January 16, 2007

Esthesioneuroblastoma

Synonyms and related keywords: esthesioneuroblastoma, ENB, olfactory neuroblastoma, esthesioneuroepithelioma, olfactory esthesioneuroma, esthesioneurocytoma


INTRODUCTION

Background: Esthesioneuroblastoma (ENB) is an uncommon malignant neoplasm of the nasal vault, believed to arise from the olfactory epithelium. In 1924, Berger and Luc first described the tumor in the French medical literature under the name esthésioneuroépithéliome olfactif. Because of the uncertainty surrounding its precise histological origin, various names have been ascribed to this tumor, but the only 2 terms used in recent publications are esthesioneuroblastoma and olfactory neuroblastoma.

Beyond the most appropriate naming, multiple opinions exist regarding its origin, diagnosis, and management. The source of this controversy stems primarily from the rarity of this unusual tumor and the fact that almost no individual clinician or even institution will treat more than a few patients each year with this diagnosis. Three other factors contribute to the controversy surrounding this neoplasm. First, this tumor exhibits varying biologic activity ranging from indolent growth with patients surviving with known tumor more than 20 years, to a highly aggressive neoplasm capable of rapid widespread metastasis and patient survival limited to a few months. The second factor involves problems with precise histologic diagnosis since esthesioneuroblastoma is often confused with other small, round cell undifferentiated neoplasms of the nasal cavity. Third, no universally accepted staging system is available (see Staging).

Pathophysiology: The exact cell of origin of esthesioneuroblastoma has been controversial. While a neuronal-neural crest origin is supported by the presence of neurofilaments in ENB, until recently, few arguments linked ENB directly to the olfactory epithelium.

The olfactory epithelium is a peculiar neurosensory organ because dying olfactory neurons are replaced by new ones, which exhibit a progressive maturation, not only during embryogenesis but also physiologically and when injured by trauma or environmental insults. The globose basal cells constitute a precursor population and express neural cell adhesion molecule (NCAM) and proteins coded by the mammalian analogue of Drosophila achaete-scute (MASH) gene. These progenitor cells differentiate in olfactory neurosensory cells, which exhibit a progressive maturation from the basal membrane to the epithelial surface. Each layer can be characterized by specific olfactory- and neuron-specific markers. Immature olfactory cells express GAP43, a 24-kd membrane-associated protein kinase C involved in polyphosphoinositide turnover. As these cells mature, they migrate toward the surface, send axons to the olfactory bulb, and express olfactory marker protein (OMP) and NCAM, but not GAP43.

In the mid 1990s, ENB was found to express HASH, the human homologue of the MASH gene, while staining negative for other olfactory epithelium markers. Further indirect evidence that ENB originates from olfactory stem cells can be derived from transgenic mice in which the SV40T oncogene was inserted under the OMP gene promoter region: these mice did not develop ENB but adrenal and sympathetic ganglia neuroblastoma. Therefore, the currently available evidence links ENB with the basal progenitor cells of the olfactory epithelium.

Inclusion of ENB within the Ewing sarcoma family of tumors or the primitive neuroectodermal tumors (PNET) was proposed because of the identification, in certain cases, of translocation t(11:22), which is regarded as a specific molecular abnormality for Ewing sarcoma. Recent studies using fluorescent in situ hybridization and reverse transcriptase polymerase chain reaction (PCR) have failed to confirm this translocation in ENB. Therefore, ENB should be seen as a distinct entity from PNET and the Ewing sarcoma family of tumors.

Most of the olfactory neuroepithelium is located at the cribriform plate; however, islands of olfactory mucosa may be found in the upper turbinates and the upper one third of the nasal septum as well as on the entire middle turbinate. On rare occasions, olfactory mucosa has been found in the inferior turbinates and in the maxillary sinus. This probably explains why a small percentage of early-stage tumors appear to be completely free of the cribriform plate.

Frequency:

  • In the US: ENB remains a rare disease.
  • Internationally: In an extensive literature review, Broich et al found about 1000 new cases reported; however, several multiple publications on the same patients were included. Most cases (80%) were reported within the last 20 years. This is certainly the result of better recognition of this disease entity by pathologists, although the possibility of a rising incidence cannot be ruled out entirely.

