Unraveling the Mysteries of Ameloblastoma in African Population: A Comprehensive Analysis of 371 Cases from Clinical, Radiological, and Histopathological Perspectives

Abstract

Objective

To analyze the frequency, clinical, histopathological, and radiological characteristics of ameloblastoma in Nigeria over the course of two decades.

Study Design

A retrospective analysis was conducted on 371 cases at a Nigerian university hospital between 2000 and 2023. Age, gender, site, histological variants, tumor size and duration were analyzed. Statistical analyses included the Shapiro–Wilk test, Mann–Whitney U test, Chi-square test, and Spearman rank correlation analysis.

Results

The median patient age was 30 years (mean age 32.2), with a male-to-female ratio of 1.12:1. 54.7% of cases occurred in young adults (age range 20–39 years). Among the lesions, 11.3% were in the maxilla and 88.7% in the mandible. Patients with mandibular lesions had a median age of 29 years, while those with maxillary lesions had a statistically significantly higher median age of 37.5 years p-value = 0.001. Median tumor size was 36 cm² for the mandible and 24 cm² for the maxilla (significant p-value of 0.002). There was no correlation between tumor size, age, or gender. However, there was a significant correlation between tumor size and the duration of the condition.

Conclusion

The study concludes that ameloblastoma is more frequent among younger individuals in Nigeria and often presents with larger tumor sizes, emphasizing the need for early detection and intervention.

Introduction

Ameloblastomas constitute a complex assortment of oral tumors. They account for an estimated 11–59% of odontogenic tumors, a range that fluctuates based on the specific report and geographic location [1–3]. The prevailing agreement that ameloblastoma derives from odontogenic sources is largely predicated on the histological similarities between the tumor and the developing enamel organ [4, 5].

In the 2022 World Health Organization (WHO) fifth edition classification of Head and Neck Tumors, ameloblastoma is categorized into five types: unicystic ameloblastoma, characterized by a single unicystic cavity; extraosseous/peripheral ameloblastoma, located beneath the gingiva and within edentulous regions of the oral cavity; metastasizing ameloblastoma, which spreads into distant sites despite its benign nature; conventional ameloblastoma, composed of ameloblast-like cells and stellate reticulum; and the recently added adenoid ameloblastoma, which is recognized as a benign epithelial odontogenic tumor [6–8]. The radiological appearance of ameloblastic tumors can be classified into unilocular, multilocular (when bone striations are visible within the lesion), mixed radiolucent/radiopaque lesions (when radiopaque materials are present against a radiolucent background), (see Fig. 1) and lesions without radiological involvement [9]. Despite its typical classification as a benign tumor, ameloblastoma carries a 2% risk of malignant transformation and metastasis [1, 10]. Ameloblastomas that are detected early generally have a favorable prognosis. However, if left untreated, ameloblastomas can lead to severe consequences such as damage to the jaw and adjacent structures, and even mortality due to rare instances of metastasis [11–13].

Fig. 1.

Distribution of Ameloblastoma by age. Most of the cases were seen between the second and fourth decade

Odontogenic tumors exhibit diverse demographic distributions [14–16]. Multiple studies have confirmed that the incidence of ameloblastoma is higher in individuals of African descent [3, 17, 18]. Although ameloblastoma has been extensively studied in various populations, research specifically targeting African populations has been comparatively limited. Moreover, two recent studies have emphasized the urgent need for research on ameloblastoma in the African population. These studies highlight a trend of earlier onset, larger ameloblastomas, and a higher recurrence rate within this demographic [17, 19]. There is a significant research gap concerning the size of ameloblastomas in general, particularly within the African population, and its correlation with age, gender, and duration of the condition.

