Neuroectodermal Lesions : Part II
Neuroendocrine Carcinoma
Carcinoid (well-differentiated neuroendocrine carcinoma) (ICD-O code 8240/3)
Atypical carcinoid (moderately differentiated neuroendocrine carcinoma) (ICD-O code 8249/3)
Small cell carcinoma (poorly or undifferentiated neuroendocrine carcinoma) (ICD-O code 8041/3)
Neuroendocrine carcinomas of a wide range of differentiation can be found in the upper aerodigestive tract, mainly in the larynx.1 However, the term has caused some confusion; it is important that they should be distinguished from olfactory neuroblastoma2 and their relationship with sinonasal undifferentiated tumors has been much debated.3 As they also represent a wide range of behavior, accurate diagnosis is important.
Definition
Tumors of neuroendocrine origin of varying differentiation and aggressiveness.
Etiology
No predisposing factors are known.
Synonyms
In addition to the degree of differentiation, they may be referred to by grade (1 to 3, with carcinoid being grade 1 and small cell being grade 3). The generic acronym “SNEC” (sinonasal neuroendocrine carcinoma) is often used. Atypical or moderately differentiated lesions are sometimes called large cell whereas the small cell type has been called “oat cell.” Small cell carcinoma has also been called small cell undifferentiated (neuroendocrine) carcinoma or poorly differentiated neuroendocrine carcinoma.4
Incidence
These lesions are rare but they may have been underdiagnosed in the past. They are more often encountered in the submucosa of the supraglottis. The first case at this site was described by Goldman in 1969.5 The atypical or moderately differentiated neuroendocrine carcinoma and small cell variants are commoner than the well-differentiated carcinoid tumors and, of the three, the small cell variant is the one that is also found in the sinonasal tract.6 Only a handful of cases of sinonasal carcinoid carcinoma have been described. Up to 2000, only around 30 small cell carcinomas had been described in the sinonasal region, but on review a number were probably sinonasal undifferentiated carcinomas (SNUCs) so the true number would have been somewhat less.7–10 By 2004 Georgiou and colleagues reported that the number of documented cases had risen to 61,11 but no series have more than 10 cases.12
Site
From the very small number of validated cases, it is impossible to generalize about the site of origin except to say that lesions have occurred in the nasal cavity, ethmoid, maxillary and, in one case, sphenoid sinuses.13–16 Specifically, they can arise on the nasal septum17 but in many other cases no site of origin can be accurately determined.2
Diagnostic Features
Clinical
(See Table 15.12.)
In the larynx these rare tumors are generally much commoner in men, occurring in the fifth to sixth decades, whereas in the nose this male predilection is not so evident and the age range is wider (26 to77 years, median 48 years). Nasal obstruction and epistaxis are the main symptoms and a smooth polypoid mass or area of granular thickening have been observed.9,10 One of the largest cohorts was reported by Babin et al in 2006,18 who considered 21 patients with neuroendocrine carcinoma presenting between 1989 and 2003 to several French hospitals. There were 12 men and 9 women, with a mean age of 55 years (range 27–79 years). The majority were T4 and three had bilateral neck nodes, the rest being N0. One patient had distant metastases at presentation. Our 7 cases comprised 5 men and 2 women, aged 32 to 70 years (mean 47.2 years). All presented with advanced disease affecting the skull base and including skin and bone metastases in 2 cases. Three presented with disease in the antroethmoid, 2 in the sphenoid, and 2 in the nasal cavity.
Some case reports describe presentation with inappropriate secretion of antidiuretic hormone associated with a neuroendocrine carcinoma in the sinonasal region.19,20
In keeping with the degrees of differentiation, cervical metastases are more common in atypical and small cell tumors, occurring in 22 to 30% at presentation with atypical tumors and 50% with small cell tumors.21,22
Imaging
Apart from a markedly enhancing soft tissue mass, there are no specific features to distinguish these tumors. Flow voids are seen on MRI, indicating the vascularity of these lesions.23 Small cell tumors show areas of necrosis (Fig. 15.15).
Histological Features and Differential Diagnosis
The histological appearances of these tumors are the same as elsewhere in the body, in particular in the lung, and their immunoreactivity to neuroendocrine markers helps distinguish SNECs from SNUCs.15
Well-differentiated neuroendocrine carcinomas are comprised of nests, ribbons, and trabeculae of small homogeneous epithelioid cells. Both the nuclei and cytoplasm of these cells may be granular. Mitoses and necrosis are rare.
