Introduction

Oral cancer accounts for about 3% of all cancers worldwide [1]. More than 300,000 new cases of oral cancer are reported worldwide every year [2]. Oral squamous cell carcinoma (OSCC) accounts for the vast majority of these instances [3]. Despite this, the 5-year overall survival rate of OSCC patients has remained below 55% over the last decade due to local aggressiveness and high rates of metastasis [4]. The presence or absence of cervical lymph node metastasis has a significant impact on OSCC patients’ prognosis [5, 6]. Clinical stage, depth of invasion, and histologic grade of malignancy are all highly linked with nodal involvement [7, 8]. Despite having a significant impact on nodal involvement, all these parameters may not necessarily represent the metastatic ability of the cancer cells. As a result, identifying a biomarker of OSCC with significant metastatic potential might be clinically advantageous. In epithelial cells, cytokeratins (CKs) are the primary structural constituents of the cytoskeleton [9, 10]. Over 20 distinct CKs have been found, with CK19 being the smallest acidic type I CK family protein. CK19 has been found to be expressed in a variety of tumour tissues, including breast cancer, colon cancer, lung cancer, head and neck cancer (HNC), and hepatocellular carcinoma (HCC) [11,12,13,14]. However, the relevance and roles of CK19 appear to change between cancers. Previous research has indicated that CK19 plays a tumour-suppressive role in breast cancer but a tumour-promoting role in colon and liver cancers. Although immunoreactivity for CK19 has been detected in OSCC tissues, the rates of positive expression and its roles in this disease are controversial. Mutation of the p53 gene is one of the most common events in oral carcinogenesis. The accumulation of p53 protein has also been detected in premalignant lesions, especially oral leukoplakia with dysplasia. This suggests that p53 gene mutation may be an early step in the malignant transformation of oral dysplastic lesions. Such patients can be identified by p53 protein specific immunohistochemistry [15]. Survival rates of oral squamous cell carcinoma (OSCC) have not improved since decades, attributed to the complex multistep process of oral carcinogenesis specifically involving oncogenes and tumour suppressor genes [16,17,18]. Point mutations in the tumour suppressor gene (TP53) are seen in 35–67% of OSCC, and overexpression is associated with poor survival rates [19, 20]. Most of the mutations prolong the half life of p53 protein (6 h) which is amenable to immunohistochemical detection, whereas a few truncated deletions escape detection [21, 22]. Cytokeratin 19 (CK19) is believed to be one of the target proteins of TP53 gene, usually expressed in the basal layer of non keratinized stratified squamous epithelium [23,24,25,26]. CK19 shows variable expression in OSCC, ranging from 29 to 100% with a role in its progression [27,28,29,30,31]. Two studies concluded that downregulation of CK19 expression with the acquisition of an invasive phenotype in OSCC. There is single study in the literature comparing CK19 and p53 on OSCC cell lines, hypothesizing that wild type p53 inhibits CK19 expression.

Thus, the present study was a maiden attempt with the aim to evaluate immunohistochemical expression of p53 and CK19 in different histopathological grades of OSCC. Secondary objective was to study the correlation (if any) between p53 and CK19 in NOM and OSCC.

Materials and Methods

This was an institutional based retrospective study approved by institutional research and ethical review committee (reference number “MUHS/PG-T/E1/3882/2015”). Formalin fixed paraffin embedded 45 biopsy specimens of OSCC (group II) and 40 of NOM (group I) were retrieved from archives. The OSCC samples were graded by Bryne’s grading system. 4 μm sections were deparaffinized (560 C for 15 min) and incubated using mouse monoclonal antibody to p53 and CK19 (Biogenx) respectively. The results were interpreted as per the method adopted by Etemad -Moghadam et al. for p53 and CK19 separately (Table 1).

Table 1 p53 and CK19 expression in Normal (group I) and OSCC (group II)
Full size table

