Histopathological and Immunohistochemical Study of the Distinction between Oral Lichen Planus and Oral Lichenoid Lesions ()
1. Introduction
Oral lichen planus (OLP) is a chronic inflammatory disease involving abnormal keratinization of the oral mucosa [1] [2]. Several clinical types of OLP have been described. The most common is the reticular form, involving the buccal mucosa with symmetrical lesions, characterized by numerous interlacing white keratotic lines or striae (so-called Wickham’s striae) and few symptoms [1]. In addition, OLP has histopathologically demonstrated hyperkeratosis or parakeratosis, a saw-tooth profile of the rete ridges, liquefaction degeneration of the basal cell layer, and compact and band-like lymphocytic (predominantly T-cell) infiltration of the epitheliomesenchymal junction [3]. As mentioned above, although OLP is diagnosed by combining gross and histopathological findings, concordance rates are low compared to other oral mucosal diseases [4]. Our previous study proposed oblique running of papillary loops to white striae as a histopathological finding allowing definitive diagnosis of OLP, because this finding is observed regardless of the stage [4]. In addition, the basal cell layer is lost due to liquefaction degeneration and lymphocytes were reportedly in contact with the spinous layer [4]. Various histopathological reports have provided descriptions of the basal cell layer in OLP, but no consensus has yet been reached [5] [6] [7].
On the other hand, clinically and histopathologically similar cases have been distinguished as oral lichenoid lesions (OLLs) in the oral mucosa. Differentiation between OLP and OLLs is extremely difficult unless the cause of the pathological condition can be identified [8]. OLLs appear clinically similar to OLP, and also present with liquefaction degeneration of the basal cell layer and an appearance of civatte bodies histopathologically [9].
In 2017, precancer lesions and precancer conditions were combined to create the new concept of oral potentially malignant disorders (OPMDs), which included OLP [10]. OPMDs are clinical presentations that carry a risk of development to cancer in the oral cavity [10]. OLLs are also termed interface mucositis or lichenoid mucositis [11]. Malignant transformation of these lesions remains controversial, but distinct clinical and histological criteria for how to differentiate OLP from OLLs have not been developed [11].
The purpose of this study was thus to elucidate those findings that can allow histopathological differentiation of OLP and OLLs using histomorphological and immunohistochemical analyses.
2. Materials and Methods
2.1. Subjects
Histopathological, and immunohistochemical evaluations were performed in 10 cases (5 males, 5 females) diagnosed with OLP and 9 cases (4 males, 5 females) diagnosed with OLLs. Subjects were selected from the pathology files of the Department of Oral Pathology in Nihon University School of Dentistry at Matsudo from 2000 to 2019. Mean age was 50.3 ± 11.4 years in OLP and 63.8 ± 14.0 years in OLLs. Subject characteristics in this study are summarized in Table 1. Cases diagnosed as OLP or OLLs by 3 dentists (including an oral surgeon) were based on gross findings of surface shape, background, and intraoral pictures of the lesion described in the electronic medical records. The conditions for the case selection of the present study are shown. At first, complete macroscopic view records were available for all patients, along with dental history of the lesion and a biopsy specimen comprising an area with epithelium and submucosal tissue. As for OLP, the cases were of unknown cause [9]. Meanwhile OLLs were considered as cases with a cause, such as, drugs, dental restorative materials, GVHD and other associated factors [9]. Patients with a history of exposure to dental materials, drugs [12], any treatment for lichen planus or drugs associated with lichenoid reaction before biopsy, any malignant or viral involvement in the mouth and pregnant women were excluded from the study sample.
Macroscopic findings were reconfirmed and classified by oral surgeons for 19 patients with well-defined looping and intersecting white lines/striae/patches with or without erosions and ulcerations [13] [14] [15] in the present study. For both lesions, clinical inspection classifications for the most prominent site were made independently by 2 oral surgeons and 3 oral pathologists based on the 6 types described by Andreasen [16] and the 2 types defined by Brant [17]. In cases of disagreement, classifications were discussed in a joint session until a consensus was reached. Biopsy specimens were obtained for all 19 patients by oral surgeons, sampling an area at the boundary of the most prominent lesion and an adjacent area of normal mucosa.
