Abstract
Purpose
Measure associations between clinicopathological and immunohistochemical human Mut-L homologue 1 (hMLH1) gene, and human Mut-L homologue 2 (hMSH2) genes, variables in recurrent AMBs.
Methods
This study consisted of a research retrospective, observational case-control study consisting of 22 cases of recurrent AMB and 22 non-recurrent cases. Cases of AMB with more than one year of follow-up were included in the study. Quantitative immunohistochemical analysis was performed considering the cellular location (nuclear) of the proteins studied. The McNemar test was used to compare variables between primary and recurrent AMBs. Recurrence-free survival was analyzed by the Kaplan-Meier method and survival functions were compared according to the variables using the log-rank test.
Results
The posterior mandible was the most affected site in the recurrent (n = 18, 81.8%) and non-recurrent groups (n = 16, 72.8%). Recurrence-free survival was 50.0 (34.5–63.6) months. The following factors were significantly associated with AMB recurrence: presence of cortical bone expansion (p = 0.01), absence of bone reconstruction (p = 0.02), conservative treatment (p = 0.02), loss of hMSH2 (p = 0.01) and hMLH1 (p = 0.04) immunoexpression, and strong Ki-67 immunoexpression (p = 0.03). The risk factors for AMB recurrence were anatomical location (OR = 3.31), locularity (OR = 1.07), cortical expansion (OR = 6.17), cortical perforation (OR = 2.10), bone resorption (OR = 1.52), tooth impaction (OR = 1.86), jaw reconstruction (OR = 6.92), and immunoexpression of hMSH2 (OR = 10.0) and hMLH1 (OR = 4.50).
Conclusion
Radiographic appearance, treatment modality, and immunoexpression of mismatch repair proteins can be used as predictors of AMB recurrence.
Supplementary Information
The online version contains supplementary material available at 10.1007/s12105-024-01743-1.
Keywords: Ameloblastoma, Recurrence, Prognosis, Immunohistochemistry, Mismatch repair proteins
Introduction
Ameloblastoma (AMB) is a benign epithelial odontogenic tumor that can be classified into conventional, unicystic, peripheral, and metastasizing variants [1, 2]. It is one of the most common odontogenic tumors in gnathic bones, with an incidence rate of 0.92 cases/million people per year and a peak incidence in the third decade of life. A slight male predilection and predominant involvement of the mandible have been reported. The follicular and plexiform types are the most prevalent morphological patterns [3]. Clinically, AMB can be classified as conventional, unicystic, peripheral, and adenoid. Conventional AMB is the most common type and predominantly affects the posterior region of the mandible. Unicystic AMB accounts for 5–22% of AMB and usually has a lower risk of recurrence. Peripheral AMB is an uncommon variant and has a lower recurrence rate. The latest classification by the World Health Organization (WHO) has included adenoid AMB, which is characterized by local infiltration and a high recurrence rate [1, 4, 5].
Clinically, AMB manifests as a painless, slow-growing swelling; however, the tumor is locally aggressive and infiltrative and postoperative recurrence rates are therefore high, ranging from 55 to 90% [6, 7]. When treated surgically, the risk of recurrence of AMB is significant, but lower recurrence rates have been observed with the use of radical therapies such as en bloc resection [3]. The infiltrative growth, therapeutic aggressiveness and high potential for recurrence of AMB are factors that contribute to the increased morbidity of patients with this tumor. AMB often compromises function and esthetics, resulting in facial deformities and dysfunctions [8].
Mutations and epigenetic events can lead to the development of pathological conditions [9]. There is a complex group of proteins organized in the so-called DNA mismatch repair (MMR) system, which are responsible for identifying and correcting mismatches and small deletion and insertion loops in DNA bases in order to prevent genetic mutations or at least to reduce their frequency. The MMR system consists of hMutS, with its heterodimers hMSH2-hMSH6 (denominated hMutSα) and hHMSH2-hMSH3 (denominated hMutSβ), and hMutL, with its heterodimers hMLH1-hPMS2 (hMutLα), hMLH1-hPMS1 (hMutLβ), and hMLH1-hMLH3 (hMutLγ) [10]. Few studies have investigated the expression of these proteins in odontogenic lesions, with hMLH1 immunoreactivity being described in odontogenic keratocysts [11]. Studies investigating the association of the expression of MMR proteins with the pathogenesis and clinical characteristics of AMB found a correlation between the immunoexpression of hMSH2 and hMSH1 and tumor recurrence [9, 12].
Studies evaluating the risk factors for AMB recurrence are scarce [8, 12]. Furthermore, few studies have performed a more in-depth assessment of the immunoexpression of proteins related to the MMR system in AMB, especially regarding the recurrent behavior of this odontogenic tumor [9, 12]. Therefore, the purpose of this study was to analyze the association of immunohistochemical expression of hMLH1 and hMSH2 and clinicopathological characteristics with recurrence-free time in individuals with AMB. The specific aims of this study are (1) to identify clinical/imaging characteristics associated with recurrence; (2) to evaluate recurrence-free survival time in recurrent and non-recurrent cases; (3) to identify treatment characteristics associated with recurrence, and (4) to evaluate immunohistochemical characteristics associated with recurrence. Our hypothesis is that a relationship exists between AMB recurrence and changes in DNA repair proteins, initial conservative treatment, anatomical location, size, and histological variant.
Materials and methods
Study Design
The study was designed to answer the following question: Are changes in the immunoexpression of DNA repair proteins risk factors for ameloblastoma recurrence?
This is a retrospective, observational case-control study. The Ethics Committee of the Federal University of Rio Grande do Norte (UFRN), Natal, Brazil, approved the study (Opinion number 5.148.782). All cases were obtained from the Pathological Anatomy Service of the Oral Pathology Unit, Department of Dentistry, UFRN, between 1975 and 2022. A period of 10 years (120 months) was established for the follow-up of recurrences. For sample calculation, a 5% alpha error rate, 95% significance level, 80% test power, and a 1:1 ratio of cases to controls were considered. Cases of AMB that exhibited one or more recurrences were included in the study. Only the first recurrence of the tumor was computed using the date of the first diagnosis (according to the histopathological report), the date of the first recurrence (confirmed by the histopathological report), and date of last follow-up.
