Research Paper
Head and Neck Oncology
Human papillomaviruses in oral
squamous cell carcinoma and
pre-cancerous lesions detected
by PCR-based gene-chip array
C.-W. Luo, C.-H. Roan, C.-J. Liu: Human papillomaviruses in oral squamous cell carcinoma and pre-cancerous lesions detected by PCR-based gene-chip array. Int. J. Oral Maxillofac. Surg. 2007; 36: 153–158. # 2006 International Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.C.-W. Luo1,2, C.-H. Roan3, C.-J. Liu4,5,6
1Division of Oral and Maxillofacial Surgery,
Department of Dentistry, Taipei Medical University Hospital, Taipei, Taiwan;2Dental
School, Taipei Medical University, Taipei, Taiwan;3Department of Oral Pathology and
Medicine Graduate School of Medicine and Dentistry, Okayama University, Okayama, Japan;4School of Dentistry,
National Yang-Ming University, Taipei, Taiwan;5Mackay Medicine, Nursing and Management College, Taipei, Taiwan;
6Department of Oral and Maxillofacial
Surgery, Mackay Memorial Hospital, Taipei, Taiwan
Abstract. Human papillomavirus (HPV) infection is a significant risk factor for uterine cervical carcinoma. Many studies have also demonstrated the presence of HPV in oral epithelia tissue, but the role of HPV infection in oral squamous cell carcinoma (OSCC) is still controversial. The aim of this study was to determine the frequency and type of HPV in OSCC and oral pre-cancerous lesions. DNA samples were collected by cytobrushing from 51 patients with OSCC, 46 with oral pre-cancerous lesions and 90 normal controls. Nested polymerase chain reaction and gene-chip arrays were used to identify the HPV types in the samples. In pre-cancerous lesions, there was a higher frequency of HPV of any type (14/46, OR = 2.844, CI = 1.186–6.816, P = 0.0216) and of low-risk HPV types (9/46, OR = 5.529, CI = 1.597–19.14, P = 0.0096) than in control samples. The prevalence of high-risk types was significantly higher in OSCC than in control lesions (11/51 vs 8/90, OR = 2.819, CI = 1.051–7.558, P = 0.0420) but this was not the case for HPV of any type (13/51 vs 12/90, OR = 2.244, CI = 0.9266–5.337, P= 0.1066). High-risk HPV types are prevalent in OSCC and may play a role in its progression, while low-risk types are associated with oral pre-cancerous lesions.
Key words: human papillomavirus; genotype; cytobrush; gene-chip; mouth neoplasm. Accepted for publication 14 September 2006 Available online 15 November 2006
Squamous cell carcinoma is the most com-mon malignant neoplasm of the oral mucosa, representing more than 90% of these malignant tumours. In Taiwan, oral cancer ranks as the seventh most prevalent cancer in both sexes and accounts for the fourth most common cancer in males3. Known risk factors for oral squamous cell carcinoma (OSCC) are long-term tobacco, alcohol and betel nut use, radiation,
viruses, and chronic irritants. While many individuals are exposed to these risk fac-tors, only a small proportion develops OSCC. This suggests that other factors may play a role in oral carcinogenesis.
Human papillomaviruses (HPV) are small, non-enveloped, icosahedral, epithe-liotropic DNA viruses. To date more than 70 different HPV genotypes have been cloned and characterized6. HPV infection
is associated with a wide spectrum of epithelial lesions, most of which are benign hyperplasia. A subgroup of HPVs is associated with lesions that have a propensity to undergo carcinogenesis. These are considered high-risk HPVs, including types 16, 18, 31, 33, 35, 39, 45 and 521,12. These types are known to play an important role in epithelial carci-nomas of the uterine cervix6. A high-risk 0901-5027/020153 + 06 $30.00/0 # 2006 International Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.
virus may become integrated into the host genome where it participates in oncogen-esis. After integration, two viral genes, E6 and E7, are consistently expressed. The E6 and E7 genes of HPV encode oncoproteins that target the p53 and retinoblastoma tumour suppressor genes, resulting in loss of control over the cell cycle1.
