Open-field fluorescence guided imaging of ameloblastoma tumors in mouse patient-derived xenograft model

Abstract

Purpose

The aim of our study is to use near infrared fluorescence imaging to demonstrate the localization of cetuximab-IRDye800 to ameloblastoma in vivo.

Methods

Ameloblastoma tumor tissue from three patients was implanted subcutaneously into athymic nude mice to form patient-derived xenografts (PDX). Animals were randomly divided into two groups; one received the epidermal growth factor receptor (EGFR) antibody, cetuximab-IRDye800, and the other received the control antibody, IgG-IRDye800. After resection of the overlying skin, the tumors were imaged on the LUNA device, an open-field, intraoperative device. Tumor/background ratios (TBRs) were calculated and statistically analyzed using a paired t test.

Results

Imaging of PDX tumors revealed the TBRs produced by cetuximab-IRDye800 (AB-20, 4.5 ± 1.57; AB-33, 1.17 ± 0.42; AB-34, 1.33 ± 0.04) were significantly higher than those produced by IgG-IRDye800 (AB-20, 2.02 ± 0.29; AB-33, 1.07 ± 0.30; AB-34, 1.11 ± 0.13; p values, AB-20, p < 0.021; AB-33; p < 0.045; AB-34; p < 0.017). Excised PDX tissues were paraffin-embedded to confirm the presence of tumor by H&E staining and EGFR expression.

Conclusion

Fluorescently labeled anti-EGFR demonstrates specificity and sensitivity for PDX tumor xenografts using an open-field, near-infrared imaging system. These open-field devices allow real-time clinical assessment of fluorescent signals during surgery and may provide future benefit in the removal of ameloblastoma tumors.

Introduction

Ameloblastoma is the most common invasive odontogenic neoplasm, with a histologic appearance mimicking basal cell carcinoma. Treatment remains largely surgical, leading to jaw and facial deformity with concomitant speech, chewing, and swallowing disturbances. Ameloblastoma is divided into three subtypes: conventional (formally known as multicystic or solid) and unicystic that occur in the jaws, and peripheral ameloblastoma that occurs in the soft tissue [1, 2]. The conventional type of ameloblastoma is more aggressive and invasive compared to the unicystic type [3]. Ameloblastomas within the posterior maxilla can be particularly difficult to resect due to proximity to the skull base, complex anatomy of the paranasal sinuses, and difficult transoral access. Difficulties in intraoperative identification of tumor margins and expansion into the jawbone limiting palpation as a guide leavinge surgeons dependent on visualization or pre-operative imaging to estimate tumor extent and select margins [4, 5]. Extending resection margins by 1 cm beyond the radiographically evident tumor edge may or may not be curative and is associated with a high rate of recurrence, however, over-resection leads to facial deformity and impaired function [6–10]. The gold standard for ensuring negative histologic surgical margins has been gross inspection, palpation, and plain film radiographs of the excised bone intraoperatively. Frozen section assessment is not practical for mineralized tissues and is not available in a number of centers, in addition to adding operative time.

Capable of detecting known tumor markers, the use of subtherapeutic doses of FDA-approved antibodies conjugated to optical probes is gaining traction as a patient- and tumor-specific method of identifying surgical margins intraoperatively [4, 11–14]. By targeting known antibody tumor markers, optical probe-based surgical navigation demonstrates high sensitivity and specificity for distinguishing tumor tissue from surrounding normal bone and soft tissue. Two FDA-approved antibodies targeting epidermal growth factor receptor (EGFR), cetuximab and panitumumab, have been reliably conjugated to optical dye, IRDye800. They have been shown, in vivo, to lack toxicity in rodent models and limited toxicity in humans in clinical trials [4, 12–21].

