Abstract
Background/Aim: The incidence of human papilloma virus (HPV)-related head and neck squamous cell carcinoma (HNSCC) has been increasing in the last decades. Analysis of oral brushing or rinsing samples for screening or stratification could potentially improve screening and prevention. Patients and Methods: Oral brushes and mouthwashes were taken from 20 patients with HPV-associated HNSCC before definite therapy. HPV genotyping was performed for the detection of 14 high-risk HPV subtypes and correlated to DNA isolated from tumor tissue. Results: Ten of 20 patients were tested HPV positive by using either method. There was a significant correlation between macroscopic visibility of tumor and positive HPV detection (p<0.001) and HPV detection and tumor size (p<0.001). HPV was detected in all macroscopically visible tumors. Half of the HPV cases who had macroscopically invisible tumors were missed by both methods. Conclusion: Both techniques are limited in the detection of macroscopically non-visible and small tumors. Therefore, the application of these techniques for screening or diagnosis of HNSCC is not recommended.
- Human papillomavirus
- HPV
- HPV testing
- oropharyngeal infection
- oral rinse
- oropharyngeal brushing
- Head and neck cancer
- HNSCC
- oropharyngeal carcinoma
- OPSCC
- sexually transmitted disease
Head and neck squamous cell carcinoma (HNSCC) includes malignancies in the five major anatomic sites oral cavity, oropharynx, nasopharynx, hypopharynx, and larynx and is therefore one of the most prevalent cancers worldwide (1, 2). Over the last decades, the epidemiology of different HNSCC phenotypes has changed tremendously (3). Incidence rates of Human Papilloma Virus (HPV) related HNSCC have been increasing despite decreasing rates of alcohol and tobacco-associated oral cancer (2). HPV-dependent cancer is five times more likely to occur in the oropharynx than in the oral cavity, larynx or hypopharynx (4). Recently, HPV-associated oropharyngeal squamous cell carcinoma (OPSCC) became a distinct HNSCC tumor entity with a specific TNM (tumor-node-metastasis) classification (5, 6). To date, approximately 202 different HPV types are known and have been isolated (7). The virus types infecting the mucosa are further classified into high- and low-risk groups based on the relative malignant potential of the lesions they cause. While low-risk HPV (HPV6, HPV11) causes benign genital warts or juvenile laryngeal papillomatosis, high-risk HPVs (e.g. HPV16, 18) cause squamous cell carcinoma in the head and neck region and/or the anogenital tract (4). Most oral HPV infections are transient and asymptomatic. Spontaneous clearance by the immune system is common. However, HPV infections can lead to cancer development without any cancer precursor (2, 8). Tonsillar crypts and the base of the tongue are reported as predominant sites of HPV in the oropharynx (9-11). HPV type 16 is the most prevalent genotype, found in 87% of HPV-induced OPSCC followed by HPV type 33 and HPV type 18 (12). HPV-induced neoplasm differs from other oral carcinomas; the viral DNA integrates into the cancer cell genome and the E6 and E7 oncogenes are expressed (13-15). Also, HPV is a strong and independent prognostic factor for survival among these patients, indicating a good prognosis and response to therapy compared to HPV-negative HNSCC (7, 8).
Because of the increasing global trend of HPV-induced cancer, it is crucial to advance screening and prevention. Whereas HPV-associated cancer of the cervix is often detected via comprehensive prevention in developed countries, there are no approved approaches for prevention and early detection of HPV-associated OPSCC. It usually takes 3-7 years for high-grade squamous intraepithelial lesion (HSIL) in cervical cells to become malignant. Cervical cancer screening may detect dysplasia before it becomes a malignant neoplasm. The screening test commonly used is called Pap test. Both, Pap test and HPV test, are applied in cells taken from the cervix and this process of sampling and testing is simple and fast (16). Women with a low-grade squamous intraepithelial lesion (LSIL) should be tested more frequently, since the associated abnormal cytology is transient with frequent clearance of HPV and the lesion. Women with HSIL can be treated by having their lesions excised. HPV-associated OPSCC patients often present with advanced stage cancers that require extended surgical procedures and multimodal therapies, culminating in serious morbidity and significant costs to the health-care system (9). Patients who present in the department of otorhinolaryngology with symptoms like dysphagia, long-lasting sore throat or chronic lymphadenopathy usually get a diagnostic workup including panendoscopy upon general anesthesia combined with histological examination, computed tomography (CT) or computed tomography in combination with positron emission tomography (PET-CT). In contrast, simple oral brushing or rinsing would be easy to accomplish and non-invasive. However, these techniques to identify HPV-positive HNSCC have not been well investigated and validated (17). The aim of this pilot study was to investigate the sensitivity of both techniques in patients with proven HPV positive HNSCC before definite therapy.