The oral cavity, like the uterine cervix, is covered by squamous epithelium with or without slight keratinization and is con-tinuously exposed to the external environ-ment. Many previous studies have demonstrated the presence of HPV in oral epithelial tissue4,5,7–25, but the relation-ship of HPV to OSCC remains unclear because of the difficulty in interpreting the variations in reported prevalence of HPV in this condition. This study was designed to investigate the prevalence and typing of HPV infection in patients with both can-cerous and pre-cancan-cerous oral mucosal lesions, and to compare these with find-ings from a group of normal control sub-jects without oral mucosal disease. Materials and methods Patients and specimens
A total of 97 patients provided specimens for this study: 51 with OSCC (48 men and 3 women, age range 33–71 years) and 46 with precancerous lesions (45 men and 1 woman, age range 20 to 74 years), includ-ing leukoplakia, oral submucous fibrosis and verrucous hyperplasia. The diagnosis in all cases was based on histological examination of haematoxylin & eosin-stained tissue sections. All patients under-went total excision of their lesions at the Department of Oral and Maxillofacial Sur-gery, Mackay Memorial Hospital, Taipei, Taiwan, during January 2002 to May 2004. In addition, 90 specimens of normal oral mucosa were obtained from 90 nor-mal subjects (52 men and 38 women,
range 15 to 75 years) seen for extraction of impacted permanent lower third molars during January 2002 to January 2004. These constituted the normal control sam-ples.
Cytological specimens were obtained from the lesions of each subject before surgery. The brush was held against the mucosa of the lesion and rotated 10 full turns. A similar procedure was performed to sample the normal mucosa of the con-trol subjects. The brush was immediately placed in HBSS solution and kept at 4 8C until further analysis. Of the 51 cases of OSCC, 10 (20%) involved the tongue, 27 (53%) the buccal mucosa, 10 (20%) the gingiva, 2 (4%) the lips and 2 (4%) the floor of the mouth. The histological types included 4 (8%) verrucous, 23 (45%) well differentiated, 20 (39%) moderately dif-ferentiated and 4 (8%) poorly differen-tiated tumours.
This study was approved by the Human Research Review Committee of Mackay Memorial Hospital, and all patients gave informed consent.
DNA extraction
The exfoliated cells from the cytology brush were vortexed for 30 s in an Eppen-dorf centrifuge tube containing 1ml of polymerase chain reaction (PBS), pelleted by centrifugation at 1500 rpm for 5 min at 20 8C, washed three times with PBS, resuspended to a final concentration of 106cell per ml in proteinase K digestion buffer (50 mM KCl, 10 mM Tris-HCl, pH 8.3, 0.5% Tween-20, 400 mg/ml of
protei-nase K), and incubated at 94 8C for 1 h. Standardized phenol-chloroform extrac-tion and ethanol precipitaextrac-tion were used for DNA purification, and the purified DNA was resuspended in sterile distilled water. To control the quality and quantity of the isolated DNA, a 300-base-pair sequence of the b-actin gene was ampli-fied by PCR as an internal control. The PCR product was electrophoresed on 2% agarose gel and stained with ethidium bromide.
HPV detection by nested PCR
Each PCR amplification reaction was car-ried out in a total volume of 50 ml contain-ing the PCR master mixture (10 mM Tris-HCl, pH 8.3; 50 mM KCl; 800 mM of each dNTP; 10 pM of each primer set; 1.25 U/L of AmpliTaq Gold DNA polymerase; Applied Biosystems, CA, USA). An opti-mal MgCl2concentration of 2.0 mM was determined to obtain specific and efficient amplification. Before amplification, the reaction mixture was treated with 1 U of uracil-N glycosylase (Roche Diagnostics, Mannheim, Germany) to prevent DNA carry-over and with 0.2 mg of TaqStartTM antibody (Clontech, Franklin Lakes, NJ, USA) to carry out a hot-start PCR.
