Eur J Oral Sci 2016; 124: 119–126 Ó 2016 Eur J Oral Sci

DOI: 10.1111/eos.12251 European Journal of

Printed in Singapore. All rights reserved
Oral Sciences

Anti-angiogenic therapy Maha M. Mahmoud1,†, Marwa M. Afifi2,†

1
Oral Medicine, Periodontology, Radiology

(bevacizumab) in the management of

and Diagnosis, Faculty of Dentistry,
Alexandria University, Alexandria; 2Oral
Pathology, Faculty of Dentistry, Alexandria

oral lichen planus

University, Alexandria, Egypt
†
The authors contributed equally.

Mahmoud MM, Afifi MM. Anti-angiogenic therapy (bevacizumab) in the management

of oral lichen planus.
Eur J Oral Sci 2016; 124: 119–126. © 2016 Eur J Oral Sci
Oral lichen planus (OLP), a mucocutaneous chronic inflammatory disease, is con-
ventionally managed using topical corticosteroid therapy. Given the fact that OLP
is strongly linked to angiogenesis, anti-angiogenic drugs, such as bevacizumab,
might be introduced as an alternative treatment for contraindicated, non-responsive
patients. The aim of the present study was to report the short-term effectiveness and
safety of intralesional bevacizumab injection in the management of atrophic/erosive
OLP. A case series study was conducted in patients with atrophic/erosive OLP in
the buccal mucosa, assigned to receive either 2.5 mg of bevacizumab, by intrale-
sional injection (n = 20, test), or topical 0.1% triamcinolone acetonide ointment
(n = 20, control). The size, score, and pain intensity of the lesions were assessed
pre- and post-treatment. Tissue biopsies were collected for histopathologic,
immunohistochemical, and ultrastructural examination. After 1 wk, the test group
had significant reductions both in lesion seize and in pain scores compared with Marwa Afifi, Champillion Street, Azarita,
controls. A marked decrease in vascular endothelial growth factor (VEGF) and Alexandria, 21521, Egypt
interleukin-8 immunoexpression was noted in tissue biopsies from bevacizumab-
E-mail: mafifi@umaryland.edu
treated lesions compared with control lesions. Furthermore, ultrastructural examina-
tion of OLP tissue specimens revealed significant healing signs associated with beva-
cizumab treatment. Short-term data suggest that intralesional bevacizumab injection Key words: angiogenesis; bevacizumab; IL-8;
effectively and safely achieved resolution of atrophic/erosive OLP lesions without oral lichen planus; VEGF
disease exacerbations during a 3-month follow-up period. Accepted for publication December 2015

