Low p27^kip1 expression in head and neck squamous cell carcinoma: association with risk factors and adverse outcomes

Abstract

Background

Squamous cell carcinoma (SCC) of head and neck is highly prevalent in South-asian countries, owing to high consumption of areca nut/gutka and chewing tobacco. p27^kip1 is a tumor suppressor gene, thought to be downregulated in oral squamous cell carcinoma. Therefore, in the present study we used immunohistochemical analysis to investigate an association between low p27^kip1 expression in SCC of the head and neck and adverse outcomes/risk factors.

Methods

Total 105 cases of SCC of head and neck excision specimens were selected from records of pathology department archives that underwent surgeries at Liaquat National hospital, Karachi from January 2008 till December 2013. Clinical and pathologic characteristics of patients were evaluated and p27^kip1 immunohistochemistry was applied on tumor blocks.

Results

In our study, low expression of p27^kip1 in SCC of head and neck was seen in 39(37.1%) cases while 66(62.9%) of the cases showed high expression for p27^kip1. Significant association of p27^kip1 expression with pan/gutka usage (p = 0.004), and recurrence (p = 0.001) was noted; however, no significant association of p27^kip1 expression with other clinicopathologic features was seen. Multivariate binary logistic regression showed cases with history of pan/gutka usage were more likely to show low p27^kip1 expression. Similarly, we also found that recurrence was more likely to develop in patients with low expression of p27^kip1 in comparison to cases showing high p27^kip1 expression.

Conclusion

Loss of p27^kip1 expression is a significant event involved in the pathogenesis of SCC head and neck especially that of oral cavity. Significant association of gutka/areca nut with low p27^kip1 expression in our study suggests that loss p27^kip1 expression is a major event involved in areca nut induced SCC of head and neck in this part of the world; however, more large scale molecular based studies are required to validate this observation. Moreover, significant association of low p27^kip1 expression with tumor recurrence suggests its importance as a prognostic biomarker in SCC of head and neck.

Introduction

Squamous cell carcinoma (SCC) of the head and neck is highly prevalent in South-Asian region [1,2,3,4]. In developing countries like Pakistan, SCC of head and neck has been reported to be epidemiologically associated with smoking, tobacco consumption and areca nut (pan/gutka). The possible mechanism that leads to the development of malignant changes in the oral cavity is the disruption of cell proliferation control [5]. Deregulation of cyclin-dependent kinases and cyclin-dependent inhibitors has been largely associated with these malignant changes [6]. p27 is a kinase inhibitor protein that inhibits the function of two cyclins complexes E–cdk 2 and A–cdk 2 through binding mechanism [7]. It plays a significant role in negative-regulation of cell cycle, mainly during the stages G0 and G1 [8]; however, it is degraded during the G1 to S phase via ubiquitin-proteasome–dependent degradation mechanism [9]. The tumor suppressor gene CDKN1B encoding for cell cycle inhibitor p27^kip1 has been identified as a driver gene mutated in many human cancers like breast carcinoma, prostate cancer and endocrine tumors. It is noteworthy that, the mutations of CDKN1B in human cancers are non-sense or small deletions/insertions resulting in the formation of truncated protein. The resulting loss of inhibitory effect of p27^kip1 on cell cycle leads to un-controlled cell proliferation and ultimately carcinoma development [6].

p27^kip1 has been reported to be actively expressed in non-proliferating cells [10]. Although genetic alterations leading to loss-of-function mutations are more commonly reported in p16 and p53 tumor suppressor proteins, deregulation of p27kip1 has been found in tumors of the breast, lungs, colorectal as well as SCC of the head and neck [3, 8]. The main mechanisms involved in p27^kip1 deregulation in human cancers include down-regulation of its expression and exclusion from the nuclear compartment. Research also highlighted that down-regulated p27^kip1 expression may result from altered proteasome-mediated degradation [11, 12]. Evidence also suggested that downregulation of p27^kip1 correlates with poor prognosis and more aggressive nature of the tumor [13, 14]. Down regulation of p27^kip1 is associated with loss of Immunohistochemical expression of p27 protein. Hence in the current study, we aimed to evaluate the frequency of low p27kip1 expression in cases of SCC head and neck in our population and its association with prognostic parameters.

Methods

Total 105 cases of head & neck SSC excision specimens were included in the study. All these patients were biopsy proven SCC of head and neck. The study was conducted in Liaquat National hospital, Karachi from January 2008 till December 2013 over a period of 6 years. The study was approved by institutional research and ethical review committee. Informed written consent was taken from the patients prior to surgery. Hematoxylin and eosin stained slides and paraffin blocks of 77 cases were retrieved and new sections were cut. Slides of all cases were re-evaluated by two senior histopathologists independently and pathologic characteristics were recorded. Clinical records of these patients were reviewed from institutional files to evaluate history of gutka/pan use history, and recurrence status. Moreover, representative tissue blocks of 105 cases were used for p27^kip1 immunohistochemistry (IHC). After retrieving tissue blocks of these 105 cases, IHC was performed manually in batches of 15 cases and positive and negative controls were run along with each batch. All the slides were similarly processed and the required IHC was completed in 5 days and then evaluated by 2 histopathologists independently and blinded with other histopathological findings.