All cases of AMB that recurred within 120 months, that were followed up for more than 1 year at the oral and maxillofacial surgery and traumatology service, and that showed no syndromes were selected. Cases of AMB treated at the same service that did not exhibit any recurrence over 120 months and no pre-existing syndromes served as controls.
Cases with incomplete clinical information and cases with insufficient material for immunohistochemistry were excluded.
Variables
In this study, the DNA repair proteins (hMSH2 and hMSH1) were the predictor variables, which were dichotomized into high and low expression based on positivity scores. The recurrence of AMB was the outcome, classified as present (yes) or absent (no).
Tumor size, locularity, and treatment were used as confounding/intervening variables. Tumor size was dichotomized into ≤ 3 cm or > 3 cm. Locularity was categorized as unilocular or multilocular, a common radiographic presentation in AMB. The treatment modality was categorized as conservative (marsupialization or decompression) or radical (enucleation, partial or complete resection of the jaws).
Several covariates were obtained by retrospectively reviewing the charts. Age and sex were used to characterize the sample. The remaining covariates were dichotomized and analyzed statistically. The radiographic findings collected included cortical perforation, tooth resorption, and tooth impaction, all categorized as absent or present. This categorization was based on the fact that AMB often promotes cortical bone resorption and expansion since it is a locally aggressive tumor.
Regarding the histological pattern (solid and unicystic), AMBs were classified according to WHO 20221. The adenoid pattern was not included since it is a recently named entity (< 10 years) and the desmoplastic pattern because there were no cases of recurrence. Finally, since AMB is a locally destructive tumor, jaw reconstruction was dichotomized into present or absent.
Data Collection Methods
The specimens were cut into 5-µm-thick sections, stained with hematoxylin and eosin, and examined under a light microscope (Olympus CX31, Olympus Japan Co., Tokyo, Japan) in order to confirm the histopathological diagnosis of each tumor.
The histopathological patterns and characteristics were classified into plexiform, follicular, unicystic (luminal, intraluminal, and mural), and others (granular cells, basal cells, desmoplastic, and acanthomatous).1 The most prominent histological pattern was classified as the predominant type.
For immunohistochemical analysis, 3-µm-thick sections were mounted on slides coated with organosilane (3-aminopropyltriethoxysilane; Sigma Chemical Co., St. Louis, MO, USA). These sections were submitted to the dextran polymer-based immunoperoxidase method using anti-hMSH2 (G219-1129, 1:200, 60’; Pharmigen, San Diego, CA, USA) and anti-hMLH1 (G168-15, 1:200, 60’; Pharmigen, San Diego, CA, USA). Diaminobenzidine was applied as chromogen.
After processing of the histological sections and immunohistochemistry, each specimen was analyzed under a light microscope by an examiner who was previously trained by an experienced pathologist. The nuclear immunoexpression of hMSH2 and hMLH1 was analyzed using adaptations of the methods described by Amaral-Silva et al. [13] and Fernandes et al. [14] to obtain the positivity index (PI). This index is used to express the percentage of positive cells indicated by nuclear brown staining in each sample and is calculated as follows: PI = (number of immunopositive cells/total number of counted cells) × 100. The total number of counted cells was set at 1,000. First, the cases were analyzed at 100x magnification to identify the most representative fields. Next, representative histological fields of each case were photographed at 400x magnification under a DFC345 photomicroscope (Leica DMR, Leica Microsystems, Wetzlar, Germany). Immunostained cells were counted using the Image J® software (version 1.51c; National Institutes of Health, Bethesda, MD, USA).
The following scores were assigned based on the percentage of positive cells per case: 0 (≤ 10% of positive cells), 1 (11–50% of positive cells), 2 (51–80% of positive cells), and 3 (> 80% of positive cells). Staining for hMSH2 and hMLH1 was then recategorized into weak (scores 1 and 2) and strong (score 3) according to a method adapted from Fernandes et al. [14].
Tissues from lymphoid tissues were employed as positive controls for hMSH2 and and hMLH1, the basal layer of normal oral mucosa was utilized. As negative controls, samples of ameloblastomas were used, with the primary antibody being substituted with 1% bovine serum albumin (BSA) in buffer solution (Supplementary material 1).
Data Analysis
The results were analyzed using the IBM SPSS Statistics freeware (version 20.0; IBM Corp., Armonk, NY, USA). Descriptive statistics were used for characterization of the sample. Normality and homoscedasticity of the data were analyzed by kurtosis and the Shapiro-Wilk test. The chi-square test and Fisher’s exact test were applied to evaluate the associations of the variables with recurrence. In the recurrent group, the variables were compared between primary tumors and tumors recurring at secondary sites using the McNemar test for paired data.
The Spearman correlation test was applied to compare the immunoexpression of hMSH2 and hMLH1 between recurrent and non-recurrent AMBs. To investigate the association of hMSH2 and hMLH1 with clinicopathological parameters, the results were categorized based on positivity scores and the chi-square test and Fisher’s exact test were applied. Ten-year (120 months) recurrence-free survival was analyzed by the Kaplan-Meier method and survival functions were compared according to the variables using the log-rank test, followed by Cox regression analysis. A level of statistical significance of 5% was adopted for all tests (p < 0.05).
Results
Clinical and Radiographic Variations
The sample consisted of 22 cases of recurrent AMB and 22 cases of non-recurrent AMB (control). Most patients were female (n = 24; 54.5%), with a female-to-male ratio of 1.2:1. The mean age at disease onset was 39.1 (± 19.8) years. There was a predominance of white patients (n = 20, 45.5%). Regarding anatomical location, the posterior mandible was the most affected site (n = 40, 90.9%), and a predominance of cases with a size greater than 3 cm (n = 39, 88.6%) (Table 1).
Table 1.