Fluorescently conjugated anti-EGFR antibodies, cetuximab-IRDye800 and panitumumab-IRDye800, are currently in clinical trials and have been shown to effectively localize to human tumors, including squamous cell carcinomas (SCC) and gliomas [4, 12, 13, 22–24]. In multiple preclinical models, research has shown that cetuximab-IRDye800 and or panitumumab-IRDye800 can successfully label HNSCC and identify tumor margins [24, 25]. Fluorescent detection of the optical probe used in these clinical trials can be achieved using an FDA-approved intraoperative imaging device, the SPY imaging system. In clinical trials, the various SPY imaging systems have been shown to be effective and safe in imaging tissue perfusion and transfer circulation in free flap reconstruction, gastrointestinal imaging, and imaging blood flow among others [26, 27].

Fluorescently imaging with cetuximab-IRDye800 allows specific targeting of EGFR-expressing tumors and can be used for clinical detection of tumors and determination of margins [23]. We and others have shown that the majority of ameloblastomas express EGFR [3, 25, 28–30]. We previously demonstrated the ability of cetuximab-IRDye800 to label and detect ameloblastoma in vivo using patient-derived xenograft (PDX) models using a closed-field device appropriate for imaging mice or resected samples [30]. For this study, we hypothesize that expression of EGFR in ameloblastoma can be used to label ameloblastoma tumor cells in vivo fluorescently and can be detected using the intraoperative, open-filed SPY Luna imaging device (Stryker, Kalamazoo MI), which better mimics the clinical setting.

Materials and methods

Tumor specimens and preparation

The study was independently approved by the University of Alabama at Birmingham (UAB) Institutional Review Board. Informed consent was obtained from all individual participants included in the study. For this type of study consent for publication is not required. 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. Tissue samples were collected from six patients with ameloblastoma undergoing surgery at UAB. All the patients had conventional ameloblastoma with mandibular bone destruction. Tumor-containing tissue samples were collected by ablative surgeons from areas not needed in determining the tumor margins. A portion of the tissue was frozen, sectioned, H&E stained and analyzed by a board-certified pathologist to confirm the presence of the tumor. For three patients, adjacent tissue samples were dissected into 2 × 2 mm pieces, rinsed in DMEM/F-12 media enriched with penicillin, streptomycin and amphotericin (ThermoFisher Waltham, MA), and implanted to establish the mouse models. The remainder of the surgical specimens were processed by the pathology department in accordance with the standard of care for surgical pathologic specimens.

Animal model

Three- to six-week-old athymic mice (Envigo, East Millstone, New Jersey) were obtained and housed in accordance with UAB Animal Resource Program and ARRIVE guidelines. All procedures and experiments were monitored and approved by UAB Institutional Animal Care and Use Committee (IACUC-22185). All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. Given the limited size of the samples, sex of mice was matched to the sex of the patient. For this study, male mice were used. Mice were anesthetized with isoflurane, dosed with carprofen preoperatively, and incision sites were prepared with alternating betadine and 70% ethanol three times. An incision was made through the left flank skin of the mice, and the tissue was implanted between the skin and leg muscle by blunt dissection. Tissue specimens were placed in the incision and then closed with staples. PDXs were permitted to implant for up to six weeks before treatment and imaging. The inclusion and exclusion criteria were the presence of a palpable tumor prior to injection.

Following the establishment of the animal models, mice were randomly selected into groups receiving fluorescently labeled anti-EGFR antibody (cetuximab-IRDye800, n = 8 for AB-20 and n = 5 for AB-33 and AB-34) or a control non-specific antibody (IgG-IRDye800, n = 4 for AB-20 and n = 5 for AB-33 and AB-34). Antibodies were conjugated to IRDye800 (Li-COR Biosciences Lincoln, NE) as previously described [17, 31–33]. Animals received an intravenous (tail vein) injection of 200ug of cetuximab-IRDye800 or IgG-IRDye800 [17]. On day 14 post-antibody injections, a skin flap overlying the tumor was removed and animals were imaged. Unconscious mice, anesthetized with isoflurane, were then euthanized using CO² and cervical dislocation according to IACUC guidelines. The implanted tumors were resected and analyzed histopathologically.

