J Oral Pathol Med (2009) 38: 72–78
ª 2009 John Wiley & Sons A/S Æ All rights reserved
www.blackwellmunksgaard.com/jopm
Molecular markers in the surgical margin of oral
carcinomas
Anders Bilde1, Christian von Buchwald1, Erik Dabelsteen2, Marianne Hamilton Therkildsen3,
Sally Dabelsteen4
1
Department of Otolaryngology – Head & Neck Surgery, Copenhagen University Hospital, Copenhagen; 2Department of Oral
Diagnostics, Faculty of Health Sciences, University of Copenhagen, Copenhagen; 3Department of Pathology, Copenhagen University
Hospital, Copenhagen; 4Department of Oral Pathology and Medicine, Faculty of Health Sciences, University of Copenhagen,
Denmark
BACKGROUND: Local or regional lymph node recur-
rence is the most common pattern of treatment failure in
oral squamous cell carcinoma (SCC). The local recur-
rence rate is 30% even when the surgical resection mar-
gin is diagnosed as tumour free. Accumulation of genetic
changes in histologically normal epithelium in the surgical
resection margin may explain the local recurrence rate.
The purpose of this study is to investigate the presence of
senescence markers, which may represent early malig-
nant changes in the margin that in routine pathological
evaluations are classified as histologically normal.
METHODS: Formalin-fixed, paraffin-embedded surgical
specimens from 16 consecutive patients with oral SCC
and a clear surgical margin were obtained. The margin
was analysed by immunohistochemistry for p53, p16,
Chk2, Laminin-5 and glycosylated oncofetal fibronectin.
RESULTS: Two patterns of p53 expression were found in
the histologically normal epithelium in the surgical
resection margin. One was characterized by no protein
expression in the majority of cells, except for small clus-
ters of basal and parabasal cells with nuclear staining. The
other was characterized by p53 expression in the nuclei
of most basal cells. The expression of p16 was confined to
small groups of cells in the basal cell layer whereas Chk2
was only seen in one case. Upregulation of the stromal
proteins, Laminin-5 or glycosylated oncofetal fibronec-
tion, was only seen at regions of invasion.
CONCLUSION: Small groups of cells expressing p53 and
p16 were found in the surgical resection margin that
appeared to be histologically normal and may represent
early malignant changes.
J Oral Pathol Med (2009) 38: 72–78
Keywords: Chk2; glycosylated oncofetal fibronectin; Laminin-5;
oral cancer; oral squamous cell carcinoma; P16; P53; surgical
margin
Correspondence: Anders Bilde, Department of Otolaryngology –
Head & Neck Surgery, Copenhagen University Hospital, Blegdamsvej
8, DK-2100 Copenhagen Ø, Denmark. Tel: +45 3545 2379, Fax: +45
3545 2267, E-mail: bilde@rh.dk
Accepted for publication July 29, 2008
doi: 10.1111/j.1600-0714.2008.00715.x
Introduction
Squamous cell carcinomas of the oral cavity (OSCC) is a
severe disease as approximately 40% of the patients fail
to respond to treatment (1). Local or regional lymph
node recurrence is the most common cause of treatment
failure and accounts for 60% of all the treatment
failures. Even when the surgical resection margin,
defined as 5 mm histologically normal mucosa, is
diagnosed as tumour free the local recurrence rate
is 30% (2–5).
It is evident from studies of carcinomas of the colon
and the bladder that cellular senescence (irreversible
proliferation arrest) may be present in the early stages of
tumour development (6). These early stages of cancer
development are characterized by the expression of
senescence markers such as p16, p53 and Chk2 (7, 8).
Genetic alteration in head and neck squamous cell
carcinoma often involves the p53 genes and p16 genes,
and altered expression of p53 and p16 has previously
been described in the histologically tumour-free surgical
resection margins adjacent to the primary tumour, which
may indicate an early malignant change (9–15). How-
ever, most of these studies are hampered by lack of
histological description of the tumour-free’ tissue that,
at least in some cases, actually shows epithelial dysplasia.
From a clinical point of view it appears to be of major
importance to establish not only the histology, but also
the molecular nature of the surgical resection margin. It
is likely that local recurrence is a result of accumulated
genetic changes present in the histologically normal
epithelium of the surgical resection margin. Thus, it was
the purpose of the present study to investigate the
presence of senescence markers that may be a sign of
early cancer development in the margin, which in
routine pathological evaluations are classified as histo-
logically normal.
Materials and methods
Patients and tissues
Following approval from the local Ethics Committee
surgical specimens of OSCC from 16 consecutive
 Surgical margin of oral carcinomas
Bilde et al.

