Neuroendocrine neoplasms (NENs) are a diverse group of cancers primarily originating from neuroendocrine cells in the gastrointestinal and bronchopulmonary systems (1). These tumors are characterized by neuroendocrine features, including the secretion of peptides through autocrine or paracrine mechanisms that may stimulate tumor growth. According to the World Health Organization (WHO) classification, NENs are subdivided into well-differentiated neuroendocrine tumors (NETs), graded as G1 (Ki-67 <3% or <2 mitoses/10 high-power fields[HPF]), G2 (Ki-67 3–20% or 2–20 mitoses/10 HPF), and G3 (Ki-67 >20% with well-differentiated morphology), and poorly differentiated neuroendocrine carcinomas (NECs), which are inherently high-grade (G3) with small- or large-cell morphology. Both NETs and NECs most commonly arise in the gastrointestinal tract, pancreas, and lungs (2). As a subset of NENs, NECs are characterized by aggressive biological behavior, poor differentiation, and an overall unfavorable prognosis. The majority of NECs arise in the pulmonary system, most commonly presenting as small-cell carcinomas (3). However, a smaller proportion originates outside the lungs, referred to as extrapulmonary NECs (EPNECs), which pose unique clinical challenges due to their rarity and heterogeneity. According to a recent comparative analysis from the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) database, 8.7% of NECs are classified as EPNECs (3).

EPNECs are rare and aggressive tumors with a poor prognosis, and their epidemiological characteristics have gained increasing attention in recent years. Population-based studies from the Netherlands and the United States have reported a rising incidence of EPNECs, most commonly originating in the bladder and gastrointestinal tract, with survival strongly dependent on disease stage (4, 5). Given their pathological similarities to small-cell lung cancer (SCLC), serum biomarkers such as CEA and NSE may also hold prognostic value in EPNECs, although evidence remains limited (6). International guidelines, including those from the European Neuroendocrine Tumor Society (ENETS) and the National Comprehensive Cancer Network (NCCN), provide recommendations for the diagnosis, staging, and treatment of NENs, emphasizing the importance of tumor differentiation, stage, and primary site in guiding therapeutic decisions (7, 8). Although significant progress has been made in the treatment of NENs, including targeted therapy and peptide receptor radionuclide therapy (PRRT), chemotherapy continues to play a central role in poorly differentiated NECs (9, 10). For EPNECs, therapeutic strategies remain highly dependent on the primary tumor site and often involve a combination of surgery, chemotherapy, radiotherapy, or targeted therapy (11).

Overall, the incidence of EPNECs is rising globally, with prognosis influenced by tumor site, stage, and biomarkers such as CEA and NSE. Given the rarity and heterogeneity of EPNECs, optimal treatment strategies remain unclear. This study aims to evaluate the clinical characteristics, prognostic factors, and impact of different treatment modalities on survival in EPNEC patients.

We retrospectively collected the medical records of patients diagnosed with EPNEC at Harbin Medical University Cancer Hospital from May 2011 to December 2023. A graphical abstract summarizing the overall study design and analysis workflow is provided in the Supplementary Materials ( Supplementary Figure A ). All included patients were 18 years or older and had no concurrent malignancies of other types. For each patient, the following data were collected: age, gender, primary tumor site, primary tumor size, lymph node metastasis, tumor staging, levels of tumor markers (CEA and NSE) at the time of diagnosis, treatment modalities, and overall survival (OS). The size of the primary tumor and the presence of lymph node metastasis were determined based on imaging data or postoperative pathology. Tumor staging was performed according to the 8th edition of the AJCC Cancer Staging Manual. The normal reference values for tumor markers CEA and NSE were 0–5 ng/ml and 0–15.2 ng/ml, respectively, according to the laboratory standards at our institution. OS was defined as the time from the date of diagnosis to the date of death or the last follow-up. Other prognostic factors were also analyzed. Patients who received chemotherapy were all treated with platinum-based or taxane-based combination regimens. Demographic and tumor characteristics were expressed as frequencies (percentages) for categorical variables, and group differences were compared using the Pearson chi-square test. Survival was analyzed using the Kaplan-Meier method with log-rank tests. Univariate and multivariate Cox proportional hazards models were used to estimate hazard ratios (HRs) for OS. Adjusted hazard ratios (aHRs) controlling for age, sex, stage, lymphatic status, and serum biomarkers (CEA and NSE) were calculated, and subgroup analyses were performed based on these adjusted models, examining different treatment strategies. A two-sided p value <0.05 was considered statistically significant. All analyses and plots were conducted in R version 4.2.2.

This study was conducted according to the Declaration of Helsinki and approved by the Medical Ethics Committee of Harbin Medical University Cancer Hospital. Given its retrospective nature, the requirement for informed consent was waived.

The 2022 WHO classification of NENs emphasizes the distinction between well-differentiated NETs and poorly differentiated NECs, which is critical for accurate diagnosis, prognostic stratification, and guiding treatment decisions (2). NECs, particularly EPNECs, represent a rare and biologically aggressive subset of NENs, characterized by poor differentiation, high proliferative indices, and generally unfavorable prognosis. Given the rising global incidence of EPNECs and their aggressive behavior, there is an urgent need for comprehensive studies to improve diagnostic strategies, prognostication, and treatment approaches. Our study provides a comprehensive analysis of 343 EPNEC patients, integrating demographic, clinicopathological, biomarker, and treatment-related variables to delineate prognostic factors and survival outcomes.