Comparison of clinicopathological and immunohistochemical features between recurrent primary tumors and non-recurrent cases
| Variable | Recurrent and non-recurrent cases | ||||
|---|---|---|---|---|---|
| Recurrent primary tumors n (%) |
Non-recurrent cases n (%) |
Odds ratio (95% CI) |
p | Time-to-recurrence (months) Mean ± SD |
|
| Total | 22 (100.0) | 22 (100.0) | 41.8 ± 22.9 | ||
| Anatomical location | |||||
| Maxilla | 1 (4.5) | 3 (13.6) | 3.31 (0.31–34.6) | 0.61b | 57.0 ± 0.0 |
| Mandible | 21 (95.5) | 19 (86.4) | 41.0 ± 23.2 | ||
| Lesion size | |||||
| ≤ 3 cm | 21 (95.5) | 18 (81.8) | 0.21 (0.02–2.09) | 0.34b | 63.0 ± 0.0 |
| > 3 cm | 1 (4.5) | 4 (18.2) | 40.8 ± 23.0 | ||
| Locularity #1 | |||||
| Unilocular | 7 (31.8) | 7 (31.8) | 1.07 (0.29–3.83) | 0.92a | 41.8 ± 24.1 |
| Multilocular | 15 (68.2) | 14 (63.6) | 41.8 ± 23.2 | ||
| Cortical expansion | |||||
| Absent | 5 (22.7) | 1 (4.5) | 6.17 (0.65–58.0) | 0.18b | 30.0 ± 16.3 |
| Present | 17 (77.3) | 21 (95.5) | 45.2 ± 23.8 | ||
| Cortical perforation | |||||
| Absent | 18 (81.8) | 15 (68.2) | 2.10 (0.51–8.57) | 0.30a | 37.8 ± 20.5 |
| Present | 4 (18.2) | 7 (31.8) | 59.5 ± 27.8 | ||
| Root resorption | |||||
| Absent | 16 (72.7) | 14 (63.6) | 1.52 (0.42–5.47) | 0.52a | 39.9 ± 20.8 |
| Present | 6 (27.3) | 8 (36.4) | 50.2 ± 33.3 | ||
| Tooth impaction | |||||
| Absent | 17 (77.3) | 19 (86.4) | 1.86 (0.38–8.98) | 0.70b | 44.9 ± 24.1 |
| Present | 5 (22.7) | 3 (13.6) | 31.2 ± 16.1 | ||
| Histological pattern | |||||
| Unicistic | 3 (13.6) | 5 (22.7) | 1.32 (0.69–2.50) | 0.69 b | 52.3 ± 15.1 |
| Solid | 19 (86.4) | 17 (77.3) | 40.1 ± 23.8 | ||
| Treatment modality | |||||
| Radical | 2 (9.10) | 9 (40.9) | 6.92 (1.28–37.2) | 0.01*a | 61.0 ± 5.65 |
| Conservative | 20 (90.9) | 13 (59.1) | 40.1 ± 23.4 | ||
| Reconstruction | |||||
| Absent | 20 (90.9) | 13 (59.1) | 6.92 (1.28–37.28) | 0.01*a | 39.9 ± 23.2 |
| Present | 2 (9.10) | 9 (40.9) | 62.5 ± 3.53 | ||
| hMSH2 | |||||
| High | 11 (50.0) | 20 (90.9) | 10.0 (1.87–53.4) | 0.01a* | 43.8 ± 21.8 |
| Low | 11 (50.0) | 2 (9.10) | 39.8 ± 24.9 | ||
| hMLH1 | |||||
| High | 12 (54.5) | 18 (81.8) | 3.75 (0.95–14.76) | 0.052a | 43.9 ± 25.9 |
| Low | 10 (45.5) | 4 (18.2) | 39.3 ± 19.7 | ||
a Pearson’s chi-square test; bFisher’s exact test. 95%CI: 95% confidence interval. #1 Missing case in non-recurrent tumors
About radiographic characteristics, the majority of cases presented a multilocular pattern (n = 29, 65.9%), presence of expansion of the cortical bone (n = 28, 63.6%), absence of cortical perforation (n = 33, 75.0%) and absence of root resorption (n = 30, 68.1%). Regarding the presence of impacted dental elements, there was a predominance (n = 36; 81.8%) of lesions that did not present impaction (Table 1).
The mean disease duration (from the time the patient noticed or knew about the existence of a lesion to diagnosis) was longer for non-recurrent cases (19.1 ± 14.4 months).
Comparison of Recurrent and Non-Recurrent Cases
Regarding treatment modality, in the recurrent group, conservative treatment predominated among primary tumors (90.9%; n = 20), while this percentage decreased to 54.5% in the group of recurrent tumors (n = 12), with a statistically significant difference between these two groups (p = 0.04). Most non-recurrent cases (n = 13; 59.1%) were submitted to conservative treatment, with the observation of a statistically significant difference compared to the group of recurrent AMBs (p = 0.01). No reconstruction technique was used in 90.9% (n = 20) of primary tumors, in 68.2% (n = 15) of recurrent tumors, and in 59.1% (n = 13) of non-recurrent AMBs. There was no statistically significant difference between recurrent AMBs (p = 0.12) but a statistically significant difference was found between this group and non-recurrent cases (p = 0.01) (Tables 1 and 2).
Table 2.