Each PCR was carried out in the Gen-eamp PCR System 9700 (Applied Biosys-tems) with the first denaturation step at 94 8C for 3 min and the final extension step at 72 8C for 5 min. The conditions and number of denaturation–annealing–exten-sion cycles were different for each set of primers. General consensus primers
Table 1. Human papillomavirus sequence primers used for polymerase chain reaction
Primer Sequence
MY11 50GCM CAG GGW CAT AAY AAT GG
GP5+ 50TTT GTT ACT GTG GTA GAT AC
GP6+ 50GAA AAA TAA ACT GTA AAT CA
Table 2. HPV-positive cases in the OSCC group
Case no. Sex Age TNM Stage Location Differention
HPV type
Risk group
Betel nut
chewing Smoking Alcohol
1 M 50 T4N0M0 IV Tongue Verrucous 33, 52 H + + +
2 F 46 T2N0M0 II Tongue Moderate 16 H No No No
3 M 61 T4N0M0 IV Mouth floor Well 18, 53 H + + +
4 M 65 T3N0M0 III Gingiva Moderate 39 H + + +
5 M 30 T2N0M0 IV Buccal Well 16 H + + +
6 M 56 T3N1M0 III Gingiva Moderate 16 H + + +
7 M 35 T2N0M0 II Tongue Moderate 72 L + + + 8 M 49 T4N1M0 IV Gingiva Well 66 L No No + 9 M 57 T2N0M0 II Buccal Well 52 H + + + 10 M 34 T2N0M0 II Buccal Moderate 11, 18 H + + + 11 M 42 T2N0M0 II Buccal Moderate 58, 16 H + + + 12 M 41 T4N0M0 IV Lip Moderate 11, 16 H + + + 13 M 66 T4N3M0 IV Gingiva Moderate 52 H + + +
MY11/GP6+ (Table 1) were used for the first PCR round to amplify the correspond-ing part of the HPV L1 gene. A nested PCR was then carried out with the primers GP5+/GP6+ (Table 1) according to pre-viously published protocols5.
HPV genotyping by gene chip
Fifteen microliters of the amplified pro-ducts were hybridized with a revert-blot of the HPV gene chip (HPV Blot1
, King Car, Taipei, Taiwan) which detects 39 genotypes of HPV DNA (6, 11, 16, 18, 26, 31, 32, 33, 35, 37, 39, 42, 43, 44, 45, 51, 52, 53, 54, 55, 56, 58, 59, 61, 62, 66, 67, 68, 69, 70, 72, 74, 82, CP8061,
CP8304, L1AE5, MM4, MM7 and
MM8) in a single hybridization process. Briefly, the HPV chip membrane was equilibrated with 2 standard saline citrate (SCC), then prehybridized using a prehybridization buffer (2 SCC, 0.5% blocking reagent, 5% dextran sulfate and 0.1% sodium dodecyl sulfate (SDS)) containing denatured salmon sperm DNA (50 mg/L) by shaking at 35 8C for 30 min, followed by hybridization with 500 ml of hybridization buffer (2 SSC, 0.5% blocking reagent, 5% dextran sulfate, 0.1% SDS and 50 mg/L denatured salmon sperm DNA) containing denatured ampli-mers (15 ml) by shaking at 35 8C for at least 3 h. The chip was washed once in washing buffer-I (2 SSC and 0.1% SDS) for 5 min at room temperature and then twice in washing buffer-II (0.2 SSC and 0.1% SDS) for 5 min at 35 8C. Following this stringent washing, the chip was equi-librated with buffer-I (1 PBS pH 7.4, 0.05% polysorbate20, 0.1% SDS) by shak-ing twice at room temperature for 5 min. It was then incubated in 500 ml of buffer-II (1 PBS pH 7.4, 0.05% polysorbate20, 0.1% SDS, 0.5% blocking reagent) con-taining Avidex-APTM (alkaline
phospha-tase conjugated and biotinylated antibodies, 1:1000 dilution; Applied Bio-systems) for 40 min. After alkaline phos-phatase conjugation, the chip was washed in buffer-I and rinsed with buffer-III (0.1 mol/L HCl) for 5 min. Then, 70 mL of substrate nitroblue tetrazolium and 5-bromo-4-chloro-3-indolyl phosphate was added and incubated for 30 min at room temperature. The reaction was stopped by aspiration of the substrate solution and addition of distilled water.
