The need for a central brain tumor registry in Africa: A review of central nervous system tumors in Africa from 1960 to 2017

Abstract

Background

Central nervous system (CNS) tumors pose a substantial health problem. Although data on specific time periods and regions of Africa have been previously reported, no study has yet to provide a systematic review of CNS tumors for the entire continent of Africa. This study aims to analyze the frequency of CNS tumors in Africa from 1960 to 2017.

Methods

A comprehensive literature search on CNS tumors in Africa was performed using multiple online scientific databases. The following keywords were queried in combination with the phrase “CNS tumors in Africa”: incidence, frequency, epidemiology, prevalence, brain, and cancer. A total of 26 articles met the inclusion criteria. Each selected article reported incidence and mortality rates from different regions of Africa in a time period between 1960 and 2017. SPSS21 statistical software was used to analyze the data.

Results

Nigeria, Egypt, and Uganda were found to have the most of the cases of CNS tumors in Africa. Males made up 54% of the 5902 cases per 100 000 population. The most common CNS tumors found were astrocytoma (24.70%), meningioma (22.22%), pituitary adenoma (8.4%), medulloblastoma (4.26%), craniopharyngioma (4.07%), and other not specified (25.17%).

Conclusions

Given the large population of Africa, the total reported cases may be underestimated when compared with other continents due to the lack of a central brain tumor registry in Africa. A comprehensive knowledge of CNS tumors in Africa is critical to population-based research and improving the current healthcare system.

Central nervous system (CNS) tumors are a group of both malignant and nonmalignant tumors of the brain and spinal cord. CNS tumors affect both adults and children and are diagnosed in all anatomical regions of the CNS resulting in a substantial amount of morbidity and mortality worldwide.¹ According to the data retrieved from the vital registration and cancer registry, the total global cases and deaths of malignant CNS tumors between 1990 and 2016 were 330 000 and 227 000, respectively.¹ The regions that these cases occurred are East Asia, South Asia, Western Europe, and America.¹ No incidence and mortality of CNS tumors from any region in Africa were included in the above-mentioned cancer registry. Furthermore, Miranda-Filho et al.² reported data from the African country of Uganda in their study on the global incidence of brain and CNS tumors; however, no other country in Africa was reported. The lower comparative rate of CNS tumors is likely an underestimate due to the limited diagnostic health infrastructure in low-income countries.³ Nonetheless, literature has proven that CNS tumors have occurred in multiple regions throughout Africa.

In Africa, the study of the incidence and mortality of all primary CNS tumors has been limited to periodic reporting in separate studies from different countries, such as Nigeria, Uganda, and Ghana.^4–6 For example, Sahabi et al.⁵ conducted a retrospective study of intracranial tumors from 1991 to 2007 using data from the Cancer Registry of Ibadan, Nigeria. Their results showed 1394 brain tumors within that time frame including 76 tumors, which were mortalities. Males exceeded females in a 1.3:1 ratio. This study was reported as the largest reported case series of CNS tumors from Ibadan, Nigeria, at that time.⁵

Additionally, Ogun et al.⁴ conducted a retrospective histopathological study of pediatric CNS tumors from 2001 to 2010 in Nigeria, which showed that 77 children between the ages of 0 and 15 were diagnosed with CNS tumors. The authors asserted that contrary to popular belief that CNS tumors are not common among black children, they are actually the fourth most common tumor in the pediatric population in Ibadan, Nigeria.⁴ Also, in Uganda, Hatef et al.⁷ used admission records, clinical registries, and surgery registries to conduct a retrospective study of CNS tumors between 2009 and 2012. The researchers were able to identify 419 intracranial and spinal cord masses including 406, which were primary CNS tumors.

Furthermore, Wiredu and Armah⁸ conducted a retrospective study that focused on cancer mortality at the Korle-Bu Teaching Hospital in Accra, Ghana. Between 1991 and 2000, a total of 144 autopsies revealed deaths secondary to CNS tumors. Another retrospective study of cancer mortality carried out at the Lagos University Teaching Hospital in Nigeria between 2000 and 2013 reported a total of 56 deaths related to CNS tumors.⁶

The reporting of very few cases of CNS tumors from Africa has resulted in the belief that CNS tumors are rare in Africa; however, the above studies from different regions in Africa have failed to show evidence to support that claim.⁵ Although there are some studies that have focused on clinically diagnosed CNS tumors and mortality from different regions of Africa, no systematic study on CNS tumor incidence and mortality in the whole continent of Africa has been performed. This manuscript intends to identify the distribution of CNS tumors by demographics and clinical diagnosis in Africa. Additionally, this study will identify key differences in the incidence and mortality of CNS tumors between the different regions of Africa as well as address the need for the development of an all-inclusive CNS tumor registry.

