Cytogenetic and micronuclei study of human papillomavirus-related oral squamous cell carcinoma.

BACKGROUND: The association of human papilloma viral infection in oral squamous cell carcinoma is well studied in the Western countries, but its correlation with DNA damage in the form of micronuclei (MN) formation, ceased apoptosis or presence of chromosomal abnormalities has not yet been studied. AIM: The aim of this study is to find any possible correlation between human papillomavirus (HPV) infection and cytogenetic damage in the oral malignant and premalignant population of West Bengal. SETTINGS AND
DESIGN: A total of 104 malignant and 103 premalignant cases were selected along with 200 controls.
METHODS: The buccal smear samples were Pap stained for the detection of MN, apoptosis frequency and koilocytes. The buccal swab samples were processed for DNA extraction followed by polymerase chain reaction for the detection of HPV DNA. The peripheral venous blood samples were processed for the detection of any chromosomal abnormality, by the method of human leukocyte culture followed by Giemsa staining. STATISTICAL ANALYSIS USED: Correlation analysis using GraphPad Prism software was used in this study.
RESULTS: About 34.6%, 42.3% and 6.73% of malignant and 6.79%, 3.88% and 20.38% of premalignant cases showed the presence of HPV DNA, koilocytes and apoptosis, respectively. The difference between the MN frequencies of premalignant and malignant oral lesions with the control group is significant with respect to various risk factors (P < 0.05). One percentage of malignant cases showed the presence of chromosomal break.
CONCLUSION: A considerable percentage of malignant cases showing the presence of koilocytes and viral DNA may indicate the effect of HPV infection leading to the malignancy, which can be correlated with the MN and apoptosis frequency.

INTRODUCTION

Oral squamous cell carcinoma (OSCC) is the 6th most common cancers worldwide[12] and 3rd most common cancers in developing world.[3] It is the most common cancer in Southeast Asia (India, Bangladesh, Pakistan and Sri Lanka). Its association with different risk factors such as various addictive habits (intake of betel quid, tobacco, alcohol, etc.), poor oral hygiene and bad oral habits has been already established worldwide. However, in recent years, many research studies suggest that there have been many cases developing this cancer without any history of habits related to these risk factors, which bring out the rise of another potent factor, namely human papillomavirus (HPV) infection.[4] Moreover, the connection of different cancers with cytogenetics or DNA damage is well studied all over the world. The preclinical genomic abnormalities include various parameters such as the presence of micronuclei (MN), apoptosis and the presence of numerical or structural chromosomal anomalies.[5] MN frequency is reportedly high in cases of malignancy than that of premalignancy,[6] apoptosis frequency is conventionally suggested to be low in cases of malignancy, and chromosomal abnormalities are variable.

METHODS

In this case–control prospective study, a stratified sampling method was used to select 407 participants (104 – oral malignant, 103 – oral premalignant and 200 – control) to be interviewed after being informed about the research. Ethical clearance for this study was obtained from the Ethics Committee of Ramakrishna Mission Seva Pratishthan and Vivekananda Institute of Medical Sciences, Kolkata. Nearly 24,550 patients were screened in the Department of ENT, Head and Neck Surgery and Oral and Maxillofacial Surgery of our hospital. Among these, 1070 came out with suspected premalignant and malignant oral lesions. Out of these, 104 patients with histopathologically confirmed cases of oral carcinoma and 103 with premalignant oral lesions and conditions were recruited for this study between June 2013 and October 2017. All cases were newly diagnosed and previously untreated. Clinical characteristics including basic medical data were obtained from medical records. All were resident of different districts of West Bengal. Two hundred controls (cancer free) were recruited simultaneously from the relative of the patients residing in the similar geographic area. Controls were selected among the relatives of the cases who accompanied them and staying in the same localities. Age distribution for the controls was comparable to that of the cases. Cases and controls were matched primarily by frequency of geographic and social origin and secondly by age distribution. They mostly belong to medium-to-low economic classes having similar lifestyle and level of education.

After signing the informed consents, participants were interviewed to collect their demographic data (age, gender and residential history), their daily lifestyle and occupation. Peripheral venous blood, buccal smear and swab samples were collected after obtaining informed consent from all the participants. About 104 malignant, 103 premalignant and 200 control peripheral venous blood samples were processed for detection of the presence of chromosomal abnormalities, by the method of human leukocyte culture followed by Giemsa staining. The buccal smears were taken on slides, Pap stained and examined under microscope for detecting the presence of koilocytes. The corresponding buccal swab samples were dissolved in phosphate-buffered saline solution and DNA was extracted from all the sample solutions following the standard Qiagen protocol, using the Qiagen DNA Mini Kit. Further, polymerase chain reaction (PCR) was performed with all the DNA samples, positive control (HPV 16 plasmid DNA) using the HPV L1 consensus primers [MY11/MY09].