Immunohistochemistry

p27^kip1 immunohistochemistry was performed using p27^Kip1(EP104) antibody purchased from Cell Marque and DAKO EnVision kit according to manufacturers protocol. Only nuclear staining for p27 was quantitatively and qualitatively evaluated. Intensity of staining was categorized into no staining (0), weak (1+), intermediate (2+), strong (3+) as follows,

0 (no staining): no nuclear staining

1+ (weak): Faint nuclear staining.

2+ (intermediate): Moderate nuclear staining.

3+ (strong): Strong nuclear staining.

Percentage of positively stained cells was measured as continuous variable from 0 to 100. Percentage of positively stained cancer cells was multiplied by intensity score to formulate an H-score. Cases with less than 200 H-score were labeled as low p27^kip1 expression, while tumors with more than 200 score were categorized as high p27^kip1 expression (Fig. 1).

Fig. 1

p27^kip1 expression in Squamous cell carcinoma. a & b. High expression of p27^kip1, showing diffuse nuclear positivity of p27^kip1; c & d. Low expression of p27^kip1, only a few tumor cells showing mild nuclear positivity for p27^kip1, majority of tumor cells revealing loss of expression . Lymphocytes serve as positive control

No set guidelines are available for p27^kip1 IHC interpretation and different studies used variable cutoffs for low vs. high expression. In studies with semi-quantitative p27 interpretation, cutoffs ranging from 5 to 30% were used [8] and [15]. Only a few studies took into account the intensity of expression. We incorporated both intensity and percentage of p27 expression to construct an H-score with 200 cutoff, similar to that used in many biomarker analysis like ER/PR, androgens etc. to evaluate p27 expression in our study.

Follow-up and recurrence

Recurrence status and follow-up were evaluated by reviewing hospital medical records. Disease free survival was taken as time between surgical excision and local recurrence or distant metastasis, death or last follow-up.

Statistical analysis

Statistical package for social sciences (SPSS 25) was used for data compilation and analysis. Mean and standard deviation were calculated for quantitative variables. Frequency and percentage were calculated for qualitative variables. Chi-square and fisher exact test was applied to determine association. Shapiro wilk test was used to check normality of data. Student t test and Mann-Whitney test were applied to compare difference in means among groups. Odds ratios were calculated by Multivariate binary logistic regression for variables with significant association. Survival curves were plotted using Kaplan- Meier method and the significance of difference between survival curves were determined using log-rank ratio. P-value ≤0.05 was taken as significant.

Results

Mean age of the patients involved in the study was 50.9 + 12.3 years with male to female ratio of 3:1. Most of the cases (i.e 50.5%) fall in T2 category (2-4 cm) with mean depth of invasion of 1.1 + 0.7 cm. Oral cavity is the most common site of the tumor (68.5%) and keratinizing squamous cell carcinoma being the most common histology. History of pan/gutka usage was revealed in 77% of cases. Detailed patient characteristics are presented in Table 1.

Table 1 Clinicopathologic characteristics of patients with Squamous cell carcinoma included in the study

In our study, low expression of p27^kip1 in SCC of head and neck was seen in 39(37.1%) cases while 66(62.9%) of the cases showed high expression for p27^kip1.

Significant association of p27^kip1 expression with pan/gutka usage (p = 0.004) and recurrence (p = 0.001) was noted; however, no significant association of p27^kip1 expression with other clinicopathologic features was seen as shown in Table 2. Patients with low p27kip1 expression were found to have poorer disease free survival compared to those with high p27^kip1 expression as shown in Fig. 2.

Table 2 Association of p27^kip1 expression with various clinicopathologic characteristics and prognostic parameters in patients with head and neck Squamous cell carcinoma

Fig. 2

Kaplan-Meier for p27^kip1 over expression (disease free survival), showing significant association of p27^kip1 expression with disease free survival. Cases with low p27^kip1 expression showing lower disease free survival compared to cases with high p27^kip1 expression

Multivariate binary logistic regression shows cases with history of pan/gutka usage were more likely to show low p27^kip1 expression. Similarly, we also found that recurrence was more likely to develop in patients with low expression of p27^kip1 in comparison to cases showing high p27^kip1 expression (Table 3).

