Oral and Maxillofacial Neuroendocrine Carcinoma: A Systematic Review

Abstract

The aim of the present study was to integrate the available data published in the literature on oral and maxillofacial neuroendocrine carcinomas concerning the demographic, clinical and histopathological features of this condition. An electronic search with no publication date restriction was undertaken in April 2021 in four databases. Eligibility criteria included reports published in English having enough data to confirm a definite diagnosis, always showing a neuroendocrine marker. Cases originating in the oropharynx, including base of the tongue and tonsils, were excluded. Outcomes were evaluated by the Kaplan–Meier method along with Cox regression. Twenty-five articles (29 cases) from nine different countries were detected. Mean patient age was 56.3 (± 17.5) years, with a slight male predilection. Symptomatology was present in 72.2% of informed cases. Regarding clinical presentation, a non-ulcerated nodule located in the gingiva with a mean size of 3.4 (± 2.0) cm was most frequently reported. Concomitant metastasis was identified in seven individuals. Histopathologically, most neoplasms were of the small cell type, and immunohistochemistry for both epithelial and neuroendocrine differentiation was used in 65.5% cases. Radical surgery was the treatment of choice in almost all cases, with or without adjuvant therapy. Mean follow-up was 20.5 (± 21.2) months, and only four patients developed recurrences. Eleven (44.0%) individuals died due to the disease. Ulcerated lesions were a prognostic factor. This study provides knowledge that can assist surgeons, oncologists, and oral and maxillofacial pathologists with the diagnosis and management of neuroendocrine carcinomas. Our findings demonstrated that the long-term prognosis of this lesion continues to be poor.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12105-021-01398-2.

Introduction

Neuroendocrine carcinoma (NEC) is a high-grade carcinoma with morphological and immunohistochemical features of neuroendocrine differentiation [1]. The term neuroendocrine refers to the fact that the tumor cells originate from the embryonic neuroectoderm and display an immunostaining profile consistent with endocrine gland cells [2]. Dr Oberndorfer [2] was one of the first investigators with expertise in neuroendocrine lesions, providing a description of carcinoid tumors at the beginning of the twentieth century [3]. The diagnosis of neuroendocrine tumors was further acknowledged by the World Health Organization (WHO), being introduced in the classification system in 2000, and then updated in 2010 and 2017 [4]. However, until the present time, the classification of neuroendocrine lesions is controversial. Within the scope of the head and neck area, there has been some debate among pathologists about how to classify NECs. In 1993, the WHO classified laryngeal NEC into three categories: typical carcinoid, atypical carcinoid, and small-cell carcinoma [5]. However, there was a later push to abandon these terms because some clinicians found their meanings ambiguous. After a time, Mills proposed the terms well-differentiated, moderately differentiated, and poorly differentiated [6].

A uniform classification—based on a consensus conference held at the International Agency for Research on Cancer (IARC) in 2017—for neuroendocrine neoplasms (NEN) at any anatomical location was proposed by Rindi and colleagues (2018) in order to reduce inconsistencies and contradictions among the various systems currently in use [7]. The classification suggested is intended to allow pathologists and clinicians to manage their patients with NENs in a consistent manner, while acknowledging organ-specific differences in classification criteria, tumor biology, and prognostic factors. In this respect, NEC is defined as a poorly differentiated neoplasm of high grade by definition, with two separate morphologic types: (1) small cell neuroendocrine carcinoma (SCNEC) and (2) large cell neuroendocrine carcinoma (LCNEC).

Head and neck NEC are extremely uncommon, representing 0.3% of cancers in this region [8]. Few cases of oral and maxillofacial NEC (OMNEC) have been published in the literature [4, 9]. However, a summary of their findings has not yet been made, and could serve as a base for surgeons, oncologists, and oral and maxillofacial pathologists dealing with OMNEC cases. In this context, the objective of the present study was to integrate the available data published in the literature on OMNEC into a systematic review of the demographic, clinical and histopathological features, immunophenotype, treatment, recurrence frequency, metastasis and survival of this condition.

Material and Methods

Information Sources and Search Strategies

Electronic searches without publication date restriction were undertaken in October 2020 and updated in April 2021, in the following databases: Embase (Elsevier), PubMed (National Library of Medicine), Scopus (Elsevier), and Web of Science (Thomson Reuters). The following search strategy was used for the databases: (“neuroendocrine carcinoma” OR “small cell neuroendocrine carcinoma” OR “large cell neuroendocrine carcinoma” OR “poorly differentiated neuroendocrine carcinoma” OR “high-grade neuroendocrine carcinoma”) AND (mouth OR “oral cavity” OR “cavitas oris” OR “vestibule of the mouth” OR “vestibule oris” OR “oral cavity proper” OR “mouth cavity proper” OR “cavitas oris propria” OR jaw OR jaws).