Carcinoid tumors are usually immunoreactive to keratins, NSE (neuron specific enolase), chromogranin, synaptophysin, and specific endocrine antigens such as serotonin.1
In the larynx, the principal differentiation is from paragangliomas, which are also very rare but are not immunoreactive to cytokeratins and have a typical sustentacular pattern of staining with S100. In the sinonasal region, neuroendocrine carcinomas are often confused with olfactory neuroblastoma.
Atypical or moderately differentiated neuroendocrine carcinomas have similar microscopic appearances as their well-differentiated counterparts but with more nuclear and cellular pleomorphism. The nucleoli may be offset to the periphery of the cell. Their immunohistochemistry is also the same, although curiously ~ 75% react with antibodies to calcitonin24 so they must be distinguished from medullary thyroid cancer by their lack of staining with thyroid transcription factor-1.
Amyloid may be seen in both well- and moderately differentiated tumors. The term “atypical” should be subject to caution as it belies an aggressive tumor with a rather poor prognosis.
Small cell carcinomas have a sheetlike growth pattern of cells with a high nuclear/cytoplasmic ratio and many mitoses and also show marked necrosis and vascular invasion. The immunoreactivity can be less specific, and even with a wide battery of agents there can be considerable overlap with SNUCs and olfactory neuroblastoma (Table 15.1).3 The other differentials include basaloid squamous cell carcinoma and lymphoma.
Natural History
The behavior of these tumors is predicted by their histological differentiation. Increasing locoregional and systemic metastases are associated with moderate and poorly differentiated tumors.25 Bone, lung, liver, and skin deposits have been reported.8,10
Treatment and Outcome
Surgery is the treatment of choice for well- and moderately differentiated tumors and some have advocated an elective neck dissection for moderately differentiated tumors occurring in the larynx. Their rarity in the nose and sinuses and paucity of lymphatic drainage make this strategy less relevant at this site, so neck dissection is confined to proven positive lymphadenopathy.
The choice of surgical approach, as ever, would be determined by the site and extent of the lesion, although it should be noted that these tumors can be quite vascular. Thus, anything from craniofacial to endoscopic resection has been described, through lateral rhinotomy and midfacial degloving.23,26
The outcome for the few cases of well-differentiated sinonasal carcinoid is reasonable if the tumor is completely excised, but no robust statistics are available. One of the largest series comprises 18 patients from the MD Anderson Cancer Center, 14 of whom had surgery and 12 chemotherapy resulting in 5-year survival of 64%.27 Five-year local failure was 21.4%, regional failure was 13%, and distant metastases occurred in 14%. However, the situation with atypical/moderately differentiated neoplasms is worse, certainly in the larynx, where 5-year survival is <50%, with two-thirds dead by 10 years.
With small cell carcinoma the outlook is even worse, with <5% alive at 5 years irrespective of treatment. The MD Anderson series included 7 patients, all with advanced disease, 5 of whom were treated with chemoradiotherapy and with only one long-term survivor. Consequently, chemotherapy and radiation employing protocols developed for pulmonary disease have been advocated rather than radical surgery.28
Cisplatin and etoposide seem to be the drugs of choice for all neuroendocrine tumors, be it induction,28 combined with radiotherapy,29 or for small cell carcinoma.30
In a series of 21 patients with SNECs reported by Babin et al, 52% had surgery, 66% radiotherapy, and 57% chemotherapy (usually cisplatin and etoposide).18 Forty-seven percent developed local recurrence, 23% had regional or systemic metastases, and 10 out of 21 (48%) died of disease within 4 years of diagnosis, with only one long-term survivor. Reviews by Their and Silva et al2,18 of SNEC patients show that in addition to intracranial extent of tumor, the presence of ectopic hormone syndrome increased mortality, whereas, interestingly, intrinsic size of tumor and number of mitoses did not correlate with recurrence, metastases, or survival.
In our small group of 7 SNEC cases, all patients received chemoradiotherapy and 5 subsequently underwent radical surgery including craniofacial resection and removal of the eye (1 case). Unfortunately, we have no long-term survivors, with death occurring between 3 and 48 months after treatment.