Fluorescent imaging and measurement

Fourteen days after injection with cetuximab-IRDye800 or IgG-IRDye800, the mice underwent imaging using the SPY Luna Imaging system, an open-field, near-infrared device to collect fluorescence images. The Luna system captures fluorescent light using a video camera at a rate of 30 frames/second and displays it on a computer monitor, allowing visualization of images in real time [17]. During image acquisition, video of the specimen in the field of view was collected for each experimental condition. DICOM files were exported from Luna and imported onto the SPY-Q Analysis software for quantitative analysis. In this program, relative fluorescence units (RFUs) could be extracted from the tumor and background regions of interest (ROI). The background ROI was an area of normal soft tissue adjacent to the tumor. The tumor ROI was then normalized by dividing it by the background ROI to obtain a tumor-to-background ratio (TBR) [17, 20]. Each analysis was done by two independent researchers No blinding of analysis was done. Animals were excluded from the analysis if no tumor was found in the PDX.

Hematoxylin and eosin and immunohistochemistry

Tissues were fixed in formalin and paraffin embedded (FFPE), sectioned and mounted onto slides (5 μm), deparaffinized, rehydrated, and stained with hematoxylin and eosin (H&E) for confirmation of tumor by a board-certified pathologist (N.J.D.). Immunohistochemistry (IHC) of EGFR in primary patient tissues and PDX tissues was performed by our lab and the UAB Pathology Core Research Lab. Slides were cured overnight at 60° C, then de-paraffinized in 3 changes of xylene and hydrated using graded concentrations of ethanol and deionized water. The tissue sections were subjected to antigen retrieval by 0.01 M Tris-1mM EDTA buffer (pH 9). Following antigen retrieval, all sections were gently washed in deionized water, then transferred into 0.05 M Tris-based solution in 0.15 M NaCl with 0.1% v/v Triton-X-100, pH 7.6 (TBST). Endogenous peroxidase was blocked with 3% hydrogen peroxide for 15 min. To reduce further nonspecific background staining, slides were incubated with 5% normal goat serum (Sigma, G9023) for 60 min at RT. All xenograft tissue samples also had a 30 min blocking step with Mouse Blocker (Jackson Immuno Res Labs Inc., 115-007-003, 1/60 dilution) followed by 10 min of neutral Buffer Formalin NBF before TBST wash. All slides then were incubated at 4° C overnight with EGFR(A-10) (Santa Cruz, Dallas, TX, sc-373746, 1/50 dilution). After washing with TBST, sections were then incubated with the Goat Anti-Mouse IgG H&L secondary antibody conjugated with HRP (Abcam ab6789, 1/200). ImmPACT DAB Peroxidase (HRP) Substrate Kit (SK4105, Vector Laboratories) was used as the chromogen and hematoxylin (no. 7221, Richard-Allen Scientific, Kalamazoo, MI, ¼ dilution) as the counterstain. Negative control was processed with secondary antibody and no primary.

Immunohistochemistry H-score

EGFR expression was semi-quantitatively scored using the H-score method, calculated as:

• H score = 1× (% cells with weak staining) + 2× (% cells with moderate staining) + 3 × (% cells with strong staining) yielding a total possible range of 0-300. H-scores were independently calculated by two different pathologists (N.J.D. and S.J.) and reported as the mean.

Sample size

We obtained the following estimates for the TBR ratios for our power calculations from our initial data: a mean of 2.104 for cetuximab-IRDye800, a mean of 1.445 for IgG-IRDye800, and a common standard deviation of 0.313. Assuming these estimates, a significance level of 5%, and a two-sided two-group t-test, we have 80% power to detect differences of 0.659 and greater between the two groups with a sample size of 5 mice per group. Sample size calculations were performed using nQuery (version 8.5; Statistical Solutions, Boston, MA).