In addition, grey literature and reference lists of selected papers were screened for relevant studies that may not have been identified through the electronic search. The retrieved references were exported to the EndNote software (Thomson Reuters, New York, NY, USA). Duplicates were removed upon identification.

Eligibility Criteria

Articles describing case reports or case series of primary NEC located in intraoral sites, maxilla, and mandible published in the English language with enough clinical, radiological and histopathological information to confirm the diagnosis were included, based on the classification proposed by Rindi et al. [7]. Moreover, only cases that demonstrated at least one neuroendocrine marker to confirm the diagnosis were included.

Exclusion criteria were letters to the editor and review articles. Cases originating in the oropharynx, including base of the tongue and tonsils, were also excluded.

Study Selection

Titles/abstracts of all references retrieved through the electronic searches were read independently by three authors (L.F.S., T.R.S., G.O.Z.). If the title/abstract met the eligibility criteria, the article was included for further analysis. Moreover, the full text of the articles with titles/abstracts providing insufficient information for a clear decision was obtained. Following the full text assessment, the references that met the eligibility criteria were included. Disagreements between authors were solved upon discussion between two oral and maxillofacial pathologists (V.P.W. and M.D.M.).

Data Extraction

The following data, when available, were extracted by three authors from each study included using a standard form: authors’ name, year of publication, and country where the case was reported, patient age and sex, risk factors (tobacco and alcohol), anatomical location of the lesion, clinical presentation, reported symptoms and duration, size of the lesion, histopathological features and immunohistochemistry (IHC) test, treatment performed, recurrence (yes/no), metastasis (yes/no and location), follow-up period (months), and outcomes status (alive/dead).

Quality Assessment

Critical appraisal of the included articles was carried out by means of the Joanna Briggs Institute—University of Adelaide tool for case reports and case series [10]. Case reports included were evaluated according to the following parameters: clear description of patient’s demographic characteristics, medical history, current clinical condition, clear description of the diagnostic tests or assessment methods, treatment, post-intervention clinical condition, adverse events, and lessons provided by the case report, i.e., histopathological analysis with representative images. For case series, ten requisites were evaluated: inclusion criteria, method of measurement of the condition, methods used to identify the condition, consecutive and complete inclusion of participants, clear description of patients’ demographic characteristics, description of outcome and follow-up, clear reporting of the presenting sites’/clinics’ demographic information, and appropriate statistical analysis. For each parameter, the included article was rated as “yes”, “no”, “unclear” or “not applicable”.

Data Analysis

The data collected were analyzed using the Statistical Package for the Social Sciences (SPSS) for Mac, version 27 (IBM Corporation, Armonk, NY). Overall survival was determined by the Kaplan–Meier method. Cumulative survival curves were constructed and compared using the log-rank test. A p value < 0.05 was considered statistically significant.

Other Information

This systematic review of case reports of OMNEC was conducted according to the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) Statement 2020 [11]. A protocol was drafted, and registration was carried out at the International Prospective Register of Systematic Reviews (PROSPERO) under the number CRD42020202450.

Results

Study Selection

The search across the electronic databases yielded 879 references. After the removal of 340 duplicates, 539 references were assessed according to the inclusion and exclusion criteria. Thirty-two hits were selected for full text evaluation. Fifteen fulfilled the eligibility criteria and were included, whereas 17 articles were excluded (Supplementary File 1). Ten articles were identified by manual search and by access to the grey literature. Thus, 25 articles reporting 29 cases of OMNEC were included [12–36]. The flowchart depicts the search and the selection process (Supplementary File 2).

Demographic and Clinical Features

Table 1 summarizes the cases of OMNEC identified in this systematic review (Supplementary Table 1 describes in detail all the information collected). Articles from 11 different countries were included. Most articles were published in Japan (n = 11), followed by China (n = 3), Iran (n = 2), Turkey (n = 2), Australia (n = 1), France (n = 1), Korea (n = 1), India (n = 1), Italy (n = 1), Morocco (n = 1), and the United Kingdom (n = 1).

Table 1.