Methods

Data Sources

A comprehensive literature search on CNS tumors in Africa was done using EBSCOhost, MEDLINE, Google Scholar, CINAHL Plus, Academic Search Complete, Cochrane databases, and Walden University Library. The following keywords were queried in combination with the phrase “CNS tumors in Africa”: incidence, frequency, epidemiology, prevalence, brain tumors, CNS tumors, Africa, and cancer. After eliminating duplicate results, 257 756 articles were found, and the articles were further narrowed down to ones containing “Africa.” A total of 986 articles remained in the search results. All the 986 articles were further reviewed for reports of patients diagnosed with CNS tumors (neoplastic or nonneoplastic). A total of 26 articles met the inclusion criteria. Each article meeting the inclusion criteria contained a description of the incidence of CNS tumors in Africa and identified the patient’s gender. The 26 selected studies (Table 1) were carried out in the following 9 African countries, with the numbers of studies notated beside each country’s name: Nigeria (11), Kenya (4), Uganda (4), Ghana (2), Sudan (1), Morocco (1), Namibia (1), Ivory Coast (1), and Egypt (1).^4,5,7,9–29 The selected articles reported total incidence and/or mortality from the different regions in the above-mentioned African nations between 1960 and 2017, covering a 57-year period. The extracted data were then analyzed to determine the total reported frequency and mortality of CNS tumors in the patient population in Africa. The recorded data included the African region or country in which studies were conducted, annual periods of studies, total reported cases and deaths, crude incidence rates and mortality rates, and frequency of each histologically identified CNS tumor.

Statistical Analysis

A systemic review was performed using data from the incidence and mortality in both adult and pediatric patients by study region for country, gender, and tumor type. The total number of cases and total mortalities was analyzed from 1960 to 2017, thus covering a 57-year period. The mean and 95% confidence interval (CI) for the total cases, gender, and mortality as well as the percentage, mean, and 95% CI for each tumor were analyzed descriptively (Tables 2 and 3). The data were tabulated and all the frequency and descriptive analysis data were analyzed using the SPSS21 statistical package. The distribution of studies by groups (adult, children, and both), African country of origin, tumor types, gender, and mortality was described.

Results

Number and Geographic Variation of CNS Tumor Studies in Africa

Of the 26 studies reviewed, 10 (38.5%) reported on a combination of adult and child CNS tumors, 10 (38.5%) reported on child CNS tumors only, and 6 (23.1%) reported on adult CNS tumors only (Table 4). The majority of the reported studies was carried out in Nigeria (42.3%), Kenya (15.4%), Uganda (15.4%), and Ghana (7.7%). The countries such as Egypt, Morocco, Sudan, Ivory Coast, and Namibia published 1 study each (3.8% per country) (Table 5).

Total Cases by Gender and Total Mortality

The total reported cases of CNS tumors in Africa from 1960 to 2017 were 5902 with a crude incidence rate of 226.98 (95% CI: 66.60-387.6) per 100 000 population. The annual average of new CNS tumors reported during this time period was approximately 103.5 cases per year. CNS tumors affected males more than females in a 1.28:1 ratio. Males had a mean incidence rate of 122.67 (95% CI: 29.27-216.07) per 100 000 population, while females had a mean incidence rate of 96.19 (95% CI: 26.24-166.15) per 100 000 population. The total reported deaths of all patients studied were 460 with an incidence rate of 17.69 (95% CI: 4.44-30.94) per 100 000 population (Table 2).

The total reported cases in Nigeria (2318) were highest when compared with the other African regions with males higher than females at a ratio of 1.54:1. Egypt was next with 1618 cases diagnosed with males having a slightly higher frequency of tumors than females by a ratio of 1.1:1. After Egypt, Uganda had the most cases of any African country with 703 cases and 66 total deaths. The frequency of tumors in males was higher than females by a ratio of 1.3:1. Morocco fell next with total reported cases of 542, with tumor distribution among males occurring slightly higher than that of females at a ratio of 1.01:1. Kenya came next with a total of 376 cases diagnosed with total mortality of 20. The male frequency was higher than that of females by a ratio of 1.5:1. The next country was Ghana with 170 cases diagnosed and total mortality of 146. In Ghana, the female frequency was higher when compared with males at a ratio of 1.15:1. Sudan had 90 total cases with males higher than females at a ratio of 1.43:1. The Ivory Coast had 57 total cases, with a higher frequency of females when compared with males by a ratio of 1.5:1. The mortality from the Ivory Coast was reported to be 13. The African country with the fewest number of cases was Namibia with only 28 cases occurring at an equal frequency between males and females. No mortality data were reported from Egypt, Sudan, Morocco, and Namibia (Table 5).