Comparison of clinicopathological and immunohistochemical features between the primary tumor and recurrent cases
| Variable | Recurrent cases | ||
|---|---|---|---|
| Primary tumor n (%) |
Recurrence n (%) | p | |
| Tumor size | |||
| > 3 cm | 1 (4.5) | 2 (9.10) | 1.00cc |
| ≤ 3 cm | 21 (95.5) | 20 (90.9) | |
| Locularity | |||
| Multilocular | 15 (68.2) | 16 (72.7) | 0.05c* |
| Unilocular | 7 (31.8) | 6 (27.3) | |
| Cortical expansion | |||
| Present | 17 (77.3) | 20 (90.9) | 0.37c |
| Absent | 5 (22.7) | 2 (9.10) | |
| Cortical perforation | |||
| Present | 4 (18.2) | 6 (27.3) | 0.75c |
| Absent | 18 (81.8) | 16 (72.7) | |
| Root resorption | |||
| Present | 6 (27.3) | 4 (18.2) | 0.69c |
| Absent | 16 (72.7) | 18 (81.8) | |
| Tooth impaction | |||
| Present | 5 (22.7) | 0 (0.0) | - |
| Absent | 17 (77.3) | 22 (100.0) | |
| Histological pattern | |||
| Plexiform | 6 (27.3) | 6 (27.3) | 1.00c |
| Follicular | 13 (59.1) | 13 (59.1) | |
| Unicystic | 3 (13.6) | 3 (13.6) | |
| Treatment modality | |||
| Conservative | 20 (90.9) | 12 (54.5) | 0.04*c |
| Radical | 2 (9.10) | 10 (45.5) | |
| Reconstruction | |||
| Present | 2 (9.10) | 7 (31.8) | 0.12c |
| Absent | 20 (90.9) | 15 (68.2) | |
| hMSH2 | |||
| Low | 11 (50.0) | 9 (40.9) | 0.39c |
| High | 11 (50.0) | 13 (59.1) | |
| hMLH1 | |||
| Low | 10 (45.5) | 11 (50.0) | 1.00c |
| High | 12 (54.5) | 11 (50.0) | |
cMcNemar test for paired samples between the first biopsy and recurrence
Statistical analysis of imaging and treatment characteristics revealed the following risk factors for AMB recurrence: anatomical location (OR = 3.31), locularity (OR = 1.07), cortical expansion (OR = 6.17), cortical perforation (OR = 2.10), bone resorption (OR = 1.52), tooth impaction (OR = 1.86), jaw reconstruction (OR = 6.92) and treatment modality (OR = 6.92). Tumor size (OR = 0.21) is protective factor, reducing the risk of tumor recurrence (Table 1).
The McNemar test for paired samples was used to compare the characteristics between recurrent cases (comparing primary and recurrent tumors) and non-recurrent cases. No statistically significant differences in the imaging or morphological variables studied were found between recurrent and non-recurrent cases (p ≥ 0.05) (Table 2).
Immunoexpression of hMSH2 and hMLH1
The immunoexpression of hMSH2 and hMLH1 in odontogenic epithelial cells was exclusively nuclear. In AMBs, there were no differences in immunostaining between the central and peripheral areas of nests/islands or cords of tumor cells (Fig. 1A − 1F). The immunoexpression results are presented in Tables 1 and 2. No statistically significant difference in hMSH2 or hMLH1 immunoexpression scores was observed between primary and recurrent tumors (p = 0.39).
Fig. 1.
Immunoexpression of hMSH2 and hMLH1 in the histological subtypes of ameloblastoma. hMSH2: follicular (A, 200x), plexiform (B, 200x), and unicystic (C, 200x). hMLH1: follicular (D, 200x), plexiform (E, 200x), and unicystic (F, 200x)
Comparison of the recurrent and non-recurrent groups showed strong expression of hMSH2 in 90.9% (n = 20) of cases, with a significant difference between recurrent and non-recurrent cases (p = 0.01). Immunoexpression of hMLH1 was strong in 54.5% (n = 12) of primary tumors, in 50% (n = 11) of recurrent tumors, and in 81.8% (n = 18) of non-recurrent cases. A statistically significant difference was observed between the recurrent and non-recurrent groups (p = 0.03) (Table 1). The immunomarkers used in this study were found to be risk factors for AMB recurrence: hMSH2 (OR = 10.0) and hMLH1 (OR = 3.75) (Table 1).
Analysis of Recurrence-Free Survival
Table 3 and Fig. 2 show the relationship between recurrence-free survival and clinical, morphological, radiographic, treatment, and immunohistochemical findings. The longest recurrence-free survival was observed in cases without cortical bone expansion (55.2%; 95% CI 38.2–69.3), cases submitted to radical treatment (81.8%; 95% CI 44.7–95.1), and when a reconstructive technique was used (81.8%; 95% CI 44.7–95.1). Regarding the immunohistochemical findings, the longest survival was found in cases with strong immunoexpression of hMSH2 (64.5; 95% CI 45.1–78.5), strong immunoexpression of hMLH1 (62.0; 95% CI 42.0–76.9).
Table 3.
Relationship between recurrence-free time and clinical, radiographic, morphological, treatment, and immunohistochemical findings
| n | Events (recurrence) | Recurrence-free survival (95% CI) | HR (95% CI) | Univariate (Log rank) | |
|---|---|---|---|---|---|
| Total | 44 | 22 | 50.0 (34.5–63.6) | ||
| Anatomical location | |||||
| Maxilla | 4 | 1 | 75.0 (12.7–96.0) | 2.63 (0.35–19.56) | 0.32 |
| Mandible | 40 | 21 | 47.5 (31.5–61.8) | ||
| Tumor size | |||||
| ≤ 3 cm | 39 | 1 | 80.0 (20.3–96.9) | 3.62 (0.48–27.01) | 0.18 |
| > 3 cm | 5 | 21 | 46.1 (30.1–60.7) | ||
| Locularity | |||||
| Unilocular | 14 | 7 | 51.8 (22.8–72.2) | 0.99 (0.89–1.10) | 0.69 |
| Multilocular#1 | 29 | 15 | 48.2 (29.4–64.7) | ||
| Cortical expansion | |||||
| Absent | 6 | 5 | 55.2 (38.2–69.3) | 3.56 (1.28–9.86) | 0.01* |
| Present | 38 | 17 | 16.6 (0.77–51.6) | ||
| Cortical perforation | |||||
| Absent | 33 | 18 | 45.4 (28.1–61.2) | 1.92 (0.65–5.69) | 0.23 |
| Present | 11 | 4 | 63.6 (29.6–84.5) | ||
| Root resorption | |||||
| Absent | 33 | 19 | 46.6 (28.3–63.0) | 1.32 (0.51–3.38) | 0.55 |
| Present | 11 | 9 | 57.1 (28.4–77.9) | ||
| Tooth impaction | |||||
| Absent | 36 | 17 | 52.7 (35.4–67.4) | 1.68 (0.61–4.59) | 0.30 |
| Present | 8 | 5 | 37.5 (8.70–67.4) | ||
| Histological patterm | |||||
| Unicystic | 8 | 3 | 62.5 (22.9–86.0) | 1.69 (0.50–5.72) | 0.38 |
| Solid | 36 | 19 | 47.2 (30.4–62.2) | ||
| Treatment modality | |||||
| Radical | 11 | 2 | 81.8 (44.7–95.1) | 4.70 (1.09–20.1) | 0.02* |
| Conservative | 33 | 20 | 39.3 (23.0–55.3) | ||
| Reconstruction | |||||
| Absent | 33 | 20 | 39.3 (23.0–55.3) | 4.86 (1.13–20.8) | 0.02* |
| Present | 11 | 2 | 81.8 (44.7–95.1) | ||
| hMSH2 | |||||
| High | 31 | 11 | 64.5 (45.1–78.5) | 3.95 (1.68–9.26) | 0.01* |
| Low | 13 | 11 | 15.3 (2.48–38.7) | ||
| hMLH1 | |||||
| High | 29 | 12 | 60.0 (40.4–74.9) | 2.37 (1.01–5.55) | 0.04* |
| Low | 15 | 10 | 28.5 (8.83–52.3) | ||
HR: Cox hazard ratio; 95% CI: 95% confidence interval. Log-rank p-value: * statistically significant results by the log-rank test. #1 One case is missing
Fig. 2.