Demographic and clinical features of oral and maxillofacial neuroendocrine carcinoma cases included in the present systematic review

Variable	n (%)
Continent (n = 25)
Asia	18 (72.0)
Europe	5 (20.0)
Afrika	1 (4.0)
Oceania	1 (4.0)
Age (years) (n = 29)
Mean (SD)	56.3 (± 17.5)
Range	17—85
Sex (n = 29)
Male	16 (55.2)
Female	13 (44.8)
Risk factors (n = 10)
Yes, current or ex	6 (60.0)
No	4 (40.0)
Anatomical location (n = 30)*
Gingiva	8 (26.7)
Tongue	5 (16.7)
Maxilla	4 (13.3)
Palate	4 (13.3)
Buccal mucosa	3 (10.0)
Mandible	2 (6.7)
Maxillary sinus	2 (6.7)
Floor of the mouth	1 (3.3)
Retromolar region	1 (3.3)
Symptomatology (n = 18)
Yes	13 (72.2)
No	5 (27.8)
Presence of ulceration (n = 20)
No	13 (65.0)
Yes	7 (35.0)
Evolution time of the lesion (months) (n = 20)
Mean (SD)	14.3 (± 37.1)
Range	0.75—168
Size of the lesion (cm) (n = 20)
Mean (SD)	3.7 (± 2.0)
Range	0.5–8.7
Presence of concomitant metastasis (n = 23)
No	16 (69.6)
Yes	7 (30.4)
Treatment (n = 29)
Surgery alone	6 (20.7)
Surgery + RT + CT	5 (17.2)
Surgery + CT	5 (17.2)
RT + CT	5 (17.2)
Surgery + RT	3 (10.4)
CT alone	2 (6.9)
Palliative care	2 (6.9)
RT alone	1 (3.5)
Follow-up (months) (n = 23)
Mean (SD)	20.5 (± 21.2)
Range	1–72
Recurrence (n = 22)
No	18 (81.8)
Yes	4 (18.2)
Status (n = 25)
Alive	14 (56.0)
Dead	11 (44.0)

CT chemotherapy, RT radiotherapy, SD standard deviation

*Multiple sites were affected in some cases

The mean age of the individuals was 56.3 (± 17.5) years (range: 17 to 85 years). Sixteen (55.2%) patients were males and 13 (44.2%) were females (male-to-female ratio: 1.2:1). Only ten patients provided information concerning risk factors. Of these, six (60.0%) reported to be ex or current smokers and/or to consume alcohol and four (40.0%) denied these habits. The cases were mostly found in the gingiva (n = 8/26.7%), followed by tongue (n = 5/16.7%) and maxilla and palate with four (13.3%) cases each. Some cases affected more than one anatomical location. Symptomatology was present in 72.2% (n = 13) of informed cases. Regarding clinical presentation, a non-ulcerated (n = 13/65.0%) nodular lesion (n = 16/69.6%) with a mean size of 3.7 (± 2.0) cm was the most common report. Twenty cases described the duration of the lesion, which ranged from 3 weeks to 14 years (mean: 14.3 ± 37.1 months).

Concomitant metastases were identified in seven (30.4%) of the informed cases. They involved cervical lymph nodes, neck, liver, and bones. TNM stage was reported only in 12 cases: 50% T3, 50% N0, and 91.7% M0. Treatment modalities were resection, chemotherapy (CT), and radiotherapy (RT). Of 29 patients, six (20.7%) were treated only by radical surgery, whereas 13 (44.8%) also received adjuvant therapy (both chemotherapy and radiotherapy in five cases, only chemotherapy in five cases, and only radiotherapy in three cases). CT and RT were used without resection surgery in eight cases: five used both, two used only CT, and one case received only RT. Two patients received only palliative care.

Mean follow-up was 20.5 (± 21.2) months. Data regarding recurrence were reported in 22 cases, and only four (18.2%) patients developed recurrences. The time to recurrence after treatment was reported in only two cases: one of 5 months and other of 6 months. Eleven (44.0%) individuals died due to the disease.

Histological Features

The description of histological findings was highly variable among studies, with many studies reporting a poor description of morphological features. More detailed information was reported in some cases, including microscopic features suggestive of neuroendocrine differentiation (Table 2). Features such as basaloid cells with scant cytoplasm, hyperchromatic nuclei, round small cells, nests with periphery palisading and extensive central necrosis, rosette-like structures, and abnormal mitotic figures were reported (Fig. 1). The histological type was small cell in 18 (78.3%) cases and large cell in only five (21.7%) cases. In six cases the authors did not specify the morphological type.

Table 2.

