Establishment of nomogram of early death in elderly pancreatic cancer patients with liver metastasis.

Abstract

Background: Many elderly patients with pancreatic cancer (PC) often have liver metastasis (LM), and these patients often have poor prognosis and early death (ED). However, few models can accurately predict ED from elderly PC patients with LM. Therefore, we aim to create nomograms to predict ED in elderly PC patients with LM.

Methods: All elderly (≥ 60 years old) PC patients with LM from 2010 to 2020 were downloaded from the Surveillance, Epidemiology, and End Result (SEER) database according to the admission criteria. The included data was randomly divided into the training set and the validation set, with a ratio of 7:3. The risk factors for ED in elderly PC patients with LM were determined by univariate and multivariate logistic regression methods, and a nomogram model was established. Lastly, the nomogram is verified by the receiver operating characteristic (ROC) curve, area under the curve (AUC), calibration curve, and decision curve analysis (DCA).

Results: A total of 1,424 elderly PC patients with LM were randomly divided into training set (n = 996) and validation set (n = 428) based on the ratio of 7:3. The independent prognostic factors for ED include T stage, N stage, surgery, chemotherapy, lung metastasis, and other metastases. These variables were used to create nomograms, where the AUC of the training set and the validation set were 0.83 (95% CI 0.80-0.85) and 0.81 (95% CI 0.77-0.85), respectively. Furthermore, the calibration curve shows that the predicted ED is in good agreement with the actual value. DCA also shows good clinical application value.

Conclusions: The developed nomogram can be used to predict the specific probability of ED in elderly PC patients with LM, which is useful in guiding the early prevention and treatment decision-making of this group of people.