Patients and Methods
Patients. Patients with proven HPV-induced HNSCC were included in the study. The investigations were carried out before the definite treatment. The histological examinations of tissue samples were carried out at the Department of Pathology at the University of Munich. All participants provided informed written consent before being included in the study. The study design received the approval by the Institutional Review Board of the University of Munich (Number: 645-16). All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
Clinical sampling. Participants were advised to rinse and clean their mouths with drinking water before sampling. First, oral superficial cells were obtained from the right and left tonsillar region and the base of the tongue by performing 5 to 10 complete backward and forward brushes at each side. The brushes were suspended in Thin prep® tubes (Hologic® Inc., Marlborough, MA, USA) and stored at 4°C until HPV detection was performed within 1 to 2 weeks.
Demographics and characteristics of the patients' cohort. The variables are subdivided in categories and specified as numbers and percentages. Age at first sexual intercourse is shown as mean and total number of sexual partners is shown as median.
Second, oral exfoliated cells were sampled using a 30-second oral rinse and gargle method with 10 ml of 0.9% sodium chloride. The patients were asked to tilt their heads down, rinse, gargle and fill up a sterile Falcon-tube. Oral lavage was centrifuged (6000 rpm, 15 min) and the resulting suspension was also filled in Thinprep® tubes stored at 4°C for further processing (16).
Human papillomavirus detection and genotyping. HPV detection was performed in oral rinse and swab samples at the same laboratory using the Cobas® Test (La Roche, Basel, Switzerland) (18). The Cobas® HPV tests are automated qualitative in vitro tests for the detection of HPV DNA. The tests include amplification of target DNA by the polymerase chain reaction (PCR) and nucleic acid hybridization for the detection of 14 high-risk HPV (hrHPV) types in a single analysis. High-risk HPV genotypes 16 and 18 were selectively tested using this PCR-based System. With the help of a third searching system the high risk HPV genotypes 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66 and 68 were unselectively detected followed by an Xpert® test (Cepheid, Sunnyvale, CA, USA) in case of a positive result to confirm each genotype (19).
This table shows 20 patients with diagnosed p16 positive oral malignancy. The table is ordered by tumor visibility during diagnostic work-up. Ten tumors (50%) were macroscopically visible. Fifteen carcinomas had their origin in the palatine tonsil (75%) and 5 at the base of the tongue (25%). The tumors were located submucosal, papillous or as a superficial ulcus. Two tumor appearances were not documented (N/A). One tumor was not operable. +: present; −: absent.
Additionally, as a control, DNA was isolated from tumor tissue. The genomic sequence of the HPV L1 gene was amplificated by PCR (MY09-MY11 and 125) and analyzed by Chipron HPV-Type 3.5C system (Chipron, Berlin, Germany). The HPV-subtypes 6, 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 are detected with this method (20). p16 expression was determined by immunohistochemistry during routine diagnostics.
Statistical analysis. All data were analyzed by chi-square test using SPSS® (SPSS Inc., Chicago, Il, USA). A p-value<0.05 was considered statistically significant.
Results
Patient characteristics. A total of 20 patients participated in this study. The demographic data are described in Table I. The median age of the study group was 64 years (range=45-79 years) and 13 patients (65%) were male. None of the patients received an HPV vaccination in the past or suffered from immunodeficiency.