Histopathological features of the included articles

Author (year of publication)	Main histological findings suggestive of neuroendocrine carcinoma	Histological type
Hayashi et al. (1987)	Diffuse proliferation of undifferentiated small cells. Cellular irregularity and mitotic figures are prominent	Small cells
Yoshida et al. (1995)	Solid sheets of small, undifferentiated cells, with hyperchromatic nuclei and scanty cytoplasm infiltrating the underlying muscle. Many mitotic figures were also observed	Small cells
Kim and Jang (2001)	Tumor arranged in solid nests and sheets, with absence of a fibrovascular stroma. Marked cellular pleomorphism associated with a crush artifact, and cellular necrosis, hyperchromatic oval to spindle shaped nuclei, and abundant mitosis were noted	Small cells
Noguchi et al. (2002)	Tumor nests of small cells surrounded by fibrous septa. The tumor cells showed a high nucleo/cytoplasmic ratio, and the nuclei were round or oval and hyperchromatic, with frequent mitotic figures	Small cells
Oku et al. (2002)	The tumor was composed of larger cells than those of small cell carcinoma and exhibited solid, trabecular, and slightly rosette-like growth patterns with a high mitotic rate in normal subepithelial interstitial tissue	Small cells
Güler et al. (2005)	Tumor cells with narrow cytoplasm and hyperchromatic nucleus	Small cells
Torazzi et al. (2007)	Diffuse and solid growth pattern, with small, monotonous, round neoplastic cells, scant cytoplasm and hyperchromatic nuclei	Small cells
Nishihara et al. (2009)	Invasive growth of the tumor; it appeared as a small ‘blue cell’ tumor that formed sheets and nests with artificially crushed cells, necrosis and stromal fibrous septa. The tumor cells had round to fusiform nuclei with fine granular chromatin and inconspicuous nucleoli. Mitotic figures were frequently observed at an average of 12 per high power field	Large cells
Mochizuki et al. (2010)	Solid nests of basaloid cells with scant cytoplasm, periphery palisading, extensive necrosis, rosette-like structures. The tumor cells showed a dot- like pattern for AE1/AE3	Large cells
Chen et al. (2012)	Scattered small foci of tumor cells with a cicatrized fibrous stroma in the muscle adipose tissue	Small cells
Cymerman et al. (2013)	Focal undifferentiated small cell like appearance. High power view of undifferentiated small cell like morphology	Small cells
Terada (2013)	Small epithelial cells with hyperchromatic nuclei, molded nuclei, scant nucleocytoplasmic ratio, and negative nucleoli	Small cells
Yoshino et al. (2013)	Densely clustered, small round cells with large nuclei	NI
Bothara et al. (2014)	NI	Large cells
Singla et al. (2014)	Frequent mitoses and apoptotic necrosis	Small cells
Wu et al. (2014)	Round cells that grew in small nests and trabeculae	NI
	Round cells that grew in sheets, with extensive necrosis	NI
Esmati et al. (2015)	Solid pattern that composed of rather uniform cells with eosinophilic cytoplasm and mild hyperchromatic nuclei, without keratinization pearl	NI
Gumusay et al. (2015)	Darkly stained nuclei with small but conspicuous nucleoli and moderate to scant cytoplasm, mitotic figures, and a central area of necrosis	Small cells
Hamamoto et al. (2015)	Poorly differentiated tumor with large alveolar structure and invasive growth to the subcutaneous tissue. The tumor cells had large, irregular-shaped nuclei, with enlarged nucleoli and eosinophilic granular cytoplasm observed, and showed partial necrosis. The tumor cells consisted of polygonal shaped cells and spindle shaped cells. The mitotic rate of the tumor was relatively low	Large cells
	The tumor consisted of polygonal cells showing solid growth. However, undifferentiated cells with alveolar configuration were also observed. The mitotic rate was very high	NI
	The tumor consisted of polygonal tumor cells, which showed a high nuclear/cytoplasmatic ratio and a high mitotic rate, with solid alveolar growth and myxomatous stroma. The tumor cells showed a partial duct-like configuration filled with Alcian Blue-periodic acid-Schiff-positive mucus materials, with neutrophils infiltrate observed inside of, and around, the tumor alveolar structure	NI
	The tumor consisted of polygonal shaped and undifferentiated tumor cells showing solid growth with cobblestone appearance, and ulcer formation was observed. Most of the tumor cells were medium sized and had oval-round nuclei with increased chromatin and a high nuclear/cytoplasmatic ratio. Some scattered large tumor cells with abnormal mitosis were observed, and the mitotic rate was very high	Large cells
Zeng et al. (2015)	Small cells with ovoid‑ to spindle‑shaped nuclei, fine granular chromatin, inconspicuous nucleoli and scant cytoplasm. The tumor cells were arranged in solid nests, sheets and cords, with frequently occurring brisk mitotic figures and areas of necrosis	Small cells
Sase et al. (2016)	Growth of small, round, atypical cells lacking in cytoplasm is seen under the epithelium	Small cells
Deyhimi et al. (2017)	Nests and sheets of small round cells with round to oval nuclei, and scant eosinophilic cytoplasm. Pleomorphism, atypia, hyperchromatism, and mitotic figures were visible to variable degrees	Small cells
Udompatanakorn et al. (2018)	Small round to oval atypical cells arranged in nests with a glandular-like-pattern, hyperchromatic molded nuclei, a high nuclear-to-cytoplasmic ratio, and a scant eosinophilic cytoplasm	Small cells
Olivetto et al. (2019)	Undifferentiated tumor proliferation, of trabeculated architecture, composed of small basophilic cells with a high nuclear-cytoplasmic ratio. The nuclei are quadrangular, hyperchromatic	Small cells
Neftah et al. (2020)	Cellular proliferation made of layers of small granular chromatin cells with nuclear molding	Small cells