Statistical analysis

The cetuximab-IRDye800 and IgG-IRDye800 antibodies TBRs were compared using unpaired student t-tests. Statistical significance was considered at p < 0.05. Error bars represent standard deviation.

Results

Expression of EGFR in ameloblastoma tumors

EGFR was detected in the primary tumor from AB-20 and AB-33 (AB-34 primary was not available) and an additional three ameloblastoma patient samples (Fig. 1). Overall, the majority of patient tissues demonstrated moderate, homogeneous EGFR expression across ameloblastic epithelial islands, consistent with prior reports of EGFR upregulation in odontogenic epithelium. Tumor presence was verified on scanned images by the pathologists.

Fig. 1.

Ameloblastoma (AB) tumor confirmation in primary patient samples. These are H&E-stained slides and EGFR IHC slides of the primary patient samples. Patients 1 through 3 are additional primary samples of AB. PT = Patient; 20x, scale bar = 100 μm

Detection of cetuximab-IRDye800 in vivo

To generate the PDX model, we implanted surgically isolated ameloblastoma tumor samples from three patients (AB-20, AB-33, AB-34) into athymic mice (week 1, workflow shown in Fig. 2). Six weeks post-implantation animals were randomly assigned to groups receiving cetuximab-IRDye800 or IgG-IRDye800, as a non-specific control. No animals were excluded. To simulate surgical excision of AB, we removed skin from over the xenograft area and imaged using the SPY Imaging device (Fig. 3A-C). The TBRs produced by cetuximab-IRDye800 for these samples were significantly higher than those produced by IgG-IRDye800 (Fig. 3D, AB-20, p = 0.021; AB-33, p = 0.045; AB-34, p = 0.017). The TBRs for AB-20 tumors ranged from 2.42 to 6.21 (mean = 4.50) for cetuximab-IRDye800 treated animals (n = 5) and 1.70–2.40 (mean = 2.02) for IgG-IRDye800 treated (n = 5). For the AB-33 tumors, the cetuximab-IRDye800 treated TBRs were 1.56 − 2.10 (mean = 1.71, n = 4) and the IgG-IRDye800 TBRs were 0.93–1.67 (mean = 1.07, n = 5). For the AB-34 tumors, the cetuximab-IRDye800 treated TBRs were 1.28–1.37 (mean = 1.33, n = 4) and the IgG-IRDye800 TBRs were 0.92–1.29 (mean = 1.11, n = 5). Three mice were excluded as no ameloblastoma tumor tissue was found in the xenografts: one from the AB-20 cetuximab-IRDye800 group, one from the AB-33 cetuximab-IRDye800 group, and one from the AB-34 cetuximab-IRDye800 group. Despite these exclusions we still saw significant differences between the cetuximab-IRDye800 and IgG-Dye800 groups. However, the lower power does increase the risk of false negatives or inflated effects size.

Fig. 2.

Diagram of workflow created in BioRender. AB = Ameloblastoma

Fig. 3.

In vivo labeling of ameloblastoma tumor tissues pre-resection. A Diagram of the orientation of the Luna images (created with BioRender). B The skin flap over the tumor was moved and animals were imaged. C Representative images from the SPY Imaging for each ameloblastomas PDX model (AB-20, IgG-IRDye800, n = 5, cetux-IRDye800, n = 5; AB-33, IgG-IRDye800, n = 5, cetux-IRDye800, n = 4; AB-34, IgG-IRDye800, n = 5, cetux-IRDye800, n = 4). D Quantification of the SPY fluorescence imaging. The Tumor-to-Background (TBR) were significantly higher in the cetuximab-IRDye800-treated animals compared to the IgG-IRDye800-treated animals (*p < 0.05, **p<0.01). Error bars are standard deviation

