Data Set for the Reporting of Malignant Odontogenic Tumors: Explanations and Recommendations of the Guidelines From the International Collaboration on Cancer Reporting Open Access

As for all pathology specimens, the kind of specimen should be listed, including debulking or curettage, excisional or incisional biopsy, and surgical resection. Not specified is used only in rare instances and is to be discouraged. Instead, there should be active personal communication with the treating physicians to obtain the required information. In case of a resection, appropriate and concise terms to be used would be mandibulectomy or maxillectomy with a prefix indicating the extent (hemi-, partial, or total). In case of a neck dissection submitted together with a tumor specimen, a separate, linked data set for Nodal Excisions and Neck Dissection Specimens for Head & Neck Tumours³¹  would be used to describe the neck lymph node findings.

The tumor site belongs to standard items listed in agreement with general rules regarding pathology reports, irrespective of the kind of lesion. Giving the exact anatomic site of involvement is most helpful, a finding that may require correlation with clinical and imaging findings.

Owing to the intraosseous nature of odontogenic lesions, reference to any imaging studies or consultation with a radiologist is recommended in order to achieve the best interpretation of maximum tumor dimension, combining macroscopy, specimen or clinical radiology, and microscopy. Size criteria for possible staging have been suggested.¹  Using data from the National Cancer Institute's SEER (Surveillance, Epidemiology, and End Results) database for the period 1973–2011, Agarwal et al²  and Lee et al³  showed that size was an independent prognostic indicator with smaller tumor size significantly associated with a better overall survival. Patients with tumors greater than 41 mm were twice as likely to die of disease as patients with smaller lesions.²  The rationale for using both core and noncore is the need to use both pathologic and imaging data to determine the lesion's dimensions, with the latter potentially not easily available to the prosecting pathologist.

The gnathic bones give rise to a variety of neoplasms, including odontogenic and bone/cartilage forming tumors; however, this data set is designed to report only the malignant odontogenic tumors (carcinomas as well as sarcoma) as classified according to the most recent edition of the World Health Organization's (WHO) Classification of Head and Neck Tumours⁴  (Table). Bone and cartilage forming tumors are classified using a different data set.

For primary intraosseous squamous carcinoma, the conventional squamous cell carcinoma grade is used.^5,6  Although other malignant odontogenic tumors are generally not graded, the pathologist can still comment on the extent of tissue changes (for these non–squamous cell carcinoma tumors, histologic tumor grade is a noncore item). This especially applies to ameloblastic carcinoma, where there may be varying degrees of cytologic atypia (Figure 1).

Tumor necrosis is not only a tool to aid in tumor grading, but also the presence of necrosis often helps to confirm a diagnosis of malignancy in odontogenic tumors in general (Figure 2). Thus, while large clinical series of these rare tumors are not available, there is strong support that reporting tumor necrosis aids in diagnosis, grade, and tumor classification.^7,8 

Perineural invasion is included as a generic descriptor that is considered to be relevant for all types of malignant tumors (Figure 3). Note, however, that the extensive perineural spread seen in sclerosing odontogenic carcinoma does not appear to be a poor prognostic feature (Figures 4 and 5, A and B).^9–11 

Lymphovascular invasion is generally a feature seen in malignancy, and thus, it is included here, although data specific to malignant odontogenic tumors are not yet reported (Figure 6).

Margin status is considered a key prognostic factor. Clear margins are commonly achieved only with a thin layer of normal tissue between the tumor and the resection margin. It is therefore of crucial importance to assess whether the excision has an adequate margin around the entire lesion, since if there is focal margin positivity, reoperation may be necessary.

The prognosis is worse when an incomplete excision is located at the infratemporal fossa and/or base of skull and thus, reporting the specific anatomic site of involved margins must be clearly specified (Figure 7).

Extent of invasion is best assessed by a combination of radiographic, macroscopic, and microscopic features (Figure 8). Use of a diagram (Figure 9) is strongly recommended, as the pertinent anatomic landmarks can be easily identified and assessed in the embedding and sectioning protocol.