Tumor characteristics. All 20 patients included in this study were finally diagnosed with a p16 positive oropharyngeal malignancy: 15 carcinomas had their origin in the palatine tonsil (75%) and five at the base of the tongue (25%). All tumors were positive for high-risk subtype HPV 16. Table II illustrates the final TNM classification of the tumors and their size and visibility during the diagnostic work up. Ten tumors (50%) were visible to the naked eye. Eight tumors were located beneath the superficial mucosa, e.g. tonsillar crypts and not visible during panendoscopy. Photographic documentation of the oropharynx as well as histologic samples of two representative cases are presented Figure 1. Four of these eight tumors were detectable by clinical palpation. Two tumors were an incidental finding in the palatine tonsil. One tumor was initially a cancer of unknown primary (CUP) origin and noticed by a p16 positive lymph node metastasis. Another tumor was detected by PET-CT. Two tumor locations were not named.
HPV detection. Ten of 20 patients were tested HPV positive using the brushing method. The same 10 patients were also tested HPV positive using the rinsing method. In summary, there were no differences in the two methods.
Endoscopic photographs of two representative cases of oropharyngeal carcinoma (A and B). In A1, the intraoperative endoscopic photograph shows a pT3 carcinoma of the right palatine tonsil. Tracheal tube is in situ. In B1, a hidden carcinoma of the left tonsil was found in the histopathological workup after cancer of unknown primary (CUP) origin was afore detected by diagnostic neck dissection. The photograph was taken during clinical examination and only a slight asymmetry of the palatine tonsils was prevalent. The corresponding histopathologic pictures (A2, A3, B2, B3) show squamous cell carcinoma (hematoxylin and eosin staining in A2 and B2) with strong, “block-like” p16 staining (immunohistochemical p16 staining in A3 and B3). Magnification 200× (A2, A3, B2, B3).
There was a significant correlation between visibility and test result (p<0.001) and between the test result and tumor size (p<0.001). None of the methods detected macroscopically non-visible tumors.
There was no significant correlation between the test result and tumor grading (p=0.061). Test result and age of patient (p=0.235) and gender (p=0.348) were independent as well.
There was also no significant correlation between smoking habits and HPV detection (p=0.338).
Discussion
A recently published study showed that the general population remains mostly unaware of the potential links between HPV infection and HNSCC (21).
Although diagnostic techniques to diagnose oral HPV infection have so far not been well investigated and validated, one may postulate that regular HPV screening could facilitate the early detection of HNSCC (22). Initial studies about HPV suggest that most of the oral infections are likely to be cleared within a year. Persistence of the infection might be the critical factor for the development of HPV-related diseases (23).
The gold standard for HPV viral load assessment is real-time PCR. Many previous studies have examined the best method to detect oral HPV. One major issue discussed in the literature is the low incidence of oropharyngeal HPV infection.
A recently published study has investigated 15,313 healthy persons, who attended dental clinics, and showed an incidence of only 1% for HPV16/18 in oral rinse (24).
As other studies have also shown, the incidence of a detectable oral HPV infection by brushing and rinsing techniques is extremely low, even in a high-risk population with proven anogenital HPV infections (16, 25).
HPV is localized in the biofilm of tonsillar crypts and at the base of the tongue. Therefore, a brush or a rinsing medium might not physically reach HPV DNA because of the depth within the lymphatic tissue and the added protection by the biofilm (26-30).
A systematic review regarding the diagnostic accuracy of oral HPV detection using oral rinses and swabs showed a good specificity of oral HPV detection (92%, 95%CI=82-97%) and a moderate sensitivity (72%, 95%CI=45-89%) for HPV-positive HNSCC (17).
Our study showed that the brushing as well as the rinsing technique are able to detect HPV-positive HNSCC that were macroscopic visible. The sensitivity for visible tumors was 100% for both methods. Compared to the use of a brush, which collects superficial mucosal cells from different anatomic sites, the oral rinse technique seemed more likely to have sampled the tumor site and the area of HPV infection. Therefore, it is remarkable that both tests showed identical results.
In a recently published study Tang et al. have shown that HPV-16 DNA viral load (>10 copies/50 ng) was significantly associated with advanced stages of HNSCC (31). In patients with recurrent, persistent locoregional (LR) disease, median baseline normalized salivary HPV DNA was nearly 20× higher compared to patients with distant disease only (p=0.01) (22). Our findings highly underline this correlation. The biological reason for that is still unclear.