patients were obtained. The patients were treated at the
Department of Otolaryngology – Head & Neck Surgery,
Copenhagen University by resection of the primary
tumour and sentinel node assisted selective neck dissec-
tion. The inclusion criterion was patients classified as
T1-2N0M0 with a clear (epithelial) surgical margin,
defined as the distance from invasive carcinoma to
surgical margin of more than or equal to 5 mm as
assessed by conventional histological examination.
The surgical margin constituting histologically nor-
mal epithelium adjacent to the primary tumour was
analysed for the expression of p53, p16, Chk2, Laminin-5
and glycosylated oncofetal fibronectin. Laminin-5 and
glycosylated oncofetal fibronection are known to be
markers of early invasion. The demographic data on
patients, classification, localization and grade of differ-
entiation of tumours are given in Table 1.
Cell lines
Immunohistochemical staining of p53 in tissue fixed and
embedded by routine procedure requires antigen
retrieval by microwave. To ensure that such a treatment
does not lead to the appearance of a cross reacting
proteins a cell line, Saos-2, lacking expression of p53,
was included in the study as a control (16). Saos-2 was
grown in a monolayer system, harvested, centrifuged
and embedded in gelatine, formalin fixed and embedded
in paraffin.
Histology and immunohistochemical procedure
Tumour tissue was fixed in neutral buffered formalin.
Before embedding in paraffin the tumour was divided
and a section representing the maximum cross-section of
the tumour was used. Routine histological sections were
used for histological evaluation.
Immunohistochemical staining for p53, p16, Chk2,
Laminin-5 and glycosylated oncofetal fibronectin was
73
performed on the paraffin embedded tissue using a
microwave antigen retrieval technique.
Sections for staining with p53 and p16 were microwave
treated for 15 min in citrate buffer, pH 6.0, and cooled by
placing the sections first in buffer and then in running tap
water. Sections for staining of with Laminin-5 and Chk2
were microwave treated for 15 min in Tris ⁄ EDTA, pH
9.0, and cooled as described above. The immunohisto-
chemical staining was performed according to routine
procedures. A DAKO (Copenhagen, Denmark) Power
Vision System with peroxidase as a marker was used as
second layer for all antibodies except for the glycosylated
oncofetal fibronectin antibody. In stainings with this
antibody a rabbit antimouse IgG (DAKO) labelled with
flourescein isothiocyanate was used.
Antibodies
p53 antibody (DO1) (Calbiochem, Darmstadt, Germany)
IgG2a, diluted 1:50, anti p16INK4a antibody (G175-405)
(Pharmingen, San Jose, CA, USA) IgG1, in dilution
1:500, and antibody against total and Thr68-phos-
phorylated Chk2 (Medinova Scientific AS, Glostrup,
Denmark), in dilution 1:500. All antibodies were mouse
monoclonal antibodies except the anti Chk2 antibody,
which was a polyclonal rabbit antibody. Laminin-5
antibody (B4-6) IgG1 (kindly provided by W.G. Carter),
in dilution 1:10. The specificity of the antibodies has
previously been established (8, 17–20).
Incubation with irrelevant primary antibody of the
same isotype was used as control for the monoclonal
antibodies. Deletion of the primary antibody served as a
control for the Chk2 polyclonal antibody.
Working dilutions of the p53, p16 and Chk2 anti-
bodies were established by titration using sections of
normal oral mucosa from four healthy young individ-
uals (21). The first dilution giving negative results in
normal epithelium was chosen as a working dilution.