References
1. Ferlito A, Barnes L, Rinaldo A, Gnepp DR, Milroy CM. A review of neuroendocrine neoplasms of the larynx: update on diagnosis and treatment. J Laryngol Otol 1998; 112(9):827–834 2. Silva EG, Butler JJ, Mackay B, Goepfert H. Neuroblastomas and neuroendocrine carcinomas of the nasal cavity: a proposed new classification. Cancer 1982;50(11):2388–2405 3. Mills SE. Neuroectodermal neoplasms of the head and neck with emphasis on neuroendocrine carcinomas. Mod Pathol 2002;15(3):264–278 4. Iezzoni JC, Mills SE. “Undifferentiated” small round cell tumors of the sinonasal tract: differential diagnosis update. Am J Clin Pathol 2005;124(Suppl):S110–S121 5. Goldman NC, Hood CI, Singleton GT. Carcinoid of the larynx. Arch Otolaryngol 1969;90(1):64–67 6. Perez-Ordonez B, Caruana SM, Huvos AG, Shah JP. Small cell neuroendocrine carcinoma of the nasal cavity and paranasal sinuses. Hum Pathol 1998;29(8):826–832 7. Raychowdhuri RN. Oat-cell carcinoma and paranasal sinuses. J Laryngol Otol 1965;79:253–255 8. Koss LG, Spiro RH, Hajdu S. Small cell (oat cell) carcinoma of minor salivary gland origin. Cancer 1972;30(3):737–741 9. Rejowski JE, Campanella RS, Block LJ. Small cell carcinoma of the nose and paranasal sinuses. Otolaryngol Head Neck Surg 1982;90(4):516–517 10. Weiss MD, deFries HO, Taxy JB, Braine H. Primary small cell carcinoma of the paranasal sinuses. Arch Otolaryngol 1983;109(5):341–343 11. Georgiou AF, Walker DM, Collins AP, Morgan GJ, Shannon JA, Veness MJ. Primary small cell undifferentiated (neuroendocrine) carcinoma of the maxillary sinus. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2004;98(5):572–578 12. Renner G. Small cell carcinoma of the head and neck: a review. Semin Oncol 2007;34(1):3–14 13. Soussi AC, Benghiat A, Holgate CS, Majumdar B. Neuroendocrine tumours of the head and neck. J Laryngol Otol 1990;104(6):504–507 14. Chaudhry MR, Akhtar S, Kim DS. Neuroendocrine carcinoma of the ethmoid sinus. Eur Arch Otorhinolaryngol 1994;251(8):461–463 15. Smith SR, Som P, Fahmy A, Lawson W, Sacks S, Brandwein M. A clinicopathological study of sinonasal neuroendocrine carcinoma and sinonasal undifferentiated carcinoma. Laryngoscope 2000;110(10 Pt 1):1617–1622 16. Westerveld GJ, van Diest PJ, van Nieuwkerk EB. Neuroendocrine carcinoma of the sphenoid sinus: a case report. Rhinology 2001;39(1):52–54 17. Galm T, Turner N. Primary carcinoid tumour of nasal septum. J Laryngol Otol 2009;123(7):789–792 18. Babin E, Rouleau V, Vedrine PO, et al. Small cell neuroendocrine carcinoma of the nasal cavity and paranasal sinuses. J Laryngol Otol 2006;120(4):289–297 19. Vasan NR, Medina JE, Canfield VA, Gillies EM. Sinonasal neuroendocrine carcinoma in association with SIADH. Head Neck 2004;26(1):89–93 20. Rossi P, Suissa J, Bagneres D, et al. [Syndrome of inappropriate antidiuretic hormone secretion disclosing a sinonasal neuroendocrine carcinoma: case report]. Rev Med Interne 2007;28(6):426–428 21. Wenig BM, Gnepp DR. The spectrum of neuroendocrine carcinomas of the larynx. Semin Diagn Pathol 1989; 6(4):329–350 22. Woodruff JM, Senie RT. Atypical carcinoid tumor of the larynx. A critical review of the literature. ORL J Otorhinolaryngol Relat Spec 1991;53(4):194–209 23. Furuta A, Kudo M, Kanai K, Ohki S, Suzaki H. Typical carcinoid tumor arising in the nose and paranasal sinuses—case report. Auris Nasus Larynx 2010;37(3):381–385 24. Woodruff JM, Huvos AG, Erlandson RA, Shah JP, Gerold FP. Neuroendocrine carcinomas of the larynx. A study of two types, one of which mimics thyroid medullary carcinoma. Am J Surg Pathol 1985;9(11):771–790 25. Lund V. Distant metastases from sinonasal cancer. ORL 2001;63:212–213 26. Lee DH, Cho HH, Cho YB. Typical carcinoid tumor of the nasal cavity. Auris Nasus Larynx 2007;34(4):537–539 27. Rosenthal DI, Barker JL Jr, El-Naggar AK, et al. Sinonasal malignancies with