There is strong evidence behind the synergetic relation-
ship between angiogenesis and chronic inflammation in
pathological conditions. The two phenomena share a
cross-talk in many chronic inflammatory conditions
with separate etiopathogenic origins, including psoria-
sis, rheumatoid arthritis, Crohn’s disease, diabetes, and
lichen planus (1–3). In chronic inflammations, inflam-
matory cells secrete growth factors, proteases, and
cytokines that stimulate extracellular matrix degrada-
tion, endothelial cell growth, and migration, thus pro-
moting angiogenesis. Conversely, blood vessels boost
chronic inflammation providing cytokines, oxygen, and
nutrients, enabling inflammatory-cell chemotaxis (4–7).
This current evolving data led to continuous pursuit of
targeted therapy against both angiogenesis and chronic
inflammation. Targeting angiogenesis can dramatically
decrease the inflow of immune cells, reducing produc-
tion of inflammatory and proteolytic moderators and
thus preventing nutrient supply to an active inflamma-
tory process (8).
Currently, non-steroidal anti-inflammatory drugs
(NSAIDs) and corticosteroids are the mainstay for
treating most chronic inflammatory disorders, including
rheumatoid arthritis and lichen planus. However, they
are not curative and have significant limitations (9–11).
Oral lichen planus (OLP) is a mucocutaneous,
chronic inflammatory disease affecting 1–2% of the
adult population (12) and is sub-classified into reticular,
atrophic, and erosive forms. In the erosive form, the
surface epithelium desquamates and the areas of ulcera-
tion and erosion persist (13). Vascular endothelial
growth factor (VEGF) is believed to be the principal
regulatory component behind endothelial cell prolifera-
tion and differentiation involved in the pathogenesis
and progression of OLP (14, 15). As a result, it seems
likely that suppression of new blood vessel formation
would delay the progression of the disease. Preventing
angiogenesis would minimize extravasation of immune
cells and abrogate pro-inflammatory cytokines (16, 17).
Thus, pairing the positive VEGF immunoreaction in
OLP and the occasional unfavorable response to rou-
tine corticosteroid drugs, angiogensisis can be an excel-
lent target for such new therapeutic modalities. This is
backed up by several reports stating that drugs target-
ing VEGF can reverse the severity and progression of
chronic inflammatory diseases through inhibiting
120 Mahmoud and Afifi
VEGF (18, 19). Several preclinical studies have
reported a targeted therapeutic approach against angio-
genesis in chronic inflammatory disorders. Thus, non-
specific anti-angiogenic compounds, such as TNP-470,
taxol, and thalidomide, have been shown to inhibit
neovascularization and pannus formation in animal
models of arthritis (20–22). Moreover, other studies
have stated that cyclooxygenase-2 (COX-2)-mediated
drugs down-regulate VEGF-induced angiogenesis in
psoriaris (23). Bevacizumab is a humanized monoclonal
antibody that competitively inhibits the VEGF-A fam-
ily in the extracellular space. It is a well-recognized
treatment for ocular neo-vascularization (ONV) and
cancer (24–27). As VEGF seems to play a central role
in the progression of pathological ONV and OLP (8,
27), it seems logical to investigate the therapeutic effi-
cacy of intralesional injection of anti-VEGF (beva-
cizumab) in the management of erosive OLP.
Therefore, the aim of the study was to investigate the
effect of bevacizumab injections on atrophic/erosive
OLP over a 3-month period and compare the results
with those of standard treatment.
Material and methods
The current study followed the guidelines of the Helsinki
Declaration, and it was approved by the Institution Ethics
Committee (Faculty of Dentistry, Alexandria University,
Egypt) for the use of human subjects in research.
Informed consent was obtained from each participant.
Patients and study design
The study included 40 patients diagnosed with atrophic/
erosive OLP lesions in the buccal mucosa, based on a pre-
viously established inclusion criteria and clinical classifica-
tion (28). All participants were experiencing their first
onset of lesions and did not have involvement of the skin,
vulvae, or other anatomic sites. Patients were diagnosed
based on clinical and histopathologic features according to
the World Health Organization (WHO) criteria (29). The
lesions were ≤4 cm2 in size, similar bilateral forms had <1
cm difference in size.
The exclusion criteria were: (i) any uncontrolled sys-
temic disease; (ii) the presence of a lichenoid reaction
caused either by drugs or amalgam restorations (and con-
firmed histologically); (iii) the use of corticosteroid for at
least 3 months before the study; (iv) pregnant or lactating
women, or women taking contraceptive pills; and (v) other
types of OLP. Demographics and baseline clinical findings,
including symptoms and sites of the lesions, were
recorded.
Of the 40 participants, 20 were chosen randomly to be
injected with intralesional bevacizumab (test group). Beva-
cizumab (100 mg/4 ml; Roche, New York, NY, USA) was
diluted with distilled water to 2.5 mg/0.5 ml (27). Intra-
lesion injections were administered into the connective tissue
below the atrophic/erosive lesions (i.e. into adjacent unaf-
fected mucosa) (30). After each injection, the participants
were observed for 30 min for any sign of acute hypersensi-
tivity reactions or other adverse events. The remaining 20
patients (control group) were instructed to use triamci-
nolone acetonide in orabase (Bristol-Myers Squibb, Devens,
MA, USA), three times a day for 3 wk. Patients were asked
to apply the topical treatment after proper mouth hygiene
(to make sure no food debris surrounding the lesions) using
a pad of washed fingers or the applicators. The lesion should
preferably have been dried first for effective adherence.
Patients were also advised to avoid food for a period of time
after treatment application.
Clinical follow up
Patients were evaluated in seven follow-up sessions: after
1 wk; and then every 2 wk for a period of 3 months. The
following parameters were assessed at every follow-up
visit.
(i) Pain intensity, using the visual analog scale from 0 to
10 cm (31).
(ii) Lesion size, which was measured using calibrated Wil-
liams periodontal probes and scored according to previ-
ously described criteria (score 5: white striae with erosive
area >1 cm2; score 4: white striae with erosive area
<1 cm2; score 3: white striae with atrophic area >1 cm2;
score 2: white striae with atrophic area <1 cm2; and score
0: no lesions). The erosive/atrophic area was calculated by
multiplying the maximum diameter (mm) of each erosive
lesion by the maximum width.
(iii) Improvement of each lesion (efficacy indices, EI),
which was measured as follows: no improvement
(EI = 0); mild improvement (0 < EI < 25%); moderate
improvement (25% ≤ EI < 75%); marked improvement
(75% ≤ EI < 100%); and healed (EI = 100%) (32, 33).
Blinding
One clinician carried out the delivery of treatment for the
study and control groups, whereas a different clinician,
who was blind to the treatments, performed pre- and post-
therapeutic, and follow-up evaluations.
Histopathology
Tissue incisional biopsies were collected from areas most
representative of the erosive OLP lesions. This was per-
formed before and 1 wk after treatment to confirm clinical
diagnosis and for further purposes.
Specimens were fixed in 10% neutral buffered formalin,
processed, and embedded in paraffin wax. Serial sections
of 4 lm thickness were cut, placed on glass slides, and
stained using hematoxylin and eosin for routine
histopathologic examination using light microscopy.
Immunohistochemistry
Immunohistochemical staining was performed using strep-
tavidin–biotin (17). Immunoexpression of VEGF was used
to compare the angiogenesis activity rates before, and
1 wk after, bevacizumab treatment (34). In addition, inter-
leukin-8 (IL-8) immunoexpression, which is directly related
to the course of OLP development, was assessed (28).
Anti-IL-8 , anti-VEGF antibodies were used to evaluate
the efficacy of the introduced treatment in comparison
with the control side. Briefly, tissue sections were deparaf-
finized, rehydrated, and treated with 3% H2O2 to quench
endogenous peroxidase activity. After washing in 10 ml of
Tris-buffered saline (TBS), sections were incubated with
10% normal goat serum (Zymed Laboratories, San
 Bevacizumab to treat oral lichen planus 121