Table 1 Immunohistochemical expression of p53, p16, Chk2 and Laminin-5 in histologically normal epithelium adjacent to the tumour in patients
with oral squamous cell carcinoma

P53 P16 Chk2 Laminin-5

Case Age ⁄ gender Site pTNM GradeSCC Nor. Dys. SCC Nor. Dys. SCC sup ⁄ pro Nor. Dys. SCC Nor. Dys. SCC
1 56 ⁄ M T T2N0M0 WD PB – ) SG – +⁄+ ) – + ) – CP
2 63 ⁄ M T T1N0M0 PD PB PB + SG – +⁄) ) + + ) BM CP
3 49 ⁄ M T T1N1M0 MD B – + – – +⁄+ ) – + ) – CP
4 57 ⁄ M FOM T2N0M0 MD B PB + SG SG +⁄) + + + ) – CP
5 62 ⁄ M T T2N2bM0 MD PB PB ) SG SG +⁄+ ) + + ) – CP
6 57 ⁄ M FOM T2N0M0 MD PB – ) SG – +⁄+ ) + + ) BM CP
7 61 ⁄ F FOM T1N0M0 WD PB – + – – +⁄) ) – + ) – CP
8 65 ⁄ M FOM T1N0M0 WD PB – + SG – +⁄+ ) – + ) – CP
9 53 ⁄ F FOM T1N0M0 MD B PB + SG – )⁄) ) + + ) BM CP
10 65 ⁄ M T T2N1M0 MD PB – + SG – +⁄+ ) – + ) – CP
11 57 ⁄ M T T2N0M0 MD PB – + – – +⁄+ ) – + ) – CP
12 66 ⁄ M FOM T2N0M0 WD PB – + SG – +⁄+ ) – + ) – CP
13 59 ⁄ F RT T1N0M0 WD PB – + SG – +⁄+ ) – + ) – CP
14 49 ⁄ M T T1N0M0 MD PB – + – – +⁄) ) – + ) – CP
15 67 ⁄ F BM T2N0M0 MD B – ) SG – +⁄) ) – + ) – CP
16 38 ⁄ M T T1N0M0 MD PB – ) – – +⁄) ) – ) ) – CP

pTNM, pathological TNM stage; Nor., normal histopathological mucosa; Dys., dysplasia; SCC, squamous cell carcinoma; sup, superficial layer;
pro, profound layer; F, female; M, male; T, tongue; FOM, floor of mouth; RT, retromolar trigone; BM, buccal mucosa; WD, well differentiated;
MD, moderately differentiated; PD, poorly differentiated; PB, parabasal and basal; B, basal; +, positive; ), negative; SG, single groups of cells; SC,
single cells; BM, basement membrane; CP, cytoplasmatic.

J Oral Pathol Med

 74

J Oral Pathol Med

A B C D

E F G H
Bilde et al.
Surgical margin of oral carcinomas

I J K

Figure 1 Immunohistochemical analysis of p53, p16, Chk2 and Laminin-5 in histologically normal epithelium adjacent to tumour. Expression of p53 was either confined to the basal cells (A, C and
H bottom) or to basal and parabasal cells as clusters surrounded by histologically normal epithelium (B and C). Small groups of p16 expressing cells were seen in the basal cell layer marked * (D) and
expression of p16 was seen in the deep layers of the tumour (E) as well as the superficial layers (F). Expression of Chk2 was seen in epithelial dysplasia (top) but not in normal epithelium (bottom) (G).
Corresponding expression of p53 was confined to basal cells (bottom) and in epithelial dysplasia to most layers of the epithelium (top) (H). Laminin-5 was expressed in normal epithelium in the
basement membrane (I) and in the cytoplasm of the tumour cells (J and K). Epithelium overlying the tumour showed accentuated expression of Laminin-5 at the basement membrane and in the
cytoplasm of the tumour cells (K).

