|Ahead of print publication
Can dual staining with p16 and Ki67 be biomarkers of epithelial dysplasia in oral lesions?
Anju Bharti1, Sumaira Qayoom2, Riddhi Jaiswal2, Preeti Agarwal2, RK Singh3, SP Agarwal3, Shalini Bhalla2, Annu Makker4, Madhu Mati Goel2
1 Department of Pathology, Institute of Medical Sciences, BHU, Varanasi, Uttar Pradesh, India
2 Department of Pathology, KGMU, Lucknow, Uttar Pradesh, India
3 Department of Maxillofacial Surgery & Otorhinolaryngology, KGMU, Lucknow, Uttar Pradesh, India
4 Department of Biochemistry, Prasad Institute of Medical Sciences, Lucknow, Uttar Pradesh, India
|Date of Submission||14-Feb-2020|
|Date of Decision||28-May-2020|
|Date of Acceptance||16-Jul-2020|
|Date of Web Publication||23-Oct-2021|
Department of Pathology, KGMU, Lucknow - 226 003, Uttar Pradesh
Source of Support: None, Conflict of Interest: None
Background: Oral carcinogenesis is a multistage process with epithelial dysplasia as a premalignant condition. There is a significant inter-observer variation in diagnosing and grading the oral epithelial dysplasia. As human papillomavirus (HPV) is believed to have à strong relationship with oral carcinogenesis, using P16 as a biomarker may help in identifying the cells which may be undergoing the malignant transformation. However, due to the low specificity of P16, dual staining test P16INK4/Ki67 might be a better promising marker for identifying the transformed cells. This study was designed to evaluate the dual expression of P16 and Ki67 as a promising biomarker for dysplasia and their correlation with clinicopathological factors.
Materials and Methods: Immunohistochemical analysis for p16 and ki67 was performed on 30 premalignant oral lesions and 36 oral squamous cell carcinoma (OSCC) by dual staining using the CINtec PLUS kit.
Results: CINtec positivity was observed only in leukoplakia with dysplasia (46.7%) and squamous cell carcinoma (25%). None of the cases of leukoplakia without dysplasia or oral submucosal fibrosis stained positive for CINtec plus staining. In leukoplakia with dysplasia, there was no significant association with any of the clinicopathological parameters studied. In OSCC cases, alcohol intake showed statistically significant association with CINtec positivity.
Conclusion: P16INK4/Ki67 assessment by dual staining is a promising biomarker for identifying dysplasia in cases with diagnostic dilemmas.
Keywords: Biomarker, CINtec PLUS, human papilloma virus, oral squamous cell carcinoma, P16INK4/Ki67
|How to cite this URL:|
Bharti A, Qayoom S, Jaiswal R, Agarwal P, Singh R K, Agarwal S P, Bhalla S, Makker A, Goel MM. Can dual staining with p16 and Ki67 be biomarkers of epithelial dysplasia in oral lesions?. J Can Res Ther [Epub ahead of print] [cited 2021 Nov 29]. Available from: https://www.cancerjournal.net/preprintarticle.asp?id=329053
| > Introduction|| |
Oral cancer is the fifth most common cancer worldwide representing the leading cause of cancer-related mortality. Although representing 2%–4% of the malignancies in the west, it accounts for almost 40% of all the cancers in the Indian subcontinent, and by 2025, more than 60% of oral cancers will be from developing world. Oral carcinogenesis is a multistage process in which invasive squamous cell carcinoma is preceded by premalignant lesions which develop after a series of genetic mutations. These premalignant changes present clinically as leukoplakia or erythroplakia and diagnosed histologically as dysplasia. A correct diagnosis of dysplasia, along with its grade is of utmost importance to predict the risk of developing invasive carcinoma. Severe dysplasia has a fivefold increased risk of developing invasive carcinoma as compared to mild dysplasia. As oral dysplasia are mainly keratinizing type, there is significant intra-observer as well as inter-observer variability in diagnosis and grading of oral dysplasia. The identification of high risk oral premalignant lesions and intervention at that stage could constitute one of the keys in reducing the mortality, morbidity, and cost of the treatment associated with oral squamous cell carcinoma (OSCC) especially in developing countries.