2.2. Histological Preparation
Histological specimens of OLP were stained using hematoxylin and eosin and selected by the following definitive histopathological criteria according to the American Academy of Oral and Maxillofacial Pathology: 1) presence of a well-defined band-like zone of cellular infiltration consisting mainly of lymphocytes in the superficial part of the connective tissue; and 2) signs of “liquefaction degeneration” in the basal layer; and 3) absence of epithelial dysplasia [18]. Conversely, specimens of OLLs stained by HE were selected by the criteria described in previous reports [9] [18]: 1) presence of associated factors; 2) diffuse and deeper distribution of lymphocytes; 3) no degeneration of basement membrane; and 4) no vascularity in basal cells. Patients with lesions that did not reflect the above strict histological criteria were not included in the study. Cases
accompanied by secondary inflammation with ulcer/erosion and evidence of malignancy were also excluded. The normal oral mucosae were obtained through polypectomy from perilesional areas of fibroma without epithelial dysplasia to constitute the control group (2 males, 1 female; 3 samples of buccal mucosae). Four oral pathologists blinded to the biopsy material made the histopathological diagnosis of OLP and OLLs, and the assessments of immunohistochemical staining. Histopathological and immunohistochemical evaluations were identical among the 4 oral pathologists. Nineteen specimens underwent Papanicolaou staining.
2.3. Immunohistochemical Preparation
Immunohistochemical studies were conducted using 10% neutral formalin solution-fixed, paraffin-embedded tissues from all cases. Five serial sections (each 4 µm thick) were prepared and deparaffinized in xylene and hydrated in graded ethanol solution for further immunohistochemical analysis to calculate microvascular irregularities. The EnVision + Polymer System (Dako, Glostrup, Denmark), which also carries secondary antibody molecules, was used for antigen detection. Cytokeratin 19 (CK19, RCK108, 1:50; DakoCytomation, Glostrup, Denmark), Ki-67 (MIB-5, 1:50; DakoCytomation) and CD3 (F7.2.38, 1:50; DakoCytomation) were used as primary antibodies to detect basal cells, proliferative activity and T-cell distribution, respectively. Perlecan (received from Prof. Saku, Osaka Dental University) [19] [20] and COX-2 (CX-294, 1:100; DakoCytomation) to evaluate epithelial intracellular arrangements and interstitial distributions of proteoglycans and enzymes. Sections were developed in a solution of 3,3'-dianibobenzidine tetrahydrochloride. Finally, all sections were counterstained with Mayer’s hematoxylin. Inflammatory oral mucosa and epithelial dysplasia were used as positive controls for primary antibodies. To evaluate the immunohistochemical staining technique, mouse and rabbit universal g-negative controls (DakoCytomation) were used as negative controls during the staining procedure instead of primary antibodies. Slides were examined under light microscopy and projected on a color monitor. All specimens of an area at the boundary part of the most prominent lesion and an adjacent area of normal mucosa were randomly photographed with ≥5 fields of view (×40), positive reactions to the various antibodies were evaluated. The thicknesses of the cornified, spinous and basal layers were measured in Papanicolaou-stained specimens. A digital camera for the microscope was used to obtain images, and CellSens Standard software (OLYMPUS, Tokyo, Japan) was used for image analysis.
2.4. Statistical Assessment
All statistical analyses were performed using SPSS for Windows version 14. OJ (IBM, Tokyo, Japan). Statistical analyses for positive rates of CK19, Ki-67 and COX-2 were performed using the Mann-Whitney U test. Comparative analyses for average thicknesses of each layer were performed using Friedman’s test and Scheff’s test. Values of p < 0.05 were considered significant.
2.5. Compliance with Ethical Standards
Informed consent was obtained from all individuals included in the study. All procedures performed in studies involving human participants were in accordance with the ethical standards of the Committee on Studies Involving Human Beings of Nihon University School of Dentistry at Matsudo (EC-18-15-14-2) and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.
3. Results
3.1. Clinicopathological Findings
The distribution of clinicopathological findings is shown in Table 1. Most of the 10 OLP patients (15 sites) showed the main lesion involving the buccal mucosa (14 sites), followed by the gingiva (1 site). Most of the 9 patients with OLLs (16 sites) likewise showed the main lesion involving the buccal mucosa (12 sites), followed by the gingiva (4 sites). The validity of the clinical diagnosis of OLP was consistent at 5 sites, but the remaining included 4 carcinomas and 1 instance of leukoplakia. For OLLs, OLP, malignant tumor and leukoplakia were 6, 2 and 1, respectively. In terms of chief complaints, 10.0% and 33.3% with OLP and 33.3% and 11.1% with OLLs presented with pain and discomfort, respectively. Figure 1 shows representative macroscopic pictures of OLP (Figure 1(a)) and OLLs (Figure 1(b)). Both show similar clinical conditions with a mixture of white striae and erosions. No specific macroscopic findings were found for either disease. Representative case where treatment was successful is shown in Figure 1(e) and Figure 1(f). OLP patient without dental metal allergies of the left buccal mucosa with white patch and erosion pre- (e) and post-treatment (f). Most of the white patch and erosions disappeared after treatment.