Kaplan-Meier curves of recurrence-free survival of patients with ameloblastomas. (A) Survival curves according to the radiographic feature analyzed. (B and C) Survival curves according to treatment used. (D and E) Survival curves according to immunohistochemical markers. (F) Recurrence-free survival. *Statistically significant variables by the log-rank test
The mean time to recurrence was 41.8 ± 22.9 months (range 12 to 93 months). The total follow-up time of recurrent cases was 133.5 ± 98.5 months after recurrence (range 15 to 435 months). In non-recurrent cases, this period ranged from 15 to 70 months, with a mean of 29.9 ± 14.3 months. The recurrence-free survival (95% CI) was 50.0 (34.5–63.6) months.
As there were 22 recurrences, one covariate was selected for every 10 events (recurrences), thus a model with two covariates is accepted for the multivariate analysis (Table 4).
Table 4.
Cox proportional hazards model for multivariate analysis of Ameloblastomas recurrence
| Parameter | HR (95% CI) | HRa (95% CI) | p |
|---|---|---|---|
| Recurrence-free survival (10-years) | |||
| Treatment modality | 4.70 (1.09–20.1) | 5.11 (1.18–22.0) | 0.029 |
| hMLH1 | 2.37 (1.01–5.55) | 2.62 (1.10–6.24) | 0.029 |
Bold values indicate statistically signifcant results; HR: hazard ratio; HRa: adjusted hazard ratio; 95% CI: 95% confidence interval
Discussion
AMB is the most prevalent epithelial odontogenic tumor in epidemiological studies, which is characterized by aggressiveness and high potential for recurrence [15, 16]. Several studies have tried to explain the biological behavior of AMB and recent studies have investigated the impact of the MMR system on the pathogenesis and risk of recurrence of these tumors but knowledge is still very limited [1, 17, 18]. Here we investigated clinical-radiographic and immunohistochemical risk factors for the recurrence of AMB. The results showed that the loss of hMSH2 and hMLH1 protein expression, as well as cortical expansion, bone reconstruction and conservative treatment, were associated with the recurrence of AMBs.
In the evaluation of clinical-epidemiological characteristics, a slight predilection for females was found, corroborating the findings of another study [18] also carried out in northeastern Brazil. On the other hand, some studies indicate a predilection for males [3, 6]. This discrepancy can be explained by some factors, such as different geographic regions studied, access to health services and consequently to diagnosis and treatment, as well as biological factors, such as hormonal changes, which are common in women.
With regard to anatomical location, both for recurrent and non-recurrent cases, the highest occurrence was in the posterior region of the mandible, in line with findings in the literature [3, 19, 20]. Mandibular or posterior maxillary lesions require larger resection margins than tumors of the mandibular symphysis, due to the greater risk of local invasion and greater difficulty in redoing the surgical procedure [16], with the posterior location being a factor that directs to choose a treatment that reduces the chances of recurrence.
Regarding radiographic features, the predominance of the multilocular pattern in the present sample was the classical radiographic finding seen in most tumors [1, 17], showing no significant association with recurrence. Bi et al. (2021) and Au et al. (2019), in their findings, reported a higher recurrence rate in unilocular AMBs; in this context, the authors suggested that unilocular AMBs were treated more conservatively, which favored the occurrence of recurrence. This may be related to the perspective that unicystic AMBs had a more favorable prognosis, which is why they were often treated more conservatively [7, 8].
The statistically significant association between cortical bone expansion and recurrence observed in the present study is expected since expansive growth is common in AMB. Thinning of the bone cortex may occur, increasing the risk of perforation and invasion of adjacent soft tissues [17]. In a systematic review, McClary et al. [21] highlighted the importance of soft tissue involvement for recurrence. Extension of the tumor into adjacent soft tissues reduces the chance of complete excision, increasing the likelihood of future recurrence [21].
Therapeutic management of AMB is a complex issue since treatment should be the least destructive because of the benign nature of this tumor but, at the same time, must be sufficiently extensive to prevent subsequent recurrences [16, 17, 21]. Recurrence rates vary according to the chosen treatment modality [17]. In our study, the use of conservative treatment was more common in primary and recurrent tumors, while radical treatment was used more frequently in the non-recurrent group. A statistically significant difference was observed between the recurrent and non-recurrent groups. This finding is consistent with studies in the literature that also reported a statistically significant difference [13, 18].