Results
Staining of the p53 in the tumour cell line (Saos-2) was
negative.
Results from staining of the tumour tissue are
summarized in Table 1.
Distribution of p53
Twelve out of 16 specimens expressed p53 in both basal
as well as parabasal cells layers in the histologically
normal epithelium (Fig. 1B,C). The positive reaction
was seen as sparsely dispersed clusters of cells often with
only a few positive clusters in each section. Four out of
16 specimens had p53 expressing cells confined to the
basal cell layer only (Fig. 1A).
Four of the five specimens containing epithelial
dysplasia adjacent to the tumour showed p53 expression
in most layers of the epithelium (Fig. 1H).
Seven tumours did not express p53, whereas the rest
showed various degrees of positive staining.
Distribution of p16
p16 was expressed in histologically normal epithelium in
11 out of 16 cases. The positive staining appeared as
small groups of cells in the basal cell layer (Fig. 1D). In
two cases (cases 9 and 10) localization of p16 corre-
sponded to the expression of p53.
In the specimens with epithelial dysplasia expression
of p16 was expressed in two out of the five cases and was
in these cases seen in most epithelial layers.
In tumours p16 expression was seen superficially as
well as in the deep layers in 10 out of 16 cases
(Fig. 1E,F). In the rest p16 expression disappeared
in the deep layer. One case showed no expression of
p16.
Distribution of Chk2
Expression of Chk2 was only seen in one case in
histologically normal epithelium. However, in the spec-
imens containing dysplastic epithelium as well as in all
tumours Chk2 stained positively (Fig. 1G).
Distribution of Laminin-5
Expression of Laminin-5 was confined either to the
basement membrane and ⁄ or to the cytoplasm. In the
histologically normal mucosa staining of laminin-5 was
either negative or appeared as a discrete staining of the
basement membrane (Fig. 1I). However, in the specimens
showing epithelial dysplasia expression of laminin-5
appeared as a broad staining of the basement membrane
(not shown). In tumours, not only the basement mem-
brane, but also primarily the cytoplasm of the tumour
cells expressed Laminin-5 (Fig. 1J,K).
There were no correlation between the expression of
Laminin-5 and p16.
Distribution of glycosylated oncofetal fibronectin in
normal mucosa, epithelia dysplasia and squamous cell
carcinoma
Glycosylated oncofetal fibronectin was only expressed in
tumour tissue and never seen in histologically normal
Surgical margin of oral carcinomas
Bilde et al.
75
mucosa or in mucosa with epithelial dysplasia. Expres-
sion of the protein was either seen as a basement
membrane staining or as a diffuse staining in the tumour
stoma as previously described (22).
Discussion
A margin of at least 5 mm of histologically normal
epithelium in the surgical specimen is traditionally
regarded as the gold standard’ in the treatment of
OSCC (23). However, molecular changes that indicate
early tumour development have been demonstrated in
surgical margins with normal histology in different
tumours, including those tumours from the larynx,
pharynx and oral cavity (24, 25).
Tumour growth and invasion involve multiple inter-
actions not only between tumour cells but also between
tumour cells, stromal cells and the extracellular stroma
(26). Laminin-5 is secreted by migratory epithelial cells
and expression of Laminin-5 is an indicator of the
migratory phenotype (27–29). Glycosylated oncofetal
fibronectin is secreted by squamous carcinoma cells or
tumour induced myofibroblasts (22). In vitro studies of
oral cell lines have shown that expression of glycosylated
oncofetal fibronectin like Laminin-5 is characteristic for
migrating cancer cells (22, 29). These structures may
form a tumour scaffold to which tumour cells may
adhere and through which they can migrate. It is
interesting that we only found these markers when there
was clear evidence of invasion. None of the areas
showing even severe dysplasia expressed these markers.
Loss of p16 is regarded as an early event in tumour
development. In approximately 80% of squamous cell
carcinomas p16 is lost either by mutation, deletion or
promoter hypermethylation (30, 31). It has previously
been demonstrated that the superficial part of most
tumours expresses p16, whereas expression of p16 is
usually not seen in the deep invasive front especially in
advance stage tumours (27). In the majority of the
tumours expression of p16 was, however, seen in both
the superficial as well as the deep layer of the tumour.
This may be explained by the fact that the present
material mainly consisted of early stage T1-T2 tumours.
A few studies have focused on the expression of p16 in
the histologically normal oral epithelium with inconsis-
tent results (27, 32–34). We found p16 to be expressed as
single groups of cells at the basal cell layer in the
histologically normal epithelium adjacent to the tumour,
probably representing activated clones (27). In a recent
paper expression of p16 has been connected to expres-
sion of Laminin-5 and a model for epithelial neoplastic
progression has been proposed (27). We found expres-
sion of p16 and Laminin-5 to be independent of each
other, whereas Natarajan et al. suggested that
up-regulation of Laminin-5 causes up-regulation of p16
(27). However, the lack of correlation between Laminin-5
and p16 found in the present study suggests that other
mechanisms are involved as well. In tumour cells
Laminin-5 was located in the cytoplasm indicating a high
synthetic level – an impaired deposition of laminin-5 – a
finding in agreement with previous results (27, 29).
J Oral Pathol Med
 Surgical margin of oral carcinomas
Bilde et al.
76
In a comprehensive study, Partridge et al. demon-
strated a p53 gene mutation rate of 72% in biopsies of
clinically normal mucosa from the tumour margins and
in the deep connective tissue margin following excision
of the tumour (11). We have in the present study used
immunohistochemistry to demonstrate expression of
early tumour markers. This technique gives, in contrast
to previous mutation analysis by Partridge et al. (11),
