A multi-centre evaluation of oral cancer in Southern and Western Nigeria: an African oral pathology research consortium initiative

Introduction Oral cancer is a leading cause of cancer deaths among African populations. Lack of standard cancer registries and under-reporting has inaccurately depicted its magnitude in Nigeria. Development of multi-centre collaborative oral pathology networks such as the African Oral Pathology Research Consortium (AOPRC) facilitates skill and expertise exchange and fosters a robust and systematic investigation of oral diseases across Africa. Methods In this descriptive cross-sectional study, we have leveraged the auspices of the AOPRC to examine the burden of oral cancer in Nigeria, using a multi-centre approach. Data from 4 major tertiary health institutions in Western and Southern Nigeria was generated using a standardized data extraction format and analysed using the SPSS data analysis software (version 20.0; SPSS Inc. Chicago, IL). Results Of the 162 cases examined across the 4 centres, we observed that oral squamous cell carcinomas (OSCC) occurred mostly in the 6th and 7th decades of life and maxillary were more frequent than mandibular OSCC lesions. Regional variations were observed both for location, age group and gender distribution. Significant regional differences was found between poorly, moderately and well differentiated OSCC (p value = 0.0071). Conclusion A multi-centre collaborative oral pathology research approach is an effective way to achieve better insight into the patterns and distribution of various oral diseases in men of African descent. The wider outlook for AOPRC is to employ similar approaches to drive intensive oral pathology research targeted at addressing the current morbidity and mortality of various oral diseases across Africa.


Introduction
Oral cancer is the sixth commonest cancer globally and the most common head and neck cancer [1]. Oral squamous cell carcinoma (OSCC) are cancers originating from the squamous epithelium in the oral cavity, accounting for more than 90% of all tumors in the head and neck region [2][3][4]. According to the World Health Organization (WHO)'s International Classification of Diseases (ICD-10), "Oral cancer" (C00, C02-C06) may be defined as any malignant neoplasm occurring on the lips (both vermillion border and oral aspect) and within the oral cavity (which includes the anterior 2/3 of the tongue, buccal and labial mucosa, gingiva, hard palate, retromolar pad and floor of the mouth) [5]. There is a marked variation with regard to the incidence of OSCC between different countries, geographic locations and ethnic/racial groups [6]. This may be attributed to exposure to different environmental factors and to ethnic-specific high-risk habits [6]. The global average incidence from the WHO/IARC global cancer statistics database (GLOBOCAN 2012) shows "oral cancer" to be the sixteenth commonest cancers type, accounting for 300,200 new cases, which consisted of 198,900 and 101,300 new cases among males and females, respectively [7].
According to this database, oral cancer related deaths has been estimated to be about 145,353 cases with about 97,900 cases for males and 47,400 cases for females in 2012; making it the twelfth and sixteenth most common cause of cancer mortality amongst males and females, respectively. In Nigeria, oral cancer has been estimated to account for about 1146 new cases, with an estimated mortality of 764 cases annually in 2012 [7].
The aetiology of oral cancer although unknown, has been historically associated with the use of tobacco products, alcohol consumption, infections with Human Papilloma virus (HPV) and environmental carcinogens. In developed regions 75% of oral cancer cases may be linked to the use of tobacco and alcohol consumption by patients [8]. However, in developing countries, other risk factor such as use of betel quid, consumption of nitrosamine rich foods (such as salted fish), infections, nutritional deficiencies and exposure to environmental carcinogens may account for the differences in epidemiology of oral cancers in different regions [9,10]. The epidemiology of Lip, and oral cavity cancers also vary across regions with India, USA and China accounting for the highest prevalence and incidence rates. In India, incidence of oral cancer has been reported to be about 11% of all cancers in 2012, while prevalence has been reported to be about 12%, making oral cancer the commonest cancers in males and the seventh in females in India [11]. However, due to underreporting, it is unclear if the burden of oral cancer is actually low in Nigeria or the low figures is as a result of poorly documented cases and lack of well-funded, state-of-the-art, population-based cancer registries across the country [12]. Most cancer registries in Nigeria are hospital-based and lack adequate coverage and reliability. Thus, there is a critical need to have a better understanding of the real distribution of oral cancer as they constitute a serious public health concern globally [11]. The African Oral Pathology Research Consortium (AOPRC) was inaugurated during the 1 st regional congress of the International Association of Oral Pathologists

Results
Age distribution: The mean age of all patients from all the study centres was 56.2 ± 16.4 years. The highest mean age was found in Port Harcourt having a mean age of 65.4 ± 11.9 years, with an age range between 49 to 88 years. While the lowest mean age was observed in Ibadan having a mean age of 54.4 ± 10.5, with age range between 45 to 75 years (Table 1).
Gender distribution: In this study, there were 92 males (57%) and 70 females (43%) with a male to female ratio of 1.3:1 ( Figure   1). The male to female ratio were 1.5:1 in Benin, 1:1 in Ibadan, 1:0.9 in Ife and 1.7:1 in Port Harcourt, however the sex distribution of patients with SCC was not significant (p = 0.5023) ( Table 2).

OSCC distribution according to age group and location:
The association of SCC with the age groups and location was found to be significant (p value = 0.0293 the third decade (n = 5, 3.3%) with all the 9 cases reported in Benin (Table 3).