After drying, the chips were assayed by genotype scanning software
(HPV-Easy-Scan, Taiwan Molecular Diagnostics Laboratories Ltd, Taipei, Taiwan) for the HPV types on the chip. The results were rechecked by direct vision to insure accurate calibration. As a duplicate test, each independent PCR product was hybri-dized with a single gene-chip and the results compared with those from the gene-chip array. If there was a discrepancy between the two results, PCR using HPV type-specific primers was performed to confirm the result. Specimens hybridizing with the generic probe but not with any of the type-specific probes were considered
Table 3. HPV-positive cases in the pre-malignance group
Case no. Sex Age Diagnosis Dysplasia
HPV type
Risk group
Betel nut
chewing Smoking Alcohol
1 M 33 Leukoplakia No CP8304 L + + +
2 M 58 Verrucous hyperplasia No 58 L + + +
3 M 58 Verrucous hyperplasia No 18, 31 H + + +
4 M 41 Oral submucous fibrosis No 11 L + + +
5 M 43 Acanthosis No 16 H + + +
6 M 45 Epithelium hyperplasia No 11, 16 H + + +
7 M 41 Acanthosis, epithelium hyperplasia No 16,66 H + + +
8 M 50 Squamous papilloma No MM8, CP8304 L + + +
9 M 26 Leukoplakia No 68, 53 L + + +
10 M 43 Oral submucous fibrosis No 56 L + + +
11 M 35 Squamous hyperplasia No 53 L + + +
12 M 29 Oral submucous fibrosis No 42 L + + +
13 M 41 Acanthosis, epithelium hyperplasia No 44 L + + +
14 M 52 Leukoplakia No 52 H + + +
Table 4. Correlation between HPV positivity and clinicopathologic date in OSCC group HPV positive (n = 13) HPV negative (n = 38) Statistically significant? Sex Male 12 36 Female 1 2 Location No (P = 0.1164) Buccal 4 22 Other 9 16 T status No T1 0 3 T2 6 5 T3 2 4 T4 5 26 N status No (P = 0.5061) N0 10 26 N+(N1, N2, N3) 3 15 Stage No (P = 0.1466)
Early (stage I, II) 6 8
Late (stage III, IV) 7 30
Differentiation No Verrucous 0 3 Well 5 19 Moderate 8 12 Poor 0 4 Habit Smoking 12 37 Alcohol drinking 12 37
to harbor unidentified an HPV type. This methodology, including the cytology brushing and a nested PCR-based gene-chip assay, has been shown to be reliable for determining HPV types2,5,9.
Statistical analysis
Rates of HPV positivity were compared between the three groups (OCSS, pre-can-cerous lesions and normal controls) by Fisher’s exact test. The same test was used to compare rates of high-risk and low-risk HPV types in the three groups. Odds ratios (OR) and 95% confidence intervals (CI) were calculated. Results were considered significant if the P value was less than 0.05.
Results
Amplification of the human b-actin gene in all DNA samples indicated that suffi-cient DNA was present for the amplifica-tion of HPV sequences. To rule out the possibility of contamination and PCR arte-facts, at least three negative control sam-ples containing all reaction components except template DNA were included for each round of PCR amplification. No PCR product was detected in the first or second round of amplification of the negative control samples, indicating the absence of contamination.
The data obtained using the nested PCR-based gene-chip method to deter-mine multiple HPV types in OSCCs, oral pre-cancer and control samples are shown
inTables 2 and 3. The presence of HPV
DNA in OSCC was not associated with personal habit, tumour location, clinical
staging or histological type (Table 4). It was also not associated with personal habit and level of dysplasia in the pre-cancerous group (Table 5). The rates of HPV posi-tivity, ORs and 95% CI for samples from OSCC and pre-cancerous lesions com-pared to control subjects are shown in
Tables 6–8. HPV DNA was detected more
frequently in pre-cancerous lesions (14/ 46, 30%) than in controls (12/90, 13%; P = 0.0216). The frequency was also higher in OSCC (13/51, 25%) than in control samples but did not reach a clear statistical difference (P = 0.106).