Table 1
Clinical signs and symptoms associated with patients with oral lichen planus (OLP) at inclusion

Site: buccal Disease duration at inclusion Size of atrophic/erosive area*

Group Age (yr) pouch Pain level (wk) (cm2)

Study group (n = 20) 43  1.3 100 6.90  0.4 13.3  0.88 0.72  0.02
Control group 42  1.1 100 6.79  0.4 13.5  1.02 0.75  0.04
(n = 20)

Values are given as mean  SD or as a percentage.

*Atrophic/erosive lesion size (cm2) = width 9 length.

Fransisco, CA, USA) to block non-specific binding. Tissue A

sections were then incubated overnight at 4 C with a 1:50
dilution of mouse monoclonal anti-IL-8 and VEGF anti-
body (PC10; DakoPatts, Carpinteria, CA , USA). Biotiny-
lated anti-mouse IgG was used and subsequently, sections
were soaked with a streptavidin–peroxidase conjugate
(Zymed Laboratories). Diaminobenzidine hydrochloride
0.02% was used as chromogen to visualize the peroxidase
activity. The preparations were lightly counterstained with
hematoxylin, mounted with Permount (Merck, Kenil-
worth, NJ, USA), and examined by light microscopy. The B
immunoreactivity of both markers was quantified using
the image analyzer computer system (Leica Qwin 500 soft-
ware, Buffalo Grove, IL, USA).