Ethics

Ethical clearance for this study was obtained from the Ethics Committee of Ramakrishna Mission Seva Pratishthan and Vivekananda Institute of Medical Sciences. The study has been independently reviewed and approved by the ethical board as well.

Statistical analysis

Descriptive analysis was conducted comparing cases with malignant and premalignant oral lesions to the control group in terms of the presence of MN in buccal smear samples. Statistical analysis was done using GraphPad Prism software. All tests were two-sided with a significant level of P < 0.05.

RESULTS

The age–sex distribution and the various risk factor associations are depicted in Table 1. The detailed number of patients and control individuals having/not having various addictions observed in this study is presented in Figure 1.

Table 1

The age and sex distribution and various addictions of malignant, premalignant and control individuals are presented here

Figure 1

The histogram shows the association of various risk factors (including nonuser) and presence of human papillomavirus DNA in malignant and premalignant groups and control individuals

Almost 42.3% (44/104) of malignant and 3.88% (4/103) of premalignant cases showed the presence of koilocytes [Figure 2], whereas none of controls showed koilocytes, when observed under the microscope. Nearly 36.4% (36/104) of malignant and 6.79% (7/103) of premalignant cases showed the presence of HPV DNA, whereas none of the control samples showed the presence of the viral DNA. There were few malignant cases (10/104 ~ 9.6%) which showed the presence of the viral DNA, in spite of not showing the presence of koilocytes. Twelve malignant nonusers showed the presence of viral DNA. The presence and absence of the viral DNA in malignant and premalignant cases and controls are depicted in Figure 3, respectively.

Figure 2

Micrograph shows human buccal cells stained by the method of Pap staining and observed at ×100, indicating a koilocyte

Figure 3

Agarose gel electrophoresis showing the presence or absence of human papillomavirus DNA in malignant, premalignant and control samples (lane 3–7), +ve control (lane 2), −ve control (lane 8)

About 6.73% (7/104) of malignant and 20.38% (21/103) of premalignant cases showed the presence of apoptosis. The range of MN frequency of different cases and controls is depicted in Table 2. The difference between the MN frequencies of premalignant and malignant oral lesions with the control group is significant with respect to various risk factors (P < 0.05), as shown in Table 3. The micrographs indicating MN and apoptosis are shown in Figures 4 and 5, respectively. Only one malignant case (~1%) showed the presence of chromosomal abnormality (break). Two metaphase plates one showing normal set of chromosomes and another showing chromosomal break are depicted in Figure 6.

Table 2

The range of micronuclei frequency among malignant, premalignant and control groups is presented here

Table 3

Significant difference of the micronuclei frequency observed in the human buccal cells (indicating the P values) between the cases and controls

Figure 4

Micrograph shows human buccal cells stained by the method of Pap staining and observed at ×100. The solid black arrow indicates micronuclei in the cell

Figure 5

Micrograph shows human buccal cells stained by the method of Pap staining and observed at ×100. The solid black arrow indicates an apoptotic cell

Figure 6

Micrographs show metaphase plates; one showing normal set of human chromosomes and the other showing the presence of chromosomal break

DISCUSSION

OSCC ranks the 15th most common cancer in males and the 11th most common cancer in females.[7] Its association with different addictions as risk factors is well established worldwide. In developed countries, the use of smoking tobacco is very common, while in developing countries, the chewing practices like intake of paan masala, guthka, zarda, khaini and snuff are more in use,[89] along with the use of smoking tobacco such as bidi and cigarette to some extent. However, recent studies stated the prevalence of another risk factor, observed in cancer cases with the history of no addictions or other risk factors. This has been attributed to the human papillomavirus (HPV) in various Western countries,[10] which has resulted in the increase of oropharyngeal carcinoma if compared to oral carcinoma, also among the young agers. The viral prevalence is being tested throughout years with the help of cytological methods for the presence of koilocytes on a primary level, thereby confirming its incidence through molecular methods such as PCR. There is a wide variation of HPV incidence in India, indicating the role of HPV infection in need to get verified more in this developing country. A study from Northern India indicates a low prevalence of HPV and oral carcinoma,[11] which is in contradiction with another study in Southern India, indicating a high prevalence.[12] A study in southern India reported the prevalence being 40.4%,[13] while another study from southern India shows HPV prevalence of 80%–90%.[14] A study from West Bengal indicates HPV positivity in head-and-neck squamous cell carcinoma tumors being 69%.[15] However, another study from Southern India has stated no role of HPV in oral carcinogenesis.[16] This is also contradicted by another study of West Bengal.[17] In India, HPV DNA has been detected less frequently in tumor specimens from individuals having habits of predisposing risk factors than the nonusers.[18] This is also consistent with another in Kerala which says that negative history of tobacco usage has shown a trend toward HPV positivity in OSCC patients.[19] This study also suggests that the combined effect of oral tobacco and HPV is also a potent factor in oral carcinogenesis, especially in case of infection by high-risk strains. Another study also reported high incidence of HPV infection in tobacco chewers.[20] Numerous studies have suggested additive effect of tobacco and alcohol consumption and HPV.[2122] However, a study from North India suggests an inclination toward tobacco habit in the oral carcinoma.[11] While HPV-16 and HPV-18 have been found as a significant risk factor for oral cancerous lesions in Western studies, in India, this role needs to be verified more, especially in case of premalignant oral lesions such as leukoplakia.[23]