NI not informed

Fig. 1.

Histopathological and immunohistochemical findings of small cell neuroendocrine carcinoma. A undifferentiated tumor proliferation of trabeculated architecture, composed of small basophilic cells (hematoxylin and eosin stain, 100×); B small, round, atypical basaloid cells with scant cytoplasm (hematoxylin and eosin stain, 200×); C high nuclear-cytoplasmic ratio, ovoid‑ to spindle‑shaped nuclei, fine granular chromatin, inconspicuous nucleoli, and abnormal mitosis (hematoxylin and eosin stain, 400×); D–I, epithelial cell immunopositivity for AE1/AE3, EMA and CK18; J–M, neuroendocrine differentiation confirmed by immunopositivity for CD56, synaptophysin and neuron-specific enolase; N and O, strong Ki67 index

Table 3 shows the immunohistochemistry markers used in the studies. AE1/AE3, CAM5.2, EMA, CAM5.2, MFN116, CK19, CK7 and CK20 were mostly used for epithelial differentiation. For neuroendocrine differentiation the authors reported the use of chromogranin A (CgA), synaptophysin (Syn), neuron-specific enolase (NSE), CD56, and TTF-1. Examples of some markers are demonstrated in Fig. 1. Nineteen (65.5%) cases used immunohistochemistry for both epithelial and neuroendocrine differentiation. The proliferative Ki-67 index was reported in 12 cases and ranged from 30 to 90%.

Table 3.

