Association of p16 as Prognostic Factors for Oropharyngeal Cancer: Evaluation of p16 in 1470 Patients for a 16 Year Study in Northeast China

Human papillomavirus (HPV) is an etiological risk factor for oropharyngeal squamous cell carcinomas (OPSCC). Our study investigates the prevalence, prognostic, and clinicopathologic features of HPV-related oropharyngeal cancer in Northeast China and elucidates the involvement of p16 in the tumorigenesis and progression of OPSCC. Specimens from 1470 OPSCC patients collected from 2000 to 2016 were analyzed using the status of HPV by polymerase chain reaction (PCR) and p16 immunohistochemistry. Overexpression of p16 was observed in 81 (5.51%) of the 1470 cases, and HPV positive was present in 78 cases (5.31%) of the 1470 cases. HPV positive and p16 overexpression have a good concordance. However, we found that the etiological fraction of HPV in cancers of the OPSCCs was obviously lower in Northeast China than other cohorts previously reported. Interestingly, nearly 89% of patients with p16 expression were smokers, and nearly 70% of patients with p16 expression had a history of alcohol. Our study also demonstrates that p16 expression is significantly associated with early stage primary OPSCCs and the patients with p16 expression tend to show better survival following surgery and radiotherapy.


Introduction
Head and neck squamous cell carcinoma (HNSCC) has been defined as the sixth leading cause of cancer in the world [1]. High recurrence rates and nodal metastases always lead to high mortality of HNSCC. Especially, 5-year survival rates of HNSCC patients with cervical lymph node metastases are reduced by approximately 50% [2]. Conventionally, patients diagnosed with early stage HNSCC would have good prognosis after surgery and adjuvant radiation [3,4].
Before HPV positive as a new risk factor for HNSCC was found, many risk factors had been reported, including tobacco, poor oral hygiene, and alcohol [5,6]. Then the prevalence of HPV-related HNSCC, especially oropharyngeal squamous cell carcinomas (OPSCC), was largely observed in many populations in Western Europe, United States, and Australia [3][4][5][6][7]. Nonetheless, the prevalence, prognostic, significance, and correlations of high-risk HPV infection in OPSCCs in China cohort, accounting for 1/4 of the global population, remain blurry. And the precise pathogenesis and clinic pathologic features of HPV-related oropharyngeal cancer in Northeast China are still unclear.
High-risk human papillomavirus (HPV) infection causes the increase of OPSCC [8]. Many studies have shown that the prevalence of HPV-related OPSCC has been evaluated to range from 45 to 90% [3][4][5][6][7]. Moreover, a dominant subtype of HPV16 is thought to represent 90% of HPV-related OPSCC. HPV-related OPSCC is identified as a unique clinical entity. Patients with HPV associated SCC are expected to have the improved survival. Thus the clinical value of exploring the role of HPV in OPSCC is also beneficial to decrease treatment related side-effect [8].
p16 protein expression has been reported to be related to HPV infection, and p16 may be used as a predictive biomarker for HPV high-risk tumors [9]. p16 as a cyclin-dependent kinase inhibitor played an important role in inhibiting CDK4 and cyclin D1 complex dependent phosphorylation of Rb (retinoblastoma), as a tumor suppressor protein [10]. Viral oncoproteins E7 is always expressed in HPV-related cancers. Studies had shown an inhibitory effect of E7 on Rb activation by HPV infection [11,12]. And inactivation of Rb by HPV-expressed E7 induced the transcription of the cyclindependent kinase inhibitor p16 [13]. Importantly, the expression of p16 was a positive indicator for improved survival. Several researches have demonstrated that p16 was a more effective independent prognostic factor for overall survival and progression-free survival than HPV status prediction [14,15]. However, whether p16 immunohistochemistry could be used as a strong discriminator of clinical outcome in patients with OPSCC has not been defined. Larger studies are necessary to determine whether p16 can be used as well established prognostic variables, including T category, depth of invasion, and nodal status of OPSCC.
In our study, we first investigated the prevalence and prognostic and clinicopathologic features of HPV-related oropharyngeal cancer in Northeast China. Furthermore, we observed that p16 expression was significantly associated with early stage primary OPSCCs and that patients with p16 expression tend to show better survival following surgery and radiotherapy. Our results suggest that p16 may be a prognostic factor of OPSCCs in China. All slides were reviewed by two pathologists and scored the pathological variables. International Collaboration on Oropharyngeal Cancer Network for Staging (ICON-S) has developed a TNM classification specific to HPV positive oropharyngeal cancer [16,17]. We followed the TNM stage from 7th edition of the UICC/AJCC TNM classification: no lymph nodes as ICON-S N0; ipsilateral lymph nodes as ICON-S N1; bilateral or contralateral lymph nodes as ICON-S N2; lymph nodes larger than 6 cm as ICON-S N3, which resembles the N classification of nasopharyngeal carcinoma except the lack of a lower neck lymph node variable. The proposed ICON-S classification is as follows: stage I is T1-T2N0-N1, stage II is T1-T2N2 or T3N0-N2, and stage III is T4 or N3. Metastatic disease (M1) is classified as ICON-S stage IV.