The objective of this study was to investigate the relative frequency and gather clinical, histopathological, and radiological data on the various forms of ameloblastoma observed in the African population in Lagos, Nigeria, over a 23-year period. Additionally, we aimed to provide, for the first time to our knowledge, an in-depth analysis of tumor size within this African cohort. This comprehensive dataset serves as a fundamental resource for pathologists and clinicians, promoting a deeper understanding of the disease. Furthermore, our findings contribute to the broader knowledge base on ameloblastoma, offering specific insights relevant to the African demographic.

Materials and Methods

Study Design

This is a retrospective observational study of histologically diagnosed cases of ameloblastoma, utilizing records obtained from a standardized proforma used during biopsy procedures. The investigation spanned cases diagnosed between 2000 and 2023 at the Lagos University Teaching Hospital (LUTH) in Lagos, Nigeria. The study was conducted in strict adherence to the World Medical Association's Declaration of Helsinki guidelines on medical research involving human subjects. Ethical approval was obtained from the Health Research Ethics Committee of the Lagos University Teaching Hospital (LUTH). Due to the retrospective nature of the study, which involved the analysis of de-identified patient records collected during hospital visits and treatments, the Institutional Review Board (IRB) waived the requirement for written informed consent.

Comprehensive information from the archives of the oral pathology department was meticulously gathered by AWA into an excel sheet. Information obtained included patient age, gender, and duration of symptoms, clinical signs, diagnosis, histological variant, radiological appearance, and tumor size in centimeters. The diameter of the tumor in an anterior posterior and superior inferior direction in centimeters, which was taken at first attendance to the out-patient clinic, was used to evaluate tumor size. These values were multiplied to obtain the surface area of the tumor in centimeters squared (cm²).

Diagnosed cases were categorized into three major types: conventional ameloblastoma, unicystic ameloblastoma, and peripheral ameloblastoma [8, 20]. Conventional ameloblastomas were further subdivided based on histopathologic features into follicular, plexiform, desmoplastic, hybrid and mixed subtypes. The mixed subtype exhibits a combination of plexiform and follicular growth patterns, making it difficult to discern which type predominates in the tissue sections. The hybrid subtype is characterized by the presence of both follicular and desmoplastic histologic variants within the same tumor. Unicystic ameloblastomas are categorized into three distinct histopathological subtypes: luminal, intraluminal, and mural patterns. Cases lacking complete patient information were excluded from the analysis.

Statistical Analysis

Statistical analysis was conducted using Python™ (version 3.12.5). The Shapiro–Wilk test was applied to numerical variables (age, tumor size, and duration), revealing that the data were not normally distributed, with an alpha level set at 0.05. The non-parametric Mann–Whitney U test was used to assess significant differences in the age of occurrence between males and females, between the mandible and maxilla, and between unicystic, peripheral and conventional ameloblastomas. Additionally, this test evaluated whether tumor size differed significantly between males and females or between the mandible and maxilla or between the clinico-histologic types. Non-parametric test was also used to evaluate the differences between tumor duration and clinico-pathologic characteristic of the cases. The Chi-square test was employed to determine statistical differences between histological variants in males and females. Finally, the Spearman rank correlation was utilized to examine the relationship between tumor size and age, as well as between tumor size and duration.

Results

Patient Demographics

A total of 371 ameloblastoma cases were retrospectively analyzed over a 23-year period. Among these cases, 53.8% were male (196 cases) and 47.2% were female (175 cases), resulting in a male-to-female ratio of approximately 1.12:1. The overall median age was 30 years (IQR: 19, mean: 32.2 years) (Table 1). The minimum age was 8 years, and the maximum age was 74 years. Overall, approximately 54.7% of cases occurred in young adults, with an age range of 20–39 years. Figure 1 shows the age distribution based on age categories. The Mann–Whitney U test indicated no statistically significant difference in median ages between male patients (median: 30 years, IQR: 19, mean: 32 years) and female patients (median: 30 years, IQR: 19.5, mean: 32.4 years), with a p-value of 0.963 (Table 1).

Table 1.