Total Cases by Tumor Type

Table 3 presents the total cases and percentage of different CNS tumors in Africa between 1960 and 2017. Tumors with the highest frequency were astrocytoma with 1462 (24.7%) cases, meningioma with 1333 (22.22%) cases, and other tumor types not specified with 1556 (25%) cases. The following remaining tumor types occurred at a significantly lower frequency: pituitary adenoma (8.04%), medulloblastoma (4.26%), craniopharyngioma (4.07%), ependymoma (3.65%), oligodendroglioma (2.01%), embryonal tumors (1.733%), glioblastoma (1.25%), glioma (Not Otherwise Specified) (0.95%), germ cell tumors (0.87%), ganglioglioma (0.37%), neuroblastoma (0.29%), pineoblastoma (0.19%), epidermoid cyst (0.099%), ependymoblastoma (0.066%), and supratentorial Primitive Neuro-Ectodermal Tumor (PNET) (0.03%).

Table 3 also presents the 95% certainty of the true case numbers of the above-mentioned CNS tumor types. Of note, the incidence rate of astrocytoma was 56.22 (95% CI: 2.69-109.75) per 100 000 population followed by meningioma with an incidence rate of 51.28 (95% CI: 5.30-107.87) per 100 000 population and pituitary adenoma with an incidence rate of 18.55 (95% CI: 1.58-38.68) per 100 000 population. Other tumor types with lower incidence rates included craniopharyngioma at 9.41 (95% CI: 2.26-16.56) per 100 000 population, oligodendroglioma at 4.65 (95% CI: 0.57-8.75) per 100 000 population, ependymoma at 7.85 (95% CI: 0.65-15.04) per 100 000 population, and medulloblastoma at 9.85 (95% CI: −2.47 to 22.16) per 100 000 population. Tumors categorized as others that were not specified had an incidence rate of 59.83 (95% CI: 26.70-92.95) per 100 000 population. Astrocytoma had the highest incidence rate per 100 000 population, whereas supratentorial PNET had the lowest reported incidence rate per 100 000 population.

Discussion

According to the vital registration and cancer registry, the mortality attributed to CNS tumors caused a substantial burden at a global level.¹ However, data on the total cases and mortality of CNS tumors from the whole continent of Africa were not included in the registry. The current study is the first of its kind to systematically report on the total number and mortality data of CNS tumors in the continent of Africa. The frequency data of CNS tumors are an important basis for determining the priorities of CNS tumor control in any population.¹⁵ There are marked differences in the distribution of CNS tumors from different regions around the globe including Africa. Prior report of rare cases in Africa was likely due to many cases going unreported since autopsy was the primary basis of tumor verification.^27,30,31 The institution of neurosurgical and neurological practice in Nigeria and other parts of Africa and the availability of improved diagnostic facilities have resulted in the documentation of more cases of CNS tumors after surgical biopsy confirmation.^5,32,33

The results of this study confirm the substantial variation of CNS tumor cases in different countries. From the years 1960 to 2017 (a 57-year time period), the total reported case number of CNS tumors (malignant and nonmalignant) in Africa was 5902 with an incidence rate of 226.98 per 100 000 population. The total cases of CNS tumors in Africa over this 57-year time interval are substantially fewer than that in the United States (101 243 cases) and Canada (11 386 cases) over a much shorter time interval (2003–2007).³³ One possible reason for the vast difference in total cases between developing and developed regions of the world could be linked to genetics and environmental factors.³³ This study found that most of the CNS tumors were from Nigeria, followed by Egypt, Uganda, Morocco, Kenya, Ghana, Sudan, Ivory Coast, and Namibia. The high case numbers in Nigeria and Egypt could be due to an improvement in investigational studies and clinical diagnosis, although this cannot be proven at this time since these data have not been systematically collected in all regions and countries in Africa.²⁷ There are also some socioeconomic and environmental factors, such as exposure to crops, ultraviolet rays or ionizing radiations, and associations with atopic conditions, which may be attributed to the differences in CNS tumor distribution in different regions of Africa.^1,15,34 Of note, although glioblastoma is the most common malignant tumor overall, it had a substantially lower incidence rate per 100 000 population in Africa compared with other tumor types including astrocytoma and meningioma. This is consistent with epidemiological data from the National Program of Cancer Registries (NPCR) and Surveillance, Epidemiology, and End Results (SEER) databases, which showed that European Americans have a glioblastoma incidence rate that is 2.5 times greater than African Americans, suggesting a genetic component to the disease.³⁵

