Background: Several immune-nutritional scores have been used to predict the prognosis of pancreatic cancer, but the efficacies of these scores have yet to be compared. This study aimed to compare various immune nutritional scores and establish a more accurate scoring system to evaluate the prognosis of pancreatic cancer. Methods: The preoperative immune-nutritional scores of 411 patients with pancreatic cancer who underwent radical surgery were retrospectively compared. The cut-off point for predicting mortality was determined using X-tile analysis. The efficacies of various immune-nutritional scores for predicting the short-and long-term outcomes of pancreatic cancer were compared. A new nomogram was established based on immune-nutritional scores.Results: Regardless of the immune-nutritional scoring method, the short-and long-term outcomes of the group with better nutritional status were better than those of the other groups. The modified Glasgow prognosis score (C-index = 0.74) and controlling nutritional status score (C-index = 0.61) were more effective for predicting the prognosis of pancreatic cancer. A nomogram based on controlling nutritional status, red blood cell distribution, modified Glasgow prognosis score, and tumor node metastasis classification stage was more accurate than any single immune-nutritional score for predicting pancreatic cancer prognosis (C-index = 0.78).Conclusion: Patients with pancreatic cancer with poor preoperative nutritional status have a poorer prognosis. We identify a new nomogram based on immune-nutritional scores that provides an accurate and individualized prediction of prognosis for pancreatic cancer.(c) 2021 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).