Transmission electron microscopy

Fresh tissue biopsies were trimmed into 1 mm3 small pieces,
fixed in 3% glutaraldehyde, washed in phosphate buffer,
and then fixed in 1% osmium tetraoxide for 2 h at 4°C. The
tissue specimens were dehydrated and embedded in araldite
capsules. Sections were mounted on copper grids; double
stained with freshly prepared uranyl acetate and lead citrate,
and finally examined using transmission electron micro-
scopy (TEM) (JEOL-100CX; Redding, CA, USA).
Statistical analysis
The IBM SPSS software package version 20.0 was used
for data analysis (35). Qualitative data were described as
number and percent, and differences between the test and
control groups were tested using the chi-square test. Quan-
titative data were described as means with standard devia-
tion. P < 0.05 was considered statistically significant.
Results
Clinical effects
Clinical signs and symptoms of the 40 enrolled patients
(35 women and five men) diagnosed with atrophic/ero-
sive OLP in the buccal mucosa are presented in
Table 1. The test and control groups did not differ in
age, OLP location, size of atrophic/erosive area, pain
level, and disease duration at inclusion.
One week after injection, the average size and pain
intensity scores of the bevacizumab-treated lesions had
decreased compared with pre-injection and with the
corticosteroid-treated control lesions (Fig. 1). Thus,
within 1 wk, the pain intensity and lesion scores were
Fig. 1. (A) Left panel, a 6 mm 97 mm atrophic/erosive area
of the buccal mucosa. The total atrophic/erosive area (mm2)
was calculated by multiplying the maximum diameter by the
maximum width of the erosive areas (circled). The erosive
areas (encircled) were surrounded with white striation. Right
panel, 1 wk following intralesional bevacizumab injection,
marked signs of healing can be seen. (B) A control patient
with oral lichen planus (OLP) who was treated using conven-
tional triamcinolone acetonide 0.1% ointment, three times a
week for 1 wk (arrow). Right panel, note the minimal
changes, after 1 wk, associated with corticosteroid therapy.
reduced by 80.3% and 78.7%, respectively (both
P < 0.001), in bevacizumab-treated lesions compared
with a 20% increase in the pain-intensity scores and no
change in lesion size for the control lesions at the same
time point (Fig. 2). Although the mean values of pain
intensity and lesion size were significantly decreased
1 wk after injection of bevacizumab, corticosteroid-
treated lesions needed 4 and 9 wk before an equivalent
reduction of pain intensity and lesion size, respectively,
was achieved (Fig. 2). Moreover, 93.3% of OLP lesions
treated with one dose of bevacizumab by injection
showed complete remission within 1 wk. The mean
improvement efficacy for bevacizumab- and corticos-
teroid-treated lesions was 97% and 27%, respectively,
after 1 wk (Fig. 2).
Histopathology
At inclusion, microscopic examination of all OLP tissue
biopsies of test and control groups revealed an alternat-
122 Mahmoud and Afifi
Fig. 2. Representative graphs showing pain intensity (A),
lesion scores (B), and improvement efficacy (C) of erosive oral
lichen planus (OLP) treated with intralesional injection of beva-
cizumab (2.5 mg/0.5 ml) vs. triamcinolone acetonide 0.1%
ointment (applied three times a week) with time. Data are pre-
sented as mean  SD. (A) A significant decrease (P < 0.001) in
pain intensity associated with OLP lesions after one injection of
bevacizumab (black squares) is apparent at week 1 compared
with the control (red circles) (1.4  2.1 for bevacizumab vs.
7.4  2.9 for control). (B) Lesion scores for bevacizumab-
injected OLP lesions (black squares) were 0.9  0.3 after 1 wk
compared with 3.88  0.9 for controls (red circles). (C) The
healing efficacy of OLP lesions was significantly improved
(92  2.3) 1 wk after an intralesional injection of bevacizumab
(black squares), whereas controls (red circles) needed 3 wk to
reach the same value.
ing area of atrophic and erosive squamous epithelium,
showing liquifactive degeneration of the basal cell layer
and a sub-epithelial lymphocytic band. The underlying
lamina propria was extensively infiltrated by blood ves-
sels. However, 1 wk after bevacizumab injection the
sub-epithelial lymphocytic infiltration in the treated
OLP lesions was significantly decreased compared with
the control lesions (Fig. 3).
Immunohistochemistry
Vascular endothelial growth factor immunoreactivity
was observed in all tissue sections extracted from OLP
lesions (study and control groups) before treatment
(Fig. 4a,c,e). The immunoreaction was seen as brown
cytoplasmic staining, mainly in keratinocytes, and it
was observed in all layers of the epithelium, including
the basal cell layer. Positive immunoreactivity was also
observed in stromal fibroblasts, lymphocytes, and
A B
C D
Fig. 3. Histopathologic images of oral lichen planus (OLP)
tissue biopsies extracted from test and control groups. (A)
Extensive sub-epithelial lymphocytic band [hematoxylin and
eosin (H&E); 9100]. (B) Lamina propria of OLP showing
heavy infiltration of lymphocytes and extensive vascularity
(H&E; 9400). (C) Liquifactive degeneration of the basal cell
layer (arrows; H&E; 9400). (D) A case of erosive OLP after
intralesional injection of bevacizumab, showing a marked
decrease in sub-epithelial lymphocytic infiltration (H&E;
9200).
endothelial cells of blood vessels. One week after beva-
cizumab injection, VEGF immunoreactivity was
