A nomogram predicting overall survival for metastatic pancreatic cancer: SEER database analysis

Abstract

Background: This study's motive was to find out the determinants of average survival for sufferers with metastatic pancreatic cancer (PC) and formulate a nomogram to evaluate overall survival (OS).Methods: A whole of 10165 sufferers with metastatic PC identified from 2010 to 2015 were recruited from the Surveillance, Epidemiology, and End Results (SEER) application. All sufferers were randomly assigned to a training set (n=7117) and an internal validation set (n=3048). A nomogram was built derived from univariate and multivariate Cox evaluation to evaluate an individual patient's 6-, 12- and 18-month OS. The nomogram's overall performance used to be assessed via concordance index (C-index), receiver operating characteristic (ROC) curves, and calibration plots.Results: A whole of 10165 sufferers with metastatic PC identified from 2010 to 2015 had been included. Our analysis revealed that age, race, grade, N stage, bone metastasis, brain metastasis, liver metastasis, lung metastasis, surgery, and chemotherapy were identified as independent predictors for overall survival. The C-index of training set was 0.681 (95%CI:0.673-0.689) and validation set was 0.689 (95%CI:0.677-0.701). The calibration curves validated that the nomogram estimated 6-, 12- and 18-month OS precisely. Kaplan-Meier curves confirmed sizable variation in OS amongst the stratified low-, and high-risk groups (P<0.0001).Conclusions: The nomogram efficiently forecasts OS in sufferers with metastatic PC. This predictive model may assist clinicians in risk stratification, prognosis predictions, and facilitating personalized treatment.