Ex vivo histopathologic confirmation of ameloblastoma and EGFR expression

To validate our PDX model, tumors were resected on day 14 post-injection, paraffin-embedded, and stained with H&E. Microscopic examination of each explanted xenograft by a board-certified pathologist (N.J.D.) confirmed the presence of ameloblastoma in our models (Fig. 4). IHC of samples from each xenograft confirmed the expression of EGFR, the target of cetuximab-IRDye800, in our xenograft models. In untreated (IgG control) PDXs, EGFR expression remained moderate to high, with H-scores ranging from 135 to 200, confirming preservation of the EGFR phenotype from the parental tumors. Staining was uniform across viable tumor regions, suggesting stable receptor expression in vivo. Following cetuximab injection, PDX tumors exhibited a wide range in EGFR immunoreactivity, with H-scores from 70 to 200, depending on the sample (Fig. 5). These findings indicate that EGFR remained detectable in all treated tumors.

Fig. 4.

Ameloblastoma tumor confirmation in implants post-resection. Upon resection, tumor tissue is processed and paraffin-embedded, stained for hematoxylin and eosin (H&E), and used for pathologic assessment. Immunohistochemistry of EGFR is performed to validate the expression of the bio-target in the in vivo model. 20x, scale bar = 100 μm

Fig. 5.

Ameloblastoma EGFR H-scores. The overall H-score for each patient sample was determined by averaging the H-scores from two pathologists

Collectively, EGFR immunostaining confirmed consistent and moderate to strong expression in both primary and xenograft ameloblastomas, supporting the biological plausibility of EGFR targeted imaging and therapeutic strategies.

Discussion

In this study, we confirm the expression of EGFR in ameloblastoma tumor tissue in our PDX model and in primary patient samples. This is consistent with our group’s previous publications which identified similar levels of EGFR expression between HNSCC and ameloblastoma [25]. A recent study agreed with our findings that 100% of ameloblastomas expressed EGFR (n = 30) and the expression of EGFR was conserved after subcutaneous implantation [34]. In our samples we saw homogenous EGFR expression. There were differences in the H score and EGFR expression in primary and PDX tissues for AB-20. The primary sample had lower levels of EGFR compared to the PDX, but all samples did have EGFR expression. The expression of EGFR has been used in clinical trials to assist in the surgical delineation of tumor margins in HNSCC and we suggest the same technology could be applied to ameloblastoma [4, 22, 35]. We hypothesize that it could be used to effectively and accurately identify tumor margins intraoperatively and potentially improve surgical outcomes by precisely removing tumor tissue while sparing normal bone and soft tissue. Here we report that cetuximab-IRDye800 labeled ameloblastoma tumor cells in vivo, compared to the control antibody IgG-IRDye800. Cetuximab-IRDye800 had significantly higher TBRs compared to IgG-IRDye800. It has been shown in our previous studies that pre-resection imaging with the skin flap removed increases TBRs [25]. This skin flap resection combined with the real-time fluorescence imaging information from the Luna allowed for guided resection of the ameloblastoma tumor tissue from the athymic mice. This could potentially translate into clinical use as the same principle using an open-field device is applied for HNSCC in clinical trials with the Stryker SPY-PHI [14, 22, 30]. (NCT03200704, NCT04719156, NCT04511078)

Currently, there is an ongoing phase II clinical at UAB that is examining the use of panitumumab-IRDye800 in head and neck cancer (NCT04511078). They recently published their findings in laryngeal and oropharyngeal squamous cell carcinoma and the use of panitumumab-IRDye800 to improve the detection of both carcinomas intraoperatively. In both cases, the fluorescence provided an additional tool for the surgeon since the palpation examination was impossible with these types of surgeries [4, 22]. Similarly, tactile examination is not useful for ameloblastoma due to its intraosseous nature. Another group is examining the use of panitumumab-IRDye800 to map metastasis and sentinel lymph nodes in HNSCC, and there are several publications on the use of this agent in glioblastoma and conjunctival squamous cell carcinoma [24, 36, 37]. In these pre-clinical and clinical papers, the agent is used in conjunction with standard-of-care practices to improve tumor delineation. We hypothesize that the use of anti-EGFR conjugated IRDye800 antibodies, such as cetuximab-IRDye800 and panitumumab-IRDye800, can be used to localize and identify ameloblastoma tumor cells in an intraoperative setting.