In HPV-positive OSCC, high-risk types were significantly more frequent than in control samples (P = 0.0420). Low-risk types were significantly more com-mon in pre-cancerous lesions than in OSCC (P = 0.0227) and control samples (P = 0.0096). The prevalence of high-risk types did not differ significantly between pre-cancerous and control specimens.
Discussion
The discrepancy in the detection of HPV between different studies may be due to differences in the detection methods used and the source of the tissue samples. According to a review by MILLERet al.12,13, assays considered to be of low sensitivity include immunoperoxidase, immunofluor-escence and in situ hybridization. Moderate sensitivity is provided by Southern-blot and dot-blot hybridization, while high sensitiv-ity assays use PCR.
The study presented by CHANG et al.5 which also used the similar methods to determine the prevalence of HPV in a Taiwanese OSCC population, revealed a higher prevalence than the present study. This may be due to the difference in sampling. They used a formalin-fixed, paraffin-embedded tissue block obtained from surgical procedure and pathological processing; this has more chance of con-tamination before DNA extraction then a fresh sample obtained by cytobrushing.
The reviews by MILLER et al. report detection of HPV DNA in 46.5% of OSCC, 22.2% of benign lesions and 10.0% of normal samples12,13. By comparison, the present data indicate HBV DNA in 26% of OSCC lesions, 30% of pre-cancerous lesions and 13% of normal samples. Factors contributing to differences from previously published results may include the study sample size and ethnic or regional varia-tion, as well as methodology. The use of fresh specimens obtained by cytobrushing in a nested PCR-based gene-chip assay is non-invasive and easy to perform, has high sensitivity, and appears to be a reliable method for investigating the prevalence of HPV in oral mucosal lesions.
While the overall prevalence of HPV DNA in OSCC samples was not statisti-cally higher than in control samples, there was a significantly higher
preva-Table 5. Correlation between HPV positivity and clinicopathologic data in pre-malignant group HPV positive (n = 14) HPV negative (n = 32) Statistically significant? Sex Male 14 31 Female 0 1 Dysplasia No (P = 0.09) Positive 0 2 Negative 14 30 Personal habits Smoking 14 31 Alcohol drinking 14 30
Betel nut chewing 14 32
Table 6. Prevalence of HPV in OSCC and control samples
HPV type OSCC (n = 51) Controls (n = 90) OR 95% CI P value
All 13 (25.49%) 12 (13.33%) 2.224 0.9266–5.337 0.1066
High risk 11 (21.57%) 8 (8.89%) 2.819 1.051–7.558 0.0420*
Low risk 2 (3.92%) 4 (4.44%) 0.8776 0.1550–4.968 1.000
*
Statistically significant.
Table 7. Prevalence of HPV in oral pre-cancerous lesions and control samples HPV type Pre-cancerous lesions (n = 46) Controls (n = 90) OR 95% CI P value All 14 (30.43%) 12 (13.33%) 2.844 1.186–6.816 0.0216* High risk 5 (10.87%) 8 (8.89%) 1.25 0.3845–4.063 0.7618 Low risk 9 (19.57%) 4 (4.44%) 5.529 1.597–19.14 0.0096* * Statistically significant.
Table 8. Prevalence of HPV in OSCC and oral pre-cancerous lesions HPV type OSCC (n = 51) Pre-cancerous lesions (n = 46) OR 95% CI P value All 13 (25.49%) 14 (30.43%) 0.7820 0.3212–1.903 0.6534 High risk 11 (21.57%) 5 (10.87%) 2.255 0.7186–7.077 0.1808 Low risk 2 (3.92%) 9 (19.57%) 0.1678 0.03419–0.82369 0.0227* * Statistically significant.
