4 Discussion

Cervical neuroendocrine tumors originate from neuroendocrine cells. The first report of this disease dates back to 1957 by Reagan et al.⁹ Primary neuroendocrine cervical tumors are exceptionally rare, accounting for only 1%∼1.5% of all primary cervical tumors.² In 1997, the American Society for Cancer Research categorized NECC into typical carcinoid, atypical carcinoid, LCNEC, and SCNEC.¹⁰ The 2014 WHO classification of female reproductive system tumors¹ recommended adopting the classification system used for the gastrointestinal tract and pancreatic neuroendocrine tumors. Cervical neuroendocrine tumors were classified as low-grade or high-grade neuroendocrine carcinoma. Low-grade tumors encompassed low-grade neuroendocrine tumors grades 1 and 2, which were equivalent to typical carcinoid and atypical carcinoid tumors, respectively.

High-grade neuroendocrine carcinomas were classified as SCNEC and LCNEC, corresponding to small-cell and large-cell carcinomas, respectively. However, in the 2020 WHO classification of female reproductive system tumors, all neuroendocrine neoplasms were consolidated into one chapter, and the distinction between low-grade neuroendocrine tumors and high-grade neuroendocrine carcinomas, except for ovarian cancer, was not recommended.⁸ Typical carcinoid and atypical carcinoid tumors are extremely rare among these four tumor types, with SCNEC being the most common variant. To date, only approximately seventeen cases of LCNEC have been reported.¹¹ In the present study, SCNEC was the predominant type, with twenty cases identified, while only one case of LCNEC was observed. No cases of typical carcinoid or atypical carcinoid tumors were found.

According to the WHO classification criteria for neuroendocrine lung tumors, if two tumor components are simultaneously detected, with each component accounting for at least 10% of the tumor, it can be diagnosed as mixed neuroendocrine carcinoma. If the neuroendocrine component is less than 10%, the diagnosis is considered SCC or adenocarcinoma with focal neuroendocrine differentiation. The mixed components may include SCC and adenocarcinoma.¹² In some cases, SCC and adenocarcinomas of the cervix, along with their precursor lesions, may coexist with neuroendocrine carcinomas. However, glandular lesions appear to be more common.

In the present review, thirteen cases of pure NECC and eight cases of mixed types of neuroendocrine cancers were examined. Among the mixed types, three cases were accompanied by SCC, and five cases were associated with adenocarcinoma. Among the pure NECCs, two were associated with HSIL, and one was associated with AIS. The clinicopathological characteristics of pure and mixed types were compared. No significant differences were observed between the two groups in terms of age, clinical FIGO stage, tumor size, LVSI, or lymph node metastasis (p ​> ​0.05). However, the mixed group had a significantly worse prognosis (p ​< ​0.05). Thus, there were no notable differences in clinicopathological features between the two groups.

The findings of this study suggest that attention should be given to the mixed type of neuroendocrine carcinoma. However, it is important to note that the number of cases in this study was limited, and larger sample sizes are required to draw more definitive conclusions.

The clinical manifestations of NECC vary and are similar to those of other types of cervical carcinomas. These manifestations include vaginal bleeding, discharge, or the detection of distinct cervical masses.¹ In our cohort of patients, the common symptoms observed were irregular vaginal bleeding and discharge, which lack specificity. Neuroendocrine-related symptoms were not observed in most cases, although they occasionally accompanied symptoms of carcinoid syndrome, which was rare. Only a small number of low-grade neuroendocrine tumors produce peptide hormones such as gastrin-releasing peptides and serotonin. In rare cases, ectopic hormone production can cause symptoms. However, most neuroendocrine carcinomas do not secrete peptide hormones, and even if they do, the activity is either negligible or insufficient to produce symptoms.¹³ In our cohort, no neuroendocrine symptoms were found.

The pathogenesis of cervical neuroendocrine neoplasms is not fully understood. Some low-grade cervical neuroendocrine neoplasms appear to originate from neuroendocrine cells located in the cervical glandular and squamous epithelium.¹⁴ NECC is associated with HPV infection. Similar to SCC, HR-HPV can be detected in the majority of NECC.^15,16 HPV infection plays a crucial role in the development of NECC.¹⁷ In a recent meta-analysis by Castle et al. HPV infection data from 403 cases of SCNEC and 45 cases of LCNEC were analyzed.¹⁶ They found that more than 80% of the cases were positive for HR-HPV. The most prevalent HPV subtypes were HPV 18 and 16. The authors concluded that the E6 protein of HPV binds to the wild-type p53 protein, causing its functional loss, which may be the underlying cause of NECC. Some cervical SCNECs have recurrent genetic alterations involving the MAPK, PI3K/AKT/mTOR, and p53/BRCA pathways.¹⁸ Genetic studies have shown that SCNECs, similar to small-cell carcinoma of the lung, frequently exhibit abnormalities in the 3p alleles and occasional deletions in 9p21,¹ suggesting shared etiological factors for these tumors. In the present study, HR-HPV tests were performed on nine patients, with seven testing positive (7/9, 77.8%).