A late diagnosis and spreading of the tumor to more than one site are associated with a poor prognosis of esthetic and functional rehabilitation. Therefore, more aggressive and radical methods for managing AMB have been advocated [22]. In the case of more radical treatment, surgical reconstruction techniques are used concomitantly or on a subsequent occasion since extensive tumor resection can cause facial deformities and dysfunctions or pathological fractures [23]. In the present study, there was a statistically significant difference in reconstruction after resection between recurrent and non-recurrent tumors, with a recurrence-free survival rate of 81% when reconstruction was performed.
Regarding the immunohistochemical markers, there was a statistically significant difference between the recurrent and non-recurrent groups, with weak expression of hMSH2 and hMLH1. An association was also found between recurrence and the loss of immunoexpression of MMR proteins.
Castrilli et al. [24] were the first to report nuclear immunoexpression of hMSH2 and hMLH1 proteins in 25 AMBs, especially in cells of the outermost layers of tumor nests and follicles. In the present study, there was no difference in cellular immunostaining between the inner and outer layers of follicles. Castrilli et al. [24] suggested that expression of these proteins is not related to the development or progression of these tumors. In contrast, Amaral-Silva et al. [13] observed significantly lower expression of hMSH2 and hMSH6 in AMBs compared to tooth germs. Another important finding was the statistically significant association between hMSH2 and hMSH3 and BRAF-V600E. Overexpression of hMutS proteins was associated with recurrence in ameloblastic carcinomas, without predicting disease-free survival of the patients. Opposite findings in terms of recurrence-free survival were obtained in the present study in which the survival rate was higher when MMR proteins were strongly expressed.
Bologna-Molina et al. [9] evaluated the expression of hMLH1 and hMSH2 in 40 solid AMBs, 40 unicystic AMBs, and 5 tooth germs. There was higher expression of hMLH1 and hMSH2 in AMBs. Another possible explanation for the findings of this study is that the decrease in the expression levels of hMLH1 and hMSH2 proteins is affected by microsatellite instability. The latter is defined as any change in the size of the repetitive DNA sequence as a result of insertions and deletions. Variations in miRNAs exist between malignant and normal cells [25].
As a strength of this study, we report important results that contribute to the understanding of the role of the MMR system (hMSH2 and hMLH1) in the biological behavior of AMB. These findings suggest that other events such as epigenetic mechanisms may be involved in the regulation of MMR gene. In some malignancies, altered expression of MMR proteins has been shown to be associated with clinicopathological variables such as tumor size, lymph node metastasis, clinical stage and resistance to chemotherapy [26, 27]. In agreement with these findings, we noted a significant association between hMSH2 and hMLH1 overexpression and disease recurrence. For Theocharis et al., (2011), lesions that present overexpression of these proteins may harbor other molecular events that would lead to a tumor with greater recurrence potential, stimulating the expression of MMR proteins as an attempt to repair other genetic errors, which may indicate events molecular [28].
Some limitations should also be mentioned. The data collected are prone to selection bias due to non-random sampling and the cross-sectional design of the study does not permit to infer the causality of the associations. It is also important to note that the results are influenced by the specificity of antibody immunoreactivity, the methods of immunostaining and immunohistochemical analysis, and the complexity of protein synthesis and degradation. For this reason, we decided to use a quantitative assessment in the present study, analyzing the proportion of immunostained cells.
In summary, the present results showed that cortical bone expansion, bone reconstruction, and conservative treatment were the clinical-radiographic features and treatment modality associated with the recurrence of AMBs. The loss of immunoexpression of hMSH2 and hMLH1 was associated with recurrence, suggesting that epigenetic events may influence the biological behavior of these tumors. Clinically, the results suggest that a more aggressive treatment strategy, particularly in cases with radiographic and clinical features indicative of a higher risk of recurrence, may be warranted. Furthermore, incorporating molecular markers such as hMSH2 and hMLH1 into the diagnostic and prognostic framework could increase the accuracy of treatment planning, ultimately improving patient outcomes.
Despite some important findings, this study has some limitations. The data collected are prone to selection bias due to non-random sampling and because the cross-sectional and retrospective design of the survey compromises the inference of causal associations. Therefore, longitudinal studies involving a larger sample and analyzing DNA mismatch repair as well as molecular techniques may be quite useful in the investigation of ameloblastoma recurrence.
Furthermore, the present results and future studies can contribute to the identification of other biological indicators of AMB, which will help to understand the clinical behavior of this tumor as well as to identify new potential therapeutic targets. Longitudinal studies with larger sample sizes and additional investigations comparing the recurrence predictive capacity with other markers (CD10, PCNA, COX2, BRAF V600E and FGFR2) [29–35] are necessary to obtain more accurate results.
Future research on ameloblastoma recurrence could harness the potential of advanced technologies, such as next-generation sequencing, to identify mutations associated with tumor progression and treatment resistance, complemented by analyses of gene expression and epigenetic alterations. The integration of multidimensional approaches, combining clinicopathological variables, DNA mismatch repair analysis, and genomic data, would enable the development of more robust predictive models and the identification of more reliable prognostic biomarkers.
Electronic Supplementary Material
Below is the link to the electronic supplementary material.
Acknowledgements
Acknowledgments: This study was supported by the Postgraduate Program in Oral Pathology of UFRN and by the Coordination for the Improvement of Higher Education Personnel (CAPES).
Author Contributions
HFP: conceptualization, data curation, formal analysis, investigation, methodology resources, software, validation, visualization, roles/writing—original draft, and writing—review and editing. GMF: data curation, formal analysis, visualization, and roles/writing—original draft. HGFM: data curation, formal analysis, visualization, and roles/writing—original draft. WRS: data curation, formal analysis, visualization, and roles/writing—original draft. RAF: formal analysis, investigation, methodology resources, visualization, and writing—review and editing. HCG: formal analysis, investigation, methodology resources, visualization, and writing—review and editing.
Funding
This study was not supported by any funding.
Data Availability
The data that support the findings of this study are available from the corresponding author, HFP, upon reasonable request.
Declarations
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 (Protocol No. 5,148,782) and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Consent to Participate
Informed consent was obtained from all individual participants included in the study.