information of both histology of the tissue investigated
and the distribution of the markers within that tissue.
Demonstration of p53 protein by the use of immuno-
histochemistry is, however, controversial. Many studies
have not been able to detect p53 protein in histologically
normal oral mucosa, whereas others demonstrate
expression in basal cells (35, 36). The apparent contra-
diction is causing uncertainty of the role of p53 in
carcinogenesis. Many factors, including methods and
duration of fixation, methods of antigen retrieval,
antibody used and choice of detection system influence
results of immunohistochemical analysis of p53 (37).
This makes comparison of results from different studies
difficult. Tissue handling and paraffin embedding were
in the present work performed in a standardized way.
Negative staining of the cell line Saos-2, which was used
as a control, indicates that the antigen retrieval method
does not induce unspecific staining of p53 unrelated
proteins. Titration of the antibodies showed that high
concentration of the antibodies stained basal cells in
normal oral epithelium from control individuals. In
order to detect an increased expression of p53, we
therefore stained the tissue with an antibody dilution,
which gave a negative result in histologically normal
oral control tissues. The demonstration of p53 in the
histologically normal mucosa in the surgical resection
margins thus points to an increased expression of the
protein in this epithelium.
In histologically normal epithelium in the surgical
resection margin we saw two patterns of protein
expression. One was characterized by no protein expres-
sion in the majority of cells, except for small clusters of
basal and parabasal cells with nuclear staining. A similar
pattern has recently been described by Farshadpour
et al. (38). The other was characterized by p53 expres-
sion in the nuclei of most basal cells resembling the
benign distribution previously described by Cruz (39).
Cruz et al. found p53 expression above the basal cell
layer in a few cases with no or mild dysplasia. They also
found that development of carcinoma from lesions with
epithelial dysplasia was more frequent in lesions that
showed p53 expression above the basal cell laye. Three
lesions with no or mild dysplasia, but with suprabasal
staining of p53, later developed carcinomas (39, 40). On
basis of this they suggested that suprabasal expression of
p53 predicts carcinoma development.
Cruz also found p53 staining in normal control tissue
without evidence of malignancy. This indicates that the
p53 antibody was not titrated and it is, therefore,
uncertain to know whether the basal cell staining they
found in the potentially malignant lesions actually
represented an upregulation of p53 or just the normal
distribution. However, in our study p53 basal cell
J Oral Pathol Med
staining represented an up-regulation of the protein
levels as demonstrated by the titration of the antibodies.
In the p53 positive epithelial dysplasias most cell layers
expressed the protein. A similar pattern was described in
an early investigation of Kushner et al. (41).
Polycyclic aromatic hydrocarbons that are one of the
major carcinogens in tobacco smoke, may cause muta-
tions in p53 and mutated p53 have been demonstrated in
the plasma from healthy smokers. As most patients with
oral carcinomas are heavy smokers and drinkers, it
may be hypothesized that there is an increase of mutated
p53 not only in oral carcinomas, but also in the
surrounding oral mucosa (38). However, there is no
indication that increased expression of p53 and the related
p63 is found in histologically normal oral mucosa from
smokers (38, 42).
According to the hierarchical stem cell concept in oral
mucosa, stem cells, located in the basal cell layer, have a
low proliferative rate when compared to amplifying cells
located in the basal and parabasal layer (43). Tumour
development requires multiple genetic changes in epi-
thelial stem cells (44). It is likely that the clusters of p16
and p53 expressing cells represent activated clones of
stem cells and amplifying cells.
Mutant p53 protein is most often stabilized and can
therefore be demonstrated by immunohistochemistry.
However, in some cases wild-type p53 may also be
stabilized and thus stainable (45). Furthermore upreg-
ulation of wild-type p53 may be seen in cells that are
stressed theoretically by paracrine factors from a neigh-
bouring tumour (45). If we assume that the pattern of
small clusters of p53 expressing cells in the basal and
parabasal cell layers contains mutated p53, the mutation
rate of p53 would be 75% (12 ⁄ 16). This mutation rate is
high, but equal to the rate previously demonstrated by
Partridge et al. (11). In order to determine whether the
expression of p53 represents mutated or wild type
protein, further investigation is required using micro-
dissection and p53 sequencing of the cells in the clusters.
The clinical significance of clusters would have to be
evaluated in a prospective study.
A previous study of early human tumours from
breast, lung, bladder and colon showed expression of
markers of activated DNA damage response such as
Chk2 (8). Chk2, an effector kinase, was found to precede
p53 gene mutations and was expressed at an early stage
before development into carcinoma (8). In our study
Chk2 was not expressed in normal epithelium adjacent
to the tumour but in epithelial dysplasia and in tumours.
Given the results of Bartkova et al. indicating that Chk2
precedes p53 gene mutations, the expression of p53
found in normal Chk2 negative mucosa at the surgical
margins suggests that this is a wild type protein rather
than mutated p53 protein. On the other hand this is in
contrast to previous studies reporting a very high rate of
p53 gene mutations in surgical margins (11).
In conclusion, histologically normal epithelium adja-
cent to oral carcinomas may show upregulation of both
p53 and p16, but with very little overlap. It is obscure
whether these changes represent early malignant
changes or are a reaction to cellular stress.