Another translatable aspect of our studies is the possible treatment of ameloblastomas that occur in canines. Ameloblastoma, practically acanthomatous ameloblastoma, accounts for 45% of odontogenic tumors in canines and can be “relatively common” [38]. It has been shown that canine acanthomatous ameloblastoma shares homology with human ameloblastoma [38]. However, this study did not examine EGFR expression. If canine ameloblastomas also express EGFR, this could be an opportunity to improve surgical outcomes as, similar to humans, surgery is the primary therapy employed for removal of ameloblastoma in dogs [39].

The Luna SPY was originally developed by NOVDAQ but acquired by Stryker in 2017. The system received FDA approval in 2007 for use during reconstructive surgeries to examine perfusion using Indocyanine green (ICG) [40]. The Luna detects the fluorescence of IRDye800-conjugated antibodies and has been used to demonstrate the pre-clinical aspects of this imaging in HNSCC, breast cancer and melanoma [15–17, 19, 41]. However, in recent years, there has been the development of more specialized imaging devices like the Stryker 1788 Tower, Stryker SPY-PHI, and the endoscope of the XI Da Vinci robot. These devices can detect the fluorescence properties of IRDye800 conjugated antibody and have improved resolution and frame rates compared to the Luna. The ability of these devices to image EGFR-expressing tumors has been demonstrated in the clinical setting [4, 13, 22]. We hypothesize that the same technology and EGFR-conjugated antibodies could be used to detect ameloblastoma tumors, allowing intraoperative imaging and precise margin delineation, reducing surgical morbidity, and preserving normal tissue. In ameloblastoma, this is crucial because it is difficult to intraoperatively distinguish intrabony tumors from surrounding healthy bone and soft tissue. As mentioned previously the intraosseous nature of the neoplasm eliminates the ability to use the tactile exam, leaving only visual clues to the surgeons to help delineate healthy from neoplastic tissue. Frozen section analysis, not only infeasible for mineralized bone, is notably unreliable in bland or benign tumors such as ameloblastoma, with a high false negative rate. This facilitates the need for wide margins to be taken and often ameloblastoma still has reoccurrence rates of 15–25% in patients with radical treatment and 55–90% in cases that receive conservative treatment [5, 10, 26]. We suggest that the use of IRDye-800 conjugated antibodies can label ameloblastoma cells in patients and provide an additional tool for accurate tumor delineation to surgeons.

One limitation of this study is the use of a limited number of subcutaneously implanted tumor xenografts [25, 30]. Although a limited number of ameloblastoma patients were recruited, we detected significant localization of cetuximab-IRDye800 to ameloblastoma tumor tissue in vivo.