reduced by 53% throughout the whole thickness of the
epithelium and the underlying lamina propria com-
pared with a reduction of only 0.29% in controls
(P < 0.001) (Fig. 4b,d,f; Fig. 6).
At inclusion, tissue sections from untreated OLP
lesions revealed IL-8 immunoexpression as brown dif-
fuse cytoplasmic immunoreactivity throughout the
whole thickness of the epithelium and underlying stro-
mal cells in all lesions in the test group (Fig. 5a,c,e). A
similar pattern was seen in all specimens from OLP
lesions treated in the control group. After 1 wk, tissue
sections from patients with OLP treated with beva-
cizumab showed a significant decrease (23%) in IL-8
immunoreactivity compared with controls (1.39%)
(P < 0.001) (Fig. 5b,d,f; Fig. 6).
Ultrastructural findings
As shown in Fig. 7, degeneration of the basal cell layer
and breaks in the basement membrane, wide intercellu-
lar spaces and discontinued desmosomal attachments,
and entrapped lymphocytes within the basal cell layer
were evident at inclusion. Apoptotic changes of the ker-
atinocytes, including nuclear shrinkage, chromatin con-
densation, cytoplasmic vacuolation, and blebbing, were
also seen (Fig. 7a–d). However, tissue biopsies
extracted from OLP lesions, 1 wk after bevacizumab
treatment, revealed narrowing of the intercellular
spaces and formation of desmosomal attachments. In
addition, basal lamina re-formation was evident, sepa-
rating the epithelium from the underlying connective
tissue (Fig. 7e–h).
122 Mahmoud and Afifi
Fig. 2. Representative graphs showing pain intensity (A),
lesion scores (B), and improvement efficacy (C) of erosive oral
lichen planus (OLP) treated with intralesional injection of beva-
cizumab (2.5 mg/0.5 ml) vs. triamcinolone acetonide 0.1%
ointment (applied three times a week) with time. Data are pre-
sented as mean  SD. (A) A significant decrease (P < 0.001) in
pain intensity associated with OLP lesions after one injection of
bevacizumab (black squares) is apparent at week 1 compared
with the control (red circles) (1.4  2.1 for bevacizumab vs.
7.4  2.9 for control). (B) Lesion scores for bevacizumab-
injected OLP lesions (black squares) were 0.9  0.3 after 1 wk
compared with 3.88  0.9 for controls (red circles). (C) The
healing efficacy of OLP lesions was significantly improved
(92  2.3) 1 wk after an intralesional injection of bevacizumab
(black squares), whereas controls (red circles) needed 3 wk to
reach the same value.
ing area of atrophic and erosive squamous epithelium,
showing liquifactive degeneration of the basal cell layer
and a sub-epithelial lymphocytic band. The underlying
lamina propria was extensively infiltrated by blood ves-
sels. However, 1 wk after bevacizumab injection the
sub-epithelial lymphocytic infiltration in the treated
OLP lesions was significantly decreased compared with
the control lesions (Fig. 3).
Immunohistochemistry
Vascular endothelial growth factor immunoreactivity
was observed in all tissue sections extracted from OLP
lesions (study and control groups) before treatment
(Fig. 4a,c,e). The immunoreaction was seen as brown
cytoplasmic staining, mainly in keratinocytes, and it
was observed in all layers of the epithelium, including
the basal cell layer. Positive immunoreactivity was also
observed in stromal fibroblasts, lymphocytes, and
A B
C D
Fig. 3. Histopathologic images of oral lichen planus (OLP)
tissue biopsies extracted from test and control groups. (A)
Extensive sub-epithelial lymphocytic band [hematoxylin and
eosin (H&E); 9100]. (B) Lamina propria of OLP showing
heavy infiltration of lymphocytes and extensive vascularity
(H&E; 9400). (C) Liquifactive degeneration of the basal cell
layer (arrows; H&E; 9400). (D) A case of erosive OLP after
intralesional injection of bevacizumab, showing a marked
decrease in sub-epithelial lymphocytic infiltration (H&E;
9200).
endothelial cells of blood vessels. One week after beva-
cizumab injection, VEGF immunoreactivity was
reduced by 53% throughout the whole thickness of the
epithelium and the underlying lamina propria com-
pared with a reduction of only 0.29% in controls
(P < 0.001) (Fig. 4b,d,f; Fig. 6).
At inclusion, tissue sections from untreated OLP
lesions revealed IL-8 immunoexpression as brown dif-
fuse cytoplasmic immunoreactivity throughout the
whole thickness of the epithelium and underlying stro-
mal cells in all lesions in the test group (Fig. 5a,c,e). A
similar pattern was seen in all specimens from OLP
lesions treated in the control group. After 1 wk, tissue
sections from patients with OLP treated with beva-
cizumab showed a significant decrease (23%) in IL-8
immunoreactivity compared with controls (1.39%)
(P < 0.001) (Fig. 5b,d,f; Fig. 6).
Ultrastructural findings
As shown in Fig. 7, degeneration of the basal cell layer
and breaks in the basement membrane, wide intercellu-
lar spaces and discontinued desmosomal attachments,
and entrapped lymphocytes within the basal cell layer
were evident at inclusion. Apoptotic changes of the ker-
atinocytes, including nuclear shrinkage, chromatin con-
densation, cytoplasmic vacuolation, and blebbing, were
also seen (Fig. 7a–d). However, tissue biopsies
extracted from OLP lesions, 1 wk after bevacizumab
treatment, revealed narrowing of the intercellular
spaces and formation of desmosomal attachments. In
addition, basal lamina re-formation was evident, sepa-
rating the epithelium from the underlying connective
tissue (Fig. 7e–h).
124 Mahmoud and Afifi