Comparison of age of cases with clinico-pathologic characteristics

Variable	Number (%)	Median (IQR) age in years	Mean (SD)age in years	P-value
Gender
Male	196 (52.8)	30 years (19.0)	32.1 (13.7)	0.835
Female	175 (47.2)	30 years (19.5)	32.4 (14.2)
Anatomic site
Maxilla	42 (11.3)	37.5 (25.0)	39.6 (15.6)	0.001
Mandible	329 (88.7)	29.0 (40.0)	31.6 (13.4)
Histologic patterns
Conventional	285 (76.8)	30.0 (18.0)	33.1 (13.5)
unicystic	76 (20.5)	25.0 (20.0)	28.0 (14.9)
Peripheral	10 (2.7)	39.0 (12.0)	39.3 (10.7)	0.088

Upon examining the lesion site, the median age of patients with mandibular lesions was found to be 29 years (IQR: 40, mean: 31.3 years). In contrast, patients with maxillary lesions had a higher median age of 37.5 years (IQR: 25.75, mean: 39.6 years). Notably, the difference in the median age between these two groups was statistically significant, with a p-value of 0.001 (Table 1).

Furthermore, the median age for patients diagnosed with peripheral ameloblastoma was 39 years (IQR: 12, mean: 39.3 years). Interestingly, the median age for patients with unicystic ameloblastomas was 25 years (IQR: 20, mean: 28 years), which was lower than that of patients with conventional ameloblastoma (median: 30 years, IQR: 18, mean: 33.1 years). However, the difference between the medians was not statistically significant, with a p-value of 0.088 (Table 1). The tumor sites were predominantly located in the mandible, accounting for 88.7% (329 cases), while a smaller proportion of 11.3% (42 cases) occurred in the maxilla (Table 1). Notably, cases occurring in the maxilla were evenly distributed between males and females.

Histological Patterns

Among the patients, 76.8% (285 cases) were diagnosed with conventional ameloblastomas, 20.5% (76 cases) had unicystic ameloblastoma, and 2.7% (10 cases) presented with peripheral ameloblastoma.

The major histologic patterns of conventional ameloblastoma included the follicular type (Fig. 2a) at 30.6%, the plexiform type (Fig. 2b) at 19.6%, and a mixed pattern of follicular and plexiform at 14.5%. Another pattern, desmoplastic ameloblastoma, was observed in 10.8% of all conventional cases. The hybrid pattern was present at 1.1%. Histologic variants within the follicular and plexiform patterns included acanthomatous (8.9%), granular (1.9%), and hemangiomatous (1.4%) (Table 2).

Fig. 2.

a Histologic section of follicular conventional ameloblastoma exhibiting islands of ameloblastomatous epithelium with peripheral tall columnar cells (red solid arrow) and stellate reticulum like cells (black star) × 40 magnification. b Plexiform ameloblastoma exhibiting interconnecting sheets of ameloblastomatous epithelium with peripheral tall columnar cells (red solid arrow) and stellate reticulum like cells (black star) some undergoing cystification (blue star) within a dense fibrous connective tissue stroma.that appear to surround connective tissue stroma (red star) × 100 magnification

Table 2.

Clinic histological types and variants of ameloblastoma: number and frequency (%)

Histologic types	Number	Frequency %
Ameloblastoma conventional	285	76.6
Follicular	114	30.6
Acanthomatous	23	6.2
Granular	4	1.1
Basal cell	2	0.5
Hemangiomatous	3	0.8
Variant not described	82	22.0
Plexiform	73	19.6
Acanthomatous	3	0.8
Granular	1	0.3
Hemangiomatous	5	1.3
Keratoacanthomatous	1	0.3
Papilliferous Keratinizing	1	0.3
Variant not described	62	16.7
Mixed	54	14.5
Acanthomatous	7	1.9
Granular	2	0.5
Hemangiomatous	1	0.3
Variant not described	44	11.8
Desmoplastic	40	10.8%
Hybrid	4	1.1
Ameloblastoma peripheral	10	2.7
Ameloblastoma unicystic	76	20.4
Luminal	7	1.9
Intraluminal	25	6.7
Mural	44	11.8
Total	371	100

For unicystic ameloblastoma, the mural pattern was the most common histologic type, comprising 11.4% of all cases. This was followed by the intraluminal type at 6.7% and the luminal type at 1.9% (see Table 2). The Chi-square test showed that there is no significant difference between histological variants in males and females (p-value = 0.899).

