnosed with state-of-the-art techniques
Mineralisation (less accurately called ‘calcification’) is the depo-
and include clinical follow-up. The correct diagnosis is therefore
sition of calcium and a minority of other minerals in tissue. In
no matter of debate. Second, the comparison between the fresh
contrast to bone tissue, the mineralised (or calcified) component
and dry bone histology enables us to focus on the mineralised tis-
in soft tissue displays as an amorphous acellular substance. Such
sue, of which the morphology is generally neglected in pathology
mineralisations may be seen in a myriad of diseases and are by no
textbooks.
means pathognomonic (Vigorita et al., 2016). They may dissolve,
persist or develop into bone tissue. The mechanisms hereof are
poorly understood. Due to the handling of paleopathological speci- 2. Materials
mens, small in vivo fragments of mineralised tissue usually get lost
(Steinbock, 1989). A total of 13 tumour specimens of recent, thoroughly diag-
Bone tissue is resorbed by large multinuclear cells from the nosed cases were selected from the collection of the Department
monocyte-lineage; so-called osteoclasts. These cells adhere to the of Pathology of the Academic Medical Centre in the Netherlands.
cortical or cancellous bone surface and dissolve the underlying The series contains the following tumour types: osteochondroma,
58 H.H. de Boer, G.J.R. Maat / International Journal of Paleopathology 21 (2018) 56–63

Table 1
Primary bone tumours included in this study.

Case no. Sex/agea Tumour location Diagnosis

1 F/7 Right radius High-grade osteosarcoma

2 M/13 Right distal femur High-grade osteosarcoma
3 F/34 Right acetabulum Grade 2 chondrosarcoma
4 M/72 Left proximal femur Grade 2 chondrosarcoma
5 M/68 Right fibula High-grade angiosarcoma
6 M/58 Right distal femur Myoepithelial carcinoma
7 M/21 Right first rib Osteochondroma
8 M/29 Right distal femur Osteochondroma
9 F/40 Frontal bone of the skull Hyperostotic meningeoma
a
F = female; M = male; age in years.

hyperostotic meningioma, high-grade angiosarcoma, grade 2 chon-

drosarcoma, myoepithelial carcinoma, high-grade osteosarcoma
and four carcinoma metastases. Care was taken to select a mix cov-
ering both the field of primary and of secondary bone tumours.
Relevant clinical information, such as medical status data of
the patients, tumour location and diagnosis can be found in
Tables 1 and 2.

3. Methods

Each tumour specimen was processed for dual diagnostic pro-

ceedings. The original specimen was kept frozen overnight at minus
80 ◦ Celsius. The frozen specimen was sawn into slices of approx-
imately 4 mm thick, after which a number of the slices were
processed according to present day routine histopathology. They
were fixated in formalin 4%, decalcified and processed into standard
haematoxylin and eosin (HE) stained histological sections. With
respect to the goal of the experiment, we will call these sections
the ‘fresh tissue sections’.
For every set of ‘fresh tissue sections’, a set of adjacent thick
slices was selected for dry bone section production. In those slices
most of the soft tissue was carefully removed. Subsequently, these
slices were macerated after which they were kept in acetone for
several days to remove any remaining fat (Fig. 1). The set was
photographed, resin-embedded and processed into undecalcified
histological sections according to the method of De Boer et al.
Fig. 1. Specimens before and after maceration. (a) An osteosarcoma of the radius.
(2013b). The result was a set of so-called ‘dry bone sections’ match-
(b) A grade 2 chondrosarcoma of the proximal femur. All slices are approximately
ing the ‘fresh tissue sections’ of every tumour specimen. 4 mm thick.
Both sets were studied, with special emphasis on the micro-
architectural features that might help to differentiate between the
various tumour types in the ‘dry bone sections’. Bright and polar- 4.1. Dry bone histology of the primary bone tumours
ized light microscopy was applied in all cases.
4.1.1. Osteochondroma and hyperostotic meningioma
4. Results The dry bone sections of the osteochondroma displayed an over-
all well differentiated bone tissue architecture. It contained within
All ‘dry bone sections’ were void of soft tissue and showed com- a dominant amount of cancellous bone, bordered by a thin cor-
parably well preserved dry bone tissue if matched with their ‘fresh tex. At higher magnification the bone tissue was of the lamellar
tissue section’ counterparts. The dry bone sections were consid- type. Only a very limited amount of remodelling activity, namely a
erably thicker (ca. 50–80 ␮m) than the fresh tissue sections (ca. low presence of Howship’s lacunae was noted. In accordance, the
8 ␮m) but this did not hamper analysis. The use of a polarizing fresh bone sections showed few active osteoblasts and osteoclasts
filter enhanced tissue architecture considerably. The histological (Fig. 2).
features will be discussed for primary bone tumours and secondary At low magnification, the dry bone sections of the hyperostotic
bone tumours separately, and are summarized in Table 3. meningioma comprised exclusively of a well-differentiated, com-