Table 3 Binary logistic regression for low expression of p27^kip1, association with history of pan/gutka usage and disease recurrence

Discussion

In the present study, we found low p27kip1 expression in 37.1% cases of SCC of head and neck. Moreover, significant association of loss of p27kip1 expression was noted with pan/gutka use and poorer disease free survival. To our knowledge, this is the first study conducted in Pakistan evaluating p27^kip1 expression in SCC of head and neck where SCC prevalence is particularly high with different risk factors compared to the west (pan/gutka use vs. HPV/alcohol). In the western countries, HPV is considered as the most important risk factor in oral cancer pathogenesis with 70–90% oral cancers showing HPV infection [16, 17]. On the other hand, the data on HPV oral infections in Pakistan is limited, however, approximately 67% of cases of oral cancers were found to be infected with HPV in one study [18]. Tobacco use is considered as an important factor in oral carcinogenesis; however chewable tobacco in the form of betal nut/ pan is more prevalent than smoking tobacco in this part of the world.

p27 is a tumor suppressor gene. Low/loss of p27^kip1 expression is believed to be associated with rapid cell growth and proliferation, thus a significant contributor of tumor development. Also demonstrated by various studies, down-regulation of p27^kip1 reflects adverse disease outcomes and poor prognosis [6, 19] as seen in our study as well. Furthermore, in the present study, clinicopathologic features of the tumors revealed that low p27^kip1 expression was irrespective of age and gender of the patients. Only 37.1% tumors in our study showed low p27^kip1 expression in our study. These findings suggest a multi-step carcinogenesis of SCC of head and neck with involvement of other oncogenes and tumor suppressor genes. Our findings acknowledge the results of another study conducted by Jocelyn Yao, et al., in which no symbolic association was established between p27^kip1 expression in colorectal carcinomas and clinicopathologic features of the tumor including age, gender, tumor stage, tumor size and histologic grade [20]. Similarly, our findings are consistent with another study conducted by Hiroyuki Mineta, et al., in which no significant association between tumor stage and loss of p27^kip1 expression was established [6].

It is noteworthy that low frequency of p27^kip1 expression was found in majority of tumors of the oral cavity, followed by tongue and we found a significant association of gutka/pan usage with loss of p27^kip1 expression. Gutka is a preparation of crushed areca nut mixed tobacco, paraffin wax, catechu and slaked lime. In our setup, areca nut/gutka chewing has been asserted as a significant risk factor attributable to leukoplakia which subsequently develops into oral carcinoma [15]. Reduced expression of p27^kip1 was also demonstrated by a study of similar nature conducted by Jordan, et al. [21]. On account of these results and antecedent studies, we suggest further research to establish the role of carcinogens like tobacco smoking and areca nut chewing in altering the cell proliferation cycle and deregulation of biomarkers including p27^kip1.

We found a significant association of p27^kip1 loss with higher recurrence/poor disease free survival. Association of loss of p27^kip1 expression with recurrence and disease free survival has been evaluated by some authors. Mineta et al. and Venkatesan et al. in their studies involving 94 and 35 cases respectively found a significant association of p27^kip1 loss with disease free survival with low expression in 44 and 20% cases respectively [9, 22]. On the other hand, no such significant association was found by Fujieda et al. in a series of 60 cases with 58% low p27 expression [23]. On the contrary, we found p27^kip1 loss in 34% cases of SCC of head and neck. The difference in frequency of low expression and association with recurrence/ disease free survival may be due to variation in antibody dilution and cutoffs used to categorize low vs. high p27^kip1 expression. We used ready to use antibody and took a cut-off of 200 to divide low vs. high p27^kip1 expression. Therefore, we suggest that proper guidelines should be defined for low p27^kip1 expression. Zhang M et al. in a study involving 110 cases of SCC of head and neck evaluated expression of p21, p27 and survivin. They found that p27 immunoreactivity was negativity correlated with 5 year disease free survival [24]. Apart from the prognostic significance of p27 in SCC of head and neck, a few investigators also explored the role of p27 staining in differentiating ordinary/ reactive squamous mucosa from malignant (dysplastic/ invasive squamous cell carcinoma. In this regard, Queiroz AB investigated the same observation and found a significantly diffuse expression of p27 in normal/ reactive squamous mucosa compared to patchy/ focal p27 expression in squamous cell carcinoma [25].

We had 3 cases in which tumor location was lip. Although, alcohol and tobacco remains the most significant risk factors in the pathogenesis, it is noteworthy that solar radiations may also play a role in lip cancer pathogenesis.

One of the limitations of our study was that we evaluated only one biomarker in the molecular cascade of squamous cell carcinoma pathogenesis; moreover, molecular studies were not performed to confirm underlying gene mutations. However, this study can serve as a platform for future studies as SCC of head and neck is a major health problem in this resource limited region of the world.

Conclusion

A significant proportion of cases in our study revealed low p27^kip1 expression which reveals that down-regulation of this protein is a major event involved in the causation of SCC of head and neck in this part of the world. Moreover, association of low p27^kip1 expression with gutka/areca nut chewing and poor prognostic parameters like tumor recurrence suggests its importance as a prognostic biomarker in SCC of head and neck.