Radiological patterns

Among radiological patterns, multilocular lesions were the most widespread (65.2%), followed by unilocular lesions (Fig. 3a) (20.5%). A mixed radiological pattern (radiopaque and radiolucent) (Fig. 3b) occurred in 11.6% of cases. Notably, 2.7% of cases showed no radiological involvement and corresponded to peripheral ameloblastoma.

Fig. 3.

Axial view of the computed tomography (CT scan) patient depicted above in Fig. 2, shows extensive facial asymmetry (multiple red arrows) with multilocular radiolucent (yellow star) and radiopaque lesion (red star)

Tumor size and duration analysis

The overall median tumor size was 35 cm² (IQR: 41, mean: 47.8 cm²). When analyzing tumor size by gender, females had a median tumor size of 32 cm² (IQR: 37, mean: 46.9 cm²), while males had a median tumor size of 38.5 cm² (IQR: 43, mean: 48.5 cm²). The p-value for the gender difference was 0.167, indicating no significant difference (Table 3).

Table 3.

Comparison of tumour size (in cm²) and tumour duration (in months) and clinico-pathologic characteristics

Variable	Gender		Anatomic site		Histologic patterns			Total
	Male	Female	Max	Mand	CA	UA	PA
Median (IQR) tumour size in cm2	38.5 (43.0)	48.5 (46.1)	24.0 (28.0)	36.0 (40.0)	40.0 ( 42.0)	22.5 (35.0)	9.5 (8.0)	35.0 (41.0)
Mean (± SD) tumour size in cm2	32.0 (37.0)	46.9 (58.9)	29.9 (24.7)	50.1 (54.6)	51.6 (54.5)	37.4 (43.8)	20.3 (35.3)
P-value1	0.600		*0.002a		*0.0001b
Median (IQR) tumour duration in months	31.0 (32.0)	36.0 (56.0)	24.0 (24.0)	36.0 (54.0)	36.0 (54.0)	24.0 (36.0)	9.5 (28.0)	34.0 (53.0)
Mean (± SD) tumour duration in months	48.9 (47.6)	49.4 (46.1)	37.8 (57.7)	50.5 (46.2)	52.9 (48.5)	38.4 (39.6)	22.1 (24.5)	49.1 (46.8)
P-value1	0.167 a		*0.002a		*0.0001b

¹p-values were computed with non-parametric statistics because the values were not normally distributed. ^aMann-whitney U, ^bKruskal-Wallis *statistically significant

max. Maxilla, mand. Mandible, CA conventional ameloblastoma, UA Unicystic ameloblastoma, PA peripheral ameloblastoma

In terms of tumor size by site, the median tumor size in the mandible was 36 cm² (IQR: 40, mean: 50.1 cm²), whereas in the maxilla, it was 24 cm² (IQR: 28, mean: 29.9 cm²). The p-value for the site difference was 0.002, indicating a significant difference (Table 3).

The overall median duration as at presentation was 34.0 months (IQR: 53.0, mean: 49.1 months). The median duration of mandibular lesions was 36.0 (IQR: 54.0, mean:) was statistically different from that of maxillary lesion at 24.0 (IQR: 24.0), (p-value = 0.002). A significant difference in the duration of the lesion before presentation was also observed among the three clinico-histologic types (p-value = 0.0001) (Table 4).

Table 4.