Consistent with previous studies, EPNECs were observed across multiple anatomical sites, with the genitourinary system, esophagus, and gastroduodenal regions being the most frequent (12). Interestingly, our study revealed that genitourinary EPNECs presented at a younger age than tumors from other sites, aligning with prior population-based analyses (13–15). Survival outcomes varied significantly by tumor site, with genitourinary tumors demonstrating the most favorable prognosis and hepatopancreatobiliary tumors the worst. These findings corroborate prior reports highlighting the prognostic relevance of primary tumor location in EPNECs (3, 15–17). Age and sex were significant predictors of survival in univariate analyses; however, neither factor retained statistical significance in multivariate models, suggesting that their prognostic effects may be confounded by tumor stage, lymphatic metastasis, biomarker status, or treatment modality. Lymph node involvement has been consistently associated with poorer prognosis across various cancers, including EPNECs (18–20). In our study, lymphatic metastasis emerged as a strong independent negative prognostic factor, in line with previous findings in both EPNECs and other high-grade neuroendocrine malignancies, including SCLC (21–23). Moreover, lymph node metastasis was significantly more frequent in tumors originating in the esophagus and gastroduodenal regions, which may reflect the more aggressive behavior of these tumor types at diagnosis. The advanced stage (III/IV) was a strong independent predictor of poor survival, reinforcing the critical importance of early detection and accurate staging.

CEA and NSE, two well-established biomarkers in SCLC (24, 25), were evaluated for their diagnostic and prognostic value in EPNECs. In our study, we observed distinct differences in tumor marker levels among EPNECs originating from various anatomical sites. CEA elevation was more frequently seen in EPNECs from the esophagus, hepatopancreatobiliary, mediastinum, gastroduodenal, and other sites. In contrast, EPNECs originating from the genitourinary, colorectal, and head/neck regions showed relatively low rates of CEA elevation, with CEA elevation being absent in head/neck tumors. On the other hand, elevated NSE was common across most anatomical sites, particularly mediastinal EPNECs. Overall, NSE is frequently elevated in EPNECs, reflecting its utility in diagnosis and disease monitoring. Nonetheless, CEA retains an important role, particularly when NSE elevation is modest or absent. Evaluating both markers concurrently offers a more comprehensive assessment of EPNEC. Our survival analysis revealed that elevated CEA and NSE levels were associated with poorer prognosis, with the worst outcomes observed in patients exhibiting elevation of both markers. This combined elevation may reflect a more aggressive disease phenotype, highlighting the value of a multifaceted biomarker approach for prognostication in EPNECs.

Therapeutic strategies for EPNECs are largely extrapolated from SCLC (26), with international guidelines from ENETS and NCCN recommending multimodal treatment based on tumor differentiation, stage, and primary site (7, 8). In line with these recommendations, our cohort demonstrated differential survival benefits across treatment modalities, which were further clarified through subgroup analyses adjusted for age, sex, stage, lymphatic metastasis, and tumor markers. Surgery conferred a significant survival advantage primarily in patients with stage I/II disease, but not in stage III/IV patients, underscoring its role as a potentially curative option in early-stage EPNECs, which was consistent with data from other studies that highlight the benefit of surgery in early-stage disease (16, 27, 28). Chemotherapy provided a significant survival benefit in stage III/IV patients, older individuals, males, patients with lymphatic metastasis, and those with normal CEA or NSE levels, supporting its selective efficacy in high-risk or advanced-stage patients (29). Regarding chemotherapy regimens, prior studies have demonstrated that platinum-based and taxane-based combinations remain standard first-line options for poorly differentiated NENs, offering meaningful response rates and progression-free survival (9, 10, 29). Our cohort exclusively received these regimens, and the observed survival benefits in advanced-stage patients align with published evidence, emphasizing the continued relevance of systemic chemotherapy despite advances in targeted therapies and PRRT (9, 10). Radiotherapy exhibited the broadest and most consistent survival advantage across multiple clinical and biomarker-defined subgroups, including those with elevated CEA and NSE levels, suggesting its utility as an integral component of multimodal therapy, particularly for unresectable or advanced tumors.

This study has several limitations. The retrospective design from a single center introduces potential selection bias and confounding. Certain subgroups had limited sample sizes, which may reduce statistical power. Prospective, multicenter studies are needed to validate these findings and refine patient-specific therapeutic strategies. Additionally, the molecular profiling of EPNECs remains an area of active investigation, and future studies should explore the genetic and molecular characteristics of these tumors to provide more targeted therapeutic approaches.

In conclusion, EPNECs are clinically heterogeneous malignancies with survival influenced by tumor site, stage, lymphatic metastasis, tumor marker status, and treatment modality. Surgery offers substantial survival benefits in early-stage disease, whereas chemotherapy and radiotherapy are particularly beneficial in advanced-stage or high-risk patients. Elevated CEA and NSE are strong negative prognostic factors, and their combined assessment can guide risk stratification and treatment planning. Our findings provide a comprehensive framework for evidence-based management of EPNECs and underscore the need for individualized, stage- and biomarker-driven therapeutic strategies for this rare and aggressive cancer.