Table 2
Secondary bone tumours included in this study.

Case no. Sex/agea Tumour location Tumour type Site of origin

10 M/68 Right femoral head Adenocarcinoma Oropharynx

11 M/59 Right distal femur Clear cell carcinoma Kidney
12 M/77 Left sixth rib Adenocarcinoma Oesophagus
13 F/28 Right proximal femur Undifferentiated large cell carcinoma Unknown
a
F = female; M = male; age in years.
 H.H. de Boer, G.J.R. Maat / International Journal of Paleopathology 21 (2018) 56–63 59

Table 3
Histological observations in dry bone per tumour group.

Tumour group Histological observations in dry bone

Benign primary bone tumours On overview, a well differentiated, compact or trabecular bone tissue architecture, reminiscent of normal
Osteochondroma (n = 2) bone.
Hyperostotic meningeoma (n = 1) On higher magnification, predominantly bone tissue of the lamellar type and a low presence of Howship’s
lacunae, suggesting a quiescent biological behaviour.
Malignant primary bone tumours On overview, a large irregular defect in the cortical and cancellous bone tissue area with surrounding
Non-osteogenic tumours osteoporosis. A variable amount of newly formed bone tissue, predominantly located at the periosteal surface,
Myoepithelial carcinoma (n = 1) adjacent to the edges of the defect.
Grade 2 chondrosarcoma (n = 2) At higher magnification, newly formed bone of rather thick woven bone trabeculas and many Howship’s
High-grade angiosarcoma (n = 1) lacunae, suggesting a highly active lesion.
The low-grade tumour displayed more subtle changes than the high-grade tumours, in keeping with its
biological behaviour.
Osteogenic tumours On overview and at higher magnification, a mixed appearance of resorptive and depositional bone tissue
High-grade conventional osteosarcoma (n = 2) changes, similar to the non-osteogenic malignant primary bone tumours.
In addition, tumorous bone deposition, with a specific, irregular, friable and ‘lace-like’ appearance.
Secondary bone tumours (metastases) Indistinguishable from the non-osteogenic, malignant primary bone tumours.
Adenocarcinoma (n = 2)
Clear cell carcinoma (n = 1)
Undifferentiated large cell carcinoma (n = 1)

Fig. 2. Osteochondroma. (a) Micrograph of the cartilaginous cap (C) of the osteochondroma covering trabecular bone tissue (arrows). Gradual enchondral ossification is seen
at the interface between the two tissue types. Fresh section. Haematoxylin and eosin stain. (b) Micrograph after removal of the cartilaginous cap and the bone marrow fat by
maceration, while the well-differentiated trabecular bone (arrows) remains. Undecalcified dry bone section. Bright field.

Fig. 3. Hyperostotic meningeoma. (a) Micrograph of the centre of the hyperostotic meningeoma shows compact bone tissue composed of thick trabeculas of lamellar
bone (LB), without any remodelling activity. The meningioma tissue is visible between the trabeculas as loosely arranged tissue with a whirling appearance. Fresh section.
Haematoxylin and eosin stain. (b) Micrograph after the removal of the soft tissue. The compact, lamellar bone structure is preserved and no Howship’s lacunae (indicating
remodelling activity) are noted. Undecalcified dry bone section. Bright field.
60 H.H. de Boer, G.J.R. Maat / International Journal of Paleopathology 21 (2018) 56–63