Regional histopathological subtype distribution of OSCC:
The histopathologic sub-types seen in all the centres were the well found to be significant (p = 0.0071) ( Table 4).

Discussion
It has been established in the scientific literature that OSCC is more prevalent in developing than developed countries [6,13]. For instance, In Israel, oral SCC is more prevalent among Ashkenazi Jews than Sephardic-Jews for reason attributed to the differences in geographic origins [14]. Similarly in England, it is more prevalent among Indian people born in the Indian subcontinent and migrated to England than among Indians born in England [15]. It has further been documented in US, that the average 5-year survival rate for men of African descent with OSCC is lower than those of their Caucasian counterparts [6,16]; and OSCC is usually at a significantly more advanced stage in blacks than in white people at the time of diagnosis [16]. The possible suggestions accrued to these geographical, ethnic and racial differences include; that the lesion is pathobiologically more aggressive in blacks than in whites; and that blacks delay longer before seeking medical advice than do whites for cultural, educational and socioeconomic reasons. As with many other types of cancer, OSCC most commonly occurs in the middle aged and elderly population [17,18]. This is in agreement with our findings as majority (69 cases, 42.59%) of the cases were seen within the age range of 60 years and above. This is followed by 37 cases (22.84%) observed between 50-60 years of age. Oral squamous cell carcinoma has long been considered to be a tumor of the elderly and has been seen only sporadically before the third decade of life. In our study, four cases (2.47%) were seen within the age bracket of less than 18 years. Male population have traditionally been known to have a higher incidence in OSCC, typically circa 2:1 compared to women [19,20]. This trend has however evened out, probably due to increased alcohol consumption and tobacco use among the female population [18].
This development agrees with our findings of male to female ratio of 1.3:1. In addition a male to female ratio of 1.4: 1 has also been reported among Iranian populations [21]. In many studies (including those conducted in Nigeria), OSCC affected more males than females [18,[22][23][24][25]. One exception is a study carried out in Ilorin (North Central Nigeria), where females were affected more than males [26]. We also observed in our study that the out of the 46 cases from Ife, 22 of them are females gender with a very close male to female ratio of 1:0.9. We observed that oral squamous cell carcinomas (OSCC) occurred mostly in the 6 th and 7 th decades of life and maxillary were more frequent than mandibular OSCC lesions.
There were more male (n = 92) than female (n = 70) OSCC cases; and regional variations were observed both for location, age group and gender distribution. Our observation is consistent with what others have reported in the literature [27,28]. For instance Dourmishev et al (1997)  Therefore a cohort of an African population with OSCC using genetic and genomic studies to identify risk loci will provide insights into founder mutations since Africa is the ancestral population. The increase in frequency of OSCC with age is likely due to increase in DNA damage with age [29,30]. This study showed that the most common histological subtype was the well differentiated, followed by the moderately differentiated in all the centres, except Ibadan where well differentiated subtype was the least common. This observation is in agreement with the finding of other researchers who have reported preponderance of well differentiated SCC in their studies [31,32]. However, Effiom et al (2008) in a study from two centres in Lagos, Nigeria observed that the most predominant subtype was the poorly differentiated and the moderately differentiated type was the least [25]. They also noted that an earlier study from one of the centres reported that that well differentiated subtype was the commonest. The reason for the intra and inter centres variations reported by various researchers is not clear and may require further study. Contrary to several other reports, the sites that were commonly affected in this study were the palate and maxilla followed by the mandible. The mandible and the tongue have been reported as the most favoured site for oral squamous cell carcinoma [25,[31][32][33], because carcinogens that dissolved in saliva readily settle down to these sites due to gravity.
Palate is not a usual site for this lesion, in fact, in the study by Udeabor et al (2014), palate was a distant third most favoured site after mandible and maxilla [33]. The pooling of data from different centres and the inclusion of some additonal oro-facial sites which are usually excluded in other studies may have influence the pattern seen in the present study. Thus, there is need for more elaborate study involving many centres in order to get the true picture of this lesion in our environment.

Conclusion
The true burden of OSCC as a public health problem among men of African descent cannot be ascertained by merely evaluating public databases. This is particularly true in Nigeria where most data is generated from defunct hospital-based registries rather than population-based repositories. Changing lifestyle patterns due to emergence of many economies in Africa must be factored into the true picture of OSCC burden in Africa. In addition, the aging population has also increased due to better health education and healthcare programs. However, the journey is still long without the ability to leverage partnership with colleagues across the globe, both for diagnostic and research services. A well-coordinated and vibrant cancer control program as it exists in many developed economies should be adopted in Nigeria and many other African countries, to improve prompt diagnosis and management of OSCC.
The trends in OSCC identified in this study demonstrated that further multi-centre collaborative research is required to gain better insight into the epidemiology of the disease.

What is known about this topic
• Oral cancer is a leading cause of cancer deaths in Africa; • The real burden of oral cancer is poorly captured in Nigeria due to under-reporting and lack of population based registries; • Collaborative team science in needed among oral pathologist to elucidate the burden of oral cancer in Africa, and advance better diagnostic and treatment modalities.

What this study adds
• This study presents a multi-centre approach to oral cancer epidemiology in Nigeria; • We have examine regional variations in oral cancer distribution across the four centres evaluated; • The multicentre approach to oral cancer study presented herein by AOPRC, would serve as a primer for further collaborative oral pathology research across Africa and globally.

Competing interests
The authors declare no conflict of interest.