Recurrent Ameloblastoma: Clinical Manifestation and Disease-Free Survival Rate

Objectives Ameloblastoma is a slow-growing epithelial odontogenic neoplasm of the jaws with a high recurrence rate. The main treatment strategies for this lesion are radical or conservative surgical approaches. The aim of the present study was to analyze clinical presentations, histological types, and treatment strategies of recurrent ameloblastoma and to define its disease-free survival (DFS) rate. Materials and Methods Twenty-four cases of recurrent ameloblastomas, treated between January 2009 and July 2021, were enrolled in this study. Medical files from each patient, including gender, age, size of the lesion, localization, patient complaints, clinical manifestation, radiographic appearance, histological type, surgical management, and treatment results were reviewed and analyzed retrospectively. Result Out of 69 operated primary ameloblastomas, the rate of recurrence was 35%. Out of 24 recurrent cases, 21 developed after conservative treatment and 3 after radical treatment. In most cases, recurrences were found in the mandible (n = 20). A unilocular pattern was predominant in radiographic examination (44%). Estimated 3-year DFS was 84.5 ± 4.8%, and the 5-year and 10-year DFS were 73.0 ± 6.3% and 43.9 ± 8.343.9 ± 8.3%, respectively. Conclusion Results obtained in the present retrospective study proved the necessity of long-term follow-up after both conservative and radical treatment approaches. The DFS median in our study was 8 years (95% CI 6 years–10 years). For recurrent cases, radical resection with histologically free margins after exact MRI determination of the ameloblastoma border within the soft tissues should be considered as the method of choice to avoid secondary recurrence.


Introduction
Ameloblastoma is a slow-growing benign odontogenic neoplasm that is locally aggressive and has a high recurrence rate [1]. Ameloblastoma has no gender predilection, but it is predominant in patients between 30 and 60 years old [2]. e mandible is affected much more frequently than the maxilla (88 vs. 12%) [3]. According to the 2017 World Health Organization (WHO) classification of head and neck tumors, ameloblastoma can be divided into three subtypes: conventional ameloblastoma (solid/multicystic variant), unicystic ameloblastoma, and extraosseous/peripheral types [4]. Conventional ameloblastoma was recognized as the most common type and is associated with significantly higher recurrence rates [2,5,6]. Primary treatment of ameloblastoma is surgical and can be classified into conservative and radical [7]. Conservative methods such as enucleation or extended curettage are less invasive and require less operation time, but these methods are associated with significantly higher recurrence rates and the need for secondary reoperations [8]. Radical surgery, including marginal resection, segmental resection, hemimandibulectomy, and maxillectomy, is associated with lower recurrence rates but often results in serious aesthetic and functional impairment, decreasing the patient's quality of life, and often requires complex reconstructive surgery [9]. With conservative treatment, 55-90% of cases recur, whereas, the recurrence rate with radical treatment is 15-25% [10,11]. In both cases, the treatment prognosis depends on the ameloblastoma type, clinical and radiological signs, and histologically confirmed diagnosis [12]. Unicystic and extraosseous ameloblastoma, according to Hertog et al., can be treated conservatively with adequate success rates, while the more aggressive conventional ameloblastoma requires radical treatment in most cases [13]. Ameloblastomas that recur after initial conservative or radical surgery are a major challenge for the surgeons as their clinical presentations and topographic relations with the surrounding anatomic structures may be distorted, and more radical interventions are needed for a positive long-term prognosis [14]. e reccurences can appear in the remained parts of affected jaws, in the soft tissues, or in the transplanted bone, used for defect replacement. e recurrences can be detected within 1 to 10 years after primary surgery [15]. Recurrent ameloblastomas that manifested 15 to 30 years after the initial surgery were also reported [16]. Most of the cases analyzed in the literature present the solid type of t ameloblastoma (particularly the follicular variant) [17]. Possible predisposal for recurrence after surgical treatment is the histologic type, location of the lesion, and its penetration into the soft tissues via a destroyed cortical layer [18]. e surgical strategy, manifestation, and prognosis for recurrent ameloblastoma are discussed in numerous scientific publications [19][20][21]. Most reports present a very small series or single clinical cases, with a marked variety of clinical symptoms and individual characteristics, making the conclusions unconvincing [22]. e aim of the present study was to analyze clinical presentations, histological types, and treatment strategies of recurrent ameloblastoma and to define its disease-free survival (DFS) rate.