lence of high-risk HPV types in OSCC. The study by CHANG et al.5 reported a similar result especially in the non-oral habit-associated patients, although not including the pre-cancerous group. The prevalence of HPV, especially high-risk types, in the overall OSCC group was higher but in the oral habit-associated group was similar to the control samples, as in the present study. In contrast, the pre-cancerous benign lesions had a higher prevalence of all HPV types than did controls. But of even more interest is the significantly higher prevalence in benign lesions of low-risk HPV types than in either control or OSCC samples. These findings of this and the previous study5support the hypothesis that speci-fic HPV genotypes may be important in malignant transformation. High-risk HPV genotypes infecting the oral epithe-lium appear to have the potential to pro-pagate new malignancies. The low-risk genotypes may in fact retard the malig-nant transformation of pre-cancerous lesions, even though the environmental risk factors and personal habits such as smoking, alcohol drinking and betel nut chewing are similar for both malignant and benign lesions.
As with most cancers, OSCC is believed to develop by a multi-step process of cumulative gene damage in oral epithelial cells23. Previous studies with in situ hybri-dization have demonstrated that HPV DNA localizes to the nuclei in the spinous and basal cell layer of the epithelium, with HPV-positive cells located in the center of the tumour nest14,15,18,23. As noted above, HPV-DNA integration into the human genome may be associated with the expression of oncogenes and the down-regulation of tumour suppressor genes in the early stage of OSCC carcinogenesis1. The specific type of infecting HPV may influence malignant progression in the host cell, with high-risk types accelerating the progression and low-risk types retard-ing it. This would account for the type prevalences in this study. The type of HPV involved may be an important prognostic factor. SCHWARTZ et al.20 and GILLISON et al.8have suggested that HPV DNA is a strong, independent predictor of out-come in oral cancer, even in advanced disease. In conclusion, the high-risk HPV types may play a role in OSCC, while low-risk types are more likely to be associated with pre-cancerous lesions. Further investigation at the molecular level of the mechanisms involved may eventually lead to clinically useful meth-ods for diagnosis, prognosis and chemo-prevention.
Acknowledgements.The authors thank Dr Rich YC Wu (Taiwan Molecular Diagnos-tics Laboratory Co. Ltd) for excellent technical assistance and Dr MJ Buttrey for her kind review and instructive criti-cism of this paper. This study was sup-ported by Foundation of Alumni Association Taipei Medical University. References
1. Alani RM, Munger K. Human papillo-maviruses and associated malignancies. J Clin Oncol 1998: 16: 330–337. 2. Ammatuna P, Campisi G, Giovannelli
L, Giambelluca D, Alaimo C, Man-cuso S, Margiotta V. Presence of Epstein-Barr virus, cytomegalovirus and human papillomavirus in normal oral mucosa of HIV-infected and renal trans-plant patients. Oral Dis 2001: 7: 34–40. 3. Cancer Registry Annual Report in
Tai-wan Area, 1999, Taipei, TaiTai-wan: Depart-ment of Health, The Executive Yuan; 2003.
4. Chang KW, Chang CS, Lai KS, Chou MJ, Choo KB. High prevalence of human papillomavirus infection and possible association with betel quid chewing and smoking in oral epidermoid carcinomas in Taiwan. J Med Virol 1989: 28: 57–61. 5. Chang JYF, Lin MC, Chiang CP. High-risk human papillomaviruses may have an important role in non-oral habits-asso-ciated oral squamous cell carcinomas in Taiwan. Am J Clin Pathol 2003: 120: 909–916.
6. Cooper K. The role of human papillo-mavirus in cervical cancer. S Afr Med J 1995: 85: 841–842.
7. Elamin F, Steingrimsdottir H, Wana-kulasuriyaS, Johnson N, Tavassoli M. Prevalence of human papillomavirus infection in premalignant and malignant lesions of the oral cavity in U.K. subjects: a novel method of detection. Oral Oncol 1998: 34: 191–197.
8. Gillison ML, Koch WM, Capone RB, Spafford M, Westra WH, Wu L, Zahark ML, Daniel RW, Viglione M, Symer DE, Shah KV, Sidransky D. Evidence for a causal association between human papillomavirus and a subset of head and neck cancers. J Nati Cancer Inst 2000: 92: 709–720. 9. Giovannelli L, Campisi G, Lama A,
Giambalvo O, Osborn J, Margiotta V, Ammatuna P. Human papillomavirus DNA in oral mucosal lesions. J Infect Dis 2002: 185: 833–836.
