Abstract
Background/Aim: This study analyzed the expression of p16 in a large cohort of patients suffering from oral squamous cell carcinoma (OSCC) who received initial surgical therapy in order to evaluate the prognostic significance of p16 expression and to analyze its value as a surrogate marker to determine human papilloma virus (HPV) status. Materials and Methods: Immunohistochemical staining of p16 was performed on tissue microarrays. Different expression levels of p16 (>25%; >50%; ≥70%) with a moderate to strong intensity were correlated with the clinical outcome. HPV DNA was analyzed by polymerase chain reaction (PCR). Results: A total of 281 patients were included in this study. The p16 expression obtained using the abovementioned three different cutoffs did not significantly influence 5-year overall survival (OS) (p=0.23; p=0.45; p=0.23) nor recurrence-free survival (RFS) (p=0.79; p=0.45; p=0.142). In univariate Cox regression analysis, the p16 expression level was not a risk factor for OS (HR=0.637; 95%CI=0.271-1.5; p=0.300) and RFS (HR=0.74; 95%CI=0.339-1.61; p=0.449). A total of 17 patients (6.0%) were p16 positive with a cutoff ≥70%. HPV DNA was found in 4/11 of these cases by PCR, resulting in a positive predictive value of 0.36. In patients receiving adjuvant radio(chemo)therapy, a significantly (p=0.042) longer OS was observed in patients with p16 expression greater than 25% vs. ≤25%. Conclusion: In comparison with OPSCC, (strong) p16 positivity is rare in OSCC; however, in patients receiving primary surgery with adjuvant radio(chemo)therapy, p16 expression is associated with a higher survival rate. In conjunction with prior studies, p16 does not seem to be a reliable surrogate marker for HPV infection in OSCC.
In 2018, the worldwide incidence of oral cancer, including cancer of the lip, was approximately 355,000 cases, accounting for 1.9% of deaths of all cancer sites (1). Despite advances in diagnostic techniques and treatment, oral squamous cell carcinoma (OSCC) remains a global burden. OSCC is mainly caused by unhealthy lifestyles, including smoking, solely or in combination with alcohol, and chewing tobacco. The percentage of attributable cancer deaths in mouth and oropharynx cancers was 16% for alcohol abuse and 42% for smoking (2), demonstrating the influence of these risk factors. Consequently, these individual factors are responsible for regional peaks of OSCC, for example in India, where it represents one third of the total cancer burden (3).
Despite these traditional risk factors, the human papilloma virus (HPV) has been identified as a potential risk factor for OSCC. However, in contrast to oropharyngeal carcinomas (OPSCC), HPV in OSCC is still not well established, and it appears to play a subordinate role (4-8). To date, only a few studies have investigated the direct influence of HPV on OSCC. Variations of the prevalence are mainly caused by regional differences, with North America showing a higher prevalence of HPV in OSCC than Asia or Europe (9). However, many studies fail to separate OPSCC from OSCC (8).
HPV-associated carcinogenesis is mediated through the influence of oncoprotein E6 and E7 expression on cell cycle regulatory pathways, which leads to p16 protein accumulation (8, 10). Therefore, this protein is used as a surrogate marker and is commonly applied to evaluate HPV association in head and neck squamous cell carcinoma (11). However, while it is suitable for some authors in OSCC, there are contradictory opinions (5, 8, 12-14). Nonetheless, even independent of HPV, p16 expression was associated with a lower stage and longer disease-free survival in OSCC (15).
The aim of this study was to analyze the expression of p16 in a large cohort of patients diagnosed with OSCC with a focus on patients receiving primary surgery in order to evaluate the prognostic significance of p16 and to analyze its value as a surrogate marker to determine HPV status in OSCC.
Materials and Methods
Ethics statement. The Ethics Committee of the Faculty of Medicine Charité Berlin approved this study (EA2/028/15).
Data collection. We retrospectively analyzed the data of 281 patients diagnosed with primary OSCC between 2005 and 2011 who received initial surgical therapy at Charité – University Hospital Berlin, Germany. Tumor staging and grading was performed using the 7th edition of the American Joint Committee on Cancer (AJCC) Staging Manual (16). Treatment and follow-up were performed according to the German medical guidelines at the time of presentation. Overall survival (OS) and recurrence-free survival (RFS) were defined as previously described (17).
p16 analysis. Formalin-fixed paraffin-embedded tissue (FFPE) and the corresponding slides from patients with primary OSCC matching the inclusion criteria were collected from the archive of the Institute of Pathology, Charité - Universitätsmedizin Berlin, Germany. All tumor diagnoses were reviewed by an experienced pathologist for head and neck pathology (K.J.). The classification of carcinomas was performed according to the current WHO (WHO classification of head and neck tumors Lyon 2017) criteria. Hematoxylin and eosin (HE) staining of all of the blocks to be examined was performed in order to determine the tumor percentage for the current examinations.