pact architecture resembling cortical bone. At higher magnification, fresh tissue sections showed that this specific type of bone tissue
all bone tissue was of the lamellar type. As in the osteochondro- was deposited by tumorous osteoblasts.
mas, hardly any Howship’s lacunae were seen. Comparison with
the fresh tissue sections confirmed the overall regular architecture
4.2. Secondary bone tumours (metastases)
and a limited remodelling activity (Fig. 3).
All four of the studied secondary bone tumours displayed a
dry bone histology predominantly characterized by an irregular
4.1.2. Angiosarcoma, chondrosarcoma and myoepithelial
and ‘moth-eaten’ appearance of the pre-existing bone tissue, with
carcinoma
defects in the fields of cortical and cancellous bone tissue. In all, the
On overview, the dry bone sections of the angiosarcoma showed
centre of the primary, largest defect was located in the medullary
a large irregular defect within the fields of cortical and cancel-
cavity. Only a restricted amount of newly deposited bone was seen,
lous bone tissue. The surrounding bone tissue was osteoporotic,
and if so, mainly at the periosteal surface. At higher magnification
whilst at the vicinity a variable amount of newly formed bone tis-
the edges of the defect revealed an abundance of Howship’s lacunae
sue was seen. The majority of this newly formed bone was located
(Fig. 7). The type of new bone tissue resembled that of the new bone
at the periosteal surface, directly adjacent to the edges of the
formed in the non-osteogenic malign primary bone tumours: with
tumorous defect. At medium magnification this newly formed bone
a trabecular architecture and being mainly of the woven type. Only
was made up of rather thick trabeculas. At higher magnification,
a limited amount displayed a lamellar architecture, being indicative
the newly deposited bone tissue was of the woven type although
of some remodelling activity. No osteoblastic activity was noted in
some small parts had a lamellar appearance. The latter indicated
the medullary cavity. Comparison with the fresh tissue specimens
(limited) remodelling activity. Furthermore, the pre-existing can-
revealed indeed abundant bone resorption by osteoclasts at the
cellous and cortical bone tissue directly adjacent to the tumorous
edges of the tumour, whilst a reactive deposition of mainly woven
defect, above mentioned as ‘the surrounding osteoporotic bone’,
bone tissue was seen at the periosteal surface and the edges of the
showed an abundance of Howship’s lacunae. A comparison with
defect.
the fresh tissue sections corroborated our findings with respect to
the osteoclastic and osteoblastic activity. The observed osteoclasts
and osteoblasts were non-tumorous, i.e. ‘wild-type’ (Figs. 4 and 5). 5. Discussion
The dry bone histology of the two chondrosarcomas was indis-
tinguishable from that of the angiosarcoma. This holds for the This study is the first comparative analysis of the fresh and dry
resorptive and for the depositional changes on overview, but also bone histology of a series of bone tumours. It enabled us to match
for the changes at higher magnification. There was abundance of the histological findings in fresh tumour specimens with those as
Howship’s lacunae, whilst the newly formed bone was generally they may be encountered in paleopathology specimens. Although
irregular and of the woven type. Only a very limited amount of this study is limited to six primary bone tumours and four carci-
remodelled, lamellar bone tissue was seen. Comparison with the noma metastases, we believe that this has deepened our insight in
fresh tissue confirmed that, despite the different tumour type, the the diagnostic value of dry bone histology in the diagnosis of bone
changes to the bone tissue were similar to that seen in the angiosar- tumours.
coma. Our findings support earlier claims that in general, histology