decreased angiogenesis and consequently in reversal of

Fig. 6. Representative plot showing a significant decrease in
the immunostaining of both vascular endothelial growth fac-
tor (VEGF) (53%) and interleukin-8 (IL-8) (23%) in lesions
injected with bevacizumab (black asterisk) vs. controls that
showed almost the same immunoexpression as biopsies
extracted from untreated lesions. Data represent mean  SD
of three independent experiments. Statistical significance is
denoted by an asterisk (P < 0.05).
the course of the disease. This was confirmed by the
significant decrease in immunoreactivity of IL-8 in OLP
tissue biopsies injected with bevacizumab compared
with controls. These findings suggest that VEGF is an
important target in the process of new vessel formation
associated with erosive OLP. Bevacizumab could help
to control the active stage of the disease in a very short
period of time compared to traditional corticosteroid
therapy by depressing numerous inflammatory media-
tors, including IL-6 and IL-8, and reducing VEGF
expression (38). Anti-VEGF therapy (e.g. bevacizumab)
is well recognized in the literature as a promising treat-
ment option for ONV and diabetic macular edema
(DME). Vascular endothelial growth factor plays a crit-
ical role in breaking down the blood–retina barrier,
with increased vascular permeability resulting in retinal
edema associated with DME (27). Interestingly, based
on our previously stated data in the current study,

A B C D

E F G H

Fig. 7. Transmission electron microscopy (TEM) micrographs of oral lichen planus (OLP) tissue sections, before and 1 wk after
injection of bevacizumab. (A) Epithelial cells show marked separation and break down of their desmosomal junctions (red aster-
isk). Marked sub-epithelial bacterial invasion can be seen (red arrow) (91000). (B) Part of the oral epithelium adjacent to lamina
propria. Basal cells undergoing apoptotic changes are indicated (red arrow). Note the discontinuity in the basement membrane
(white arrow) (94000). (C) The epithelial cells show marked, wide intercellular spaces (red asterisk). Lymphocytes are seen among
the separated disorganized epithelial cells (arrow) (92000). (D) Apoptotic basal cells are in close proximity to red blood cells
(RBC) (92000). (E) Well-organized prickle cells of the oral epithelium are evident, with normal cytoplasmic electron density, nor-
mal nuclear appearance, and normal density of the desmosomal junctions with the adjacent cells (9500). (F) Formation of base-
ment membrane (arrow) (94000). (G) Two well-organized prickle cells showing normal nuclear density and normal desmosomal
attachments (92000). (H) Well-organized prickle cell of the oral epithelium, with normal cytoplasmic electron density, normal
nuclear appearance, and normal density of the desmosomal junctions with the contacting cells (92500).