Immunohistochemical staining was performed on tissue microarrays (TMAs). TMA construction was performed as previously described (17). The immunohistological staining was performed with the p16 antibody from Ventana (clone E6H4, CINtec Histology V-Kit) ready to use on Ventana platform BenchMark ULTRA IHC/ISH staining module with UltraView DAB procedure (Ventana Medical Systems Inc., Tuscon, AZ, USA). The intensity of staining (negative, weak, moderate, and strong) and the percentage of stained tumor cells were scored. In case of more than one tumor slide per patient, arithmetic means were calculated from the intensity and positivity values of these slides. Different expression levels of p16 (>25%; >50%; ≥70%) with a moderate to strong intensity were defined and correlated with clinical parameters. For surrogate marker analysis, the cutoff value for p16 was defined as at least 70% of all tumor cells showing nuclear expression with a moderate to strong intensity.
HPV detection and typing. If available, DNA was extracted from consecutive sections of formalin-fixed paraffin embedded tumor specimens using Maxwell RSC DNA FFPE Kit, (Promega). HPV typing was performed using the HPV 3.5 LCD-Array Kit (Chipron GmbH, Germany) according to the manufacturer´s instructions. In brief, a primer mixture was used to generate biotinylated PCR products from a region of the L1 gene of the HPV genome. The PCR was performed under the following conditions: initial denaturation for 5 min at 95°C; followed by 45 cycles for 1 min at 94°C, 2 min at 45°C and 2 min at 72°C; and final elongation at 72°C for 7 min. PCR products were hybridized to HPV type-specific capture probes on the HPV 3.5 LCD-Array Kit, allowing the discrimination of HPV types 06, 11, 16, 18, 31, 33, 35, 39, 42, 44, 45, 51, 52, 53, 54, 56, 58, 59, 61, 62, 66, 67, 68, 70, 72, 73, 81, 82, 83, 84, 90, and 91.
Statistical analysis. The data were collected in Microsoft Excel (Microsoft Corporation, Redmond, WA, USA) and analyzed using IBM (Armonk, NY, USA) SPSS Statistics Version 27 and R Version 4.0.3 (18). Categorical variables are presented as a frequency and percentage, while continuous variables are shown as mean values including standard deviations (SDs). The 5-year overall survival (OS) and relapse-free survival (RFS) were examined by the Kaplan– Meier method and log-rank test. Cox regression was used to identify prognostic factors for OS and RFS. All p-values are exploratory and reported without adjustment for multiple testing. p-Values smaller than α=0.05 were denoted as significant.
Results
A total of 281 patients (101 females; 180 males) with a mean age of 61.4 years (SD ±11.6; range=27-96 years) were included in this study. The mean follow-up for OS was 63.9 months (SD ±38.4). The cumulative 5-year OS of all patients was 63.3%. RFS was 54.1%.
p16 expression was analyzed in all patients and grouped according to three different cutoffs: >25% (N=27; 9.6%), >50% (N=22; 7.8%), and ≥70% (N=17, 6.0%). Typical pictures of p16 IHC staining are shown in Figure 1. The p16 expression using these cutoffs did not significantly influence 5-year OS (p=0.23; p=0.45; p=0.23) nor RFS (p=0.79; p=0.45; p=0.142) (Figure 2). In univariate Cox regression analysis, the p16 expression level was not a risk factor for OS (HR=0.637; 95%CI=0.271-1.5; p=0.300) and RFS (HR=0.74; 95%CI=0.339-1.61; p=0.449).
Immunohistochemical staining images of p16 in oral squamous cell carcinoma (OSCC). A) Weak cytoplasmic staining, the nuclei of the tumor cells are negative (black arrow); B) Strong nuclear staining (Table III, patient no. 2/HPV DNA PCR positive); C) Strong nuclear staining (Table III, patient no. 9 / HPV DNA PCR negative).
Kaplan–Meier curve showing overall survival of patients in relation to p16 expression ≥70% compared to <70%.
A total of 94 OSCC originated from the floor of the mouth (55 limited to the floor of mouth; 39 as the primary site with multilocular expansion). In these patients, RFS was significantly (p=0.045) longer in patients with an expression of p16 ≥ than 70% (N=10) compared to <70% (N=84). For OS, a strong tendency (p=0.072) was seen.