Immunohistochemical features related by the included articles

Author (year of publication)	Positive Epithelial Markers	Positive Neuroendocrine Markers	Other positive markers	Negative Markers	Proliferation index (Ki67)
Hayashi et al. (1987)	NI	Vasoactive intestinal polypeptide (VIP)	S100, CEA, myosin	NI	NI
Yoshida et al. (1995)	Keratin, EMA	NSE	CEA	CK, desmin, vimentin, alfa-fetoprotein, ACTH, hCG, and S-100	NI
Kim and Jang (2001)	AE1/AE3	NSE	NI	NI	NI
Noguchi et al. (2002)	NI	NSE, CgA	NI	S100, vimentin and CEA	NI
Oku et al. (2002)	CAM5.2	CgA and CD56	NI	NI	NI
Güler et al. (2005)	AE1/AE3, CK8-18	NSE (focal)	NI	LCA, CD20, CD3, CA15-3, vimentin, and S-100	NI
Torazzi et al. (2007)	AE1/AE3	Syn, NSE	EMA	CD45 Ro, CD20, CD3, and CD138	NI
Nishihara et al. (2009)	AE1/AE3	NSE, CD56	EMA	CK20, vimentin, S100, HMB-45, CgA and Syn	83%
Mochizuki et al. (2010)	CK7, CK19	CgA, NSE, Syn, CD56	NI	CK1, CK2, CK4, CK6, CK8, CK9, CK10, CK13, CK16, CK18 and CK20	80%
Chen et al. (2012)	AE1/AE3	Syn, CgA and CD56	GFAP	myogenic differentiation 1, EMA, S100, CD3, CD20, CD43, CD79a, CD99, LCA (CD45), BcL-2, HMB-45, melan-A, myeloperoxidase, and TTF-1	NI
Cymerman et al. (2013)	AE1/AE3, MNF116, CK7, CK20	Cg, CD56, Syn	NI	Melanoma and lymphoid markers	NI
Terada (2013)	Pan-CK MNF-116, CK34BE12, CK5/6, CK14	Syn, CD56	vimentin, KIT (CD117), p53,	AE1/3, CAM5.2, CK7, CK8, CK18, CK19, CK20, EMA, NSE, Cg, PDGFRA, CD45, CD45RO, CD3, CD20, CD30, CD79a, and bcl-2	70%
Yoshino et al. (2013)	NI	NSE, Syn, CD56	NI	CgA, TTF-1	70%
Bothara et al. (2014)	NI	NSE, Cg	NI	NI	NI
Singla et al. (2014)	Pan-CK, CAM5.2	CD56, Cg, Syn, TTF-1	NI	NI	80–90%
Wu et al. (2014)	NI	NSE, Syn	S100	CgA, NF, TTF-1, vimentin, LCA, CD68, CK, AE1/AE3, CK7, CK20 and HMB-45	NI
	AE1/AE3, EMA	Syn	NI	CK20, CgA, vimentin, S100, calponin, NSE, HMB-45, LCA, NF and SMA	NI
Esmati et al. (2015)	CK	NSE, Syn, Cg	NI	LCA	60%
Gumusay et al. (2015)	CK	TTF-1, CgA, Syn	NI	CD3, CD20, CD30 and p63	NI
Hamamoto et al. (2015)	NI	CD56, Syn, CgA	NI	p53	40%
	NI	CD56, Syn	p53	CgA	70%
	NI	CD56, CgA	NI	Syn, p53	60%
	NI	CD56	NI	Syn, CgA, p53	80%
Zeng et al. (2015)	AE1/AE3	CgA, Syn, NSE	NI	S100, vimentin, SMA, HMB-45, LCA, cluster of differentiation CD99, CD56, TTF-1 and CK20	70%
Sase et al. (2016)	Pan-CK	CD56, Syn	Neural markers	-	NI
Deyhimi et al. (2017)	CK, EMA, Pan-CK, CK20	NSE, Syn, Cg	S100	LCA, CD99, vimentin, desmin	NI
Udompatanakorn et al. (2018)	NI	Cg, Syn, CD56	P16	34βE12 and p63	NI
Olivetto et al. (2019)	AE1/AE3	Syn, Cg, CD56 (focal)	TTF1, CD3, CD20	NI	NI
Neftah et al. (2020)	Pan-CK	Cg	NI	Syn	30%

NI not informed, CK cytokeratin, NSE neuron specific enolase, TTF-1 thyroid transcription factor‑1, EMA epithelial membrane antigen, LCA leukocyte common antigen, SMA smooth muscle actin, ALK anaplastic lymphoma kinase, HMB-45 human melanoma black-45, CEA carcinoembryonic antigen, hCG human chorionic gonadotropin, ACTH adrenocorticotropic hormone, NF neurofilament, Cg chromogranin, Syn synaptophysin, NCAM neural cell adhesion molecule, HMWK high-molecular-weight keratin

Survival and Prognostic Factor Analysis

Cox regression analysis for disease-specific survival is presented in Supplementary Table 2. The presence of an ulcerated lesion (HR, 8.71; 95% CI 1.53–49.56; p = 0.01) was significantly associated with poor survival. No other variables tested showed significant associations in survival analysis, although some tendencies were noted in Kaplan–Meier cumulative survival curves, such as: males, more than 66 years of age, a history of alcohol and/or tobacco consumption, asymptomatic lesions, lesion size of more than 3 cm, presence of metastasis (regional or distant), individuals without surgical treatment, and presence of recurrences (Fig. 2). Moreover, tendencies to a poor prognosis were also noted regarding the large cell type of NEC (Fig. 3). Survival analysis revealed a considerably poor prognosis with 1-year and 3-year survival rates of 69% and 44%, respectively (Fig. 4).

Fig. 2.

Independent variables (clinical-demographic features) associated with cumulative survival

Fig. 3.

Independent variables (histopathological features) associated with cumulative survival

Fig. 4.

Survival curve analysis of individuals with oral and maxillofacial neuroendocrine carcinoma

Quality Assessment

Supplementary Tables 3 and 4 depict the results of the quality assessment of case reports and case series included in this systematic review, respectively.

Discussion

The present systematic review confirmed that OMNEC are rarely published in the literature. In fact, NEC is extremely rare, representing 0.3% of all head and neck cancers and showing a highly aggressive behavior [8]. We identified only 29 cases reported in the English literature, published between 1987 and 2020. A narrative review was published in short communication form by Huang et al. [37] after the customary steps of registry, reading articles, extracting data and writing the article [37]. Although some of the articles included in the present paper are the same as those reported by Huang et al., our sample differs by being systematized, following specific inclusion criteria (i.e., use of a neuroendocrine marker to confirm the diagnosis) and exclusion criteria (oropharyngeal cases, such as base of the tongue). A summary of the most relevant data described in those studies concerning demographic and clinical features, treatment, outcomes (recurrence and survival) along with histopathological and immunohistochemical features was presented in our study. This comprehensive analysis allowed us to identify the demographic and clinical profile of OMNEC based on preferred oral site and mean age at diagnosis, survival trend and possible prognostic factors, as well as the main pitfalls in reporting those cases. Deficient morphological description and lack of differential diagnosis emerged as the main shortcomings of the reports included in the present study.

In the present review, the mean age of patients with OMNEC was 56.3 years, similar to what has been reported for larynx NEC [38]. Yet is important to stress that we found reports in almost all decades of adult life, ranging from the second to the ninth decade. Interestingly, men with OMNEC appear to be affected as older (63.2 ± 13.8 years) individuals, while women are more affected at an earlier age (47.8 ± 18.3 years). Males were slightly more affected than females, with a male-to-female ratio of 1.2:1. These findings agree with a study conducted by Perez-Ordoñez (2018) about NEC of the larynx and other head and neck regions [38]. Moreover, 95% and 91% of patients with SCNEC and LCNEC, respectively, had a history of smoking habit [38]. Unfortunately, in our systematic review, risk factors were poorly described. Of the 29 cases, only ten included these data, with six patients reporting ex or current smoking habit or alcohol consumption.

Neuroendocrine cells are specialized cells and can be present in any part of the body—known as the disseminated/diffuse neuroendocrine system. Therefore, NEC can virtually arise in any anatomic location. In the oral cavity, neuroendocrine cells are mainly distributed along the masticatory mucosa and tongue since they are thought to have mechanoreceptor functions. Our findings revealed that the gingiva was the most affected site, followed by the tongue. Herein, NEC located in the base of the tongue and tonsils were not included since these sites are considered to be part of the oropharynx. Moreover, Sinno et al. [39] have recently reviewed the literature concerning oropharyngeal NEC and our aim was to focus on oral cavity and jawbone cases [39].

Our findings demonstrated that most cases presented as a nodular lesion. Although most lesions were non-ulcerated, survival analysis demonstrated that the presence of an ulcer was a tendency to a poor prognosis. Moreover, except for one case, in which the patient reported an evolution time of 14-years (with a significant growth in one month), the mean duration of the lesion was 6.2 ± 8.5 months (ranging from 3 weeks to 36 months). In this respect, OMNEC is a lesion with a more duration time in relation to another malignant lesion, but faster compared to benign neoplasia. According to Strojan et al. [40], symptoms of NEC varied regarding the subsite involved and the extent of the disease and included swallowing difficulties, breathing problems and/or hoarseness of varying degrees and in different combinations [40]. More than 70% of our sample reported some symptomatology such as tenderness, pain, bleeding, limited mouth opening, or odynophagia.

Histologically, the distinction between a small or large cell can be made through comparison with lymphocytes. For small cell tumor, cell size should be less than three lymphocytes. For large cell, the cells must be more than three times the size of a normal lymphocyte [41]. SCNEC exhibit a submucosal cellular proliferation usually associated with surface ulceration and arranged in solid nests, sheets, or ribbons with the absence of a prominent fibrovascular stromal component. The neoplastic cells are small to medium sized and have scant cytoplasm with oval to spindled hyperchromatic nuclei lacking nucleoli and showing nuclear molding [42]. The diagnosis of LCNEC is based on certain criteria. First, the tumors should show a typical neuroendocrine architecture including organoid nesting, trabecular growth, rosettes, and peripheral palisading. Second, the tumor cells should be of medium to large size with moderate to abundant cytoplasm. Marked cellular pleomorphism, vesicular chromatin, and small to prominent nucleoli are characteristic features [42].

The final diagnosis of OMNEC requires immunohistochemistry to demonstrate neuroendocrine differentiation, which can be achieved with a range of markers including CgA, Syn, and NSE, among others [43]. We observed that many authors relied only on neuroendocrine makers to present the final diagnosis, and both epithelial and neuroendocrine markers were presented in 65.5% of cases. Interestingly, some studies have associated antibodies as a prognostic factor. Nielsen et al. [44] found that negative CgA was associated with a significantly worse overall survival in high grade gastroenteropancreatic neuroendocrine neoplasms. Moreover, this marker was significantly related to the Ki67-index. In our analysis, Syn, Cg and Ki67 did not show a significant association with prognosis. However, this result should be interpreted with caution because of the limited informed cases.