Like all other carcinoma, oral cancer also results from various mutations and chromosomal abnormalities in genes controlling the cell cycle and DNA repair.[24] In addition to the potential for metastasis, this can also be characterized by the loss of the ability of cells to evolve to death when genetic damage occurs. The induction by carcinogens of micronucleated cells, both in vivo and in vitro, is a sign of the genotoxic effect of such substances. Hence, various cytogenetic tests including the micronucleus test in addition to tests of various degenerative alterations indicative of apoptosis (karyorrhexis, pyknosis and condensed chromatin), which examine the presence of MN, apoptosis and any chromosomal anomaly under microscope, give a promising method to detect the occurrences of cytogenetic alterations in the oral epithelium due to this carcinoma.[4] However, many phytochemicals including essential plant oils have shown their benefits in various cancer cell lines.[25] The authors observed MN frequencies in exfoliated buccal cells in normal mucosa, precancerous lesions and squamous cell carcinoma of Indian participants and concluded that the MN frequency may be a marker of epithelial carcinogenic progression.[26] Here, we have focused on the MN formation, apoptosis frequency and detection of chromosomal abnormalities and observed in various cases of premalignant and malignant stages of OSCC and their association with various risk factors in participants of West Bengal, which will help us to analyze the extent of carcinogenic invasion and compare these with the control groups. The prevalence of HPV in oral cancer has been well documented in several studies from different geographical regions of the world. This incidence has been studied earlier in various Western countries stating the infection of HPV as an independent risk factor leading to this malignancy in participants not having any kind of history of patent risk factors such as intake of tobacco (chewed or smoked), alcohol and betel quid.[10] The microscopic indication of koilocytes can serve as a valuable biomarker for the detection of HPV incidence,[27] further leading to this malignancy in the cases obtained from West Bengal. Hence, in order to find the effect of HPV infection in this zone, we have carried out this study which indicates that although our numbers are small, it would appear that a considerable percentage of the presence of koilocytes states the prevalence of HPV working as a factor in this carcinogenesis.

CONCLUSION

In this study, the significant difference of MN frequency between the cases and control groups with respect to various risk factors indicates the already established fact of using this parameter as an important biomarker in estimating the cancer progression. The usual pattern of reduced apoptosis frequency in premalignant and malignant cases in comparison to the controls also supports the previous studies in this field. Moreover, a considerable percentage of malignant cases showing the presence of HPV DNA indicate a possible correlation between the HPV infection and the development of oral malignancy in this population too, as already established in the Western countries. However, its independent role is not yet depicted since association with other risk factors are taken into consideration, and a very small percentage of nonusers have shown the malignancy with an impact of the viral infection. Hence, a higher sample size is definitive for bringing out this role. Furthermore, a very small percentage of malignant cases showing the presence of chromosomal abnormality (chromosomal break) cannot exert the fact of this cytogenetic damage created as a result of this viral infection. This observation cannot imply a possible relation between these factors, rather just implementing the creation of DNA damage as a result of any other risk factor in association with the HPV infection, since this malignant case also showed the presence of viral DNA. This correlational study also focuses on the possible link between the viral infection and the caused DNA damage in the form of MN, ceased apoptosis, chromosomal damage, etc., A considerable number of malignant cases showing both the factors into play may indicate a possible impact of the viral infection on the DNA of the affected patients as an outcome of MN formation and further cancer progression with the usual trend of stoppage of the programmed cell death or may be the created environment of the viral factor playing a positive role in the interplay of different risk factors finally leading to the DNA damage and malignancy. A very small percentage of malignant cases showing the presence of chromosomal abnormality in the form of chromosomal break cannot confirm the correlational link between this viral infection and this cytogenetic damage, yet assessing the damage created by any other risk factor in association with HPV infection, since this malignant case also showed the presence of the viral DNA. To bring out such correlation, a much higher sample size with extensive statistical analysis is required.

Financial support and sponsorship

This study was financially supported by DST Inspire Fellowship, New Delhi.

Conflicts of interest

There are no conflicts of interest.