This case describes the initial detection of cervical small cell neuroendocrine carcinoma on conventional endometrial and cervicovaginal liquid-based cytology. Given its rarity and high aggressiveness, early recognition and rapid, intensive treatment are crucial for improving patient outcomes. Neuroendocrine carcinomas (NEC) of the gynaecological tract are extremely rare and highly aggressive malignancies 1. When NEC affects gynaecological organs, the uterine cervix is the most frequently involved site, with small cell neuroendocrine carcinoma (SCNEC) being the most prevalent subtype 2-4. Cervical SCNEC accounts for approximately 2%–3% of all cervical malignancies and carries a significantly poorer prognosis (5-year survival, 20%–37%) than conventional squamous cell carcinoma (74.3%) and adenocarcinoma (64.6%) 4, 5. Given its aggressive clinical course and strong tendency for lymph node involvement and distant metastases, accurate and prompt diagnosis is essential, as early detection combined with intensive multimodal therapy is crucial for achieving long-term survival 1, 4, 5. Cytological examination plays a vital role in the initial screening for cervical cancer. However, despite its importance, accurate diagnosis of cervical SCNEC using cytology remains challenging 3, 5. This difficulty stems from the rarity of the entity and the tendency of SCNEC to exhibit non-specific cytomorphological features that overlap with other small round cell malignancies, including small cell variants of squamous cell carcinoma and adenocarcinoma, malignant lymphoma, metastatic carcinoma, large cell NEC, and melanoma 5, 6. Furthermore, given that cervical SCNEC is usually associated with human papillomavirus (HPV)-related epithelial neoplasms, it is frequently misdiagnosed as more common lesions, such as squamous intraepithelial lesions, atypical glandular cells, squamous cell carcinoma, or adenocarcinoma during the initial cytological examination 1, 3, 4. We present a rare case of SCNEC identified in both a conventional endometrial cytology smear and cervicovaginal liquid-based preparations (LBP). A 71-year-old postmenopausal woman with a history of osteoporosis, controlled hypertension, and hyperlipidaemia presented with a several-month history of persistent vaginal bleeding. Colposcopic examination revealed a large cervical mass. Pelvic magnetic resonance imaging revealed a large, lobulated cervical mass measuring 4.0 cm, with definite posterior parametrial invasion, and Positron emission tomography-computed tomography confirmed intense localised uptake in the uterine cervix. The tumour was staged as FIGO 2018 Stage IIB based on clinical and radiologic evidence of parametrial involvement. The patient underwent cervicovaginal and endometrial cytological sampling as part of the initial diagnostic workup for the cervical mass and vaginal bleeding. A preliminary diagnosis of poorly differentiated carcinoma was made due to the limitations of cytology. HPV polymerase chain reaction (PCR) testing of the cytological specimen was positive for HPV type 18. Serologic studies showed mildly elevated carcinoembryonic antigen levels (11.0 ng/mL reference range, 0–5 ng/mL) and normal levels of squamous cell carcinoma-associated antigen, CA125, and CA19-9. A subsequent cervical biopsy confirmed a high-grade malignancy characterised by nests and sheets of small-to-medium-sized blue cells with nuclear moulding (Figure 1A). Immunohistochemistry confirmed the neuroendocrine phenotype with diffuse positivity for synaptophysin, chromogranin, and CD56, and showed block-type overexpression of p16 as a surrogate marker of pathogenic HPV infection 7. The Ki-67 labelling index was high (approximately > 90%) in the tumour cells (Figure 1B–F). Other immunohistochemical markers included p53 and the four mismatch repair proteins, which showed wild-type and intact nuclear expression, respectively. The patient received definitive concurrent chemoradiation therapy based on the FIGO stage IIB classification (60.0 Gy external beam pelvic radiation with concurrent weekly cisplatin, followed by 22.5 Gy intracavitary brachytherapy). Follow-up imaging demonstrated a decrease in the size of the pre-existing lobulated mass in the uterine cervix (from 4.0 cm to 1.4 cm), resolution of parametrial invasion, and no evidence of regional or distant metastasis over a period of 8 months. Cervicovaginal cytology was processed using the EASYPREP automated LBP platform (Youngdong, Korea), and for endometrial sampling, direct conventional smears were prepared and stained using the Papanicolaou method. Biopsy specimens were fixed in 10% buffered neutral formalin, embedded in paraffin, and sectioned to a thickness of 4 μm for haematoxylin and eosin staining and immunohistochemical analysis. Immunohistochemical staining was performed on a Ventana BenchMark Ultra (Roche) using synaptophysin (MRQ-40), chromogranin A (LK2H10), CD56 (MRQ-42), p16 (E6H4), Ki-67 (MIB-1), and p53 (DO-7), under vendor-validated conditions. HPV genotyping was conducted using the Seegene Allplex HPV28 real-time PCR kit. Cervical cytology remains a cornerstone of screening for cervical lesions, as it is readily accessible, minimally invasive and relatively cost-effective 3. Nevertheless, SCNEC