neuroendocrine differentiation: patterns of failure according to histologic phenotype. Cancer 2004;101(11):2567–2573 28. Rischin D, Coleman A. Sinonasal malignancies of neuroendocrine origin. Hematol Oncol Clin North Am 2008; 22(6):1297–1316, xi 29. Fitzek MM, Thornton AF, Varvares M, et al. Neuroendocrine tumors of the sinonasal tract. Results of a prospective study incorporating chemotherapy, surgery, and combined proton-photon radiotherapy. Cancer 2002; 94(10):2623–2634 30. González-García R, Fernández-Rodríguez T, Naval-Gías L, Rodríguez-Campo FJ, Nam-Cha SH, Díaz-González FJ. Small cell neuroendocrine carcinoma of the sinonasal region. A propose of a case. Br J Oral Maxillofac Surg 2007;45(8):676–678Sinonasal Undifferentiated Carcinoma
Definition
(ICD-O code 8020/3)
Sinonasal undifferentiated carcinoma (SNUC) is a rare aggressive malignancy of the sinonasal region that is generally regarded as being of neuroendocrine origin.1–4 The World Health Organization subsequently defined SNUC as a highly aggressive and clinicopathologically distinct carcinoma of uncertain histiogenesis.5
Etiology
Sinonasal undifferentiated carcinoma was originally described by Frierson in 1986, but its precise etiology is unknown.6
Synonyms
Anaplastic carcinoma.
Incidence
This tumor is increasingly diagnosed and was undoubtedly overlooked in the past. Between 1986 and 2009, 28 papers with 152 patients were published, so close to 200 cases have been reported since the condition was recognized in 1986, with numbers increasing every year.7
Site
The rapid growth of this tumor can make determination of the exact site of origin difficult as more than one area of the nose and sinuses is usually involved, often with intracranial and orbital extension at presentation. Generally the nasal cavity, ethmoid, and maxilla are affected (Table 15.13).6
Diagnostic Features
Clinical Features
The mean age at diagnosis is the sixth decade8 but a range of third to ninth decades is reported and the condition can occur at any age from childhood onward.1,9
In the literature SNUCs are reported to be slightly commoner in males (2:1 to 3:1).1 In our cohort of 24 patients, there are 14 men and 10 women, and their ages at presentation ranged from 21 to 79 years (mean 53 years) (Table 15.13).
Patients usually present with advanced disease as the tumor is associated with rapid growth into adjacent structures such as the orbit and cranial cavity. Patients initially have the usual symptoms of nasal obstruction and epistaxis or serosanguinous discharge, followed by signs of orbital involvement such as proptosis, diplopia, epiphora, and ultimately chemosis and visual loss. Cranial nerve deficits may also occur though intracranial extension is often silent.
Up to one-third may present with cervical lymphadenopathy, although systemic metastases are less common initially.2,10–13 Only 8% of our cohort (2/24) had cervical lymphadenopathy.
Imaging and Staging
CT and MRI appearances have been described by Philips et al in a series of 11 cases.14 On CT a homogeneous opacification was evident with significant bone destruction and without calcification. MRI differentiated tumor from obstructed sinuses and presented generally an isodense mass on T1W images and a range of density on T2 weighting, tending to the hyperdense. Use of contrast medium produced heterogeneous enhancement (Figs. 15.16 and 15.17).
The TNM, AJCC, and Kadish staging systems originally designed for olfactory neuroblastoma have been applied to SNUCs. The majority of reported cases fall into T4, Stage IV or Stage C, respectively,7,10,15,16 as was the finding in our own cohort.
Screening of the neck by ultrasonography and imaging of the neck, thorax, and abdomen should be considered in this particular tumor due to the propensity for early spread, and PET-CT is increasingly used to stage disease.