The etiology of OSCC is believed to be multifactorial, with tobacco and alcohol being established etiological agents. Recently, several studies have shown that human papillomavirus (HPV) is clearly involved in the pathogenesis of a subgroup of OSCC. This distinct subgroup of head and neck cancers is characterized by distinctive clinical and histopathological features which include younger patients, basaloid SCC, and poorly differentiated yet they respond better to induction chemotherapy and have an overall better outcome as compared to HPV-negative OSCC. In the current era of personalized medicine, these facts would make reliable identification of HPV-induced OSCC important for HPV stratified therapy protocols in clinical practice. A review of the literature shows the variable prevalence of HPV in OSCC, ranging from 27% to 74%. P16 immunohistochemistry has been widely used for the detection of HPV in carcinoma cervix. So by analogy, it could be accepted as a biomarker for HPV infection in OSCC also. However, certain studies have reported decreased expression of P16 in immunohistochemistry in premalignant and malignant oral lesions., In addition, the fact that p16 can be overexpressed in nonmalignant cells and 5%–10% of HPV-Negative OSCC impedes its unrestricted use for HPV detection. Moreover, mere overexpression of P16 does not detect the malignant transformation of the cell. Recently, the detection of p16 along with proliferation marker Ki67 in the same cell has been used in the detection of cancer cells in the cervix with encouraging results. We hypothesized that dual expression of p16 and ki67 would indicate the transformed cells on the oral squamous epithelium. The present study was designed to observe the dual expression of p16 and Ki67 as a promising biomarker for dysplasia and their correlation with clinicopathological factors.
| > Materials and Methods|| |
This was a retrospective case–control study conducted in the Department of Pathology, KGMU, Lucknow, India, and approved by the Institutional Ethics Committee. A total of 71 biopsy tissues were included in the study comprising of 5 healthy controls, 30 cases of premalignant oral lesions, and 36 cases of OSCC diagnosed between January 2013 and December 2014.
Evaluation of cases and immunohistochemical staining
Sections from formalin-fixed paraffin-embedded tissue were stained with H and E staining. All the cases were reviewed by a senior pathologist (MMG). Immunohistochemical analysis for p16 (clone E6H4) and Ki67 (Clone 274-11AC3) was performed on all the study sections by dual staining protocol as per the manufacturer's instruction of CINtec PLUS kit (Roche MTM Laboratories Manheim Germany). P16INK4a was visualized by horseradish peroxidase mediated conversion of 3, 3-diaminobenzidine chromogen leading to brown staining and Ki67 by alkaline phosphatase-mediated conversion of fast red chromogen leading to red staining at the target antigenic sites.
Positive control for P16 and KI67 was sections taken from cases that were known positive for high-risk HPV16/18 by real-time polymerase chain reaction and also positive for immunohistochemical staining for p16 and ki67 (single staining individually as well as by dual staining protocol). With each batch of slides positive and negative control was also put up. The slides were examined at ×40 magnification. The presence of brown-colored end product at the site of the target antigen was indicative of positive reactivity for p16, and the red-colored end product was indicative of Ki 67. The dual staining was considered positive when both brown and red staining was positive in the same cell, meaning that the cell was positive for both p16INK4A and Ki67.
The statistical analysis was performed using GraphPad Software, San Diego, CA, USA. Data were presented as percentages for categorical variables. Two-sided Chi-square test was performed for assessing the association of clinicopathological variables with immunohistochemical expression. A P ≤ 0.05 was considered as statistically significant.