is frequently misdiagnosed or even missed as benign on initial cervical Pap cytology examinations. In a study by Liu et al., none of the 11 histologically confirmed SCNEC cases were correctly diagnosed on initial cytology examination. However, targeted re-review with cognition of SCNEC cytomorphologic features may improve diagnostic accuracy 4. Cervicovaginal LBP (Figure 2A–D) and conventional endometrial examination (Figure 2E–G) revealed moderate and high cellularity, respectively (Figure 2A,E). We observed characteristic features and diagnostic clues of this tumour, which included clustered and isolated tumour cells showing strikingly hyperchromatic nuclei, a high nuclear-cytoplasmic ratio with scant cytoplasm, nuclear moulding, finely granular chromatin, barely visible nucleoli, diathesis and rosette-like structures (Figure 2B,C,F,G). Small to medium-sized SCNEC cells with dark, round to ovoid, and polygonal nuclei with irregular contours clearly stood out from the background of intermediate and superficial cells. The SCNEC nuclei were larger, darker staining, had more irregular membranes, and showed closer inter-nuclei distances than the nuclei of intermediate cells. In addition, tumour cells had less cytoplasm and were smaller than intermediate cells (Figure 2C). Interestingly, several cells exhibiting cercariform or tadpole shapes were detected in LBP specimens (Figure 2D). A few small, dense tumour clusters showed peripheral hobnail-like cells resembling feathering. However, these cercariform-like cells appeared more ovoid and with bare nuclei, thinner tadpole tails, and sometimes multipolar tapering cytoplasmic tails than those in benign or other carcinomas. Although these cells have rarely been reported in the literature, we could easily identify them in the patient's LBP 5. The cytological features of cervical SCNEC reported in the literature are summarised (Table 1), and the differential diagnoses with key distinguishing cytological features are outlined (Table 2) 1-5. Clusters composed of small, large, cohesive, sheet-like, loose cohesive, or crowded Without specific architectural features Occasional pseudoglandular or rosette-like arrangements Scattered singly dispersed cells Relatively monomorphic small and round neoplastic cells High nuclear-to-cytoplasmic ratio Nuclear moulding (the most significant feature) Oval-to-round Smooth nuclear membranes Small-sized naked Nuclear pleomorphism (atypical nuclei) Occasional mitotic figures Smaller cell size than adenocarcinoma Chromatin stippling Fine granular salt-and-pepper type chromatin Nuclear moulding Nuclear smearing Scanty cytoplasm Caterpillar-like linear clusters Small amount of rosette-like or pseudoglandular arrangements possibly seen Block-type positivity of p16 in cervical biopsies is a surrogate marker for the molecular signature of oncogenic HPV. In addition to p16 diffuse expression, HPV type 18 was detected using PCR in this study. Several studies have demonstrated that among the high-risk HPV types, HPV-18 is a more dominant subtype of SCNEC than HPV-16 8. This finding provides evidence of a primary cervical tumour and helps differentiate it from a potential metastatic small-cell carcinoma of pulmonary origin. Although TTF-1 can be expressed in cervical NEC (approximately 52.4%), p16 expression or HPV status remains the most definitive tool for establishing cervical origin in the absence of primary extragenital SCNEC 9. The aggressive clinical course and poor prognosis associated with SCNEC emphasise the critical need for prompt and accurate diagnosis and treatment 4. This patient received rapid concurrent chemoradiotherapy owing to the prompt diagnosis of SCNEC, resulting in a markedly regressed tumour. If SCNEC is not considered in the differential diagnosis based on morphology alone, it may lead to a missed diagnosis. Even if NEC is present as only a small component of mainly conventional carcinoma, its prognosis is as poor as that of pure NEC 1, 2, 9. Therefore, it is essential to pay careful attention to cytological and histological features and accurately diagnose SCNEC through prompt tissue biopsy and immunohistochemical staining to improve a patient's prognosis. Yesul Jeong: conceptualisation, study design, data collection, literature review, first draft and revision. The author thanks Professors Sung Hak Lee and Sangjeong Ahn, the Pathfinder Laboratory, and Professor Joohee Yoon (Department of Obstetrics and Gynecology) for their valuable support. The author has nothing to report. Ethical approval was waived by the Institutional Review Board of The Catholic University of Korea, St. Vincent's Hospital for this single-patient case report (waiver determination No. VIRB-면20251029-001). The requirement for informed consent was waived by the Institutional Review Board of the Catholic University of Korea. St. Vincent's Hospital due to the minimal risk nature of this single-patient case report and the use of fully de-identified data and images. The author declares no conflicts of interest. All relevant patient data and images supporting the findings of this study are included in this article. Data are available from the corresponding author upon reasonable request.
Yesul Jeong (Thu,) studied this question.