Subgroup analysis revealed a significantly (p=0.042) longer 5-year OS in patients receiving adjuvant radio(chemo)therapy with a p16 expression greater than 25% (N=6) compared to ≤25% (N=69) (Figure 3). There was no significant difference in RFS in this subgroup (p=0.059); however, a strong tendency was seen.
Kaplan–Meier curves showing overall survival of patients who received primary surgery and adjuvant radio(chemo)therapy in relation to p16 expression greater than 25% compared to ≤25%.
The clinicopathological characteristics of all patients receiving adjuvant radio(chemo)therapy with p16 expression greater than 25% (and moderate or strong intensity) are summarized in Table I.
Clinicopathological characteristics of all patients receiving adjuvant radio(chemo)therapy with p16 expression greater than 25%.
For surrogate marker analysis, the cutoff value for p16 was defined as at least 70% of all tumor cells showing expression with a moderate to strong intensity. As mentioned, a total of 17 patients of the 281 analyzed cases (6.0%) were p16-positive with a cutoff ≥70%. The clinicopathological characteristics are summarized in Table II. No significant correlation was found between patient- and tumor-related factors and p16 positivity.
Clinicopathological characteristics of all patients and patients with primary surgery grouped by p16-status (cutoff ≥70%).
A total of 14 p16-positive patients with a cutoff ≥70% were available for HPV DNA analysis using PCR (Table III). No amplification was observed in 3 cases. HPV DNA was found in 4 of 11 cases resulting in a positive predictive value (PPV) of 0.36.
Clinicopathological characteristics of all p16 positive (cutoff ≥70%) patients.
Discussion
Nowadays, it is broadly accepted that the proportion of head and neck cancers caused by HPV is increasing and that different molecular mechanisms are responsible for HPV- and non-HPV-related forms of the disease (19). However, head and neck cancers consist of a heterogeneous group of malignant tumors, and this connection appears to vary among different entities and might not be applicable to all tumors. Therefore, the present study focused on the homogeneous group of patients suffering from OSCC to evaluate the prognostic significance of p16 and its role as a surrogate marker for HPV in oral squamous cell carcinoma.
Studies evaluating the prevalence of HPV DNA in oral cancer show wide differences. A large meta-analysis published in 2014 found a pooled overall HPV DNA prevalence of 24.2% for the oral cavity (20). A more recent large international study published in 2016 including 1,264 oral cancer patients found a prevalence for HPV DNA of 7.4% (21). These authors observed geographical differences among the prevalence of HPV DNA, which might be one reason for the discordance with the aforementioned study. A large study from Germany published by Götz et al. evaluating HPV DNA showed a prevalence of 3.5% (8). Recently, a study from the Netherlands showed a prevalence of 2.2% in a cohort of 940 samples (7). Taken together, these lines of evidence indicate that, for OSCC, HPV appears to play a subordinate role in most patients.
In OPSCC, p16 expression in 70% or more of tumor cells with strong intensity shows a strong correlation with HPV (11, 22). Other authors use a cutoff level of greater than 75% expression with at least moderate intensity (23). Similar cutoffs are used by authors for non-oropharyngeal HNSCC; however, it is not clear whether the p16 scoring system is fully transferable (24) since the oral squamous epithelium is different from other sites of the head and neck area, including the oropharyngeal epithelium, as mentioned by other authors (7).
In the present study, p16 positivity for surrogate marker analysis was defined as at least 70% of all tumor cells showing nuclear expression with a moderate to strong intensity. A total of 17 patients (6.0%) of the evaluated 281 cases were p16 positive with a cutoff ≥70% in the present study. HPV DNA was found in 4 of 11 cases, resulting in a positive predictive value (PPV) of 0.36. This is comparable to other studies analyzing p16/HPV DNA exclusively in OSCC, although the exact criteria for p16 positivity/overexpression are not always mentioned and/or comparable.