| > Results|| |
In our study, 30 cases of premalignant lesions comprised 6 cases of oral submucosal fibrosis (OSMF), 9 leukoplakia (LKP), and 15 leukoplakia with dysplasia (LKPD). Most of the patients with OSMF (4/6; 67%) and LKP (6/9; 67%) were <40 years of age while the majority of patients with LKPD and OSCC were more than 40 years of age. The study participants were 11 females and 60 males. Males were more commonly affected by OSMF (5/6; 83.33%), LKP (8/9; 88.89%), and LKPD (13/15; 86.67%) as compared to females. The most prevalent site for premalignant lesions was buccal mucosa (20/30; 66.67%) and for malignant lesion was tongue (21/36; 58.33%). The history of tobacco chewing, smoking, and alcohol intake was considered for both cases and controls. None of the controls was exposed to the habit of tobacco intake in any form. Overall, 2/6; 33.33% cases with OSMF, 6/15; 40% cases with LKPD and 27/36; 75% cases with OSCC had the habit of tobacco chewing. Based on histological differentiation, 19/36 (53%) cases of OSCC were well-differentiated, and 17/36 (47%) were moderately differentiated. Simultaneous qualitative immunohistochemical localization of p16INK4A and Ki67 protein was performed on all cases and controls using the CINtec PLUS kit. P16INK4A positivity (alone) was observed only in LKPD (5/15; 33.33%) and OSCC (7/36; 19.44%) cases. None of the normal tissue, OSMF, or LKP case stained positive for p16 alone. Ki67 positivity alone in other than basal layer was seen in cases with OSMF (1/6; 16.67%), LKP (3/9; 33.33%), LKPD (10/15; 66.67%), and OSCC (28/36; 77.78%). Dual positivity (both P16INK4A and KI67) or CINtec positivity was observed only in cases with LKPD (7/15; 46.67%) and OSCC (9/36; 25%), as shown in [Table 1] and [Figure 1] and [Figure 2]. The mean expression of ki67 was 5% in OSMF, 8% in LKP, 25% in LKPD and 40% in SCC [Table 2]. Patients with LKPD and OSCC were categorized based on their CINtec PLUS immunoexpression into CINtec-positive and CINtec-negative groups. The association of these two subgroups was analyzed with clinicopathological parameters which included age, gender, site, and risk factors. In LKPD, no significant association was observed with any of the above-mentioned parameters [Table 3]. In OSCC, alcohol intake showed a statistically significant association with CINtec positivity (P = 0.002) [Table 4]. The follow-up was available in 12 cases of OSMF and LKP, and none of the patients developed any malignant lesion at the same site during 4 years of follow-up.
|Figure 1: Leukoplakia without dysplasia staining negative for dual stain (p16INK4A/Ki67) (a, ×20) Leukoplakia with dyspalsia showing dual stain positivity (p16INK4A/Ki67) in dyspastic epithelium (b, ×20)|
Click here to view
|Figure 2: Case of squamous cell carcinoma staining negative for dual stain (p16INK4A/Ki67) (a, ×20) Squamous cell carcinoma showing dual stain positivity with moderate proliferative index (p16INK4A/Ki67) (b, ×20) and high proliférative index (c, ×20)|
Click here to view
|Table 2: Quantitative result of Ki67 labelling index other than basal layer on Immunohistochemistry|
Click here to view
|Table 3: Association of CINtec Plus immunoexpression with clinicopathological parameters in patients with leukoplakia with dysplasia|
Click here to view
|Table 4: The association of CINtec plus immunoexpression with clinicopathological parameters in patients with oral squamous cell carcinoma|
Click here to view
| > Discussion|| |
Immunohistochemical overexpression of p16INK4A has been reported as an established surrogate marker of Human Papilloma Virus (HPV) infection in cervical cancer. However, in head and neck cancers, the reports have been contradictory with some studies showing decreased expression of p16 while others reporting increased expression on immunohistochemistry in oral dysplasia and squamous cell carcinoma. Wu et al. demonstrated reduced expression of p16INK4A during malignant transformation of oral leukoplakia to OSCC. Cases of ambiguous p16INK4A overexpression are observed in nonmalignant tissue such as tonsillar crypt epithelium and proportion of branchial cleft cysts, and overexpression is also seen in 10% OSCC HPV negative. These aspects impede the unrestricted application of p16INK4A as a molecular biomarker in head and neck squamous cell carcinoma. The expression of Ki67 alone although significant in differentiating normal mucosa, benign lesions, and mild dysplasia on the one hand, from moderate and severe dysplasia on the other hand, is less useful in differentiating these conditions from invasive carcinomas. Various studies have been reported on p16INK4A, and Ki67 immunostaining in various HPV-related cancers including head and neck but most of them have been done with individual antibodies and very few for combined detection of the two antigens.,, Takkem et al. reported mean Ki67 expression in normal oral epithelium, oral epithelium with dysplasia, and OSCC as 10.6, 36.83, and 42.87, respectively. There was a statistically significant difference in expression of ki67 between normal epithelium and oral epithelium with dysplasia.
The simultaneous detection of the cell cycle regulator p16INK4A and the proliferation marker Ki67 within the same cell has shown good results in identifying malignant cells in cytology preparations from the uterine cervix. The combined staining method has a unique advantage of indicating proliferative activity in HPV transformed cells. Reports in the literature have shown a strong correlation between HPV positivity and dysregulation of p16INK4A suppressor protein which results in overexpression of p16INK4A.