Comparison of tumour size (in cm²) and clinico-pathologic characteristics

Variable	Median (IQR) tumour size in cm2	Mean (± SD) tumour size in cm2	P-value1
Gender
Male	38.5 (43.0)	48.5 (46.1)	0.167 a
Female	32.0 (37.0)	46.9 (58.9)
Anatomic site
Maxilla	24.0 (28.0)	29.9 (24.7)	*0.002a
Mandible	36.0 (40.0)	50.1 (54.6)
Histologic patterns
conventional	40.0 ( 42.0)	51.6 (54.5)	*0.0001b
Unicystic	22.5 (35.0)	37.4 (43.8)
Peripheral	9.5 (8.0)	20.3 (35.3)
Total	35.0 (41.0)	47.8 ( 52.5)

¹p-values were computed with non-parametric statistics because the values were not normally distributed. ^aMann-Whitney U, ^bKruskal-wallis, *statistically significant

Lastly, the correlation between age and tumor size is almost non-existent, with a correlation coefficient (− 0.0043), and P value: 0.9335. However, the tumor size slightly increased with age in females and decreased in males, with no significant correlation (Fig. 4). In our study, the Spearman rank correlation between duration and tumor size was correlation coefficient: 0.4019 indicating a moderate positive correlation, which is statistically significant with a p-value of < 0.05. This suggests that as the duration of the condition increases, the tumor size tends to increase as well. Finally, Fig. 5 displays a graphical representation of data where individual values contained in a matrix are represented as colors.

Fig. 4.

Interactive scatter plot showing the relationship between age and tumor size for different genders, including a linear regression trend line to indicate the overall trend for each gender (Tumour size analysis)

Fig. 5.

Heatmap depicting the Correlations among all Study variables. Each square of the grid represents a specific interaction between two variables, with the color of the square indicating the strength or magnitude of that interaction. Darker colors represent higher values, and lighter colors represent lower values

Discussion

This article presents a single-center study conducted at a university hospital in Nigeria’s largest city, utilizing data collected over nearly two decades. To the best of our knowledge, this study represents the largest single-center cohort on ameloblastoma in the African continent. Additionally, it is the first to explore tumor size analysis in this context. Notably, ameloblastomas are the most prevalent odontogenic tumors with significant clinical consequences among the African population [18, 20, 21]. In fact, Africans are five times more likely to develop ameloblastoma than Caucasians [1, 3, 22].

Our data showed that the mean age of ameloblastoma in this cohort is 32.25 years, which is considerably lower than the average age (36–38 years) in Southeast Asia and North America [23]. Furthermore, it is even lower than the mean age reported in African Americans [3]. This falls precisely within the range found in a multi-center study in Nigeria [20]. It should be noted that about 74.2% of the cases occurred in patients aged between 8 and 39 years, with 54.7% in patients aged between 20 and 39 years of age. The mean age for peripheral ameloblastoma diagnosis was 39.3 years, which is lower than that of the largest study on Nigerians at 44.8 years [20]. Meanwhile, the mean age in the literature ranges from 40 to 63 years [24, 25].

On the other hand, our data revealed a male-to-female ratio of approximately 1.12:1, indicating an almost equal distribution with a slightly higher prevalence in males. This finding is consistent with studies conducted in South Africa and Nigeria [20, 26], and demonstrates a lower ratio than the 1.4:1 [27, 28] and 2.2:1 [29] observed in the Asian population. Our data underscores a relatively balanced gender distribution in Nigeria. In our dataset, ameloblastomas were more frequently found in the mandible, which is entirely in line with previous literature [20, 30]. However, tumors in the maxilla appeared approximately 8 years later than those in the mandible, which is about 5 years later than typically reported in Nigeria [20]. The reason for this time difference is unknown but speculation include the lack of prominent early growth of the tumor in the maxilla [31] and a difference in genetic driver mutation in maxillary ameloblastoma as they have been shown to have more of smoothen (SMO) gene mutation compared to the mitogen-activated protein kinase (MAPK) pathway BRAF V600E gene mutations in most mandibular ameloblastoma [32]. Peripheral ameloblastoma occurred in only 10 instances, constituting 2.7% of all occurrences, which falls within the lower range of frequencies reported in the literature (1–10%) [25, 33]. Based on the histological distribution, the conventional pattern dominated, followed by the unicystic and peripheral patterns, respectively. This aligns with previous research [3].