Conflicts of interest statement
None of the authors have any conflict of interest
regarding the present work.
References
1. Hansen HS, Overgaard J, Jorgensen KE. DAHANCA
(Danish head and neck cancer) – 20 years of cooperation.
Ugeskr Laeger 1998; 160: 821–6.
2. Looser KG, Shah JP, Strong EW. The significance of
‘‘positive’’ margins in surgically resected epidermoid
carcinomas. Head Neck Surg 1978; 1: 107–11.
3. Byers RM, Bland KI, Borlase B, Luna M. The prognostic
and therapeutic value of frozen section determinations in
the surgical treatment of squamous carcinoma of the head
and neck. Am J Surg 1978; 136: 525–8.
4. Loree TR, Strong EW. Significance of positive margins in
oral cavity squamous carcinoma. Am J Surg 1990; 160:
410–4.
5. Chen TY, Emrich LJ, Driscoll DL. The clinical signif-
icance of pathological findings in surgically resected
margins of the primary tumor in head and neck carci-
noma. Int J Radiat Oncol Biol Phys 1987; 13: 833–7.
6. Collado M, Serrano M. The power and the promise of
oncogene-induced senescence markers. Nat Rev Cancer
2006; 6: 472–6.
7. Bartkova J, Rezaei N, Liontos M, et al. Oncogene-
induced senescence is part of the tumorigenesis barrier
imposed by DNA damage checkpoints. Nature 2006; 444:
633–7.
8. Bartkova J, Horejsi Z, Koed K, et al. DNA damage
response as a candidate anti-cancer barrier in early human
tumorigenesis. Nature 2005; 434: 864–70.
9. Kato K, Hara A, Kuno T, et al. Aberrant promoter
hypermethylation of p16 and MGMT genes in oral
squamous cell carcinomas and the surrounding normal
mucosa. J Cancer Res Clin Oncol 2006; 132: 735–43.
10. Brennan JA, Mao L, Hruban RH, et al. Molecular
assessment of histopathological staging in squamous-cell
carcinoma of the head and neck. N Engl J Med 1995; 332:
429–35.
11. Partridge M, Li SR, Pateromichelakis S, et al. Detection
of minimal residual cancer to investigate why oral tumors
recur despite seemingly adequate treatment. Clin Cancer
Res 2000; 6: 2718–25.
12. van Houten VM, Leemans CR, Kummer JA, et al.
Molecular diagnosis of surgical margins and local recur-
rence in head and neck cancer patients: a prospective
study. Clin Cancer Res 2004; 10: 3614–20.
13. Fearon ER. Genetic alterations underlying colorectal
tumorigenesis. Cancer Surv 1992; 12: 119–36.
14. Braakhuis BJ, Leemans CR, Brakenhoff RH. A genetic
progression model of oral cancer: current evidence and
clinical implications. J Oral Pathol Med 2004; 33: 317–22.
15. Slaughter DP, Southwick HW, Smejkal W. Field cancer-
ization in oral stratified squamous epithelium; clinical
implications of multicentric origin. Cancer 1953; 6: 963–8.
16. Chandar N, Billig B, McMaster J, Novak J. Inactivation
of p53 gene in human and murine osteosarcoma cells. Br J
Cancer 1992; 65: 208–14.
17. El-Deiry WS, Harper JW, O’Connor PM, et al. WAF1 ⁄ -
CIP1 is induced in p53-mediated G1 arrest and apoptosis.
Cancer Res 1994; 54: 1169–74.
18. Natarajan E, Saeb M, Crum CP, Woo SB, McKee PH,
Rheinwald JG. Co-expression of p16(INK4A) and laminin
Surgical margin of oral carcinomas
Bilde et al.
77
5 gamma2 by microinvasive and superficial squamous cell
carcinomas in vivo and by migrating wound and senescent
keratinocytes in culture. Am J Pathol 2003; 163: 477–91.
19. Thorup AK, Dabelsteen E, Schou S, Gil SG, Carter WG,
