Data Set for the Reporting of Carcinomas of the Nasal Cavity and Paranasal Sinuses: Explanations and Recommendations of the Guidelines From the International Collaboration on Cancer Reporting

Patients with locally advanced sinonasal carcinomas may be treated with preoperative chemotherapy or concurrent chemoradiotherapy protocols that may result in a significant improvement in survival in selected patients.^5–9  Such treatments may significantly alter the gross and microscopic appearance of the tumor and result in difficulties in tumor typing and grading. Although quantification of the extent of response is currently considered not relevant for clinical purposes, specimens should be extensively sampled, and changes presumably induced by treatment should be reported. Type of (chemo) therapy, number of cycles, interval between last cycle of chemotherapy, and local regional treatment initiation can also be annotated if available.

Different options are currently available for the surgical treatment of sinonasal malignancies, which may be chosen according to histopathology, extent of the lesion, and experience of the surgeon. Surgical approaches include craniofacial resections (open or endoscopic), endoscopic endonasal resections, and combined approaches.^10–12  This results in a wide range of surgical specimens submitted for histopathologic analysis (Figure 1).

Specimens from endoscopic surgery typically consist of fragmented material that should be properly labeled at the time of surgery, including a description of the anatomic site and type of tissue submitted (tumor, margin, or other). Because of the inherent orientation difficulty in samples, separately submitted margins, properly identified and labeled, are encouraged. Surgical resection specimens consist most often of the maxillary bone and adjacent anatomic structures removed according to the extent of the primary tumor (Figure 2).¹³ 

The sinonasal tract consists of the nasal cavity and the paired paranasal sinuses (maxillary, ethmoid, frontal, and sphenoid). The nasal cavity can be further subdivided into the nasal septum, floor, lateral wall, and vestibule. Among sinonasal tract carcinomas, the most common site of tumor origin is the maxillary sinus, followed by the nasal cavity and ethmoid sinus. It is rare for carcinomas to arise from the frontal or sphenoid sinuses.^14–18 

The precise tumor site within the sinonasal tract is important to record. This is not always easy to determine, because some exophytic tumors may almost entirely fill the nasal fossa without any infiltration, and only involve a relatively limited portion of the mucosa (Figure 3). Second, there is prognostic importance to the tumor location. For example, carcinomas primary to the nasal cavity have been shown to have an improved prognosis compared with carcinomas primary to the paranasal sinuses, likely because nasal carcinomas give rise to symptoms (eg, nasal obstruction or epistaxis) and reach clinical attention earlier in the disease course. Among maxillary sinus carcinomas, those arising from the anterior-inferior portion have a better prognosis than those arising from the superior-posterior portion, likely because the latter group more easily involves the skull base and/or orbit.^19,20  Interestingly, primary maxillary carcinomas have a behavior different from carcinomas originating in the oral cavity and secondarily involving the maxilla. Further, there are separate staging systems for the 2 major anatomic sites: maxillary sinus and nasal cavity and/or ethmoid sinus.⁴  Finally, certain carcinomas are closely associated with specific sinonasal subsites. For example, intestinal-type adenocarcinomas and neuroendocrine carcinomas occur most often in the ethmoid sinuses, whereas squamous cell carcinoma occurs most often in the maxillary sinuses.^21–23 

It is recognized that some carcinomas, particularly highly aggressive types like sinonasal undifferentiated carcinoma or nuclear protein in testis (NUT) carcinoma, usually affect more than 1 subsite. In this case, every affected site should be specified in the report.

The maximum diameter of the tumor should ideally be assessed on the unfixed specimen to avoid size underestimation resulting from formalin fixation–induced shrinkage. Care should be taken not to overestimate tumor size by including areas of adjacent nonneoplastic tissue. The gross assessment of tumor size should be confirmed microscopically, and in cases where nonneoplastic tissue has been mistakenly incorporated into the tumor measurement, tumor size should be adjusted accordingly. If tumor dimensions are estimated only microscopically, then “at least” should be added to indicate that the measurement is an underestimation resulting from fixation and tissue processing. The option “cannot be assessed” may be used when the tumor is submitted in fragments, as in endoscopic resections. In these cases, radiographic imaging may also be considered to determine tumor dimensions.

