Nomogram model for predicting early recurrence for resectable pancreatic cancer: A multicenter study.

Area of Science:

• Oncology
• Surgical Oncology
• Medical Informatics

Background:

• Pancreatic cancer is aggressive with high recurrence rates post-surgery, impacting patient prognosis.
• Early recurrence (ER) significantly affects outcomes in resectable pancreatic adenocarcinoma.
• Predictive tools are needed to identify high-risk patients preoperatively.

Purpose of the Study:

• To develop and validate an accurate preoperative nomogram model for predicting early recurrence in patients with resectable pancreatic adenocarcinoma.
• To identify key preoperative risk factors associated with early recurrence after pancreatectomy.

Main Methods:

• Retrospective analysis of 826 patients with pancreatic ductal adenocarcinoma undergoing pancreatectomy (2011-2020).
• Development of a predictive nomogram model using a training set (604 patients) and validation in a separate set (222 patients).
• Statistical analysis including Kaplan-Meier curves and logistic regression to identify risk factors and assess model performance (AUC).

Main Results:

• Identified preoperative risk factors for ER: Charlson age-comorbidity index ≥ 4, tumor size > 3.0 cm, clinical symptoms, elevated carbohydrate antigen 19-9 (> 181.3 U/mL), and elevated carcinoembryonic antigen (> 6.01 ng/mL).
• The nomogram model achieved an AUC of 0.711 in the training group and 0.730 in the validation group, indicating good predictive accuracy.
• The model demonstrates potential for predicting postoperative ER risk in resectable pancreatic ductal adenocarcinoma.

Conclusions:

• The developed preoperative nomogram model effectively predicts early recurrence risk in resectable pancreatic cancer.
• This tool can aid in optimizing preoperative decision-making and tailoring treatment strategies for individual patients.
• Improved prediction of ER may lead to better patient management and potentially improved outcomes for pancreatic cancer.