Reibel J. Differential expression of integrins and laminin-5
in normal oral epithelia. APMIS 1997; 105: 519–30.
20. Matsuura H, Takio K, Titani K, et al. The oncofetal
structure of human fibronectin defined by monoclonal
antibody FDC-6. Unique structural requirement for the
antigenic specificity provided by a glycosylhexapeptide.
J Biol Chem 1988; 263: 3314–22.
21. Dabelsteen E. Quantitative determination of blood group
substances A of oral epithelial cells by immunofluores-
cence and immunoperoxidase methods. Acta Pathol
Microbiol Scand [A] 1972; 80: 847–53.
22. Wandall HH, Dabelsteen S, Sorensen JA, Krogdahl A,
Mandel U, Dabelsteen E. Molecular basis for the presence
of glycosylated onco-foetal fibronectin in oral carcinomas:
the production of glycosylated onco-foetal fibronectin by
carcinoma cells. Oral Oncol 2007; 43: 301–9.
23. Batsakis JG. Surgical excision margins: a pathologist’s
perspective. Adv Anat Pathol 1999; 6: 140–8.
24. Califano J, van der RP, Westra W, et al. Genetic
progression model for head and neck cancer: implications
for field cancerization. Cancer Res 1996; 56: 2488–92.
25. Tabor MP, Brakenhoff RH, van Houten VM, et al.
Persistence of genetically altered fields in head and neck
cancer patients: biological and clinical implications. Clin
Cancer Res 2001; 7: 1523–32.
26. Park CC, Bissell MJ, Barcellos-Hoff MH. The influence of
the microenvironment on the malignant phenotype. Mol
Med Today 2000; 6: 324–9.
27. Natarajan E, Omobono JD, Jones JC, Rheinwald JG.
Co-expression of p16INK4A and laminin 5 by keratino-
cytes: a wound-healing response coupling hypermotility with
growth arrest that goes awry during epithelial neoplastic
progression. J Investig Dermatol Symp Proc 2005; 10: 72–85.
28. Dabelsteen E, Gron B, Mandel U, Mackenzie I. Altered
expression of epithelial cell surface glycoconjugates and
intermediate filaments at the margins of mucosal wounds.
J Invest Dermatol 1998; 111: 592–7.
29. Pyke C, Romer J, Kallunki P, et al. The gamma 2 chain of
kalinin ⁄ laminin 5 is preferentially expressed in invading
malignant cells in human cancers. Am J Pathol 1994; 145:
782–91.
30. Rocco JW, Sidransky D. p16(MTS-1 ⁄ CDKN2 ⁄ INK4a) in
cancer progression. Exp Cell Res 2001; 264: 42–55.
31. Reed AL, Califano J, Cairns P, et al. High frequency of p16
(CDKN2 ⁄ MTS-1 ⁄ INK4A) inactivation in head and neck
squamous cell carcinoma. Cancer Res 1996; 56: 3630–3.
32. Bradley KT, Budnick SD, Logani S. Immunohistochem-
ical detection of p16INK4a in dysplastic lesions of the oral
cavity. Mod Pathol 2006; 19: 1310–6.
33. Shintani S, Mihara M, Nakahara Y, et al. Expression of
cell cycle control proteins in normal epithelium, prema-
lignant and malignant lesions of oral cavity. Oral Oncol
2002; 38: 235–43.
34. Pande P, Mathur M, Shukla NK, Ralhan R. pRb and p16
protein alterations in human oral tumorigenesis. Oral
Oncol 1998; 34: 396–403.
35. Nylander K, Dabelsteen E, Hall PA. The p53 molecule
and its prognostic role in squamous cell carcinomas of the
head and neck. J Oral Pathol Med 2000; 29: 413–25.
36. van Oijen MG, van de Craats JG, Slootweg PJ. p53
overexpression in oral mucosa in relation to smoking.
J Pathol 1999; 187: 469–74.
J Oral Pathol Med
 Surgical margin of oral carcinomas
Bilde et al.