A study published in Germany by Götz et al. showed that 12 of 202 cases (5.9%) were positive for p16 (with strong intensity). Of these, only 4 (33%) were positive for HPV (8). A large, international study published by Castellsagué et al. showed that 44/91 (48.4%) HPV DNA-positive oral cancers were p16 positive (21). In another study published in Thailand by Nopmaneepaisarn et al., only 9 out of 260 cases (3.5%) showed p16 positivity, with a poor predictive value for detecting HPV in OSCC (25). A higher expression of p16 was observed in oral cancer of a US population (26). A study from Hernandez et al. found an expression in 30% of tumors with a moderate or strong staining of p16 in equal to or greater than 70% of tumor cells (26). However, p16 was only positive in 9 out of 38 HPV DNA-positive tumors. A study from China published by Ni et al. showed p16 positivity in 14.3% (21/147) of cases using 25% as the cutoff value and 4.8% (7/147) using 75% as the cutoff value. However, only one patient was found to be HPV16 positive in the p16-positive patients (27). A recently published study by Nauta et al. showed that 46 of 580 (7.9%) OSCC cases were positive for p16. The study chose a cutoff value of ≥70% with strong and diffuse nuclear and cytoplasmic immunostaining. Of these, 21 cases (46%) were positive for HPV (7). These and other studies are summarized in Table IV. Taken together, these lines of evidence indicate that p16 seems not to be a reliable surrogate marker for HPV infection in OSCC, which is in line with the results of the present study.
Studies evaluating p16/HPV in oral squamous cell carcinoma (OSCC).
The p16 protein has prognostic significance in various cancers, including gastrointestinal stromal tumors (28), lung cancer (29, 30), vulvar (31), and cervical cancer (32). For instance, low p16 expression is associated with poor prognostic significance for survival in patients with non-small cell lung cancer (29). In contrast, p16 over-expression is associated with early relapse in patients with prostate cancer treated with radical prostatectomy (33). Among head and neck cancers, loss of p16 is associated with poor overall survival in nasopharyngeal carcinoma (34), and p16 over-expression leads to a significantly better prognosis in oropharyngeal cancer (35).
In the present study exclusively analyzing OSCC, we observed that p16 expression (cutoff >25%) in patients receiving primary surgery with adjuvant radio(chemo)therapy was associated with a better survival. The exact reason cannot be determined. Similar results for OSCC were described by Satgunaseelan et al. where p16 expression among the group of patients who received adjuvant radiotherapy was associated with a significantly longer disease-free and overall survival (15). This connection is also seen in oropharyngeal cancer (36), although the exact mechanism remains unclear.
Analyzing patient cohorts in general, Chen et al. found a poorer prognosis and a higher risk of tumor recurrence in OSCC without p16 expression in a Taiwanese cohort (37). A study from Germany published by Gröbe et al. found a significant correlation between p16 expression (differentiated whether p16 expression was negative, predominantly cytoplasmatic or nuclear) and recurrence-free survival (38). For OSCC of the buccal mucosa, Ni et al. found a longer RFS in p16 negative patients (27). Improved survival in p16 negative patients was also found in a Brazilian patient cohort (39). In the present study, p16 expression had no significant influence on RFS or OS concerning the whole patient cohort. Other studies have also shown no prognostic value of p16 for OSCC patients (7, 8, 26).
Many studies that have evaluated p16/HPV in head and neck cancer failed to discriminate the different entities of this heterogenous group of tumors. A particular strength of this study is the analysis of a homogeneous group of patients suffering from OSCC with a comparatively high number of patients included. The evaluation of HPV DNA by PCR in cases in which p16 was over-expressed showed that p16 is not a good surrogate marker for HPV infection in OSCC. However, our study had a few limitations. The data were collected retrospectively, and p16 expression was evaluated on tissue microarrays. The percentage of positive tumor cells of the TMA might not accurately reflect the whole tumor. Due to the fact that not all patients were evaluated for HPV DNA, the prevalence of HPV in our OSCC samples cannot be concluded from this study, and only the positive predictive value can be presented.
In conclusion, the results of this study show that p16 positivity is rare in OSCC; however, in patients receiving primary surgery with adjuvant radio(chemo)therapy, p16 expression is associated with supposedly better survival, as could also be shown with oropharyngeal tumors. Further studies with a larger cohort are necessary to confirm our results and to evaluate this pathomechanism. Our results support the literature showing that p16 does not seem to be a reliable surrogate marker for HPV infection in OSCC.
Acknowledgements
This study is part of the doctoral thesis of Maximilian Richter. The Authors thank Birgit Milluks for the great support.
Footnotes
Authors’ Contributions
Author Contributions: Conceptualization/Design: JDR and KJ; Data collection/Investigation: MR, CD, AL; Statistical analysis: KN, AP and CD; Supervision: MH and BBB; Writing – original draft: CD and CS; Writing – review & editing: PL, CS and AC. All Authors contributed to the article and approved the submitted version.
Conflicts of Interest
The Authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
- Received February 13, 2022.
- Revision received March 12, 2022.
- Accepted March 14, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