While we do only use PDX samples from three patients, we feel they do represent conventional ameloblastomas, which have been shown in multiple studies to express EGFR. There are significant logistical challenges which make ameloblastoma PDXs difficult to scale. We are limited to the original tissue samples available as ameloblastoma PDXs may be established in vivo, but do not grow and cannot be passaged into new mice. Given the rarity of the disease this can represent a significant time delay between samples and not every sample has adequate tissue for implantation. To date, no orthotopic models of ameloblastoma have been successfully established; however, this would be a logical next step in developing new treatments. Although subcutaneous PDXs provide useful models and proof-of-principle data, these models do not replicate the tumor environment as ameloblastomas occur within the bones of the jaw. Another possibility is using canine “patients” whose tumors would be in the appropriate microenvironment. This depends on the 91% homology of the human and canine EGFR receptors and the ability of cetuximab to bind to canine EGFR [42]. Canines with ameloblastoma could be given cetuximab-IRDye800 prior to surgery and the surgical specimen imaged ex vivo to confirm margin status. Using fluorescent imaging in a bone environment may be challenging as healthy bone tends to be autofluorescent [43]. However, this may not be the case with near- infrared (NIR) fluorescence like employed in our model. Studies using NIR imaging in bone have accurately identified prostate and osteosarcoma cells within the tibia of mice, glioblastoma intracranially, and breast cancer cells within the spine of mice [44–46]. However, the NIR agents used in the studies lack the specificity and targeting associated with panitumumab-IRDye800, which specifically binds the EGFR expressed by tumor cells providing an advantage over previous NIR agents used in vivo. Due to limited to tumor availability, we were unable to implant tumors pieces into both male and female mice. We matched the sex of the mouse to the sex of the patient, and in this study all patients and mice were male. This could limit the generalizability of our study; however, ours and other studies have reported that the majority of ameloblastomas, from both male and female patients, express EGFR, the target of our imaging approach [3, 25, 28–30]. Although animals were randomized into treatment groups the analysis was not blinded. This could be a potential source of bias within our study. The imaging time point was based on a previous report of being able to detect fluorescently labeled EGFR in PDXs through the skin of mice [19]. However, we found that we could only detect cetuximab-IRDye800 in xenografts with removal of the skin over the xenograft. We do not find this to be a limitation as the surgical site would be open for fluorescence detection. Other reports have shown detection of fluorescently labeled antibodies in mice within 48–96 h post-injection. This is more comparable to what would be used clinically. Another limitation in our study, as discussed previously, is the use of the LUNA Spy device rather than newer technologies such as the Stryker SPY-PHI. While newer technologies exist, the LUNA Spy device still provides data regarding localization of IRDye800 labeled antibodies and the proof of principle data we present here.

In conclusion, our group has demonstrated that EGFR is expressed in ameloblastoma tissue, and this expression can be used to label the cells with anti-EGFR conjugated antibodies for imaging. We discussed how these fluorescent agents are being used in both clinical trials and pre-clinical setting with other head and neck tumor types. We hypothesize with the use of intraoperative imaging devices we can leverage near infrared labeled EGFR antibodies to detect and treat invasive ameloblastoma.

Acknowledgments.

Authors’ contributions

Hope M. Amm – Conceptualization, Investigation, Data Curation, Methodology, Resources, Visualization, Supervision, Writing – original draft, review & editing, Funding acquisition; Beta C. Idigo – Investigation, Data Curation, Visualization, Writing – original draft; Logan D. Stone – Investigation, Data Curation, Visualization, Writing – original draft; Revati Suryawanshi – Investigation, Data Curation, Writing – review & editing; Sherin James – Pathologic analysis, Investigation, Writing – review & editing; Nisha J. D’Silva – Pathologic analysis, Investigation, Writing – review & editing; Eben L. Rosenthal – Conceptualization, Methodology, Writing – review & editing; Anthony B. Morlandt – Patient care, Conceptualization, Methodology, Writing – original draft, review & editing, Funding acquisition.

Funding

This study represents independent research supported by the National Institute of Health (NIH) and the Oral Maxillofacial Surgery Foundation. The work was supported by the National Institute of Dental & Craniofacial Research (NIDCR, K99/R00-DE023826, R21-DE029603, T90-DE022736) and the Oral Maxillofacial Surgery Foundation (Research Support Grant 2015). Imaging was conducted in the UAB Small Animal Imaging Facility and supported by the National Cancer Institute (NCI, P30CA013148) and the O’Neal Comprehensive Cancer Center. No sponsor was involved in the study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.

Data availability

Data is available upon request from corresponding authors.

Declarations

Ethics approval and consent to participate

The study was independently approved by the University of Alabama at Birmingham (UAB) Institutional Review Board. Informed consent was obtained from all individual participants included in the study. For this type of study consent for publication is not required. 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. Approval given, University of Alabama at Birmingham (UAB) Institutional Review Board, IRB-300001887 Approval given, UAB Institutional Animal Care and Use Committee, IACUC-22185.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.