The present study was carried out to demonstrate dual immunostaining for p16INK4 and ki67, as a surrogate of HPV infection in premalignant and malignant oral lesions and their correlation with clinicopathological factors. The detection of oral HPV infection has been reported in individuals in their 20 s, i.e., at the peak time of sexual activity; however, it has an even higher prevalence among older age groups. In the present study, we found that dual p16INK4 and ki67 staining was positive in 5/10 LKPD patients in the age group below 40 years and 2/5 LKPD patients above 40 years. Whereas in OSCC group, 1/10 cases were below 40 years and 8/26 patients were above 40 years of age. Dual p16INK4 and KI67 staining in the present study was higher in males both in cases with LKPD (7/15; 46.66%) and OSCC (8/30; 26.66%). Werness et al. have reported a statistical correlation of male predominance in gender-wise distribution of HPV-positive OSCC cases. Findings of investigators such as Cruz et al. and Koppikar et al. have also concluded that men are more likely to be positive for HPV than females. In the present study, no significant association of dual stained positive cases was observed with any clinicopathological parameter in the LKPD group. In the OSCC group, alcohol intake showed a statistically significant association with combined p16/ki67 positivity. Gillison et al. have suggested that HPV infection may interact with alcohol and tobacco exposure in tumor promotion. The OSCC cases in the present study were categorized according to histological differentiation into moderate and well differentiated. A significant association between dual staining p16INK4A nd ki67 staining and histopathological grade of differentiation was also observed. The dual positivity was higher in well-differentiated SCC (53%). The histological subtype of nonkeratinizing morphology is significantly more likely to be HPV and p16INK4 positive than keratinizing SCC. Contrary to the reported literature, dual staining of p16INK4A and ki67 was seen more in OSCC showing keratinization, and only some of them showed areas of papillary transitional and basaloid morphology but not pure basaloid or non-keratinizing types. Our findings were supported by Elango et al. 2011 who have also reported p16ink4 expression and HPV positivity in well-differentiated carcinoma of tongue with low recurrence rate; these findings were statistically significant.
In the present study, p16INK4A positivity alone was observed in cases with LKPD (5/15; 33.33%) and OSCC (7/36;19.44%) while Ki67 positivity alone was seen in cases with OSMF (25%), LKP (33.33%), LKPD (66.67%), and OSCC (77.78%). The frequency of p16INK4A positivity observed in our series is similar to the findings of Gaurav et al. 2013. p16INK4A overexpression in 5%–10% of HPV-negative cases can occur in non-transformed head and neck squamous epithelium., preventing a potential application of p16INK4A immunostaining as a marker of HPV-associated transformation in head and neck cytopathology.
In our study, dual positivity was higher in cases with LKPD (7/15; 46.67%) and OSCC (9/36; 25%) as compared to that seen with p16INK4A alone and support the hypothesis by Reuschenbach et al. which stated that p16INK4A positive cells in epithelia displaying only focal p16INK4A expression do not co-express ki67, and focal expression of p16INK4 was seen in metaplastic epithelium while epithelia with dysplasia had a more diffuse expression of p16ink4 and co-expressed ki67. None of our cases without dysplasia stained positive for dual staining. One of the recent studies by Prigge et al. has demonstrated that combined expression of p16INK4A and the proliferation marker Ki67 (in a single IHC procedure) occurred exclusively in transformed squamous epithelial cells in the head and neck. Co-expression was never detected in the nontransformed epithelium. All HPV oncogene expressing OSCC demonstrated combined p16INK4A/Ki67 expression. Co-expression was stringently associated with a diffused p16INK4A expression pattern. Their findings are in strong support of our study. However, these workers performed dual IHC only on cases of oropharyngeal carcinomas and compared their results with nonoropharyngeal nonmalignant tissues to nullify the effect of nonspecific p16INK4A positivity in tissues other than OSCC.