On low magnification, the dry bone sections of the myoepithe- of dry bone tissue may be a helpful tool to assess the biological
lial carcinoma displayed a similar defect within the fields of cortical behaviour of bone tumours (Eshed et al., 2002; Plenk Jr., 1999;
and cancellous bone tissue as was noted in the angiosarcoma and Strouhal et al., 1996). The benign tumours in our sample displayed
the chondrosaromas. However, only a small amount of new bone well-differentiated lamellar bone tissue. Conversely the malign
formation was seen at the periosteal surface and at the edges of bone tumours featured deposition of irregular, woven bone, com-
the defect. On higher magnification, the presence of Howship’s bined with abundant destruction or resorption of bone tissue.
lacunae was generally less than seen in the angiosarcoma and the Non-destructive, well-differentiated bone architecture is the dom-
osteochondromas. The relatively limited amount of new bone for- inant feature in the morphology of benign tumours. Nevertheless
mation was of the woven type and showed no remodelling activity. caution with respect to this conclusion should be kept, since var-
Thus the myoepithelial carcinoma displayed similar features as the ious benign tumours (not studied in this series) may, like malign
angiosarcoma and the chondrosarcomas, be it in a more subtle way. tumours, also present with woven bone deposition and bone tis-
Comparison with the fresh tissue section corroborated the relative sue resorption, especially when it concerns a fast-growing lesion
limited amount of resorptive and depositional activity. (Czerniak, 2016; Fletcher, 2013; Vigorita et al., 2016).
The malign tumours in our series showed little variation in
the extensiveness of the resorption and formation of bone tissue.
4.1.3. High-grade conventional osteosarcoma The myoepithelial carcinoma, clinically designated as a low-grade
On low magnification, the dry bone sections of the two stud- malignant tumour, displayed overall less abundant osteoclastic
ied osteogenic malign primary bone tumours (osteosarcomas) had and osteoblastic activity than the other (high-grade) malignancies.
a mixed appearance of resorptive and depositional bone tissue Although based on a single myoepithelial carcinoma in our series,
changes. Large cortical and cancellous bone defects were alter- this finding seems to suggest that the degree of the dry bone tissue
nated by irregular, brittle and ‘coral’-like’ bone depositions at the changes might be used as an indicator of the biological behaviour of
periosteal surface and within the medullary cavity. The periosteal the tumour. Since radiography is generally non-destructive we sug-
bone depositions had an irregular, trabecular architecture, oriented gest reserving histology for cases in which the biological behaviour
more or less perpendicularly to the cortical surface. of a tumour remains uncertain despite radiological analysis.
On high magnification, the edges of the resorptive defects were The benign tumours in our series did not present with pathog-
lined by Howship’s lacunae. All newly formed bone tissue was of nomonic features, i.e. a definitive diagnosis on the basis of histology
the woven type. In addition to trabecular bone depositions simi- only was impossible. The regular, lamellar architecture of both
lar to that seen in the non-osteogenic primary bone tumours, both tumours in our series can be found in all relatively quiescent bone
osteosarcomas displayed a specific type of newly formed bone tis- tumours. Prior reports have shown that a limited degree of irregular
sue with an irregular, friable and ‘lace-like’ appearance (Fig. 6). The architecture with patches of woven bone may be seen in osteochon-
 H.H. de Boer, G.J.R. Maat / International Journal of Paleopathology 21 (2018) 56–63 61