The overall median tumor size observed in our study is 35 cm², with an interquartile range (IQR) of 41 cm², and a mean of 47.8 ± 52.5 cm². Our data reveal a significantly larger mean tumor size in comparison to White patients (11.8 cm² ± 13.0 cm²) and Afro-Americans (15.7 ± 17.0 cm²) as reported by Vila et al. [3]. These findings suggest that larger ameloblastomas are more prevalent among Africans. It has been established that late presentation by patients gives rise to larger tumors often due to lack of pain as a major symptom, poor socioeconomic status associated with lack of funds, restriction of movement and access to specialist care for the patients [3, 12, 26]. However, there is a notable lack of comprehensive studies documenting the relationship between the size of ameloblastoma and race. An urgent need for further research to examine the tumor size across various ethnic groups is important because the time of discovery of the lesion might be an associated factor in the large sizes of ameloblastoma seen in Africans (Fig. 6). This factor can be easily modified compared to possible inherent genetic factors that can predispose to the development of large tumors.

Fig. 6.

a A patient with a big lesion on the left side of the face that was diagnosed as ameloblastoma. b The patient had a recurrence 7 years after the first tumour. Recurrence is one of the risk factors of presenting with large tumours

Conversely, our study found no significant difference in tumor size between males and females. Additionally, age does not appear to play a significant role in the size of ameloblastomas. However, we identified a significant correlation between tumor size and the duration since detection. The duration of lesion before patient present for treatment was 34 months (mean 49.1 months). The duration of most ameloblastoma is between 1 and 5 years with an average of 2.8 years for conventional ameloblastoma [18]. Large tumors were reported to have an average duration of symptoms before presenting for treatment of 9.04 years in a systematic review [34]. Some individuals may delay therapy due to the benign nature of ameloblastomas and the extensive surgical procedures often recommended. In the African population, delayed treatment can be attributed to factors such as fear of surgery, lack of knowledge, and cultural norms [35].

The goals of treating ameloblastoma are to excise the tumor, restore functionality and aesthetics, and prevent recurrence. These goals are often not met in many cases, particularly when patients present late. Delayed presentation continues to be the primary challenge in the detection and management of maxillofacial tumors in the African Continent, impacting postoperative quality of life and exacerbating functional and cosmetic issues. The factors for delay have to be investigated comprehensively with an intention to provide intervention to mitigate it.

Conclusion

In the African population, ameloblastoma is observed more frequently among younger individuals, and these cases often present with larger tumor sizes. The association between larger tumor sizes and younger age in the African population highlights the critical need for early detection, diagnosis, and intervention.

Authors' Contributions

AA, GS, JE, OE, JA, AR, AC contributed to the conception and design of the study, AA, GS, JE, OE acquired the data, and AA, GS, JE, OE, JA did the analysis and interpretation of data. All authors participated in drafting the article or revising it critically for important intellectual content. All authors read and approved the final version of the manuscript submitted.

Funding

This study was not supported by any funding.

Availability of Data and Material

The data that is used in this study is available with the corresponding author on reasonable request.

Code Availability

Not applicable.

Declarations

Conflict of interest

The authors declare no competing interests.

Ethics Approval

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. Ethical approval was obtained from the Health Research Ethics Committee of the Lagos University Teaching Hospital (LUTH).

Consent to Participate

Not required. This is a study of de-identified cases in the archives. The IRB waived this requirement.

Consent for publication

For this type of study consent for publication is not required.