Histological Features and Differential Diagnosis
Diagnosis is often one of exclusion as the tumor is undifferentiated by definition and falls into a large group of potential tumors (Table 15.14). SNUCs are composed of medium-sized cells that grow in nests, ribbons, sheets, and wide trabeculae, with many mitoses and apoptosis. There is a high nuclear-to-cytoplasmic ratio with prominent nucleoli. Necrosis and vascular invasion are prominent features and there is both mucosal surface growth and extension into surface mucosal glands.7,17
• Sinonasal undifferentiated carcinoma |
• Olfactory neuroblastoma |
• Small cell neuroendocrine carcinoma |
• Malignant melanoma |
• Rhabdomyosarcoma |
• Lymphomas, e.g., T/NK lymphoma |
• Primitive neuroectodermal tumor |
• Ewing′s sarcoma |
• Nasopharyngeal carcinoma |
• NK—natural killer |
Source: After reference 3. |
Adjunctive analyses such as immunohistochemistry, molecular biology, and electron microscopy are usually required to confirm the diagnosis. Indeed, immunohistochemistry is regarded as mandatory to confirm the diagnosis (Table 15.1)18 and specifically shows positivity to parakeratins, cytokeratin 7 and 8,19 and neuroendocrine markers such as NSE and synaptophysin but not as intensely as olfactory neuroblastoma (which is negative to cytokeratin). It is for this reason that it is now classified with other neuroendocrine tumors. Epstein-Barr virus status and melanocytic and lymphoid antigens should not be present to differentiate the tumor from NPC, melanoma, and lymphomas.1,9,20 The other important differentiation is from basaloid squamous carcinoma and adenoid cystic carcinoma (solid variant). In our entire cohort of nearly 1,000 malignant sinonasal tumors, apart from the 24 proven SNUCs, there is an additional group of 20 patients who, prior to the immunohistochemistry available from the early 1980s, were diagnosed as having anaplastic forms. It is quite possible that these were also SNUCs but we have excluded them from this discussion and they demonstrate the difficulties of estimating accurate incidence.
It has also been suggested recently that some SNUCs may be part of an unusual group of aggressive midline carcinomas characterized by genetic translocations that involve the gene nuclear protein in testis (NUT), a novel gene on chromosome 15.21 These tumors often involve the upper aerodigestive tract in younger people and appear undifferentiated, although may have some focal squamous differentiation. This may explain some of the debate whether SNUCs are of epidermoid or neuroendocrine origin.
Natural History
In addition to the usually large tumor mass at presentation, cervical lymphadenopathy is reported at between 10% and 30%.2,10–13,22 While distant metastases are uncommon at diagnosis, subsequent spread can occur and the cerebrospinal fluid may be seeded, leading to “drop metastases.”23 Time to metastasis ranges from 2 to 30 months.24
Treatment and Outcome
All permutations of surgery, radiotherapy, and chemotherapy have been used for this aggressive tumor. However, two broad strategies have been adopted. If the tumor is resectable, a craniofacial resection followed by radiotherapy has been employed with some success16,25,26 and has been our approach in the past 20 years. Endoscopic excision has also been described, either alone or combined with craniotomy plus chemoradiotherapy27 in a small number of patients7 with at least comparable results. Overall and disease-free survival of 57% with a mean follow-up of 32.3 months was reported. This would be in keeping with our own experience but our numbers are too small to undergo robust statistical analysis. However, the majority of patients who present with advanced disease are now managed by chemoradiotherapy after biopsy, usually using cisplatin-based regimes, with surgery reserved for residual or recurrent disease.12,16 It could be argued that this is one tumor for which endoscopic debulking as opposed to oncological resection is an acceptable approach.
A case can also be made for irradiation of the N0 neck.26 Intensity-modulated radiotherapy and/or proton beam therapy may offer some benefit in the future but this is essentially speculative. High-dose chemotherapy combined with autologous marrow transplantation has also been advocated but lack of numbers precludes comment.28
In a meta-analysis of the literature up to 2009, results are often compromised by relatively short follow-up which has ranged from 1 to 195 months (mean 23.2 months, median 14 months).29 However, overall the tumor has a poor prognosis with 5-year disease-free survival less than 25% despite dramatic initial responses. Twenty-two percent were reported as alive with disease and 52.6% dead of disease. Lin et al13 reviewed 19 patients treated over 13 years in a single institution who at 2 years had local and regional control rates of 83% and regional control of 50%. However, at 5 years, overall survival was 22%. Indeed, one study suggested a median survival of 4 months, others less than a year.1,22 Death occurs from extensive local disease and distant metastases to liver, bone, brain, and elsewhere. However, in our own series there are several long-term survivors (Table 15.13) and Rosenthal and colleagues, who reviewed 72 individuals with neuroendocrine tumors of which 16 were SNUCs, suggested that survival may be improving, either due to better histological differentiation from small cell carcinoma or/and the addition of chemotherapy.15 The difficulty remains that with small numbers treated with a range of therapeutic options it has been difficult to prove that one strategy was superior to another. However, the meta-analysis by Reiersen in 200929 comparing similarly staged disease (irrespective of the system of staging) demonstrated that the best patient survival was obtained with combined surgery, radiotherapy, and chemotherapy.