Furthermore, our study strengthens the literature data, which show a high false negative rate (32). In our patients who already have HNSCC, oral HPV detection misses one-half of HPV-related cases, especially those tumors, who are located beneath the superficial mucosa.
Overall these findings indicated a limitation of these techniques to screen for HPV-positive HNSCC in a healthy population (17). As there is no approved biomarker that can be used to screen for HNSCC, most carcinomas are diagnosed in the advanced stages of the disease (III and IV) when patients show symptoms. The diagnosis at an early stage of the disease followed by early intervention with the recommended therapy would increase the rate of survival up to 5 years and also reduce mortality rates (33). Several methods are available to diagnose HPV-dependent cancer. While immunohistochemistry (IHC) analysis of p16INK4A in HNSCC tumor biopsies has be shown to be a surrogate marker to identify HPV infection, IHC staining against specific HPV proteins in tumor tissue should be performed in routine clinical practice (34). Therefore, the development of IHC staining for the oncogenic E6 and E7 proteins would prove that HPV oncogenes are present (20). In our study group all samples were stained positive for p16 in routine clinical practice. In the additional DNA-testing of the pathological specimens, we only found the HPV 16 subtype, which was matched with the PCR results for HPV infection. This finding is consistent with several studies on HPV subtypes found in HNSCC (12, 35). Furthermore, it is reported in the literature that recurrent HNSCCs are only detected with whole body imaging. These findings might also indicate that the tested methods have limited prospects in HPV detection in recurrent and metastasized disease. Nevertheless, this should be discussed in further studies (36).
The development of a Pap test equivalent for the oral cavity is certainly challenging. The problem seems to be how to identify and obtain a sample of the relevant epithelium, where persistent HPV infection can lead to dysplasia or cancer.
We could show, that a small tumor size is a predictor for missing relevant HPV infection. At present, small or hidden tumors in the depth of the tonsil crypts or the base of the tongue seem to be unapproachable for HPV detection using the brushing and rinsing techniques.
Conclusion
The presented data highlights the risk of missing HPV in a cohort with HPV-associated OPSCC, especially when the HPV-DNA-containing mucosa lesions are not macroscopically visible. Half of the HPV-associated tumors were not macroscopically visible, including tumors in the depth of the palatine tonsils or the base of tongue. The oropharyngeal brushing and rinsing methods of the oral cavity and oropharynx detected only visible tumors. There is no significant difference regarding the accuracy of both methods. The crucial limitation of both methods is missing hidden tumors. Therefore, a regular screening for persisting HPV infection and HPV-associated OPSCC using the described methods is not recommended.
Acknowledgements
The study was funded by the medical faculty of the LMU Munich.
Footnotes
↵* These Authors contributed equally to this study.
Authors' Contributions
TKE, CD, SM, KS, and JKSG protocol/project development. TKE, KS, PB, and JKSG data collection and management, data analysis. CD, CJT, JK, UJ cytopathologic/HPV evaluation. DM and ES histopathology evaluation, TKE, PB, KS and JKSG manuscript writing/editing. TN, TKE, KS, JKSG statistical analysis. All Authors interpreted the results and approved the final manuscript.
Conflicts of Interest
TKE declares that she has received speaker honoraria from companies Aristo Pharma and Roche. JKSG and CJD declare that they have received speaker honoraria from companies MSD and Roche. SM reports grants and personal fees from AstraZeneca, personal fees from Clovis, grants and personal fees from Medac, grants and personal fees from MSD. He also reports personal fees from Novartis, grants and personal fees from PharmaMar, grants and personal fees from Roche, personal fees from Sensor Kinesis, grants and personal fees from Tesaro, grants and personal fees from Teva, outside the submitted work. KS, PB, CJT, UJ, ES, DM, JK, declare no conflicts of interests. TKE, KS and JKSG state that they have full control of all primary data and that they agree to allow the Journal to review their data if requested. Additional potential conflicts of interest relevant to this article were not reported.
- Received February 21, 2020.
- Revision received March 9, 2020.
- Accepted March 11, 2020.
- Copyright© 2020, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.