Immunohistochemical Analysis

From FFPE blocks of 45 OSCC specimens, 5 micrometre thick sections on poly L lysine coated slides were subjected to IHC analysis of p53 RTU (Ready To Use), Primary antihuman rabbit antibody, Leica Biosystems, Japan), and CK19 (RTU, Primary antihuman rabbit monoclonal antibody, Leika). Tissue sections were deparaffinized in xylene (twice), treated with a graded series of alcohol (100%, 95%, 85%, and 75% ethanol), and then incubated in phosphate-buffered saline (PBS, pH 7.4) for 5 min. Heat-induced antigen retrieval was done by immersion in 10 mM Tris-ethylene diamine tetra-acetic acid with pH 9 at 600 W in a pressure boiler until two whistles. Endogenous peroxidase was inactivated by 3% hydrogen peroxide for 10 min. The tissue sections were incubated with primary antibodies against p53 and CK19 for 40 min in humidifying chambers followed by incubation with a secondary polyclonal conjugate (Dako, Glostrup, Denmark) for 30 min. Lastly, tissue sections were treated with diaminobenzidine as a substrate chromogen and counterstained with hematoxylin. As negative controls, tissue sections were treated with PBS instead of the primary antibody. Skin sections and lung SCC were taken as positive controls for CK19 and p53 respectively. The slides were then mounted, observed, and evaluated using a research microscope (Nikon Eclipse Ni-U) using NIS Basic research software.

Immunohistochemical Evaluation

The positivity for p53 and CK19 was evaluated using a research microscope (Nikon Eclipse Ni-U). The expression was quantitatively assessed on five randomly selected fields under 400X by grid aided image analysis using NIS Basic research software. Positivity for p53 was observed in the nucleus of the cell and in cytoplasm of the basal layer for CK19. The percentage positivity for both markers was calculated by the number of positive cells per 1000 in the specimen.

Statistical Analysis

The analysis was performed using SPSS version 20.0 (IBM Corp.) software. The mean number of positive cells was compared in two groups using the t-test; staining intensity and staining index were compared using Fisher‘s exact test.

Results and Discussion

The histological findings of OSCC may range from one location to the next within the same tumour, necessitating a meticulous examination for an appropriate diagnosis. A precise histological grading of OSCC is critical because it provides an indication of the severity of the lesion in addition to therapeutic therapy and predicts the likely clinical course of the disease. Several histological grading schemes are in use to predict OSCC clinical behaviour. Despite the fact that biopsy is still the gold standard in diagnosis, no single histological grading system has been matched with its molecular behaviour till date to explain the severity of the lesion. Incisional biopsies can provide early hint on tumour behaviour with minimum invasiveness, preventing overtreatment. The samples in the present study showed male predominance, with the alveolobuccal complex (42.22%) constituting the predominant site in OSCC, which is in accordance with the previous literature (Table 2). The NOM group showed p53 immunopositivity in 85% of samples with expression confined to nuclei of the basal layer of the epithelium [mean ± standard deviation (SD) 16.00 ± 5.20] (Fig. 1). The Staining index was predominantly low (mean ± SD 1.87 ± 0.34) as compared to OSCC (mean ± SD 5.63 ± 3.02). Despite wild type p53 being hard to detect by immunohistochemistry due to its short half life (20 min), the positivity may be attributed to physiological stabilisation caused by UV radiation, hypoxia and viral proteins leading to genotoxic stress, an increased half-life and thus being amenable to detection [32, 33]. This explains the variable expression of p53 in NOM in different studies ranging from no expression to focal patchy positivity and 18.3–30% in other studies. In our study, OSCC showed 66.67% p53 immunopositivity in the nuclei of the basal and suprabasal layers of epithelium and in invasive carcinoma tissue (mean ± SD 45.57 ± 26.47). The immunopositivity in OSCC correlates with accumulated mutant p53 protein [34]. However, immunonegativity in the remaining OSCC samples (33.3%) does not certify the total absence of defects, as some deletion mutations produce non-detectable truncated proteins [35,36,37].

Table 2 Demographics of study groups
Full size table
Fig. 1
figure 1

p53 and CK19 expression in NOM and WDSCC. (A) NOM showing parakeratinized stratified squamous epithelium overlying fibrous stroma (H&E, objective 10x). (B, C) p53 protein immunostaining of NOM restricted to nuclei basal layer of epithelium (Objective 10x, 40x), (D) CK19 protein staining cytoplasm of basal layer of epithelium (Objective 40x). (E) WDSCC showing islands of atypical epithelial cells (H&E, objective 10x), (F, G) p53 nuclear immunopositivity in invasive cancerous tissue (Obective10x, 40x) with high staining index. (H) CK19 staining invasive islands with high staining index (Objective 40x)

Full size image

The percentage of overall p53 positive cells in OSCC samples was (66.67%) in our study. Montebugnoli et al. (57%) [38]; Yang et al. (62.5%) [40, 41]; Swaminathan et al. (65%) [42] found similar results for p53 positively stained cells in concordance with the present study. Hashmi AA et al. (66.1%) [43]; Verma et al. (66.6%) [37]; Kannan et al. 70%) [44], Bidaud P et al. (88.5%) [39]; and Kerdpon et al. (94%) [45] study results were contradictory to present study showing increased number of p53 positively stained cells .

