Abstract This study evaluates the diagnostic value of serum carcinoembryonic antigen, carbohydrate antigen, neuron-specific enolase, and lipid metabolism markers (total cholesterol, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol) in pancreatic neuroendocrine tumors. A total of 175 pancreatic neuroendocrine tumor patients (experimental group) and 88 healthy individuals (control group) were enrolled. Serum levels of carcinoembryonic antigen, carbohydrate antigen, neuron-specific enolase, total cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, and body mass index were analyzed. A logistic regression analysis model was constructed, and the receiver operating characteristic curve was utilized to evaluate the efficacy of individual indicators with statistically significant differences, as well as the combined detection model, in predicting the occurrence, pathological grading, and metastasis of pancreatic neuroendocrine tumors. Pancreatic neuroendocrine tumor patients showed higher levels of carcinoembryonic antigen, neuron-specific enolase, low-density lipoprotein cholesterol and lower levels of high-density lipoprotein cholesterol than the control group. High-grade pancreatic neuroendocrine tumor patients showed higher carcinoembryonic antigen, carbohydrate antigen, neuron-specific enolase, total cholesterol, and low-density lipoprotein cholesterol levels than low-grade cases, with increased metastasis risk. Elevated carcinoembryonic antigen, carbohydrate antigen, and neuron-specific enolase levels were linked to metastasis. The results are statistically significant. High neuron-specific enolase and low high-density lipoprotein cholesterol were pancreatic neuroendocrine tumor risk factors, with area under the curves of 0.899 and 0.666, respectively, improving to 0.911 when combined. Elevated carbohydrate antigen and neuron-specific enolase levels predicted high-grade (area under the curves: 0.728, 0.645; combined area under the curve: 0.765) and metastatic pancreatic neuroendocrine tumors (area under the curves: 0.693, 0.651; combined area under the curves: 0.689). Increased neuron-specific enolase is a strong predictor of pancreatic neuroendocrine tumor occurrence, grading, and metastasis. Lower high-density lipoprotein cholesterol predicts pancreatic neuroendocrine tumor occurrence but not grading or metastasis. Combined markers enhance diagnostic accuracy for pancreatic neuroendocrine tumor detection and grading.
Zhang et al. (Mon,) studied this question.