Fig. 4. Grade 2 Chondrosarcoma. (a) Micrograph of new bone formation at the periosteal surface, at some distance of a grade 2 chondrosarcoma. The newly deposited bone
consists of cancellous bone trabeculas made up of woven bone. A string of active osteoblasts that deposit osteoid is visible in the upper right corner of the image (arrows).
Fresh section. Haematoxylin and eosin stain. (b) Micrograph of the corresponding dry bone section shows that the irregular microarchitecture of the cancellous woven bone
trabeculas stays visible after removal of the soft tissues. Undecalcified dry bone section. Bright field.

Fig. 5. Grade 2 Chondrosarcoma. (a) Micrographs of the edge of the grade 2 chondrosarcoma shows how several multinucleated osteoclasts (arrows) line and resorb the
pre-existing lamellar bone tissue directly adjacent to the tumorous cartilaginous tissue (CT). Note the numerous Howship’s lacunae at the edge of the bone tissue (*). Fresh
section. Haematoxylin and eosin stain. (b) Micrograph of the bone tissue at the edge of the tumour, after removal of the soft tissue by maceration. Only a non-specific image
of lamellar bone tissue lined by multiple Howship’s lacunae (*) remains. Undecalcified dry bone section. Bright field.

Fig. 6. High-grade conventional osteosarcoma. (a) Micrographs of the high-grade conventional osteosarcoma cells that deposit osteoid in a delicate, irregular and ‘lace-like’
pattern. Subsequent mineralisation of the osteoid results in a coarse, dark-pink stained meshwork of woven bone. Fresh section. Haematoxylin and eosin stain. (b) Micrographs
of the tumorous bone after removal of the soft tissue by maceration. The mineralised parts of the meshwork are still recognisable. The ‘laces’ show no lamellation. Undecalcified
dry bone section. Bright field.
62 H.H. de Boer, G.J.R. Maat / International Journal of Paleopathology 21 (2018) 56–63

Fig. 7. Metastasis of an undifferentiated large cell carcinoma of the lung. (a) Micrographs of the edges of the tumour show pre-existing lamellar bone tissue (LB) with an
irregular, jagged aspect due to the resorption of bone tissue by multinucleated osteoclasts (arrows). Note the presence of nests of large tumour cells (*) in the top of the image.
Fresh section. Haematoylin and eosin stain. (b) Micrograph of the bone tissue directly adjacent to the tumour after removal of the soft tissue by maceration. A non-specific
image of excessive bone resorption, indicated by the abundance of Howship’s lacunae (*) is seen. The histology is identical to that of the resorption activity seen in the
non-osteogenic primary bone tumours of our series. Undecalcified dry bone section. Bright field.

dromas and hyperostotic meningeomas (Campillo, 1991; Plenk Jr., 6. Conclusion

1999; Vyhnánek et al., 1999), which further limits the value of
histology in these tumour types. The comparative analysis of the fresh and dry bone histology of
In the malign primary tumours of our series, the deposition of the various benign and malign bone tumours in our series shows
irregular, friable ‘lace-like’ or ‘tumorous’ bone by malignant cells that dry bone histology could be a valuable adjunct in the diagnosis
was seen only in the high-grade conventional osteosarcomas. Two of a high-grade conventional osteosarcoma or when the biological
of three other of such cases reported in paleopathology litera- behaviour of a tumour is to be evaluated. Nonetheless, the differ-
ture showed similar bone formations (Campillo and Marcí-Balcells, ential diagnosis on any of the bone tumours in our series should