bevacizumab can now be considered as the treatment
of choice for chronic inflammatory angiogenesis-depen-
dent diseases, including atrophic/erosive OLP.
Ultrastructural examinations of the untreated
atrophic/erosive OLP biopsies revealed morphologic
features that were similar to previously reported ultra-
structural features associated with OLP (39). Basal lam-
ina exhibited breaks, and degenerative changes in the
basal cell layer were evident. The prickle cell layer
showed marked widening of the intercellular spaces and
loss of desmosomal attachments. Apoptotic changes
were seen, including increased nuclear chromatin, cyto-
plasmic vacuolization, and blebbing. As a result of the
breaks in the basement membrane, lymphocytes were
trapped within the basal and prickle cell layers. There
was a marked bacterial infection adherent to the disor-
ganized layer of epithelial cells. This is likely to be seen
due to the superimposed inflammation. Both the breaks
in the basement membrane and sub-epithelial bacterial
invasion can explain the increased pain intensity associ-
ated with atrophic/erosive forms of lichen planus.
Transmission electron microscopy photomicrographs of
OLP tissue biopsies revealed outstanding findings in
OLP tissue biopsies injected with bevacizumab, namely
that pretreatment bacterial infection, discontinuity of
the basal lamina, apoptosis, and intercellular spaces
were reversed. This strongly confirms the healing effi-
cacy of bevacizumab after only one injection. The
decreased bacterial infection was probably responsible
for the reduced signs of inflammation, including pain.
In addition, our TEM images illustrate the remarkable
angiogenic changes associated with OLP (Fig. 7). This
was previously described by others (8, 17). Thus, anti-
angiogenic drugs seem to be a promising approach in
the treatment of OLP, especially of erosive types.
The optimal dose of intralesional bevacizumab for
treating OLP is still undetermined. Doses of beva-
cizumab used in the literature have ranged from 1.25 to
2.5 mg in treating diabetic retinopathy (40, 41). We
chose the higher dose of 2.5 mg in the current work in
the hope of achieving a prolonged therapeutic effect
and minimizing the number of repeated injections.
However, a bevacizumab dose–response evaluation over
time is clearly needed in future studies.
Although there were no adverse effects at the site of
bevacizumab injection during the observation period,
longer follow-up periods are required to assess long-
term safety and efficacy of the proposed treatment.
These observations are in agreement with previous
in vitro studies that showed no significant side effects in
cells of the anterior segment of the eye after application
of bevacizumab (42). Moreover, other studies have
reported that bevacizumab showed no significant side
effects when injected intravitreally in the treatment of
choroidal and retinal neovascularization in healthy sub-
jects and patients with diabetes (42).
To conclude, in the current study, the intralesional
bevacizumab injection reduced the signs and symptoms
in 100% of patients with atrophic/erosive OLP. Further-
more, our preliminary results support that angiogenesis
is one of the main etiopathogenic factors in OLP lesions
Bevacizumab to treat oral lichen planus 125
and that targeting VEGF could rapidly terminate the dis-
ease process. Intralesional bevacizumab injections seem
to be a relatively safe, effective, and well-tolerated option
for the treatment of progressive OLP. Intralesional beva-
cizumab therapy may be a favorable alternative in con-
traindicated, non-responsive, or resistant patients.
Future controlled prospective randomized trials are nec-
essary to learn more about the long-term safety, tolera-
bility, and efficacy of bevacizumab. Additional research
will be needed, including a larger sample size, an
extended study period, and a panel of different beva-
cizumab doses, to validate these results.
Acknowledgements – This study was financially supported by
Alexandria University, Egypt.
Conflicts of interest – The authors declare no conflicts of interest.
References
1. SKLAVOUNOU-ANDRIKOPOULOU A, CHRYSOMALI E, IAKOVOU M,
GARINIS GA, KARAMERIS A. Elevated serum levels of the
apoptosis related molecules TNF-alpha, Fas/Apo-1 and
Bcl-2 in oral lichen planus. J Oral Pathol Med 2004; 33:
386–390.
2. ARUNKUMAR S, KALAPPANAVAR AN, ANNIGERI RG, KALAPPA
SG. Relative efficacy of pimecrolimus cream and triamci-
nolone acetonide paste in the treatment of symptomatic oral
lichen planus. Indian J Dent 2015; 6: 14–19.
3. SANATKHANI M, MOSANNEN MOZAFARI P, AMIRCHAGHMAGHI
M, NAJAFI FATHI M, SANATKHANI M, SARJAMI N, AZARIAN
AA. Effect of cedar honey in the treatment of oral lichen pla-
nus. Iran J Otorhinolaryngol 2014; 26: 151–161.
4. COSTA C, INCIO J, SOARES R. Angiogenesis and chronic
inflammation: cause or consequence? Angiogenesis 2007; 10:
149–166.
5. SCARDINA GA, RUGGIERI A, MESSINA P, MARESI E. Angiogen-
esis of oral lichen planus: a possible pathogenetic mechanism.
Med Oral Patol Oral Cir Bucal 2009; 14: e558–e562.
6. MARRELLI A, CIPRIANI P, LIAKOULI V, CARUBBI F, PERRI-
CONE C, PERRICONE R, GIACOMELLI R. Angiogenesis in
rheumatoid arthritis: a disease specific process or a com-
mon response to chronic inflammation? Autoimmun Rev
2011; 10: 595–598.
7. SCARDINA GA, RUGGIERI A, MARESI E, MESSINA P. Angiogen-
esis in oral lichen planus: an in vivo and immunohistological
evaluation. Arch Immunol Ther Exp 2011; 59: 457–462.
8. CARVALHO JF, BLANK M, SHOENFELD Y. Vascular endothelial
growth factor (VEGF) in autoimmune diseases. J Clin Immu-
nol 2007; 27: 246–256.
9. CHEN X, LIU Z, YUE Q. The expression of TNF-alpha and
ICAM-1 in lesions of lichen planus and its implication.
J Huazhong Univ Sci Technolog Med Sci 2007; 27: 739–741.
10. HUSSEIN MR. Evaluation of angiogenesis in normal and
lichen planus skin by CD34 protein immunohistochemistry:
preliminary findings. Cell Biol Int 2007; 31: 1292–1295.
11. MCCAUGHEY C, MACHAN M, BENNETT R, ZONE JJ, HULL CM.
Pimecrolimus 1% cream for oral erosive lichen planus: a 6-
week randomized, double-blind, vehicle-controlled study with
a 6-week open-label extension to assess efficacy and safety. J
Eur Acad Dermatol Venereol 2011; 25: 1061–1067.
12. DI STASIO D, GUIDA A, SALERNO C, CONTALDO M, ESPOSITO
V, LAINO L, SERPICO R, LUCCHESE A. Oral lichen planus: a
narrative review. Front Biosci 2014; 6: 370–376.
13. NICO MM, FERNANDES JD, LOURENCO SV. Oral lichen planus.
An Bras Dermatol 2011; 86: 642–633.
14. MONACO C, ANDREAKOS E, KIRIAKIDIS S, FELDMANN M, PALE-
OLOG E. T-cell-mediated signalling in immune, inflammatory
and angiogenic processes: the cascade of events leading to
124 Mahmoud and Afifi