In this study, the total number of male cases was noted to be higher than the total number of female cases. Also, in Nigeria, the frequency of male cases was higher than female cases by a ratio of 1.5:1. A similar finding of gender difference was published from Japan and the United States.^36,37 Possible reasons for the observed gender difference are sex hormones and genetic features in addition to exposure to chemicals, pesticides, biologic agents, and overall adverse work conditions among men.^38,39

A notable finding in this study is the comparison between the number of studies between adults and pediatric populations. Of the 26 studies analyzed, 6 reported on adult CNS tumors only, 10 reported on pediatric cases only, and 10 reported on both adult and pediatric patients. This finding is consistent with prior literature, which asserted that the collection of brain tumors in children from developing countries is usually not well segregated from those of adults.^4,22,27 Population studies are usually impossible to conduct in developing countries due to logistic limitations from the way the hospital collects these data.⁴ Thus, hospital-based studies are usually the only source of available data that can be used to describe the pattern of childhood CNS tumors in the population.⁴ Studies by Olasode et al.²⁷ and Aghadiuno et al.²⁹ illustrate this point as their pediatric data were generalized for the population of Ibadan, Nigeria.

This study identified 17 primary CNS tumors with astrocytoma as the most common form followed by meningioma, although meningioma was found to occur at a higher rate when compared with astrocytoma in other studies.^16,22 The current high frequency of astrocytoma is in contrast with other studies written over 50 years ago, which found gliomas to be uncommon in Africa.^31,40 Recent studies, however, have found astrocytic tumors to be high in Asia and America, especially among children.³³ The high rate in children is also consistent with some of the studies analyzed in this current study.⁴ With regard to meningioma, radiation exposure has been linked to its increase in incidence.^33,41,42

In this study, the only reported total mortality of CNS tumors in Africa from 1989 to 2017 was 460 with an incidence rate of 17.69 per 100 000 population. Nigeria had the highest mortality, while Ivory Coast had the lowest. The total reported deaths are significantly fewer than the global estimate by the GLOBOCAN study possibly because of the decreased amount of reported cases patients residing in Africa.^1,6 Furthermore, a published report of the International Agency for Research and Cancer (IARC) by GLOBOCAN acknowledges the lack of quality mortality data from Africa and so asserted that countries with little or no data could be due to an extrapolation from limited samples, a mixture of real data, and/or informed guesses.⁶ It has also been suggested in another study that the reporting of low mortality to total case numbers can be attributed to a timely accurate diagnosis and improved treatment.¹

Muir et al.⁴³ asserted that determining the incidence rates of CNS tumors has been complicated by multiple factors, including variations in registration practice and efficiency over place and time. The Central Brain Tumor Registry of the United States (CBTRUS) is a research organization that provides quality statistical data on population-based primary brain and CNS tumor incidence in the United States.⁴⁴ Using the CBTRUS, Surawicz et al.⁴⁵ were able to obtain 5 years of data on the primary brain and CNS tumors in the United States between 1990 and 1994. These researchers pulled a total of 20 765 CNS tumor cases from the CBTRUS database from 11 collaborating state cancer registries.⁴⁵ Surawicz et al.⁴⁵ asserted that their study represented the largest compilation of data on the primary brain and CNS tumors in the United States. They emphasized the importance of universal adoption of a standard reporting definition and practice that could improve the quality and use of the cancer data registry.⁴⁶ It is important for Africa to adopt a centralized population-based database of primary nonmalignant and malignant CNS tumors in order to improve surveillance and reporting of CNS tumors. Using the CBTRUS model, it may be possible to establish a Central Brain Tumor Registry of Africa.

In summary, CNS tumors pose a substantial public health problem due to their high morbidity and mortality, and the data from our analysis provide additional evidence that CNS tumors in Africa may have some epidemiological differences when compared with other continents. There are limitations to our study, which are systemic to the broad healthcare system found across countries in Africa, including lack of an effective registry system, limited diagnostic health infrastructure, and publication bias. Nonetheless, this study, to our knowledge, represents the first meta-analysis on the descriptive epidemiology of CNS tumors in Africa from 1960 to 2017. Further studies to support the establishment of a population-based CNS tumor registry system in Africa are warranted.

Conflict of interest statement. All authors report no conflicts of interest.

References