1984; Strouhal et al., 1997; Suzuki, 1987). Thus, these findings seem predominantly be based on a combination of physical anthropolog-
to indicate that ‘lace-like’ bone depositions are highly characteris- ical patient data (age, sex), gross anatomy (e.g. tumour morphology
tic for high-grade conventional osteosarcomas. This feature can be and location) and radiography.
especially helpful in cases in which radiology and gross anatomy is
equivocal, for instance when one needs to differentiate between an Conflict of interest
osteosarcoma and an osteoblastic carcinoma metastasis. We have
to take into account that in case of paleopathological specimens a The authors declare to have no conflict of interest.
great deal of these delicate depositions might be lost during the
recovery and processing of the specimens. Thus, whereas tumor- Acknowledgements
ous bone depositions seem limited to high-grade conventional
osteosarcomas, their absence does not necessary plea against the Kerri Colman MSc is acknowledged for her assistance in mac-
diagnosis. erating the tumour specimens. We are also indebted to Dilara
In all other aspects, the histology of the high-grade conven- Savci-Heijink MD and the laboratory personnel of the Department
tional osteosarcomas was identical to that of the other malign of Pathology of the Academic Medical Centre.
primary bone tumours of our series. In addition, no specific his-
tological differences were noted between the malign primary and References
secondary bone tumours (carcinoma metastases). Such similarity is
understandable, since with exception of the tumorous bone deposi- Alt, K.W., Adler, C.-P., 1992. Multiple myeloma in an early medieval skeleton. Int. J.
tion in high-grade conventional osteosarcomas, all the bone tissue Osteoarchaeol. 2, 205–209.
Anderson, T., Wakely, J., Carter, A., 1992. Medieval example of metastatic
changes in these tumours result from (normal) ‘wild-type’ cells carcinoma: a dry bone, radiological, and SEM study. Am. J. Phys. Anthropol. 89,
only. 309–323.
Our series did not include non-neoplastic mimickers, such as Campillo, D., Marcí-Balcells, V.J., 1984. Microscopy of osteal tumours in
palaeopathology. 6th European Meeting of the Palaeopathology Association.
Paget’s disease of bone, so-called ‘brown tumours’ related to hyper-
Sienna, Italy, 35–49.
parathyreoidism, bone cysts, or excessive callus formation. Only Campillo, D., 1991. The possibility of diagnosing meningiomas in palaeopathology.
few of these present with pathognomonic histological features, Int. J. Osteoarchaeol. 1, 225–230.
such as the disorganized bone histomorphology of the sclerotic Czerniak, B., 2016. Dorfman and Czerniak’s Bone Tumors, 2 ed. St. Louis: Mosby.
De Boer, H.H., Maat, G.J.R., 2012. The Histology of Human Dry Bone (a Review).
phase in Paget’s disease of bone, or the ‘dissecting osteitis’ in Cuadernos de Prehistoria y Arqueología de la Universidad de Granada 22,
hyperparathyreoidism (De Boer et al., 2013a). These features may 49–65.
however be obscured by taphonomic processes. We therefore rec- De Boer, H.H., Van der Merwe, A.E., Maat, G.J.R., 2013a. The diagnostic value of
microscopy in dry bone palaeopathology: a review. Int. J. Paleopathol. 3,
ommend that in general, for all bone tumours or non-neoplastic 113–121.
mimickers, the differential diagnosis is primarily based on a com- De Boer, H.H., Aarents, M.J., Maat, G.J.R., 2013b. Manual for the preparation and
bination of patient data, gross anatomy and radiography. staining of embedded natural dry bone tissue sections for microscopy. Int. J.
Osteoarchaeol. 23, 83–93.
 H.H. de Boer, G.J.R. Maat / International Journal of Paleopathology 21 (2018) 56–63 63