    In view of the lack of precise epidemiologic studies, the authors' data suggest an incidence of 1 case per 1,000,000 per year. Similar incidence figures were recently obtained in Denmark. Thus, the authors tend to think that ENB represents about 5% of all nasal malignant tumors.

Race:

  • ENB does not show familial prevalence and has been reported in all races and on all continents.

Sex:

  • ENB affects males and females with similar frequency.

Age:

  • ENB occurs in all age groups.

CLINICAL

History: Esthesioneuroblastoma (ENB) does not cause specific symptoms, similarly to most other nasal and paranasal malignancies. The symptoms can be classified into nasal, facial, oral, ophthalmic, neurological, and cervical. The unilateral nature of symptoms should raise the suspicion of a possible neoplasia. Symptoms of nasal and paranasal cancers are as follows:

Physical: A nasal examination, particularly if aided by endoscopy, reveals a reddish-gray tumor arising in the upper nasal fossa, which bleeds easily during instrumentation. Although the aspect is strictly different from the white glistening appearance of benign nasal polyps, little differentiates ENB from other nasal malignancies.

Causes: No clear etiologic agent or exposure has been documented in humans. However, ENB can be consistently induced by nitrosamine compounds in rodents. In cats with spontaneous ENB or in transgenic mice developing ENB, type C retroviral particles have been demonstrated and classified as feline and murine leukemia virus, respectively. The role of retrovirus sequences in human ENB remains to be evaluated.


DIFFERENTIALS

Lymphoma, Non-Hodgkin
Malignant Melanoma
Metastatic Cancer, Unknown Primary Site
Plasmacytoma, Extramedullary


Other Problems to be Considered:

Nasal and paranasal squamous cell carcinoma
Sinonasal polyposis
Choanal polyp
Juvenile angiofibroma
Neuroendocrine carcinoma
Embryonal rhabdomyosarcoma
Undifferentiated sinonasal carcinoma
Ewing sarcoma


WORKUP

Lab Studies:

Imaging Studies:

Other Tests:

Procedures:

Histologic Findings: Well-differentiated ENBs exhibit homogenous small cells with uniform round-to-oval nuclei with rosette or pseudorosette formation and eosinophilic fibrillary intercellular background material. True rosettes (ie, Flexner-Wintersteiner [FW]) refer to a ring of columnar cells circumscribing a central oval-to-round space, which appears clear on traditional pathological sections. Pseudorosettes or Homer Wright (HW) rosettes are characterized by a looser arrangement and the presence of fibrillary material within the lumen. Fibrils have been shown by electron microscopy to represent cellular cytoplasmic processes.

In undifferentiated ENB characterized by anaplastic hyperchromatic small cells with numerous mitoses and scant cytoplasm, differentiation from other small-cell nasal neoplasms via light microscopy becomes difficult. Immunohistochemical staining and electron microscopy are essential for establishing the pathological diagnosis of sinonasal small cell neoplasms, which include malignant melanoma, embryonal rhabdomyosarcoma, malignant lymphoma, extramedullary plasmocytoma, and, especially, sinonasal undifferentiated carcinoma and sinonasal neuroendocrine carcinoma. No specific immunocytologic stain identifies ENB, but a typical staining profile can be very helpful. With ENB, the stain is positive for S-100 protein and/or neuron-specific enolase, while the stain usually is negative for cytokeratin, desmin, vimentin, actin, glial fibrillary acidic protein, UMB 45, and the common leukocytic antigen. For difficult cases, electron microscopy can be useful.

ENB can be distinguished from melanoma by the lack of UMB 45 immunopositivity and the pattern of S-100 staining, which is scattered and peripheral in ENB, but, in melanoma, the staining is diffuse and strong. ENB can be distinguished from rhabdomyosarcoma by the lack of cross-striated cells (ie, rhabdomyoblasts) and an absence of immunostaining with desmin, vimentin, and actin. ENB can be distinguished from lymphoma by the lack of common leukocyte antigen immunostaining.