3.2. Histopathological Findings
Figure 1 shows representative histopathological pictures of OLP (Figure 1(c)) and OLLs (Figure 1(d)). In OLP, at sites where basal cells were lost due to degeneration, spinous cells were in direct contact with band-like lymphocyte infiltration in the subepithelial connective tissue (Figure 1(c)). OLLs showed slight band-like or diffuse infiltration of lymphocytes into subepithelial connective tissue, although no loss of basal cells was observed (Figure 1(d)). In high-power fields (Figures 2(a)-(c)), melanin pigmentation just below the basal membrane was observed in all 10 OLP specimens (100%) and 2 of 9 OLL specimens (22%).
3.3. Immunohistochemical and Papanicolaou Findings
The results of immunohistochemical staining are shown in Figure 3. For CK19 (Figure 3(a)), positive findings were continuously expressed in the basal layer of normal mucosa. In OLLs, CK19-positive findings were discontinuous in the
![]()
Figure 1. Gross and histopathological findings of oral lichen planus (OLP) and oral lichenoid lesions (OLLs). (a, b): OLP (a) and OLLs (b) of the right buccal mucosa with white striae and erosive lesion; (c): Liquefaction degeneration in the basement membrane and loss of basal cells in OLP (Hematoxyline and Eosin staining, x20); (d): No degeneration in the basement membrane and vascularity in basal cells in OLLs (Hematoxyline and Eosin staining, x20); (e, f): OLP patient without dental metal allergies of the left buccal mucosa with white patch and erosion pre- (e) and post-treatment (f).
![]()
Figure 2. High power field of basal layer and subepithelial tissue of oral lichen planus (OLP) and oral lichenoid lesions (OLLs) and normal tissue (Hematoxyline and Eosin staining, x20) (a): Melanin pigment and melanophores are scattered in OLP; (b): Many mixed inflammatory cells are infiltrated, but no melanophores in OLLs; (c): Very few lymphocytes are scattered in normal mucosa.
![]()
Figure 3. Photographs of each representative immunohistochemical and Papanicolaou staining results. *: the most prominent lesion; **: an adjacent area of normal mucosa; Pap.: Papanicolaou staining; Main different findings were highlighted by red circle. (a(1-5), b(1-5)), c(1-5)), d(1-5)) shows CK19, COX-2, CD3 and Perlecan staining in oral lichen planus (OLP) and oral lichenoid lesions (OLLs), respectively. CK19 showed almost no positive cells at prominent area in OLP and scattered in OLLs. COX-2 reaction was apparent at prominent and adjacent area in OLLs. Many CD3 positive cells were appeared directly under spinous layer in OLP. Perlecan reacted epithelial cells and stroma in OLLs. (e(1-5)) present Papanicolaou staining. Thickening of the spinous layer was observed in all areas of OLLs.
prominent area and almost continuous in the site adjacent the lesion. As for CK19, no staining was noted in contact with the basement membrane in the prominent area and weakly positive in the site adjacent the lesion in OLP. Both were reduced compared to OLLs (Figures 3(a-1, 2)).
For COX-2 (Figure 3(b)), positive cells were slightly scattered in the stroma of normal mucosa. In OLP, COX-2 was expressed in the basal to deep spinous layers of the prominent area and scattered in the deep spinous layer at the site adjacent to the lesion (Figures 3(b-1, 2)). In OLLs, numerous positive cells were seen in the basal to spinous layers of the prominent area and moderate expression was identified in the basal to deep spinous layers at the site adjacent to the lesion. Many positive cells were observed in the submucosal connective tissue in OLP and OLLs (Figure 3(b-1), Figure 3(b-3)).
For CD3 (Figure 3(c)), it exhibited some positive findings in normal mucosa. OLP showed positive findings in most of the band-like infiltrated lymphocytes, and invasion into the epithelium was also observed. In the site adjacent to the lesion, positive cells appeared in a slightly band-like infiltration (Figures 3(c-1, 2)). In OLLs, positive cells were diffusely present at the subepithelial junction and were scattered at the site adjacent to the lesion (Figures 3(c-3, 4)).
Perlecan was negative in normal mucosa. In OLP, negative results were seen in both the prominent area and the site adjacent to the lesion (Figures 3(d-1, 2)). In OLLs, epithelial cells were mildly positive in the prominent area and were slightly positive in basal cells at the site adjacent the lesion. Strong positive reactions were shown in submucosal tissue at the site adjacent the lesion (Figures 3(d-3, 4)). Many Ki-67-positive cells were seen in both OLP and OLLs at prominent areas and the adjacent sites (data not shown).