De La Rúa, C., Baraybar, J., Etxeberria, F., 1995. Neolithic case of metastasizing Schultz, M., 1993. Microscopic investigations on tumorous lesions from Christian
carcinoma: multiple approaches to differential diagnosis. Int. J. Osteoarchaeol. Sayala (Egyptian Nubia). Anthropol. Anz., 117–121.
5, 254–264. Schutkowski, H., Fernandez-Gil, M., 2010. Histological manifestation of
Eshed, V., Latimer, B., Greenwald, C.M., Jellema, L.M., Rothschild, B.M., Wish-Baratz, mycobacterial bone lesions. 18th European Meeting of the Paleopathology
S., Hershkovitz, I., 2002. Button osteoma: its etiology and pathophysiology. Association. Vienna, Austria.
Am. J. Phys. Anthropol. 118, 217–230. Šefčáková, A., Strouhal, E., Němečková, A., Thurzo, M., Staššíková-Štukovská, D.,
Fletcher, C.D., Bridge, J.A., Hogendoorn, P., Mertens, F., 2003. WHO Classification of 2001. Case of metastatic carcinoma from end of the 8th–early 9th century
Tumours of Soft Tissue and Bone. 2013. International Agency for Research of Slovakia. Am. J. Phys. Anthropol. 116, 216–229.
Cancer, Lyon, pp. 305–310. Steinbock, R., 1989. Studies in ancient calcified soft tissues and organic concretions.
Fletcher C.D.M, 2013. Diagnostic histopathology of tumors. I: a review of structures, diseases, and conditions. J. Paleopathol. 3, 35–38.
Grupe, G., 1988. Metastasizing carcinoma in a medieval skeleton: differential Strouhal, E., VhynÁ.Nek, L., HorÁ.Č.KovÁ, L., BeneŠ.OvÁ, L., N.Ě.MeČ.KovÁ, A., 1996.
diagnosis and etiology. Am. J. Phys. Anthropol. 75, 369–374. Two unusual benign tumours in skulls from the ossuary at Křtiny (Czech
Mangham, D.C., Athanasou, N.A., 2011. Guidelines for histopathological specimen Republic). Int. J. Osteoarchaeol. 6, 289–299.
examination and diagnostic reporting of primary bone tumours. Clin. Sarcoma Strouhal, E., Vyhnánek, L., Horáčková, L., Benešová, L., 1997. A case of osteosarcoma
Res. 1, 6–6. in a late Medieval–early modern skull from Kyjov (Czech Republic). Int. J.
Merczi, M., Marcsik, A., Bernert, Z., Józsa, L., Buczkó, K., Lassányi, G., Kelemen, M.H., Osteoarchaeol. 7, 82–90.
Zádori, P., Vandulek, C., Biró, G., Hajdu, T., Molnár, E., 2014. Skeletal metastatic Strouhal, E., 1991. Myeloma multiplex versus osteolytic metastatic carcinoma:
carcinomas from the roman period (1st to 5th century AD) in Hungary. differential diagnosis in dry bones. Int. J. Osteoarchaeol. 1, 219–224.
Pathobiology 81, 100–111. Strouhal, E., 1993. A case of metastatic carcinoma from Christian Sayala (Egyptian
Miller, T.T., 2008. Bone tumors and tumorlike conditions: analysis with Nubia). Anthropol. Anz., 97–115.
conventional radiography. Radiology 246, 662–674. Suzuki, T., 1987. Paleopathological study on a case of osteosarcoma. Am. J. Phys.
Molnár, E., Marcsik, A., Bereczki, Z., Schmidt-Schultz, T.H., Schultz, M., Pálfi, G., Anthropol. 74, 309–318.
2009. Malignant tumors in osteoarchaeological samples from Hungary. Acta Tkocz, I., Bierring, F., 1984. A medieval case of metastasizing carcinoma with
Biol. Szegediensis 53, 117–124. multiple osteosclerotic bone lesions. Am. J. Phys. Anthropol. 65, 373–380.
Plenk Jr., H., 1999. Differential diagnosis on ancient skeletal remains: conventional Van der Merwe, A., Maat, G., Steyn, M., 2010. Ossified haematomas and infectious
methods and novel application of the BSE-mode in SEM on a skull tumor of the bone changes on the anterior tibia: histomorphological features as an aid for
early Bronze Age. Coll. Antropol. 23, 483–494. accurate diagnosis. Int. J. Osteoarchaeol. 20, 227–239.
Priolo, F., Cerase, A., 1998. The current role of radiography in the assessment of Vigorita, V.J., Ghelman, B., Mintz, D., 2016. Orthopaedic Pathology, 3 ed. Lippincott
skeletal tumors and tumor-like lesions. Eur. J. Radiol. 27 (Suppl. (1)), S77–S85. Williams & Wilkins, Philadelphia.
Rühli, F.J., Galassi, F.M., Haeusler, M., 2016. Palaeopathology: current challenges Vyhnánek, L., Strouhal, E., Nemečková, A., 1999. ‘Kissing’osteochondroma: a case
and medical impact. Clin. Anat., e-pub ahead of print. from ancient Egypt. Int. J. Osteoarchaeol. 9, 361–368.
Rothschild, B.M., Rothschild, C., 1995. Comparison of radiologic and gross Wakely, J., Anderson, T., Carter, A., 1995. A multidisciplinarian case study of
examination for detection of cancer in defleshed skeletons. Am. J. Phys. prostatic (?) carcinoma from Mediaeval Canterbury. J. Archaeolog. Sci. 22,
Anthropol. 96, 357–363. 469–477.
Schultz, M., Schmidt-Schultz, T.H., 2015. Is it possible to diagnose TB in ancient Wakely, J., Strouhal, E., Vyhnánek, L., Němečková, A., 1998. Case of a malignant
bone using microscopy? Tuberculosis 95 (Suppl. (1)), S80–S86. tumour from Abingdon, Oxfordshire, England. J. Archaeolog. Sci. 25, 949–955.
Schultz, M., Parzinger, H., Posdnjakov, D.V., Chikisheva, T.A., Schmidt-Schultz, T.H., Waldron, T., 2009. Palaeopathology. Cambridge University Press.
2007. Oldest known case of metastasizing prostate carcinoma diagnosed in the Weston, D.A., 2009. Brief communication: paleohistopathological analysis of
skeleton of a 2,700-year-old Scythian king from Arzhan (Siberia, Russia). Int. J. pathology museum specimens: can periosteal reaction microstructure explain
Cancer 121, 2591–2595. lesion etiology? Am. J. Phys. Anthropol. 140, 186–193.