Consent for Publication
Consent for publication was obtained for every individual person’s data included in the study.
Conflict of Interest
The authors declare that they have no conflict of interest.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Vered M, Wright JM (2022) Update from the 5th Edition of the World Health Organization Classification of Head and Neck tumors: odontogenic and maxillofacial bone tumours. Head Neck Pathol 16:63–75. 10.1007/s12105-021-01404-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Gültekin SE, Aziz R, Heydt C, Sengüven B, Zöller J, Safi AF, Kreppel M, Buettner R (2018) The landscape of genetic alterations in ameloblastomas relates to clinical features. Virchows Arch 472(5):807–814. 10.1007/s00428-018-2305-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hendra FN, Van Cann EM, Helder MN, Ruslin M, de Visscher JG, Forouzanfar T, Vet HCW (2020) Global incidence and profile of ameloblastoma: a systematic review and meta-analysis. Oral Dis 26(1):12–21. 10.1111/odi.13031 [DOI] [PubMed] [Google Scholar]
- 4.Kreppel M, Zöller J (2018) Ameloblastoma-Clinical, radiological, and therapeutic findings. Oral Dis 24(1–2):63–66. 10.1111/odi.12702 [DOI] [PubMed] [Google Scholar]
- 5.Effiom O, Ogundana O, Akinshipo A, Akintoye S (2017) Ameloblastoma: current etiopathological concepts and management. Oral Dis 24(3):307–316. 10.1111/odi.12646 [DOI] [PubMed] [Google Scholar]
- 6.Hendra FN, Natsir Kalla DS, Van Cann EM, de Vet HCW, Helder MN, Forouzanfar T (2019) Radical vs conservative treatment of intraosseous ameloblastoma: systematic review and meta-analysis. Oral Dis 25(7):1683–1696. 10.1111/odi.13014 [DOI] [PubMed] [Google Scholar]
- 7.Bi L, Wei D, Hong D, Wang J, Qian K, Wang H, Zhu H (2021) A retrospective study of 158 cases on the risk factors for recurrence in ameloblastoma. Int J Med Sci 18(14):3326–3332. 10.7150/ijms.61500 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Au SW, Li KY, Choi WS, Su YX (2019) Risk factors for recurrence of ameloblastoma: a long-term follow-up retrospective study. Int J Oral Maxillofac Surg 48(10):1300–1306. 10.1016/j.ijom.2019.04.008 [DOI] [PubMed] [Google Scholar]
- 9.Bologna-Molina R, Pereira-Prado V, Sanchez-Romero C, Tapia-Repetto G, Soria S, Hernandez M, Gonzalez-Gonzalez R, Molina-Frechero M, Mikami T (2018) Expression of hMLH1 and hMSH2 proteins in ameloblastomas and tooth germs. Med Oral Patol Oral Cir Bucal 23(2):e126–e131. 10.4317/medoral.22210 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Pećina-Šlaus N, Kafka A, Salamon I, Bukovac A (2020) Mismatch repair pathway, genome stability and cancer. Front Mol Biosci 26:7:122. 10.3389/fmolb.2020.00122 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Monteiro BBV, Cavalcante RB, Nogueira RLM, Miguel MMC, Nonaka CFW, Silveira ÉJD (2015) Participation of hMLH1, p63, and MDM2 proteins in the pathogenesis of syndromic and nonsyndromic keratocystic odontogenic tumors. Oral surgery, oral medicine. Oral Pathol Oral Radiol 120(1):52–57. 10.1016/j.oooo.2015.03.011 [DOI] [PubMed] [Google Scholar]
- 12.Yang R, Liu Z, Peng C, Cao W, Ji T (2017) Maxillary ameloblastoma: factors associated with risk of recurrence. Head Neck 39(5):996–1000. 10.1002/hed.24720 [DOI] [PubMed] [Google Scholar]
- 13.Amaral-Silva GK, Sánchez-Romero C, Wagner VP, Martins MD, Pontes HAR, Fregnani ER, Soares FA, Almeida OP, Rocha AC, Santos-Silva AR, Fonseca FP, Vargas PA (2017) Prognostic significance of hMSH2, hMSH3, and hMSH6 expression in ameloblastoma. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 124(3):286–295. 10.1016/j.oooo.2017.05.511 [DOI] [PubMed] [Google Scholar]
- 14.Fernandes GS, Girardi DM, Bernardes JPG, Fonseca FP, Fregnani ER (2018) Clinical benefit and radiological response with BRAF inhibitor in a patient with recurrent ameloblastoma harboring V600E mutation. BMC Cancer 18(1):887. 10.1186/s12885-018-4802-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Singh T, Wiesenfeld D, Clement J, Chandu A, Nastri A (2015) Ameloblastoma: demographic data and treatment outcomes from Melbourne, Australia. Aust Dent J 60(1):24–29. 10.1111/adj.12244 [DOI] [PubMed] [Google Scholar]
- 16.Laborde A, Nicot R, Wojcik T, Ferri J, Raoul G (2017) Ameloblastoma of the jaws: management and recurrence rate. Eur Ann Otorhinolaryngol Head Neck Dis 134(1):7–11. 10.1016/j.anorl.2016.09.004 [DOI] [PubMed] [Google Scholar]
- 17.Jaeger F, Saturnino M, Silva FB, Soraya M, Alencar E, Cássia M, Mesquita RA (2017) Prevalence profile of odontogenic cysts and tumors on Brazilian sample after the reclassification of odontogenic keratocyst. J Craniomaxillofac Surg 45(2):267–270. 10.1016/j.jcms.2016.12.011 [DOI] [PubMed] [Google Scholar]
- 18.de Medeiros W, da Silva L, Santos P, Pinto L, de Souza L (2018) Clinicopathological analysis of odontogenic tumors diseas 22 years period: Experience of a single center in northeastern Brazil. Med Oral Patol Oral Cir Bucal, 1;23(6):e664-e671. 10.4317/medoral.22618 [DOI] [PMC free article] [PubMed]