Comparing the distributions of positive expressions for CK19 and CD3, CD3-positive cells showed a band-like distribution in contact with the spinous-layer cells that lost CK19 expression in OLP (Figure 3(a-1), Figure 3(c-1)). In OLLs, more CK19-positive cells were found in contact with the stroma, and CD3-positive cells were scattered in the stroma (Figure 3(a-3), Figure 3(c-3)).
Increased keratinization, which stains orange, was observed at the prominent area and adjacent sites of both OLP and OLLs (Figures 3(e-1, 2, 3, 4)). Thickening of the spinous layer was observed in all areas of OLLs.
3.4. Statistical Analysis
Statistical results are presented in Figures 4-6. For CK19, positive cells were significantly found in OLLs at both the prominent area and site adjacent to the lesion comparison with those of OLP’s (p < 0.05, Figure 4). The number of COX-2-positive cells was significantly higher in spinous (p < 0.05) and basal layers (p < 0.01) in OLLs of the prominent area. Conversely, OLP showed significantly more positive reaction than OLLs at the site adjacent to the lesion (p < 0.05, Figure 5). Concerning Ki-67, no significant difference was seen between OLP and OLLs in either prominent areas or sites adjacent to the lesion (data not shown). Figure 6 shows the results for thickness of the epithelium from the
![]()
Figure 4. Results of CK19 immunohistochemical staining. Black, light gray bars indicate oral lichen planus (OLP) and oral lichenoid lesions (OLLs), respectively. 1: the most prominent lesion; 2: an adjacent area of normal mucosa; *: p < 0.05.
![]()
Figure 5. Results of COX-2 immunohistochemical staining. Black, light gray bars indicate oral lichen planus (OLP) and oral lichenoid lesions (OLLs), respectively. 1: the most prominent lesion; 2: an adjacent area of normal mucosa; *: p < 0.05; **: p < 0.01.
![]()
Figure 6. Result of epithelial layer thickness by Papanicolaou staining. Black, light gray bars indicate oral lichen planus (OLP) and oral lichenoid lesions (OLLs), respectively. 1: the most prominent lesion; 2: an adjacent area of normal mucosa; *: p < 0.05; **: p < 0.01.
results of Papanicolaou staining. Spinous and basal layers at both the site adjacent to the lesion and the prominent area in OLLs were thicker than those in OLP (p < 0.05).
4. Discussion
Lichenoid reactions are observed in various lesions, but the malignant transformation rate of OLLs should be noted to be higher than that of OLP [9]. Examining these histopathological differential findings is thus of great significance.
Cytokeratins are keratin-containing intermediate filament proteins found in the intracytoplasmic cytoskeleton of epithelial tissue. Among the 20 epithelial cytokeratins that have been identified, CK19 is the lowest molecular weight acidic keratin (40 kDa) and is a specific cytoskeletal structure of simple epithelium and the basal cells of stratified squamous epithelium [21]. In the OLP lesion, T-cell infiltration was induced just below the epithelium after recognition of an unknown antigen and involved a cell-mediated immune response and might have attacked CK19-positive basal cells in the present study. On the other hand, in OLLs, basal cells remained discontinuous, and have been speculated to facilitate histopathological differentiation. OLP is a lymphocyte-mediated immunological disorder in which the basal cells seem to be targeted by T lymphocytes [22] [23] [24] [25]. Basal cells disappeared in areas of OLLs in this study, and common events in the lichenoid tissue reaction are speculated to include activation of dendritic cells and keratinocytes, and recruitment and activation of T cells, followed by cytotoxic damage to keratinocytes with release of keratinocyte antigens [26]. The positive findings for CD3 and CK19 in OLP and OLLs were consistent with the above speculation in the present study.
Ki-67 is often used as an adjunct marker to assess the proliferative activity of potentially malignant lesions [27]. In the present study, no significant difference was seen between OLP and OLLs with Ki-67-positive rates in either the prominent areas or adjacent sites. This result was concordant with the results of a previous study [28]. Because both lesions represent chronic inflammation, the increase in Ki-67-positive cells might show that the epithelium is required for enhanced proliferation and healing [22]. Concerning epithelial thickness as determined by Papanicolaou staining, spinous and basal layers at the adjacent site and prominent area in OLLs were thicker than those in OLP. This implied that growth of the epithelium was promoted by something like a chronic regeneration reaction or epithelial growth factor in OLLs, despite the lack of significant difference in Ki-67 positivity between OLP and OLLs.