There are several immunohistochemical and molecular studies that may be clinically relevant,¹²  with some to have potential but unproven therapeutic benefit. Examples include EWSR1 rearrangements in clear cell odontogenic carcinoma and BRAF V600E point mutation in ameloblastic carcinoma.^13,14  Such tests may also increase diagnostic certainty. Thus, if any of these tests are performed, they should be recorded.

This data set is based almost exclusively on professional judgement owing to the lack of high-quality evidence to support individual data items. Malignant odontogenic tumors are rare and published series are often not homogeneous by tumor type, classification, stage, or treatment, making conclusions about the value of individual items difficult. In general, the tumors that show aggressive histologic features are more likely to be associated with poor survival, but this tumor group is characterized by unpredictable behavior; low-grade tumors may recur or metastasize many years after excision. For all types, large size, local recurrence, and metastasis are poor prognostic features^2,3,15,16  and outcomes are relatively poor after local recurrence.^17–19  Published mortality rates are generally limited by short follow-up. Ameloblastic carcinomas appear to carry a better prognosis than other tumor types for reasons that are still unclear,²  although maxillary tumors behave worse than mandibular neoplasms,²⁰  with up to one-third of maxillary tumors yielding lung metastases. Odontogenic sarcomas are overall of low grade and tend to locally recur rather than to metastasize, but are still associated with significant mortality rates^21–23  owing to local involvement of vital structures.

There are no validated grading systems for odontogenic tumors, although primary intraosseous squamous cell carcinomas are graded by conventional squamous cell carcinoma grading schemes, ostensibly showing some value.^5,6  Tumor necrosis is included as a histologic core feature but is also considered useful in confirming a diagnosis of malignancy in general. Perineural invasion is an included element, although it must be emphasized that in sclerosing odontogenic carcinoma, despite extensive perineural spread, this carcinoma nevertheless carries a relatively good prognosis.^9,10,11 

Margin status after surgical excision is thought to be the key prognostic feature^15,24–27  and the best evidence relates to ameloblastic carcinoma,^2,28  primary intraosseous carcinoma,^5,17  and clear cell carcinoma.²⁴  Surgical margin clearances may be very small or inadequate and extension into soft tissues beyond the periosteum is usually associated with a significant risk of local recurrence. The prognosis is worse when incomplete excision is seen at the infratemporal fossa and/or base of skull and therefore the anatomic site of involved margins must be clearly specified and marked on a diagram (Figure 9) during macroscopic and microscopic examination.

Tumor dimensions (size) and site are also important prognostic features. Further, carcinomas arising in or limited to cysts carry a better prognosis than those with widespread infiltration.²⁹ 

The role for adjuvant or salvage radiotherapy remains to be defined. The literature does not provide useful information on specific radiotherapy indications or the intent when it has been used. Despite its use to locally control incompletely excised malignant odontogenic tumors, its value appears limited,^24,25  although it has been supported in some large series²  and is usually considered most effective as part of a multimodality treatment approach. This is true particularly for maxillary lesions in which vital structures may prevent surgical management.

Ancillary studies still play a relatively limited role^12–14 ; they are sometimes helpful in diagnosis, while molecular findings may help to guide targeted therapeutic options. However, availability and cost constraints limit widespread adoption, especially in developing nations.

Finally, while ameloblastomas are benign, they often require an oncologic (surgical) approach to management, and thus, the panel recommends discussion of ameloblastomas at a multidisciplinary team meeting (tumor board) to ensure optimal management. Furthermore, this tumor would also benefit from a standardized reporting data set. While this data set was prepared for malignancies, the panel also strongly advocates its use for ameloblastoma.

The authors would like to express their appreciation to the sponsoring societies and organizations and give special thanks to Fleur Webster and Hannah B. Canlas for their exceptional organizational and editing contributions. The views expressed are those of the authors solely.