The present study was limited by factors such as comparatively small sample size and HPV typing not performed in the study population. It has been proposed that p16INK4A/ki67 co-expression appears as a general marker of head and neck malignant transformation rather than an exclusive marker of HPV transformed. Some authors suggest alternative HPV independent mechanisms that allow for p16INK4A expression in the context of proliferation, such as disruption of the p16INK4A downstream effector pathway. Increasing efforts are being made to understand the molecular biology of this distinct p16INK4A positive, HPV-negative group., However, the underlying mechanisms of p16INK4A expression in these cancers have not been discerned sufficiently to date.
| > conclusion|| |
The results of our study imply a promising impact of combined p16INK4A ki67 immunostaining in the assessment of ambiguous p16INK4A expression in oral cancer by specifically identifying p16INK4A expressing cells with proliferating activity, and the technique may thus be used as a surrogate for identifying HPV infection in oral premalignant and malignant lesions. As none of our cases without dysplasia showed dual staining, this fact can be used in cases of dilemma about dysplasia. The combined evaluation of two phenotypic alterations in a single cell might provide more prognostic information on oral cancer as well as progression from a precancerous lesion to cancer. This requires evaluation of these markers on a longitudinal cohort of oral cancer and precancerous lesions and a long-term follow-up on a larger sample size in a well-designed study on histological and cytological samples.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| > References|| |
Mehrotra R, Singh M, Kumar D, Pandey AN, Gupta RK, Sinha US. Age specific incidence rate and pathological spectrum of oral cancer in Allahabad. Indian J Med Sci 2003;57:400-4.
] [Full text]
Vargas-Ferreira F, Nedel F, Etges A, Gomes AP, Furuse C, Tarquinio SB. Etiologic factors associated with oral squamous cell carcinoma in non-smokers and non-alcoholic drinkers: A brief approach. Braz Dent J 2012;23:586-90.
Hardisson D. Molecular pathogenesis of head and neck squamous cell carcinoma. Eur Arch Otorhinolaryngol 2003;260:502-8.
Bradley TK, Budnick SD, Logani S. Immunohistochemical detection of P16INK4a in dyplastic lesions of the oral cavity. Modern Pathol 2006;19:1310-6.
Brothwell DJ, Lewis DW, Bradley G, I Leong RC Jordan, Mock D, et al
. Observer agreement in the grading of oral epithelial dysplasia. Commun Dent Oral Epidemiol 2003;31:300-5.
Liu SC, Klein-Szanto AJ. Markers of proliferation in normal and leukoplakic oral epithelia. Oral Oncol 2000;36:145-51.
Sritippho T, Chotjumlong P, Iamaroon A. Roles of human papillomaviruses and p16 in Oral Cancer. Asian Pac J Cancer Prev 2015;16:6193-200.
Klaes R, Friedrich T, Spitkovsky D, Ridder R, Rudy W, Petry U, et al
. Overexpression of p16(INK4A) as a specific marker for dysplastic and neoplastic epithelial cells of the cervix uteri. Int J Cancer 2001;92:276-84.
Reed AL, Califano J, Cairns P, Westra H, Jones RM, Koch W, et al
. High frequency of p16 (CDKN2/MTS-1/INK4A) inactivation in head and neck squamous cell carcinoma. Cancer Res 1996;56:3630-3.
Ai L, Stephenson KK, Ling W, Zuo C, Mukunyadzi P, Suen JY, et al
. The p16 (CDKN2a/INK4a) tumor-suppressor gene in head and neck squamous cell carcinoma: A promoter methylation and protein expression study in 100 cases. Mod Pathol 2003;16:944-50.
Lewis JS Jr., Thorstad WL, Chernock RD, Haughey BH, Yip JH, Zhang Q, et al
. p16 positive oropharyngeal squamous cell carcinoma: An entity with a favorable prognosis regardless of tumor HPV status. Am J Surg Pathol 2010;34:1088-96.
Wu W, Wang Z, Zhou Z. Role of the human papillomavirus in malignant transformation of oral leukoplakia distinct from oropharyngeal squamous cell carcinoma: A study of 76 patients with internal-control specimens. Oral Surg Oral Med Oral Pathol Oral Radiol 2019;128:273-9.
Begum S, Cao D, Gillison M, Zahurak M, Westra WH. Tissue distribution of human papillomavirus 16 DNA integration in patients with tonsillar carcinoma. Clin Cancer Res 2005;11:5694-9.
Cao D, Begum S, Ali SZ , Westra WH. Expression of p16 in benign and malignant cystic squamous lesions of the neck. Hum Pathol 2010;41:535-9.