In contrast to present study, some studies of p53 immunopositivity in OSCC such as ; Shiraki et al. (46%) [46] and Siegelmann-Danieli et al. (43%) [47] observed low expression for p53 positive cells. Wong et al. has suggested ubiquitin mediated rapid degradation of p53 by the E6 protein of human papilloma virus (HPV). The variable results of p53 expression in OSCC in different studies may be due to different techniques, types of antibodies, differences in ethnicity and risk factors associated with OSCC pathogenesis in heterogeneous populations.

In our study, within different grades of OSCC, the percentage of positive cells showed a linear increase from WDSCC to PDSCC (Table 3). The comparison of NOM with different grades of OSCC for staining intensity and staining index was highly significant (Table 4). However, all these parameters were statistically non-significant within the intra-group grade comparison of OSCC.

Table 3 Comparison of p53 immunoexpression between normal group with pathological grades of differentiation
Full size table
Table 4 Calculation of percentage of positive cells, staining intensity and staining index
Full size table

Similar to our study, a study by Siegelmann- Danieli [47], Abbas NF et al. [48] and Verma et al. [37] found non-significant association of p53 expression and tumor grading. However, Jain et al. (2008) [49] and Hashmi AA et al. (2018) [43] found a significant association between p53 over-expression with tumour grading and survival rates. This difference can be due to the small sample size in our study or p53 being an independent prognostic factor that is not dependent on histological grades [50]. For CK 19 immunoexpression (Table 1; Figs. 1 and 2) 17.5% samples of NOM showed immunopositivity restricted to the cytoplasm of the basal layer of epithelium [mean ± SD 18.29 ± 4.11]. 10% of samples showed moderate and 7.5% low staining intensity and low staining index [mean ± SD 1.57 ± 0.53] (Fig. 3). OSCC showed immunopositivity only in 2 (4.44%) samples of WDSCC [mean ± SD 41.00 ± 7.07] with high staining intensity and index (mean 6). A Statistical comparison between two groups with regard to staining index comparison was statistically significant (p 0.0019). Staining intensity comparison between two groups and within different histopathological grades of OSCC was not possible because of inadequate positive samples.

Fig. 2
figure 2

p53 and CK19 expression in MDSCC and PDSCC. (A, E) MDSCC (H&E, objective 10x), and PDSCC (H&E, objective 40x), showing sheets of malignant epithelial cells, (B-C, F-G) p53 staining nuclei of malignant epithelial cells with high staining index in MDSCC and PDSCC (objective 10x,40x), (D, H) CK19 immunonegativity in MDSCC and PDSCC (objective 40x)

Full size image
Fig. 3
figure 3

p53 immunohistochemistry showing medium and low staining index. (A-C) p53 immunopositivity with medium staining index (1 × 3 = 3) (Objective 4x,10x,40x). (D-F) p53 immunopositivity with low staining index (1 × 1 = 1) (Objective 4x,10x,40x)

Full size image

Similar results were seen by Crowe et al. who observed consistent downregulation of CK19 expression in OSCC cell lines as compared to normal epithelium. Similarly, Kale AD et al., also found immunonegativity in all OSCC samples. The low (17.5%) expression of NOM in our study can be explained as most of our samples were of keratinized mucosa, and downregulation of CK19 is essential for terminal differentiation of keratinocytes. Immunonegativity in the majority of samples of OSCC in our study is justified as over-expression of CK19 decreases the invasive potential by diminishing the migratory capability or formalin fixation masking CK19 antigenic sites in most of the samples [7, 15].

However, most of the studies showed variable expression of CK19 in OSCC, such as Nie et al.(100%) [51] Zhong LP et al.(90.9%), Hamakawa et al.(66.7%) [52], Babiker AY et al. (58%) [53], Frohwitter G et al. (40.1%) [27], Fillies T et al. (40.6%) [28] and Vora HH et al. (29%) [54] and most studies showed a positive correlation of CK19 overexpression with increasing grade of OSCC and poor prognosis. The contrasting results in our study may be due to a different study population with heterogeneous risk factors, a small sample size, formalin fixation, which damages CK19 antigenic sites, different antibodies and antigen retrieval techniques used.

Conclusions

This study concluded that p53 is a marker of oral carcinogenesis independent of histopathological grade and CK19 expression. The switching of CK19 expression may serve as marker for invasion in OSCC; however, this criterion cannot be specified as only 4.44% of samples were immunopositive which indicates that CK19 is neither a reliable nor sensitive marker in oral carcinogenesis.