TCT for NECC is not sufficiently sensitive or specific. In the present study, the detection rate of NECC using TCT was significantly lower than that of SCC. Among the nine patients who underwent TCT before the biopsy, five were negative, and four cases were positive and diagnosed with HSIL. NECC cells were not detected by TCT. Cytologically, the findings typically included single detached cells or clusters of several cells. The cells exhibited size variation, but they tended to be small, with a high nuclear-to-cytoplasmic ratio. The nuclei appeared hyperchromatic and coarse-grained, and sometimes a “squeezed” phenomenon was observed. These findings suggest that in some cases, HSIL is diagnosed. Due to the small size of the HSIL cells and the high nuclear-to-cytoplasmic ratio, the chromatin of the nucleus appears coarse. Another possible explanation is that TCT only sampled the superficial part of the lesion, while neuroendocrine carcinoma can coexist with SCC or HSIL lesions, which can only be detected using microscopy.

Among the four positive cases, the final pathological diagnosis revealed that two cases were associated with SCC, and two cases were associated with HSILs. The five negative cases were pure NECC, which were not detected by TCT. Since our focus was primarily on identifying infiltrating stromal elements of pure neuroendocrine cancer cells, only cases that presented with simultaneous pathological changes could potentially improve the cytological detection rate. The cases in this particular group fell into the latter category. These findings suggest that regardless of TCT results, increasing the detection of HR-HPV (especially in symptomatic patients) will significantly improve the diagnostic rate of early NECC. Combining TCT with HR-HPV testing could significantly enhance the detection rate of NECC before resorting to a biopsy. In this study, 17 patients were ultimately diagnosed with NECC through cervical biopsy. Four cases were initially missed by the biopsy but were later diagnosed based on subsequent radical resection specimens. This discrepancy could be attributed to the superficial sampling site during colposcopy, insufficient sampling specimens, or limited experience of the pathologists involved.

Immunohistochemical staining is a valuable supportive diagnostic method for NECC. It involves the use of epithelial tissue markers such as Ckpan and EMA, as well as neuroendocrine markers including CgA, Syn, CD56, and NSE. Among these markers, CD56 has the highest sensitivity, with positivity rates reaching up to 80%.¹⁹ However, its specificity is relatively low, as CD56 staining can also be observed in non-neuroendocrine carcinomas, including lymphoma.²⁰ Syn is generally more sensitive than CgA, while CgA and NSE exhibit lower sensitivities. In the present study, Syn and CD56 were positive in 18 cases (85.7%, 18/21), while CgA was positive in nine cases (42.9%, 9/21), with more diffuse positive staining observed than focal positive staining. These findings align with previously reported positivity rates.⁵ CgA showed strong specificity but limited sensitivity, resulting in a relatively low positivity rate in this study.

p16^INK4a staining was diffusely or focally positive in 17 cases (81%, 17/21), consistent with the findings of some literature.²¹ Their study indicated that p16^INK4a overexpression was significantly associated with the presence of HR- HPV strains in NECC and was not correlated with p53 or NRAS mutations. TTF-1 was frequently positive in 11 cases (52.4%, 11/21). The positive rate of TTF-1 is 85% in small-cell lung cancer and ranges from 40% to 50% in large-cell lung cancer and extrapulmonary NECC.²² However, the TTF-1 positivity rate in NECC is typically lower than that in pulmonary NECC. The differential immunohistochemical expression of TTF-1 in cervical and pulmonary NECC may be attributed to the distinct pathogenesis of these diseases.²³ Nonetheless, TTF-1 staining does not hold diagnostic value in differentiating between cervical and pulmonary primary tumors.^6,24

Loss of Rb expression was observed in 95.2% (20/21) of cases, while p53 mutated expression was found in 38.1% (8/21) of cases. The loss of Rb expression and mutated p53 expression at this early stage suggest genetic and epigenetic damage contributing to the development of NECC.²⁵ Loss of Rb expression by IHC may serve as an ancillary marker supporting the diagnosis of NECC in routine practice, as it is a molecular genetic hallmark of small cell lung cancer and appears to hold true for NECC as well.²⁶

The main differential diagnoses for NECC include poorly differentiated adenocarcinoma, SCC of the cervix, and metastatic neuroendocrine carcinoma. Occasionally, poorly differentiated carcinomas may exhibit focal regional neuroendocrine differentiation, but this should not exceed 10% of the tumor. Otherwise, mixed NECC should be considered. Neuroendocrine markers (such as Syn, CgA, and CD56) in poorly differentiated carcinomas often show variable local positivity or negativity.²⁴ Since NECC is histologically indistinguishable from other neuroendocrine tumors, it is crucial to rule out the possibility of metastasis from another site. Similar to pulmonary neuroendocrine tumors, NECCs are aggressive tumors with a high propensity for metastasis.⁷ Due to their rarity, they are often misdiagnosed, and standardized treatment regimens are lacking. Therefore, treatment options have been adapted from lung neuroendocrine tumors. Early-stage patients (Stage I-IIA) are typically treated with radical surgery and chemotherapy, while intermediate-to-advanced stage patients (Stage IIB-IV) receive radiation and chemotherapy.²⁷ Factors such as advanced clinical stage, lymph node metastasis, and depth of tumor infiltration have been identified as independent prognostic factors.²⁸ NECC is associated with poor prognosis, with six patients in this study succumbing to the disease. The literature showed that the 5-year survival rate is extremely low, much lower than that of SCC.

In conclusion, NECC is an extremely rare primary tumor characterized by aggressive behavior and poor prognosis, particularly in the case of mixed types. The tumor is associated with HPV infection. The combined use of TCT and HPV testing can significantly improve the detection rate before resorting to a biopsy. The pathological diagnosis relies on histological examination and immunohistochemistry.