All sinonasal tract carcinomas should be given a type based on the most recent edition of the WHO's Classification of Head and Neck Tumours,²⁴  currently from the 4th edition (Table 1). Importantly, because of an effort to reduce duplication and only including an entity once in the book, not every histologic sinonasal tract tumor is described.

The sinonasal tract gives rise to a very large and diverse group of carcinomas, which may arise from the surface epithelium or the underlying seromucinous glands. Squamous cell carcinoma is, by far, the most common sinonasal malignancy, and it is subdivided primarily into keratinizing and nonkeratinizing subtypes (Figure 4, A). Additional subtypes (eg, spindle cell, basaloid, adenosquamous) are rare but should be noted if present. Uncommon carcinoma variants (such as sinonasal undifferentiated carcinoma, NUT carcinoma) are also included, whereas adenocarcinomas are separated into salivary gland types, intestinal types (Figure 5), and nonintestinal types, among other less common variants. Correct classification results in appropriate treatment, correlated with overall patient outcome. These classifications will take additional importance as targeted, molecular-based therapies become more widespread.^25–27 

A 3-tiered grading system based on degree of differentiation is used for squamous cell carcinoma, and it is also applied to some sinonasal adenocarcinomas, as well as selected salivary gland neoplasms (eg, adenoid cystic carcinoma, mucoepidermoid carcinoma, etc). Undifferentiated tumors should be considered high grade. Minor salivary gland neoplasms have grading systems unique to selected tumors,^28,29  whereas other adenocarcinomas are assigned a grade based on necrosis and mitotic activity.^30,31  Nearly all sinonasal tract neuroendocrine carcinomas are high grade.³² 

Bone invasion is a frequent finding in sinonasal carcinomas. Both bone erosion and destruction are reported, but sometimes findings from radiographic studies may need to be incorporated into this evaluation because limited biopsies may not include enough material to yield a meaningful interpretation. In resections, it is strongly recommended that histologic sections be taken from the areas of maximum bone involvement (after appropriate decalcification).

The frequency of perineural invasion (including named nerves) in sinonasal carcinomas is lower than it is in other head and neck sites, and it varies according to the histologic subtype, identified most frequently in adenoid cystic, sinonasal undifferentiated, squamous cell, and NUT carcinomas, respectively.^33,34  In sinonasal carcinomas, perineural invasion is associated with a high rate of positive margins, with maxillary origin, and with previous surgical treatment, but it is not an independent prognostic factor.³³ 

Lymphovascular invasion includes neoplastic cells within an endothelial-lined space, either lymphatic or venous, and it must be distinguished from retraction artifacts. Immunohistochemical staining for an endothelial marker may help in this distinction in selected cases. Lymphovascular invasion is reported in up to 60% of sinonasal squamous cell carcinomas, but its clinical significance remains elusive.³³ 

Complete tumor resection with negative surgical margins poses a significant challenge in sinonasal carcinomas, given the proximity to critical anatomic structures. The presence of residual microscopic disease has been reported with high frequency in cases managed both by open and by endoscopic surgical techniques.^35,36  In a large series of sinonasal squamous cell carcinomas treated with surgery, 16% had microscopic residual disease and 13% had macroscopic positive resection margins.³⁷  A negative surgical margin is associated with improved overall survival in retrospective studies for both open and endoscopic approaches.^37,38 

Ideally, the resection specimen should be oriented by the surgeon, including sutures or other annotations, best achieved by direct communication between the surgeon and pathologist, supplemented with specimen photography before and after sectioning. Failing this, the margins should be labeled by the surgeon and/or illustrated with a diagram. Specimens from endoscopic tumor resections should also be labeled and oriented, if possible. If the margins are sent separately, for frozen section or otherwise, identification of their site in relation to the resection specimen should be clarified by the surgeon. The surgical margins—mucosal, soft tissue, bone, and deep—should be thoroughly sampled, because additional therapy is often predicated on margin status. Because of limited data, a clear margin ranges from 3 to 7 mm, with 5 mm generally associated with a better prognosis,^39–43  although recurrence is seen in up to 25% of patients with a clear margin.^41,42  Depending on stage and other factors, a narrower margin may be adequate.^44,45  When the complex anatomy is taken into consideration, a “pushing” border into the periorbital tissues may not require orbital exenteration to achieve a 5 mm margin, if it is a low-grade neoplasm (Figure 6). Thus, margin status is one parameter of many used in treatment and prognostication.^46–48  Most studies also consider carcinoma in situ/high-grade dysplasia as a positive margin.⁴⁰  The presence of dysplasia at the margin is associated with a significant risk of local recurrence⁴⁹  and development of a second primary.⁵⁰  Thus, this information, including distance to the margin from invasive and in situ/dysplasia, is reported.