Perlecan, a heparan sulfate proteoglycan of about 470 kDa in size, is one of the major basement membrane macromolecules and plays an important role in cellular growth, differentiation, adhesion, and motility through its interactions with growth factors and cytokines [19] [20]. Recently, perlecan has been localized in the intercellular space of the oral epithelium and is overexpressed in dysplastic epithelial cells and deposited in their interepithelial space, resulting in the histological finding of reduced of cellular cohesion [20]. OLLs show mild to moderate expression in the basal to spinous layers of epithelial cells of the lesion. Positive expression was shown in the adjacent area and weak to moderate positivity was seen in subepithelial tissue. Conversely, negative results were seen for epithelial cells in the prominent and adjacent area, and in the stroma of the adjacent area in OLP of the present study. For epithelial cells in OLLs, perlecan is speculated to be synthesized by germ cells with parabasal cell-like appearances, as major constituent cells of epithelial dysplasia, and is deposited in the intercellular space of dysplastic epithelial cells to facilitate their proliferation [20]. Another possible function of intraepithelial perlecan was considered to be in providing a space for the migration of intraepithelial cells, such as lymphocytes and macrophages (including Langerhans cells), which are usually distributed in the epithelial layer of the oral mucosa and are thought to patrol for immune stimulations from the oral cavity [29]. Concerning perlecan in stroma, the myxoid or edematous matrices of immature granulation tissues were shown to be immunopositively simultaneous with epithelial cells [19]. Perlecan is expressed in the process of tissue remodeling of inflammatory lesions [30]. Tissue regeneration involving perlecan is speculated to be taking place in the stroma of OLLs, but the relationship with the acquisition of neoplastic characteristics of epithelial cells needs further investigation.
Cyclooxygenases (or prostaglandin H synthases), commonly referred to as COXs, are a family of myeloperoxidases located on the luminal side of the endoplasmic reticulum and nuclear membrane [31] and catalyze the rate-limiting step of prostaglandin biosynthesis from arachidonic acid [31]. The COX-2 enzyme is often expressed during inflammation. COX-2 aids in tissue repair, angiogenesis, cell proliferation and differentiation, but chronic, persistent inflammation as noted in OLP and OLLs can lead to detrimental effects [32]. In addition, inflammation is strongly associated with carcinogenesis, including the development and progression of oral cancer [33] [34]. Further, COX-2 expression has been correlated with higher grades of oral epithelial dysplasia and could represent an early event in oral carcinogenesis [35] [36]. Several studies, including that of Arreaza, et al. [32], found a higher COX-2 expression in OLP than in OLLs, and Chankong, et al. [37] observed a direct correlation between COX-2 expression in OLP and the clinical severity of this pathology. Conversely, Cortes et al. found higher expression of COX-2 in OLLs than in OLP [38]. In the present study, the number of COX-2-positive cells in epithelium was significantly higher in OLLs of the prominent area. However, OLP showed significantly more positive reactions than OLLs at the adjacent sites. COX-2 overexpression by epithelial cells in OLLs could inferred oncogenic alterations, because was described to modulate cell proliferation and apoptosis against oncogenic alterations [39] [40] [41].
Accordingly, COX-2 and perlecan were found in the basal to spinous layers of the prominent area and adjacent site, and of the prominent area in OLLs, respectively, in the present study. These findings suggest the acquisition of neoplastic characteristics in the epithelial cells of OLLs [35]. Conversely, of all OPMDs, malignant transformation of OLP has been described as the most controversial [38]. The authors also concluded that this finding suggested a different etiology and molecular pathophysiological pathways for OLP and OLLs. Patil, et al. [42] and Fitzpatrick [43] also observed features of dysplasia in OLP and OLL in their case series, reiterating the malignant potential of OLP and OLL.
It is therefore very important for the pathologist to differentiate OLP from epithelial dysplasia with lichenoid features, as the distinction has significant implications for diagnosis, research and controversies surrounding OLP with regard to its malignant potential.
5. Conclusions
The following could be concluded in the present study:
1) Almost no basal cells were noted in the prominent area in OLP and were significantly reduced compared to OLLs.
2) COX-2 was found in the prominent area in OLLs with significance.
3) OLLs showed mild to moderate expression for perlecan in the basal to spinous layers and in subepithelial tissue.
4) Although there are restrictions, these suggested the possibility of helping to distinguish between OLP and OLLs.
Acknowledgements
This work was supported by JSPS KAKENHI Grant Number 18K07000.