Antibodies and Immunohistochemistry (IHC).
Formalinfixed samples and paraffin-embedded sections (4 m thick) were first blocked with 1% H 2 O 2 . Then the samples were treated by antigen retrieval in trypsin for 30 min at 37 ∘ C, followed by immersion in citrate buffer (pH 6.0; Mitsubishi Chemical Medience, Tokyo, Japan) for 20 min at 120 ∘ C in an autoclave. Protein Blocking Agent (Streptavidin-Biotin Universal Detection System; Beckman Coulter, Marseille, France) was used to block the sections. And then the sections were incubated with the following primary antibodies overnight at 4 ∘ C: rabbit anti-human P16 (1 : 100, INK4a, IgG, Zhongshan, China). After that, sections were incubated with secondary antibodies from the Streptavidin-Biotin Universal Detection System (Beckman Coulter) and visualized by DAB. The negative controls were specific isotype control antibodies and phosphate-buffered saline (PBS; omitting primary antibodies).
For calculating the p16 [INK4a] expression, nuclear and cytoplasmic positivity were identified as positive reactions and were scored semiquantitatively as described by previous study [15]: negative score was <1% of positive cells; sporadic score was that isolated cells were positive but <5%; focal score was small cell clusters but <80% of positive cells; and diffuse score was >80% of positive cells. Positive cells with p16 expression were defined as strong and diffuse nuclear and cytoplasmic staining in at least 80 percent or more of the tumor cells.

Statistical Analysis.
Student's -test was performed to estimate the significant difference between HPV positive OPSCC patients and HPV negative OPSCC patients. We also analyzed the correlation among clinical presentation, HPV state, and P16 by Spearman's correlation analysis and Pearson's correlation analysis (SAS Institute Inc., Cary, NC, USA). values less than 0.05 were considered as significant differences.

p16 Expression Was Significantly Associated with Improved
Survival. Among all 1470 cases, positive p16 expression was linked with markedly improved overall survival (OS,     = 0.05), but this result was not significant in multivariate analysis (Table 5).

Discussion
This study including 1470 patients over a period of 16 years has a few inherent biases typical of a retrospective cohort. The principles of surgical management and patient selection for radiotherapy have essentially remained the same during this period, although there has been increasing use of adjuvant chemotherapy and highly conformal radiation techniques. Consequently, we have used HPV status detection as the gold standard to evaluate the potential clinical value of other prognostic markers for HPV-related OPSCC.
The percentage of HPV positive OPSCC in northeast Chinese patients calculated by our research was substantially lower than that published by recent meta-analysis data in oropharyngeal cancers [18]. The percentage of HPV positive OPSCC was 11.7% in an Eastern Chinese population and 21.7% in Southern Chinese patients as previously reported [19,20]. The discrepancy may be related to the differences in the geographic origin of patients, heterogeneous laboratory procedures, or different methods used to detect HPV status.
We also found that HPV status was significantly correlated to sex and age of patients. About sex, HPV positive estimates were substantially higher in men than in women in our cohort, different from European cohort in other studies [3][4][5][6][7]. Additionally, we also found that HPV positive estimates were substantially higher in 46-55 ages. Finally, the incidence of tobacco smoking was 88.16% ( = 1296) in our cohort and nearly 89% of patients with p16 expression were smokers. Sixty (60.54%) of 890 patients were alcohol consumers, and nearly 70% of patients with p16 expression had a history of alcohol. The carcinogenic effects of smoking and alcohol mediated through p53 mutations are notable.
In our results, we observed that OPSCC patients with p16 overexpression had significantly longer disease-specific survival than p16 negative patients following surgery as well as postoperative adjuvant radiotherapy, which was consistence with published data about the potential prognostic maker of p16 in oropharyngeal cancer. It suggested that p16 may be a biomarker for predicting the prognosis of OPSCC patients in China.
HPV and p16 as biomarkers or therapeutic targets in the treatment of HNSCC have the growing consensus of the importance [21]. In our study, p16 positive patients had significantly longer disease-specific survival on univariable analysis, which was essentially equivalent to that published by previous reports [22][23][24]. However, p16 expression was not only an independent predictor of survival on multivariable analysis. As discussed above, it may be reasonable to assume that p16 expression may also mediate survival of OPSCC patients by controlling the proliferative capacity and invasive potential of the primary tumor.

Conclusion
Our study demonstrates that p16 expression is significantly associated with early stage primary OPSCCs and that patients with p16 expression tend to show better survival following surgery and radiotherapy. p16 expression, as well as HPV status, may be a prognostic maker of OPSCCs in China. Furthermore, the etiological fraction of HPV in cancers of the OPSCCs is substantially lower in Northeast China than that in United States and Western Europe. Thus, the real prevalence of HPV in OPSCCs is still the future burden. Further researches will define the more detailed mechanisms underlying HPV involvement in OPSCCs.