Olfactory neuroblastoma (ONB, esthesioneuroblastoma) is a rare malignant tumor of neuroectodermal origin. It arises from the olfactory epithelium that lines the superior one-third of the nasal septum, cribriform plate, and superior turbinates

Some good reviews: here, here, here, here, here 

See the review article below and a discussion of radiation here.

CLINICAL PRESENTATION AND DIAGNOSIS  ONB occurs over a broad age range (three to 88 years) with a mean age of 45 years. Approximately one-fifth of cases occur between the ages of 11 and 20. There is no gender predilection.

Signs and symptoms  The most common symptom is unilateral nasal congestion or obstruction. In one series of 22 patients, the most common symptoms were nasal congestion (64 percent), anosmia (55 percent), and recurrent epistaxis and pain (36 percent each). Other symptoms included frontal headache and diplopia. The symptoms are related both to the site and local extent of the tumor:

  • Anosmia is caused by involvement of the cribriform plate, while epistaxis reflects the marked vascularity of the tumor
  • Pain, proptosis, and excessive lacrimation are secondary to orbital extension
  • Ear pain and otitis media result from obstruction of the eustachian tube
  • Frontal headache suggests involvement of the frontal sinus

Physical examination usually reveals a red-brown, polypoid mass located high in the nasal cavity.

Early diagnosis is uncommon due to the nonspecificity of symptoms. In a review of 553 cases published in the literature over a 70 year period, 18 percent presented as stage A, 32 percent as stage B, and 49 percent as stage C . The incidence of cervical lymph node involvement ranges from 8 to 25 percent.

Radiographic studies  Plain radiographs may reveal an intranasal soft tissue density, at times with bone destruction and opacification of paranasal sinuses. Computed tomography (CT) and Magnetic resonance imaging (MRI) can help to differentiate tumor from other causes of nasal obstruction and are useful for tumor staging.

  • CT provides the best information about local invasion into surrounding bony structures and permits detailed assessment of bony erosion or destruction, particularly of the cribriform plate.
  • MRI is more accurate for defining the margins of intracranial tumor extension into adjacent soft tissue areas, such as the anterior cranial fossa and the retromaxillary space.

However, CT and MRI alone are insufficient to make a definite diagnosis and the radiographic findings cannot differentiate ONB from other tumors such as squamous cell carcinoma, undifferentiated lymphoma, amelanotic melanoma, and embryonal rhabdomyosarcoma. Definitive diagnosis requires examination of histology and confirmation with electron microscopy and immunohistochemistry.

Staging  There is no uniformly accepted staging system for ONB. The Kadish clinical staging system, a three tier classification based upon disease extent, has traditionally been used

  • Stage A confined to the nasal cavity. 5 year survival: 90%
  • Stage B one or more paranasal sinuses. 5 year survival: 71%
  • Stage C extends beyond the nasal cavity and paranasal sinuses: 5 year survival: 47%

Although the initial report, and some later series demonstrate prognostic value to this staging system, other studies fail to confirm an association of Kadish stage with disease-free survival. One modification of the Kadish staging system has classified patients with lymph node or distant metastases as stage D

Another staging system in use, the Dulguerov system, is based upon a TNM type of classification and uses CT scanning and MR imaging to ascertain the local disease extent.

PATHOLOGY  On gross examination, the tumor is polypoid, soft, and hemorrhagic. Microscopically, it is composed of sheets or discrete nests or lobules of small round cells slightly larger than lymphocytes, often compartmentalized into nodules by thin fibrous septa. The tumor cells have hyperchromatic nuclei with uniform chromatin distribution, small inconspicuous nucleoli, and sparse cytoplasm. The stroma is typically pink, neurofibrillary or edematous, and well-vascularized. Mitoses are rare.

Homer Wright pseudorosettes, which are composed of tumor cells surrounding a central pink fibrillar material, are seen in one-half of ONBs; true (Flexner type) rosettes, composed of tumor cells surrounding a central lumen, are infrequent. Necrosis, dystrophic calcification, and vascular or lymphatic invasion are not uncommon. In rare instances, a few admixed ganglion cells may be seen. Cytoplasmic glycogen is absent and reticulin fibers surround the tumor lobules, rather than individual tumor cells. Electron microscopy of ONBs demonstrates cytoplasmic neurofilaments, neurotubules, mitochondria, and dense-core neurosecretory granules (100 to 200 nm in diameter)

The Hyams histologic grading system grades tumors from I to IV based upon pathologic features such as mitotic activity and necrosis.