Although we acknowledge the essential role of immunohistochemistry, we believe morphological analysis is a key procedure since it will guide the differential diagnosis and determine the panel of markers that need to be investigated. We observed that both steps—morphological description and differential diagnosis discussion based on histological features—were poorly described. Other malignant small round cell neoplasms should usually be considered in the differential diagnosis, including olfactory neuroblastoma, sinonasal undifferentiated carcinoma, and basaloid squamous carcinoma [9]. Especially in the oral cavity, it is important to consider the main differential diagnoses of NEC such as non-keratinizing squamous cell carcinoma, non-Hodgkin lymphoma and solid adenoid cystic carcinoma. Immunostaining can help with this distinction, since non-keratinizing squamous cell carcinomas are positive for CK 5/6, p63 and p40 [1], whereas the cytokeratin pattern in NEC is often in the form of dots [45]. Regarding adenoid cystic carcinoma, although there is controversy concerning the immunostaining of p63 and p40, the solid pattern is negative for p40 [46] and still positive for C-kit [47, 48], while non-Hodgkin lymphomas are positive for LCA and lymphocyte cell markers [49]. In the absence of these tumors, metastasis must be considered, but the presence of neuroendocrine differentiation markers is essential for a more conclusive diagnosis [50].

It is especially imperative to distinguish poorly differentiated NECs from the well- and moderately differentiated subtypes, and also from squamous cell carcinomas because surgery is not the mainstay of treatment for poorly differentiated NEC [9]. The most effective treatments have yet to be determined; at present, it appears that only early and aggressive therapies employing different treatment modalities may potentially be considered curative [40]. Most patients included in this systematic review received surgical resection associated with adjuvant therapy. In the Kaplan–Meier curve, we observed a tendency to a better prognosis of cases with surgical treatment, with or without adjuvant therapy. Individuals that received only CT and/or RT died before two years of follow-up. In fact, some studies demonstrated that the prognosis of NEC depends on treatment. Ding et al. [51] published a retrospective study of 106 cases of esophageal SCNEC submitted to different treatments, with surgery or radiotherapy alone leading to 0% 5-year survival rates. In comparison, surgery combined with adjuvant therapy achieved a combined 5-year survival rate of 27%.

Our results identified regional and distant metastases in only 30.4% of cases. In contrast, Gnepp et al. [52] reported that 90% of patients with poorly differentiated laryngeal NEC would experience a metastasis at some point in their illness, most likely to the cervical lymph nodes, liver, lung, or bone [52]. More recently, Yan et al. [53] studied 493 head and neck NEC individuals, and almost 80% of them had distant metastases [53]. On this basis, we may infer that OMNEC may exhibit a lower potential metastasis rate than other anatomical locations. However, the prognosis of patients with OMNEC is generally poor. The results of the present systematic review showed that 44% of the patients died due to their disease. The 1- and 3-year survival rates were 69% and 44%, respectively.

Our study has some limitations due to the low number of studies on this topic and the heterogeneity of tumor origin and treatment regimens. The results of a systematic review of case reports or case series depends on the detailed information described in the included cases. However, some authors did not report important data such as symptomatology, clinical presentation, evolution time, lesion size, follow-up or recurrence rate. Unfortunately, some cases were excluded because of the lack of a neuroendocrine marker to confirm the diagnosis. Thus, although we made an effort to perform a more representative compilation of OMNEC, some failures may have occurred.

Conclusion

In summary, OMNEC represents a rare lesion, affecting with more frequency middle aged adult males. The histopathological diagnosis is challenging, requiring the use of immunohistochemical markers (always accompanied by some neuroendocrine marker) to define the origin of the lesion. The findings presented in this systematic review demonstrated that the long-term prognosis of this lesion remains poor. Ulcerated lesions were significantly associated with a poor prognosis. Moreover, some variables exhibited tendencies to cumulative survival. In this respect, it is important to highlight that individuals who received surgical treatment had a better survival rate. This study provides knowledge that can assist surgeons, oncologists, and oral and maxillofacial pathologists with the diagnosis and management of neuroendocrine carcinomas.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Study concepts and Study design: LFS, MDM; Data acquisition: LFS, TRS, GOZ; Quality control of data and algorithms: LBK, FMS; Data analysis and interpretation and Manuscript preparation: LFS, VPW; Statistical analysis: VPW; Manuscript editing: MATM, RC, JNS, FPF; Manuscript review: PAV, MSM.

Funding

No funding obtained.

Data Availability

None.

Code availability

None.

Declarations

Conflict of interest

No conflict of interest to disclosure.

Ethical Approval

Not applicable.

Consent to Participate

Not applicable.

Consent for Publication

Not applicable.