decreased angiogenesis and consequently in reversal of

Fig. 6. Representative plot showing a significant decrease in
the immunostaining of both vascular endothelial growth fac-
tor (VEGF) (53%) and interleukin-8 (IL-8) (23%) in lesions
injected with bevacizumab (black asterisk) vs. controls that
showed almost the same immunoexpression as biopsies
extracted from untreated lesions. Data represent mean  SD
of three independent experiments. Statistical significance is
denoted by an asterisk (P < 0.05).
the course of the disease. This was confirmed by the
significant decrease in immunoreactivity of IL-8 in OLP
tissue biopsies injected with bevacizumab compared
with controls. These findings suggest that VEGF is an
important target in the process of new vessel formation
associated with erosive OLP. Bevacizumab could help
to control the active stage of the disease in a very short
period of time compared to traditional corticosteroid
therapy by depressing numerous inflammatory media-
tors, including IL-6 and IL-8, and reducing VEGF
expression (38). Anti-VEGF therapy (e.g. bevacizumab)
is well recognized in the literature as a promising treat-
ment option for ONV and diabetic macular edema
(DME). Vascular endothelial growth factor plays a crit-
ical role in breaking down the blood–retina barrier,
with increased vascular permeability resulting in retinal
edema associated with DME (27). Interestingly, based
on our previously stated data in the current study,

A B C D

E F G H

Fig. 7. Transmission electron microscopy (TEM) micrographs of oral lichen planus (OLP) tissue sections, before and 1 wk after
injection of bevacizumab. (A) Epithelial cells show marked separation and break down of their desmosomal junctions (red aster-
isk). Marked sub-epithelial bacterial invasion can be seen (red arrow) (91000). (B) Part of the oral epithelium adjacent to lamina
propria. Basal cells undergoing apoptotic changes are indicated (red arrow). Note the discontinuity in the basement membrane
(white arrow) (94000). (C) The epithelial cells show marked, wide intercellular spaces (red asterisk). Lymphocytes are seen among
the separated disorganized epithelial cells (arrow) (92000). (D) Apoptotic basal cells are in close proximity to red blood cells
(RBC) (92000). (E) Well-organized prickle cells of the oral epithelium are evident, with normal cytoplasmic electron density, nor-
mal nuclear appearance, and normal density of the desmosomal junctions with the adjacent cells (9500). (F) Formation of base-
ment membrane (arrow) (94000). (G) Two well-organized prickle cells showing normal nuclear density and normal desmosomal
attachments (92000). (H) Well-organized prickle cell of the oral epithelium, with normal cytoplasmic electron density, normal
nuclear appearance, and normal density of the desmosomal junctions with the contacting cells (92500).