78
37. McCluggage WG, Connolly LE, McGregor G, Hyland
PL, Hall PA. A strategy for defining biologically relevant
levels of p53 protein expression in clinical samples with
reference to endometrial neoplasia. Int J Gynecol Pathol
2005; 24: 307–12.
38. Farshadpour F, Hordijk GJ, Koole R, Slootweg PJ. Head
and neck squamous cell carcinoma in non-smoking and
non-drinking patients with multiple tumors: etiologic
significance of p53 and Ki-67 in non-tumorous epithelium.
J Oral Pathol Med 2008; 37: 549–554.
39. Cruz IB, Meijer CJ, Snijders PJ, Snow GB, Walboomers
JM, van Der Waal I. p53 immunoexpression in non-
malignant oral mucosa adjacent to oral squamous cell
carcinoma: potential consequences for clinical manage-
ment. J Pathol 2000; 191: 132–7.
40. Cruz IB, Snijders PJ, Meijer CJ, et al. p53 expression
above the basal cell layer in oral mucosa is an early event
of malignant transformation and has predictive value for
developing oral squamous cell carcinoma. J Pathol 1998;
184: 360–8.
41. Kushner J, Bradley G, Jordan RC. Patterns of p53 and
Ki-67 protein expression in epithelial dysplasia from the
floor of the mouth. J Pathol 1997; 183: 418–23.
42. Thurfjell N, Coates PJ, Uusitalo T, et al. Complex p63
mRNA isoform expression patterns in squamous cell
carcinoma of the head and neck. Int J Oncol 2004; 25: 27–35.
43. Tudor D, Locke M, Owen-Jones E, Mackenzie IC.
Intrinsic patterns of behavior of epithelial stem cells.
J Investig Dermatol Symp Proc 2004; 9: 208–14.
44. Costea DE, Tsinkalovsky O, Vintermyr OK, Johannessen
AC, Mackenzie IC. Cancer stem cells – new and poten-
tially important targets for the therapy of oral squamous
cell carcinoma. Oral Dis 2006; 12: 443–54.
45. Vogelstein B, Lane D, Levine AJ. Surfing the p53 network.
Nature 2000; 408: 307–10.
Acknowledgements
The authors would like to express their gratitude to DMD, PhD Karin
Nylander for culturing the cell line used and for help and advice on the
manuscript. Also to DMD Meghna Bhatt and Hanne Lykke Hansen for
help with the immunohistochemical stainings.

J Oral Pathol Med