ENB can be distinguished from sinonasal undifferentiated carcinoma through histology (ie, absence of numerous mitotic figures, areas of necrosis, vascular invasion, glandular differentiation) and immunohistochemistry (ie, lack of staining with cytokeratin antibodies). ENB can be distinguished from neuroendocrine sinonasal carcinoma by the lack of cytokeratin immunopositivity, which, in neuroendocrine carcinoma, tends to assume a punctate paranuclear distribution upon electron microscopy. ENB can be distinguished from Ewing sarcoma (rare within the paranasal sinuses) by the lack of myc-2 protein immunostaining.

In summary, the pathological distinction of poorly differentiated small neoplasms of the nasal cavity is difficult and is based on a panel of immunohistochemical stains and, if necessary, electron microscopy. The reactions used should include S-100 protein, neuron-specific enolase, chromogranin and/or synaptophysin, cytokeratin, desmin, actin, UMB 45, common leukocytic antigen, and myc-2 protein.

The demonstration of HASH gene expression, although still investigational, could become the diagnostic procedure of choice. Mhawech et al were able to distinguish ENB from other sinonasal, poorly differentiated tumors by real-time PCR assay of hASH1 mRNA levels. In addition, as an inverse association was noted between the grade of ENBs and hASH1 mRNA levels.

Histopathologic Grading According to Hyams (1988)

Grade Lobular Architecture Preservation Mitotic Index Nuclear Polymorphism Fibrillary Matrix Rosettes Necrosis
I + Zero NoneProminent HW rosettes None
II +Low Low Present HW rosettes None
III +/- Moderate ModerateLow FW rosettes Rare
IV +/-High HighAbsent None Frequent

Staging: From a limited series of 17 patients, Kadish et al were the first to propose a staging classification for ENB. Patients were divided into 3 categories: groups A, B, and C. Group A is limited to tumors of the nasal fossa; in group B, extension is to the paranasal sinuses; group C is defined as extension beyond the paranasal sinuses.

The Kadish classification has several shortcomings. Recognizing these inadequacies, a proposed classification is based on the Tumor Node Metastasis (TNM) system, which is predicated on CT scan and MRI findings that can be elucidated before treatment. Thus, patients treated by primary or preoperative radiation can be staged as reliably as surgical candidates. Although this classification system has gained popularity, attempts have been made to modify the Kadish system for ENB.


TREATMENT

Medical Care: The role of an accurate histopathological diagnosis before initiating treatment for ENB cannot be overemphasized. Classic treatment strategies of ENB are based on surgery or radiotherapy as unique modalities or a combination of surgery and radiation therapy. More recently, chemotherapy has been introduced in the therapeutic armamentarium. The optimal treatment sequence varies in different institutions. Because of the lack of any randomized trial comparing these treatment protocols, the available data can be summarized as follows:

Surgical Care: Endocranial extension and the close relationship to the ethmoidal roof and cribriform plate require a combined transfacial and neurosurgical approach. A craniotomy probably is not justified for T1 tumors, where clear radiological evidence of a normal cribriform plate and upper ethmoid cells exist. All other patients should be treated by a transfacial approach combined with a bifrontal craniotomy. In the authors' data, craniofacial resection resulted in much better local control than other surgical resections (ie, 100% vs 40%), and other series tend to support these findings.

Consultations: Usually, patients with head and neck tumors are discussed in university centers, within multidisciplinary tumor boards. Head and neck tumor boards usually include a head and neck surgeon and radiation and medical oncologists. The treatment protocol is decided in common, and each specialist performs his or her own specific task as described above.

Diet: No specific postoperative dietary restrictions are required for patients with ENB.

Activity:


MEDICATION

Routine chemotherapy for ENB is not recommended.


FOLLOW-UP

Further Inpatient Care:

Further Outpatient Care:

In/Out Patient Meds:

Complications:

Prognosis:


MISCELLANEOUS

Medical/Legal Pitfalls:


No comments: