1.1. Well-Differentiated Neuroendocrine Tumors (NETs)
The mean age at diagnosis for cervical carcinoid tumors is approximately 50 years. Carcinoid tumors of the cervix almost never feature the classic paraneoplastic syndromes usually associated with neuroendocrine neoplasms, carcinoid syndrome being extremely rare
[6][2].
In spite of the most recent medical advancements, the absolute scarcity of cervical carcinoids leads to underdiagnosis or misdiagnosis and inappropriate therapeutic management. Ancillary techniques including plasma assays like chromogranin A (CgA) and 5-hydroxyindoleacetic acid (5-HIAA) or immunostaining for neuroendocrine markers may prove to be helpful in some cases, but in order to be requested and actually performed, these methods require increased awareness and a high level of suspicion from both the clinician and the pathologist. Although not correlated with the intensity of flushing, the urinary level of 5-HIAA appears to have increased specificity for carcinoid tumors in comparison with 5-HIAA serum levels
[6][2]. Widespread implementation of Octreoscan significantly enhanced the diagnosis of neuroendocrine tumors, with a median detection rate of approximately 89%, as reported by Modlin et al.
[7][3]. This method uses synthetic radiolabeled somatostatin receptor analogs, which after being injected into the bloodstream of the patient, attach to tumor cells with somatostatin receptors and can be detected via gamma camera scanning.
Microscopically, typical carcinoids (neuroendocrine tumors grade 1) usually feature organoid nested growth pattern, with trabecular, insular or perivascular rosette-like structures. The neoplastic cells have abundant, finely granular, argyrophilic cytoplasm and round to oval nuclei with conspicuous nucleoli, featuring salt and pepper chromatin. There is no or mild cytologic atypia, no necrosis and extremely rare mitoses. Atypical carcinoid tumors (neuroendocrine tumors grade 2) have similar architectural patterns as typical carcinoids, but feature increased cellularity, mild to moderate cytologic atypia, increased mitotic activity (5–10 mitoses/10 high power fields) and foci of necrosis. Lymphovascular space invasion is usually absent.
Carcinoid tumors are usually indolent neoplasms. Most cases of cervical carcinoid reported in the scientific literature were histopathologically diagnosed postoperatively and due to their absolute scarcity, the most effective therapeutic management still remains indefinite. Some authors have reported extremely favorable results in patients diagnosed with typical carcinoid treated solely with radical hysterectomy
[8][4], while others revealed that cisplatin carboplatin and taxol, which are key drugs in gynecological cancer treatment
[9][5], were ineffective for primary atypical carcinoid tumors with metastases
[10][6]. Burzawa et al. achieved complete response using arterial chemoembolization with streptozotocin and 5-fluorouracil in a case of atypical carcinoid of the uterine cervix with multiple liver metastases
[11][7]. In a multi-center retrospective study of neuroendocrine tumors of the uterine cervix, Ishikawa et al. concluded that locally-advanced and extra-pelvic disease are independent prognostic factors
[12][8].
1.2. Poorly Differentiated Neuroendocrine Carcinomas (NECs)
Small cell neuroendocrine carcinoma (SCNEC) is a high-grade tumor consisting of small to medium-sized cells with scant cytoplasm and neuroendocrine differentiation. Although rare across the entire gynecological tract, SCNEC is most frequent in the cervix, where it represents approximately 2% of all cervical carcinomas. This tumor has been historically considered a rare subtype of squamous cell carcinoma, but there is now tremendous molecular, immunohistochemical and even serologic evidence for neuroendocrine differentiation.
These tumors affect a wide age range (21–87 years), but tend to be younger than those affected by squamous cell carcinoma. Patients may be asymptomatic, present with vaginal bleeding or abnormal Papanicolaou (PAP) smear. Occasionally, patients may have paraneoplastic manifestations caused by ectopic hormone production, including: Cushing syndrome (corticotropin), syndrome of inappropriate antidiuretic hormone (vasopressin), hypoglycemia (insulin), carcinoid syndrome (serotonin), hypercalcemia (parathormone) or myasthenia gravis. Obviously, the golden standard for diagnosing SCNEC is histopathological examination and immunohistochemical demonstration of neuroendocrine differentiation.
Upon gross inspection, tumors may range from small and inconspicuous to large, ulcerated, sometimes polypoid masses, which may cause marked distortion resulting in a barrel-shaped cervix. Microscopically, SCNEC is typically hypercellular and usually features diffuse, insular, corded or trabecular nested growth pattern with occasional rosette-like or acinar formation. The neoplastic cells are monotonous, small, round or oval to spindle, with scant cytoplasm and hyperchromatic nuclei, featuring finely dispersed chromatin and inconspicuous nuclei. Similar to small cell neuroendocrine carcinoma of the lung, SCNEC may show nuclear molding, crush artefacts, increased mitotic activity (>10 mitoses/10 HPF), numerous apoptotic bodies, extensive areas of necrosis and frequent lymph-vascular space involvement.
Tumor cells show punctate staining for low-molecular weight cytokeratin are also positive for epithelial membrane antigen (EMA), carcinoembryonic antigen (CEA) and neuroendocrine markers. Due to being frequently associated with high-risk HPV infection, particularly HPV 16 and 18, these tumors are generally strongly and diffusely positive for p16. Interestingly, some others have also reported focal TTF-1 positivity in cervical primary SCNEC. Alejo et al. identified HPV in 86% of all cervical NENs, with 55% being positive for HPV 16 and 41% being positive for HPV 18
[13][9] and a subsequent systematic review and meta-analysis achieved similar results
[14][10]. This particular HPV association is also useful in the differential diagnosis between SCNEC primary to the cervix and metastatic SCNEC, since the latter would not be HPV-positive.
Additionally, Inzani et al. has proved that cervical neuroendocrine carcinomas can express SST2-SST5, which could enable the opportunity for a therapy using somatostatin analogues
[15][11].
The clinical outcome of cervical SCNEC is radically different in comparison to squamous cell carcinoma or adenocarcinoma of the cervix and features more similarities to pulmonary small cell carcinoma. SCNEC is much more likely to feature lymphovascular space invasion, regional lymphatic spread as well as local and distant relapses, with a grim 5-year overall survival rate of approximately 30%
[16,17][12][13]. Currently available prognostic data suggests that the extent of disease is the most important prognostic factor.
Due to the lack of prospective clinical trials, the therapeutic management of SCNEC is difficult and associated with increased uncertainty. The current National Comprehensive cancer Network (NCCN) Clinical Practice Guidelines in Oncology for SCNEC of the cervix are aligned with the management algorithm proposed by Chen et al. in 2003. Patient workup starts with histopathological examination and imaging data. For tumors ≤4 cm and confined to the cervix, primary treatment consists of radical hysterectomy with pelvic lymphadenectomy and para-aortic lymph node sampling, followed by chemotherapy with cisplatin/etoposide or carboplatin/etoposide in the adjuvant setting. For tumors >4 cm and still confined to the cervix, patients should undergo neoadjuvant chemotherapy first, then consider interval hysterectomy. After radical hysterectomy with bilateral pelvic lymphadenectomy, adjuvant radiotherapy or chemoradiation should be considered. Non-surgical candidates may be referred to chemoradiation and brachytherapy.
Patients with locally advanced disease require chemoradiation, brachytherapy and adjuvant chemotherapy before assessment of treatment response. Tumors with response to treatment undergo surveillance, while persistent or recurrent local disease should receive systemic therapy or consider pelvic exenteration.
Due to the scarcity of these neoplasms and the lack of randomized clinical trials, an individualized treatment approach is appropriate.
Large cell neuroendocrine carcinoma (LCNEC) is a high-grade tumor composed of large cells with neuroendocrine differentiation. LCNEC is rare in the female genital tract but may occur in the cervix and the endometrium. Microscopically, LCNEC is composed of diffuse, insular, organoid, trabecular or cord-like structures, with prominent peripheral palisading and frequent glandular differentiation. The neoplastic cells are medium to large, with abundant eosinophilic/argyrophilic cytoplasm and large vesicular nuclei with prominent nucleoli and brisk mitotic rate (>10 mitoses/10 HPF).
The main differential diagnoses for LCNEC include undifferentiated carcinoma, poorly differentiated squamous cell carcinoma, lymphoepithelioma-like carcinoma, adenocarcinoma with neuroendocrine features, lymphoma and melanoma
[18,19][14][15]. A broad immunohistochemical panel including p63, CD45 and Melan A, among others, is recommended. Also, before establishing the diagnosis of cervical LCNEC, metastases from other organs should also be taken into consideration.
Most NENs feature positive immunoreaction for at least one neuroendocrine marker: chromogranin A, synaptophysin, neuron specific enolase (NSE), CD56, Leu-7 or PGP9.5. Some SCNECs may be negative for all neuroendocrine markers. Synaptophysin and CD56 are considered the most sensitive markers, although CD56 is nonspecific. If only one neuroendocrine marker is positive, diagnosis of neuroendocrine tumor should be rendered only with a strong suspicion based on morphologic features. Focal immunoreactivity for neuroendocrine markers within a neoplastic epithelial proliferation may be due to the presence of isolated neuroendocrine cells and should not be interpreted as a neuroendocrine neoplastic component. Further studies may ultimately reveal even more specific markers, such as INSM1 (insulinoma-associated protein 1), which has been reported by Kuji et al. as being positive in 95% of NECs
[20][16]. It is important to emphasize that pure neuroendocrine tumors of the cervix are extremely rare, while focal neuroendocrine differentiation may coexist with carcinoma in situ, invasive squamous cell carcinoma or adenocarcinomas. In such cases, p63 can be extremely useful in distinguishing between a NEN and a non-NEN variant of squamous cell carcinoma. If performed, electron microscopy will reveal intracytoplasmatic neurosecretory granules.
LCNEC has an extremely aggressive evolution and most patients do not survive more than 2 years after the initial diagnosis. Various authors propose different therapeutic strategies in the attempt to improve patient outcome. Tangjitgamol et al. studied the expression of estrogen and progesterone receptors in neuroendocrine tumors, but identified that only a minority of patients actually expressed significant amounts of tumor cells positive for hormone receptors
[21][17]. Another novel approach proposed by Kajiwara et al. involved octreotide, a somatostatin type 2A analog, for treating tumors expressing somatostatin type 2A receptors within tumor cells
[22][18].
The therapeutic management of poorly differentiated neoplasms can include, if diagnosed early, a surgical approach, followed by chemotherapy. If the tumoral stage is I or IIA, than a surgical approach can be attempted in patients with tumor size smaller than 4 cm, which can be followed by chemotherapy (etoposide or cisplastin) and radiation therapy
[23][19]. Cases with an early stage, but larger than 4 cm, usually undergo neoadjuvant chemotherapy, followed by surgical excision, once the tumor has shrunken. Patients diagnosed at an advanced stage of disease (IIB-IV), can benefit from combined chemo-therapy and chemoradiation and usually do not undergo a surgical intervention
[24][20]. Both small and large cell neuroendocrine carcinomas of the cervix can benefit from an improved outcome if the therapeutic scheme includes platinum and etoposide
[25][21]. The latter is preferred due to its low toxicity levels
[26][22]. Regarding primary or postoperative chemoradiation, most studies have not been able to show a clear outcome benefit, when comparing to those cases that have not received it
[23,27][19][23]. Moreover, patients who underwent adjuvant pelvic radiation have also developed subsequent reccurences
[28,29][24][25].
In a multi-center retrospective study of neuroendocrine tumors of the uterine cervix, Ishikawa et al. discovered that chemotherapy cycle count is associated with patient outcome
[12][8]. Patients with NEC appear to respond well to etoposide & platinum or irinotecan & platinum regimens, while TC shows poor efficacy
[12][8]. According to Hou et al., treatment by radical surgery or definitive radiation therapy with external beam radiation therapy yields equally poor survival
[30][26].
Tempfer et al. suggested that immune checkpoint inhibitors may prove beneficial, but controlled evidence for their efficacy is lacking
[31][27]. In two case reports, nivolumab offered persistent remissions in patients with recurrent disease, as did the MEK-inhibitor trametinib in a woman with recurrent SCNEC and KRAS mutation
[32,33][28][29].