Deng Z, Hasegawa M, Aoki K, Matayoshi S, Kiyuna A, Yamashita Y, et al
. A comprehensive evaluation of human papillomavirus positive status and p16INK4a overexpression as a prognostic biomarker in head and neck squamous cell carcinoma. Int J Oncol 2014;45:67-76.
Angiero F, Berenzi A, Benetti A, Rossi E, Sordo RD, Sidoni A, et al
. .Expression of P16, P53 and Ki67 proteins in the progression of epithelial dysplasia of the oral cavity. Anticancer Res 2008;28:2535.
Takkem A, Barakat C, Zakaraia S, Zaid K, Najmeh J, Ayoub M, et al
. Ki-67 Prognostic value in different histological grades of oral epithelial dysplasia and oral squamous cell carcinoma. Asian Pac J Cancer Prev 2018;19:3279-86.
Ikenberg H, Bergeron C, Schmidt D, Griesser H, Alameda F, Angeloni C, et al
. Screening for cervical cancer precursors with p16/Ki-67 dual-stained cytology: Results of the PALMS study. J Natl Cancer Inst 2013;105:1550-7.
Gillison ML, Broutian T, Pickard RK, Tong ZY, Xiao W, Kahle L, et al
. Prevalence of oral HPV infection in the United States, 2009-2010. JAMA 2012;307:693-703.
Werness BA, Levine AJ, Howley PM. Association of human papillomavirus types 16 and 18 E6 proteins with p53. Science 1990;248:76-9.
Cruz IB, Snijders PJ, Steenbergen RD, Meijer CJ, Snow GB, Walboomers JM, et al
. Age-dependence of human papillomavirus DNA presence in oral squamous cell carcinomas. Eur J Cancer B Oral Oncol 1996;32B: 55-62.
Koppikar P, devilliers EM, Mulherkar R. Identification of human papillomavirus in tumors of the oral cavity in an indian community. Int J Cncer 2005;113:946-50.
Chernock RD, El-Mofty SK, Thorstad WL, Parvin CA, Lewis JS. HPV related non keratinizing squamous cell carcinoma of the oropharynx: Utility of microscopic features in predicting patient outcome. Head Neck Pathol 2009;3:186-94.
Elango KJ, Suresh A, Erode EM, Subhadradevi L, Ravindran HK, Iyer SK, et al
. Role of human papilloma virus in oral tongue squamous cell carcinoma. Asian Pac J Cancer Prev 2011;12:889-96.
Agarwal GP, Joshi PS, Agrawal A. Role of HPV-16 in pathogenesis of oral epithelial dysplasia and oral squamous cell carcinoma and correlation of P16INK4A expression in HPV-16 positive cases. Immunohistochem Study 2013;2013:7.
Lewis JS, Thorstad WL, Chernock RD, Haughey BH, Yip JH, Zhang Q, et al
. P16 positive oropharyngeal squamous cell carcinoma: An entity with a favourable prognosis regardless of tumor HPV status. Am J Surg Path 2010;34:1088-96.
Lingen MW, Xiao W, Schmitt A, Jiang B, Pickard R, Kreinbrink P, et al
. Low etiologic fraction for high-risk human papillomavirus in oral cavity squamous cell carcinomas. Oral Oncol 2013;49:1-8.
Reuschenbach M, Seiz M, von knebel Doeberitz C, Vinokurova S, Duve A, Ridder R, et al
. Evaluation of cervical cone biopsies for co-expression of P16INK4A and Ki67 in epithelial cells. Int J Cancer 2012;130:388-94.
Prigge ES, Toth C, Dyckhoff G, Wagneret S, Müller F, Wittekindt C, et al
. P16INK4A/Ki67 co-expression specifically identifies transformed cells in the head and neck region. Int J Cancer 2015;136:1589-99.
Perrone F, Gloghini A, Cortelazzi B, Bossi P, Licitra L, Pilotti S. Isolating p16-positive/HPV-negative oropharyngeal cancer: An effort worth making. Am J Surg Pathol 2011;35:774-7.
Rietbergen MM, Snijders PJ, Beekzada D, Braakhuis BJ, Brink A, Heideman DA, et al
. Molecular characterization of p16-immunopositive but HPV DNA-negative oropharyngeal carcinomas. Int J Cancer 2014;134:2366-72.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4]