  • Grade I tumors are characterized by a prominent fibrillary matrix, tumor cells with uniform nuclei and absence of nuclear pleomorphism, mitotic activity, or necrosis.
  • Grade II tumors have some fibrillary matrix and exhibit moderate nuclear polymorphism with some mitotic activity. There is no necrosis.
  • Grade III tumors have minimal fibrillary matrix and Flexner type rosettes are present. There is prominent mitotic activity and nuclear polymorphism and some necrosis may be seen.
  • Grade IV tumors have no fibrillary matrix or rosettes and show marked nuclear pleomorphism and increased mitotic activity with frequent necrosis.

Most studies report a good correlation of Hyams grade with prognosis. As an example, in one series of 49 patients, the only significant predictor for overall survival, disease-free survival and local control was the Hyams grade. Low grade lesions were associated with a significantly better five-year survival than high-grade tumors (73 versus 38 percent).

TREATMENT  The rarity of ONB and its prolonged natural history  complicate the assessment of therapeutic efficacy, particularly for those with advanced disease. Management has evolved but there are no randomized trials to guide treatment decisions. As a result, there is no consensus as to the optimal treatment either for localized or advanced disease.

Localized disease

Surgery plus radiotherapy  A combined otolaryngologic and neurosurgical anterior craniofacial resection followed by postoperative radiotherapy is the most widely used approach for patients with localized ONB. Many series have reported lower recurrence rates, although several did not observe a benefit from the addition of radiation when the surgical margins were negative.

Newer radiotherapy techniques, such as proton beam therapy and intensity modulated radiation therapy (IMRT) may be more effective at sparing normal structures

In a review of 898 cases in the published literature, 68 percent were treated with surgery and postoperative radiotherapy. Five-year follow-up was available for 234 cases: 68 percent were alive and recurrence-free. Similar results were seen in a more contemporary review of 390 cases that were published between 1990 and 2000. Surgery plus radiation resulted in a five-year survival of 65 percent, compared to 48 percent with surgery alone.

These observational data are subject to selection and publication bias and randomized trials have not been performed. For most physicians taking care of such patients, these results support the benefit of adding radiation to surgery compared to resection alone.

Endoscopic surgery  A minimally invasive approach, combining endoscopic sinus surgery with stereotactic radiosurgery, appears promising in patients with localized ONB. In a series of 14 patients, for example, local tumor control was achieved within the treated area in all cases, although four patients required a second radiosurgical procedure at 6 to 79 months. In one patient, extensive tumor spread led to a subsequent craniotomy.

Chemotherapy  Chemotherapy may also be of value, either with radiation therapy following resection or for induction therapy prior to resection. However, the contribution of chemotherapy in this setting is difficult to assess, and its role is unclear.

The possible value of preoperative chemotherapy is illustrated by retrospective review of 50 patients treated at a single institution. Over a 28 year period, 32 patients with Kadish stage C disease were treated with chemoradiotherapy prior to craniofacial resection, while the remaining patients received radiotherapy prior to surgery. The actuarial five- and fifteen-year disease-free survival rates were 87 and 83 percent, respectively; these values were higher than those cited above for the combination of radiation and surgery alone (65 to 68 percent at five years). Seven of the 17 patients who relapsed underwent successful salvage surgery.

Advanced disease  The benefit of systemic chemotherapy for advanced ONB is difficult to ascertain because chemotherapy is rarely used as a single treatment modality. Furthermore, although anecdotal reports of successful treatment are often reported, treatment failures are frequently not. The total number of patients who were treated with chemotherapy and did not respond cannot be determined with reasonable certainty from any report.

Cisplatin-based combination regimens (particularly cisplatin and etoposide) have often been chosen, predominantly because of their success in the treatment of head and neck squamous cell cancer. A number of protocols have been described that have objective antitumor activity in a limited number of patients. Nonplatinum combinations, such as irinotecan plus docetaxel, may also be active.

Benefit is usually of short duration (one to nine months). In at least one of these reports, chemotherapy was only effective in patients with high grade tumors.

PROGNOSIS  The most important factor influencing outcome in patients with ONB is the extent of disease at diagnosis. This was illustrated in a series of 311 patients identified from the Surveillance, Epidemiology and End Results (SEER) database. Improved disease-specific survival at ten years correlated with limited disease according to the modified Kadish staging system (83, 49, 39, and 13 percent for stages A, B, C, and D, respectively).

Other factors have been identified that appear to influence long-term outcome:

  • Patients with low histologic grade appear to have a better survival than those with higher grade tumors.
  • DNA ploidy as determined by flow cytometry has been a useful prognostic indicator in some  but not all studies. DNA ploidy may simply represent a surrogate for histologic grade.
  • Point mutations of the p53 tumor suppressor gene do not appear to play a role in the early development of ONB. However, hyperexpression of wild type p53 may correlate with local aggressive behavior and a tendency to recurrence.