Materials and Methods
Twenty-four cases of recurrent ameloblastoma, treated at the Kyiv regional clinical hospital and the National Cancer Institute (Kyiv, Ukraine) from 1 January 2009 to 31 July 2021, were retrospectively analyzed. ese cases were set aside for further analysis from the cohort of 69 ameloblastoma patients with not less than three years of follow-up. e study was approved by the Bioethics Committee of Bogomolets National Medical University, Kyiv, Ukraine (protocol no. 107). e inclusion criteria were the following: histologically confirmed diagnosis of recurrent ameloblastoma with definite surgical treatment performed; welldocumented cases with clinical, radiological, and histological data. For each patient, the data concerning gender, age, personal history (alcohol, smoking, and drug addiction), size of the lesion, localization, patient complaints, clinical manifestation, radiographic appearance, surgical management, histological type, and complications were collected from medical records reviewed and analyzed retrospectively. All recurrent ameloblastoma cases were histologically confirmed and classified according to 2017 WHO classification [4]. If a mixture of patterns was observed in single ameloblastoma, the predominant pattern was considered for subclassification. In all cases, CT with 3-dimensional visualization was applied and carefully reviewed.
All patients enrolled in the study underwent conservative or radical surgical treatment depending on the size, location of the lesion, its clinical and radiological manifestations, and histological diagnosis. Conservative treatment included enucleation and extended bone curettage, while radical treatment consisted of bone resection (segmental or marginal) and hemimandiblectomy according to the recommendations of Hendra et al. [9].
Analysis of the results of the study was performed using the statistical software EZR v. 1.54 (graphical user interface for R statistical software version 4.0.3, R Foundation for Statistical Computing, Vienna, Austria) [23]. e mean value (Y) and standard error (m) were calculated for the quantitative data. Frequency (%) was calculated for the qualitative data. e survival analysis (disease-free survival rate) was performed by the Kaplan-Meier method. Risk ratios (HRs) with 95% confidence intervals (95% CI) were calculated for risk of disease recurrence analysis. To evaluate the effect of risk factors on disease-free survival (the calculation of adjusted HR), a Cox proportional hazard regression model was utilized. A stepwise method was used to select the independent factors of the multivariate models. A P value less than 0.05 was considered statistically significant.

Results
Within the cohort of 69 primary ameloblastoma in our study, the frequency of recurrence was 35%. In some patients, the recurrences develop even after secondary surgical interventions: Out of 24 patients with recurrent ameloblastoma, 35 episodes of recurrence were diagnosed (mean 1.45 ± 0.88 per patient). e recurrent ameloblastoma was presented mostly by conventional (solid/multicystic) ameloblastoma (n � 23) and slightly less frequently by unicystic ameloblastoma (n � 1) (p � 0.657). e following histological subtypes were identified as follows: 9 follicular (37.5%), 1 basaloid (4.1%), 2 plexiform (8.3%), 1 unicystic (4.1%), and 11 not specified cases (45.8%) (Figure 1). Twenty cases of recurrent ameloblastoma were detected in the mandible. e most frequent location for recurrence development in the mandible was the body (62.5%), then the angle (37.5%), and the ramus (33.3%). In 8 cases, recurrences developed when the primary tumor affected more than one anatomical area of the mandibular. In the maxilla, all recurrences were located at the premolar or molar region. e radiographic examination detected 16 cases of unilocular and eight multilocular patterns with dimensions ranging from 2 to 10.5 cm.
ere was a small male predominance among the patients 54% male vs. 46% female. e average age of the patients with recurrent cases was 41.08 ± 12.66 years, ranging from 15 to 62 years. e performed multifactorial analysis showed the absence of any kind of correlations between the clinical features, histological type of ameloblastoma as well, the applied treatment method, and the probability of recurrence in the evaluated cohort of patients. Journal of Oncology e DFS median in our study was 8 years (95% CI 6 years-10 years). Estimated 3-year DFS was 84.5 ± 4.8%, the 5-year DFS and 10-year DFS were 73.0 ± 6.3% and 43.9 ± 8.343.9 ± 8.3%, respectively ( Figure 2). However, the difference between DFS curves, defined for each group separately, in favor of radical treatment was not statistically significant in this cohort of patients (p � 0.172) (Table 1). e DFS median for conservatively treated patients was 8 years (95% CI 6 years-10 years). e DFS median for surgically treated patients was not reached in 13-year observation period (Figure 3). e radical treatment of recurrent ameloblastoma was applied to 14 patients and included the extended bone (and, if necessary, soft tissue) resection with a 1.5 cm clear margin around the radiologically determined borders. Secondary recurrence after such treatment was noted in four cases. In three of them, the secondary recurrence developed in the soft tissues of the infratemporal fossa with no connection to the bone. In one case , episodes of recurrence in soft tissues were noted over the observation period of 7 years, resulting in multiple surgeries. e other ten patients were treated conservatively. Secondary recurrence in these patients developed in four cases.