10. Mao EJ, Schwartz SM, Daling JR, Oda D, Tickman L, Beckmann AM. Human papilloma viruses and p53 muta-tions in normal, pre-malignant and malig-nant oral epithelia. Int J Cancer 1996: 69: 152–158.
11. Mao EJ, Wash S. Prevalence of human papillomavirus 16 and nucleolar organi-zer region counts in oral exfoliated cells from normal and malignant epithelia.
Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1995: 80: 320–329. 12. Miller CS, Johnstone BM. Human
papillomavirus as a risk factor for oral squamous cell carcinoma: a meta-analy-sis, 1982–1997. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2001: 91: 622– 635.
13. Miller CS, White DK. Human papillo-mavirus expression in oral mucosa, pre-malignant conditions, and squamous cell carcinoma: a retrospective review of the literature. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1996: 82: 57–68.
14. Niv A, Sion-Vardi N, Gatot A, Nash M, Fliss DM. Identification and typing of human papillomavirus (HPV) in squa-mous cell carcinoma of the oral cavity and oropharynx. J Laryngol Otol 2000: 114: 41–46.
15. Pillal MR, Phanidhara A, Kesari AL, Nair P, Nair MK. Cellular manifesta-tions of human papillomavirus infection in the oral mucosa. J Surg Oncol 1999: 71: 10–15.
16. Pintos J, Franco EL, Black MJ, Ber-geronJ, Arella M. Human papilloma-virus and prognoses of patients with cancers of the upper aerodigestive tract. Cancer 1999: 85: 1903–1909.
17. Premoll-De-Percoco G, Ramirez JL. High risk human papillomavirus in oral squamous carcinoma: Evidence of risk factors in a Venezuelan rural population: preliminary report. J Oral Pathol Med 2001: 30: 355–361.
18. Premoli-De-Percoco G, Ramirez JL, Galindo I. Correlation between HPV types associated with oral squamous cell carcinoma and cervicovaginal cytology: an in situ hybridization study. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1998: 86: 77–81.
19. Rensburg EJV, Heerden WFPV, Ven-terEH, Raubenheimer EJ. Detection of human papillomavirus DNA with in situ hybridisation in oral squamous carcinoma in a rural black population. S Afr Med J 1995: 85: 894–896.
20. Schwartz SR, Yueh B, Mcdougall JK, Daling JR, Schwartz SM. Human papillomavirus infection and survival in oral squamous cell cancer: a population-based study. Otolaryngol Head Neck Surg 2001: 125: 1–9.
21. Smith EM, Hoffman HT, Summersgill KS, Kirchner HL, Turek LP, Haugen TH. Human papillomavirus and risk of oral cancer. Laryngoscope 1998: 108: 1098–1103.
22. Summersgill KF, Smith EM, Kirchner HL, Haugen TH, Turek LP. p53 poly-morphism, human papillomavirus infec-tion in the oral cavity, and oral cancer. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2000: 90: 334–339. 23. Uobe K, Masuno K, Fang YR, Li LJ,
WenYM, Ueda Y, Tanak A. Detection of HPV in Japanese and Chinese oral
carcinomas by in situ PCR. Oral Oncol 2001: 37: 146–152.
24. Ward KA, Napier SS, Winter PC, Maw RD, Dinsmore WW. Detection of human papilloma virus DNA sequences in oral squamous cell papillo-mas by the polymerase chain reaction. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1995: 80: 63–66.
25. Wong DTW, Munger K. Association of human papillomaviruses with a subgroup of head and neck squamous cell carcino-mas. J Natl Cancer Inst 2000: 92: 675–677. Address:
Chung-Ji Liu
Department of Oral and Maxillofacial Surgery
Mackay Memorial Hospital No. 92
Sec. 2
Chung-Shan North Road Taipei 104
Taiwan R.O.C
Tel: +886 2 2543 3535x2216 E-mail:[email protected]