- 19.Aregbesola B, Soyele O, Effiom O, Gbotolorun O, Taiwo O, Amole I (2018) Odontogenic tumours in Nigeria: a multicentre study of 582 cases and review of the literature. Med Oral Patol Oral Cir Bucal 23(6):e761–e766. 10.4317/medoral.22473 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Dandriyal R, Lal V, Giri KY, Bavanthabettu NI, Chaurasia A, Pant S (2022) Ameloblastoma: Retrospective Study and Analysis of 102 cases over 10 years, single centre, institutional experience. J Oral Maxillofac Surg 21(2):730–738. 10.1007/s12663-022-01694-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.McClary AC, West RB, McClary AC, Pollack JR, Fischbein NJ, Holsinger CF, Sunwoo J, Colevas AD, Sirjani D (2015) Ameloblastoma: a clinical review and trends in management. Eur Arch Otorhinolaryngol 273(7):1649–1661. 10.1007/s00405-015-3631-8 [DOI] [PubMed] [Google Scholar]
- 22.Milman T, Ying G-S, Pan W, LiVolsi V (2016) Ameloblastoma: 25-year experience at a single institution. Head Neck Pathol 10(4):513–520. 10.1007/s12105-016-0734-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Al-Moraissi EA, Dahan AA, Alwadeai MS, Oginni FO, Al-Jamali J, Alkhutari AS, Al-Tairi NH, Almaweri AA, Al-Sanabani JS (2017) What surgical treatment has the lowest recurrence rate following the management of keratocystic odontogenic tumor? A large systematic review and meta-analysis. J Craniomaxillofac Surg 45(1):131–144. 10.1016/j.jcms.2016.10.013 [DOI] [PubMed] [Google Scholar]
- 24.Castrilli G, Piantelli M, Artese L, Perfetti G, Rubini C, Fioroni M, Ranelletti FO, Piattelli A (2001) Expression of hMSH2 and hMLH1 proteins of the human DNA mismatch repair system in ameloblastoma. J Oral Pathol Med 30(5):305–308. 10.1034/j.1600-0714.2001.300508.x [DOI] [PubMed] [Google Scholar]
- 25.Piantelli GCM, Artese L, Perfetti G, Rubini C, Fioroni MRFO, Piattelli A (2001) Expression of hMSH2 and hMLH1 proteins of the human DNA mismatch repair system in ameloblastoma. J Oral Pathol Med 30(5):305–308. 10.1034/j.1600-0714.2001.300508.x [DOI] [PubMed] [Google Scholar]
- 26.Kato M, Takano M, Miyamoto M, Sasaki N, Goto T, Tsuda H, Furuya K (2015) DNA mismatch repair-related protein loss as a prognostic factor in endometrial cancers. J Gynecol Oncol 26(1):40–45. 10.3802/jgo.2015.26.1.40 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Wagner VP, Webber LP, Salvadori G, Meurer L, Fonseca FP, Castilho RM, Squarize CH, Vargas PA, Martins MD (2016) Overexpression of MutSα Complex proteins predicts poor prognosis in oral squamous cell carcinoma. Medicine 95(22):e3725. 10.1097/MD.0000000000003725 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Theocharis S, Klijanienko J, Giaginis C, Rodriguez J, Jouffroy T, Girod A, Point D, Tsourouflis G, Sastre-Garau X (2011) Expression of DNA repair proteins, MSH2, MLH1 and MGMT in mobile tongue squamous cell carcinoma: associations with clinicopathological parameters and patients’. Survival J Oral Pathol Med 40(3):218–226. 10.1111/j.1600-0714.2010.00945.x [DOI] [PubMed] [Google Scholar]
- 29.Ahlem B, Wided A, Amani L, Nadia Z, Amira A, Faten F (2015) Study of Ki67 and CD10 expression as predictive factors of recurrence of ameloblastoma. Eur Ann Otorhinolaryngol Head Neck Dis 132(5):275–2799. 10.1016/j.anorl.2015.08.016 [DOI] [PubMed] [Google Scholar]
- 30.Kim J, Yook JI (1994) Immunohistochemical study on proliferating cell nuclear antigen expression in ameloblastomas. Eur J Cancer B Oral Oncol 30B(2):126–131. 10.1016/0964-1955(94)90064-7 [DOI] [PubMed] [Google Scholar]
- 31.Montezuma MAP, Fonseca FP, Benites BM, Soares CD, do, Amaral-Silva GK, de Almeida OP, Soares FA, Pagano RL, Fregnani ER (2018) COX-2 as a determinant of lower disease-free survival for patients affected by ameloblastoma. Pathol Res Pract, 214(6), 907–913. 10.1016/j.prp.2018.03.014 [DOI] [PubMed]
- 32.Fregnani ER, Perez DE, Paes de Almeida O, Fonseca FP, Soares FA, Castro-Junior G, Alves FA (2017) BRAF-V600E expression correlates with ameloblastoma aggressiveness. Histopathology 70(3):473–484. 10.1111/his.13095 [DOI] [PubMed] [Google Scholar]
- 33.Kurppa KJ, Catón J, Morgan PR, Ristimäki A, Ruhin B, Kellokoski J, Elenius K, Heikinheimo K (2014) High frequency of BRAF V600E mutations in ameloblastoma. J Pathol 232(5):492–498. 10.1002/path.4317 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Pereira NB, Pereira KM, Coura BP, Diniz MG, de Castro WH, Gomes CC, Gomez RS (2016) BRAFV600E mutation in the diagnosis of unicystic ameloblastoma. J Oral Pathol Med 45(10):780–785. 10.1111/jop.12443 [DOI] [PubMed] [Google Scholar]
- 35.Brown NA, Rolland D, McHugh JB, Weigelin HC, Zhao L, Lim MS, Elenitoba-Johnson KS, Betz BL (2014) Activating FGFR2-RAS-BRAF mutations in ameloblastoma. Clin Cancer Res 20(21):5517–5526. 10.1158/1078-0432 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from the corresponding author, HFP, upon reasonable request.