Discussion
Ameloblastoma is a benign neoplasm with aggressive behavior, local invasiveness, and a high recurrence rate [22]. Recurrent ameloblastoma is always a major challenge for the maxillofacial surgeon, as it requires additional extended resection in distorted topographic conditions. Such cases are usually associated with significant aesthetic and functional deficiency and require multistage reconstructive operations. In most cases, recurrent ameloblastoma is associated with incorrect treatment tactics and failure of first surgery [24][25][26][27][28][29][30][31][32]. Last meta-analyses demonstrated that the incidents of recurrence in ameloblastoma treatment are within the range of 15% to 29% [1, 6, 33, 34]. Laborde reported a higher recurrence rate of 40% [19]. e present study analyzed 69 patients treated over 13-year period. e recurrence rate after treatment of primary ameloblastoma was 35%. We did not find any statistically significant correlation between the risk of recurrence development and gender, age, type of ameloblastoma, location, radiological, or histopathological. e same results were presented in Almeida et al., systematic review and meta-analysis [6]. e pooled recurrence rate for solid ameloblastoma was 4.7% after radical and 32.8% after conservative treatment (n � 21). For unicystic ameloblastoma, these numbers were 25% after conservative treatment and no recurrence after radical surgeries. ese findings show that the solid or multicystic type behavior is slightly more aggressive than the unicystic type and, however, was not statistically significant in this study (p � 0.330). In our series, the conventional (solid) ameloblastoma was observed in 23of 24 of the recurrent cases, the follicular type being predominant (37.5%) for this group.  [2,6,34,35]. e systemic review by Reichart et al. demonstrated that the risk of recurrence correlated significantly with the histopathological type [36]. ese findings are the same as in the research of Hong, who also demonstrated a correlation between histopathology and recurrence rate [2]. However, the association between histologic patterns and recurrent ameloblastoma is still controversial. Au et al. found no association between the histological pattern and recurrence rate as well as the other authors came to the same conclusions [37].   According to Muller and Slootweg, most cases of recurrent ameloblastoma are diagnosed within 5 years after surgery [29]. In the earlier series by Olaitan et al., more than 80% of recurrent cases were diagnosed within the first 5 years after surgery, with the longest period before the manifestation of the recurrent ameloblastoma of more than 13 years [38]. In our study, 34% of cases were diagnosed in terms more than 5 years after surgery, with the mean period between surgery and the clinical or radiological manifestation of the recurrence 7.75 ± 1.75 years and the longest being 12 years. e expected DFS median in our study was 8 years (95% CI 6 years-10 years). e analysis of DFS with the use of the Kaplan-Meier method in our cohort of patients demonstrated comparable results of 3-year survival rate with similar studies of Yang et al. [39,40]. However, 5-and 10year disease-free results of abovementioned studies differed from those obtained in our cohort of patients. e DFS rate, defined in our study decreased dramatically comparing 5and 10-year recurrent free results. is suggested that the results were dependent on underestimated conditions because the main clinical and pathological conditions of our cohort were similar to those shown by Yang et al. [39].
According to the literature, recurrence can occur in remote postoperative periods. Adebayo et al., Hayward, Collings, and Harrison have reported recurrence after 21, 30, and 49 years, respectively [41][42][43]. A possible source of recurrence is the remaining cells at the osteotomy site or the retained soft tissue ameloblastoma islands during the surgical procedures, especially in complex regions like the infratemporal fossa. e morphological study of Carlson and Marx uses radiograph samples in 82 cases after resection to demonstrate that the ameloblastoma histologically extends beyond 2-8 mm, and the average spread of it to the bone is 4.5 mm [21]. It proves the necessity of extended resections of 1.5-2 cm margins around the radiologically determined ameloblastoma borders. Additionally, the worst prognosis in recurrent ameloblastoma is associated with the spread of the lesion outside the bone with the involvement of the soft tissues [17]. Such cases cannot be seen on the CT, and MRI is often required to plan the extent of resection. In the present study, 14% of the recurrences were asymptomatic and revealed by X-ray or CT examination. It proves the importance of long-term radiologic follow-up for early diagnostics and adequate treatment of recurrent cases.
In our study, three recurrent ameloblastomas occurred exceptionally in the soft tissues, after the spread of the primary neoplasm outside the bone. In these 3 cases, the recurrence developed in the soft tissues of the infratemporal fossa with no connection to the bone. However, the frequency of ameloblastoma that recur in the soft tissues after radical treatment can be more significant. In the series of 26 cases of recurrent ameloblastoma reported by Olaitan et al., the tumor spread into the soft tissues in 4 cases (15%) [38]. In a later multicenter analysis by Arotiba et al., 23.3% of 30 cases of recurrent ameloblastoma were located in the soft tissues [44]. Yang et al. defined the perforation of the cortical bone as a risk factor for recurrence [39]. At the same time, Eckardt et al. have reported recurrences in bone grafts after the long-term follow-up [31,[45][46][47][48]. In our study, only one recurrence developed in patients with bone grafting procedures after radical jaw resections.
e main limitation of our study was the relatively restricted number of well-documented cases with a predominance of conservatively treated ones. Prevalence of more conservative treatment resulted in preference for less complicated surgery, associated with a low rate of postoperative complications and aesthetic and functional deficiencies that can influence the patient's quality of life. At the same time, the radical approach often requires several complex and expansive surgical procedures with an unclear prognosis.

Conclusions
Results obtained in the present retrospective study proved the necessity of long-term follow-up after both conservative and radical treatment approaches. e DFS median in our study was 8 years (95% CI 6 years-10 years). For recurrent cases, radical resection with margins should consider the method of choice to avoid secondary recurrence, with MRI determination of the ameloblastoma border within the soft tissues.

Data Availability
e data used to support the findings of this study are included in the article.

Conflicts of Interest
e authors declare that they have no conflicts of interest.   Journal of Oncology