This study utilizes the SEER database to investigate the impact of age on overall survival and disease management in elderly patients with non-functional pancreatic neuroendocrine tumors (PanNETs), a topic of significant concern for both doctors and patients. To ensure the reliability of the results, samples with missing variables were excluded, and a sufficiently large sample size was included. The study found that age is an independent risk factor for overall survival, with younger elderly patients benefiting more from surgical treatment. Additionally, tumor differentiation and N stage were identified as independent risk factors for overall survival, while tumor location, size, and T stage were not associated with overall survival.
Regarding the relationship between age and PanNETs, studies have shown that younger patients (< 50 years) with PanNETs tend to have fewer comorbidities, higher tumor staging, and larger tumors, but they have better overall survival after radical surgical resection[6]. Our study indicates that younger elderly patients benefit more from surgery, consistent with previous research. For younger elderly patients, more aggressive surgical treatment may be considered. A meta-analysis showed that compared to non-surgical treatments, surgical resection provides better overall survival, even for small non-functional neuroendocrine tumors. Studies have also shown that enucleation has shorter operation time and less intraoperative bleeding compared to standard surgery, but a higher incidence of postoperative pancreatic fistula[7]. Another study comparing endoscopic ultrasound-guided radiofrequency ablation (EUS-RFA) with surgical resection for insulinomas found that EUS-RFA is safer and more effective, potentially becoming a first-line therapy[8]. For elderly patients with locally growing neuroendocrine tumors, EUS-RFA may replace surgery and see broader clinical application.
Our research indicates that gender is an independent prognostic factor. A retrospective study from Italy revealed that compared to male patients, female patients have a higher incidence and earlier onset of disease, and when combined with type 2 diabetes, they present with more advanced tumor staging[9]. Another study confirmed that female patients have a longer overall survival, suggesting that prognosis and treatment should consider gender factors[10]. Regarding underlying mechanisms, a study investigated the correlation between female estrogen exposure and clinical characteristics of pancreatic neuroendocrine tumors, finding that estrogen exposure correlates with tumor size and may inhibit tumor growth[11]. Patients with better prognosis showed higher expression of ERβ in tumor tissues[12]. These studies collectively support the influence of gender on the prognosis of pancreatic neuroendocrine tumors.
This study shows that N stage is an independent risk factor for prognosis. A meta-analysis indicated that lymph node metastasis is common even in small, well-differentiated neuroendocrine tumors and is associated with poor prognosis, suggesting that the wait-and-see strategy for small tumors should be reconsidered, and lymph node dissection should be performed during surgery[13]. Another retrospective study found that tumors larger than 1.5 cm, primary tumors in the pancreatic head, Ki-67 index greater than 20%, and lymphovascular invasion are associated with a higher likelihood of lymph node metastasis, and patients with lymph node metastasis have a shorter median disease-free survival. The study recommends lymph node dissection during surgery even if preoperative variables do not reliably predict a low probability of lymph node involvement[14].
Studies have shown that the size of non-functional PanNETs is closely related to malignant phenotype, and incidentally discovered tumors smaller than 2 cm are recommended for non-surgical management[15]. Our study found that in univariate Cox analysis, tumor size and T stage were related to prognosis, but in multivariate Cox analysis, tumor size was not associated with prognosis, suggesting an indirect or spurious relationship rather than a true independent risk factor. As one study showed, tumor size is positively correlated with the Ki-67 index in well-differentiated neuroendocrine tumors[16], but not necessarily in poorly differentiated or undifferentiated tumors. Therefore, making treatment decisions based solely on tumor size may delay appropriate treatment.
This study also shows that tumor differentiation is an independent risk factor for prognosis. Studies have indicated that patients with G3 neuroendocrine tumors have significantly better overall survival if the tumors are well-differentiated compared to poorly differentiated ones[17]. In clinical practice, there can be discrepancies between mitotic count and Ki-67 index-based grading in PanNETs, with clinical outcomes being worse when the mitotic count is G2 and the Ki-67 index is G3. More importantly, G3 well-differentiated PanNETs are significantly less aggressive than true poorly differentiated neuroendocrine carcinomas (NECs)[18]. Tumor differentiation plays a crucial role in prognosis. A limitation of our study is the lack of Ki-67 index grading data. The Ki-67 index grading in well-differentiated neuroendocrine tumors is important for predicting prognosis, and our prognostic model needs further improvement.