Staging of sinonasal cancer remains a difficult task because of the complexity of the anatomic site and the heterogeneity of the tumor types. Nevertheless, an accurate staging is essential for treatment and prognosis implications.⁵¹  The data set is based on the T category of the 8th edition of the Union for International Cancer Control staging system, which is identical to that of the previous edition (Table 2). Tumors located in the maxillary sinus are separated from those arising in the nasal cavity and ethmoid sinus. The pT stage is based on the involved anatomic sites, as well as invasion of the bone, orbit, dura, and other structures. Most sinonasal carcinomas present in advanced T category (T3 and T4). It is worth noting that a number of studies have reported significant discrepancies between clinical and pathological T categorization.^52,53  In particular, imaging is not sufficiently accurate in the assessment of invasion of the orbit, the skull base, and frontal or sphenoid sinuses.⁵²  Thus, careful histopathologic assessment is essential for a correct staging of sinonasal carcinomas. Finally, when neck lymph node dissections are included as part of the surgical specimen, a separate linked data set for Nodal Excisions and Neck Dissection Specimens for Head & Neck Tumours has to be completed.⁷⁰ 

Sinonasal carcinomas may present with multiple synchronous lesions located in different anatomic subsites.³⁶  Similar to other data sets for reporting carcinomas of the head and neck, multiple different histologic primaries are separately reported. The term “multifocal” may be used for microscopic foci of in situ or invasive carcinoma adjacent to the primary.

A number of histopathologic findings may be associated with sinonasal tract carcinomas. The presence of squamous dysplasia/carcinoma in situ can be used as evidence for histologic classification of the tumor. This is especially true with spindle cell carcinoma or other variants of squamous cell carcinoma that arise from and are often associated with overlying squamous dysplasia/carcinoma in situ.⁵⁴ 

It is well recognized that a subset of sinonasal squamous cell carcinomas, and less frequently other histologic types, originate from preexisting sinonasal papillomas (formerly Schneiderian papilloma). These tumors represent a group of sinonasal carcinomas characterized by specific genetic alterations and oncogenic mechanisms.^55–57 

Finally, foci of intestinal metaplasia of the adjacent sinonasal respiratory epithelium can be detected in about 25% of intestinal-type adenocarcinomas.^58,59 

Ancillary studies are variably needed for the diagnosis of specific entities at this site. For example, NUT carcinoma is recognized by the presence of nuclear protein in testis (NUTM1) gene rearrangement or positivity with the C52 monoclonal antibody against NUT protein.⁶⁰  The diagnosis of HPV-related multiphenotypic sinonasal carcinoma requires HPV-specific testing as part of the tumor definition (Figure 4, B),⁶¹  whereas for the diagnosis of SMARCB1 (INI1)–deficient carcinoma, loss of nuclear immunohistochemical staining for INI1 is required.^62,63 

In poorly differentiated malignancies, immunohistochemical markers can be used to assign a tumor to a specific category. Useful markers of squamous differentiation are p40, p63, and cytokeratin 5/6, whereas markers of intestinal differentiation, such as cytokeratin 20 and CDX2, help in the diagnosis of intestinal-type adenocarcinoma. Neuroendocrine carcinomas can be diagnosed with the support of positive reactivity with at least 1 neuroendocrine marker.

A subset of sinonasal carcinomas appear to be related to high-risk HPV, including nonkeratinizing squamous cell carcinoma, basaloid squamous cell carcinoma, papillary squamous cell carcinoma, adenosquamous carcinoma, and conventional keratinizing squamous cell carcinoma.^64–68  However, the clinical significance of these findings is still debated, and HPV testing is currently considered investigational in this context.

In summary, the contents of the ICCR data set for reporting carcinomas of the nasal cavity and paranasal sinuses were reviewed. In such a highly heterogeneous group of rare tumors, an internationally agreed-upon pathology data set is important to facilitate data collection and comparison, ensuring that all clinically relevant information is included.

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.