AUTHOR=Liu Ling , Lei Chu , Li Li , Peng Xi , Gong Haiyan , Hu Anhong 

TITLE=A prognostic nutritional index-based nomogram for predicting postoperative survival in stages I–III rectal cancer patients

JOURNAL=Frontiers in Nutrition

VOLUME=Volume 12 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/nutrition/articles/10.3389/fnut.2025.1680287

DOI=10.3389/fnut.2025.1680287

ISSN=2296-861X

ABSTRACT=IntroductionRectal cancer (RC) is a common malignancy of the digestive system with both high incidence and mortality. Its prognosis is influenced by multiple factors, with nutritional status playing a pivotal role. However, current prognostic models rarely incorporate this factor.MethodsTo address this gap, we have developed a novel prognostic nomogram. The newly constructed Prognostic Nutritional Index (PNI)-incorporated nomogram incorporates preoperative pathological tumor-node-metastasis (pTNM) stage, preoperative PNI, preoperative serum carcinoembryonic antigen (CEA) levels, intraoperative blood loss (IBL), and postoperative serum CEA levels.ResultsOur analysis showed that preoperative PNI ≤ 47.15, preoperative CEA > 14.13 ng/mL, IBL > 130 mL, postoperative CEA > 4.8 ng/mL, and advanced pTNM stage were independent risk factors for poor survival in patients with stage I-III rectal cancer. Compared with the non-PNI nomograms (combining preoperative CEA, postoperative CEA, pTNM and IBL, but without PNI) and the conventional pTNM staging models, the C-index of the PNI-incorporated nomogram is 0.721, compared to 0.710 for non-PNI nomograms and 0.636 for pTNM staging models, demonstrating improved predictive performance. Furthermore, the PNI-incorporated nomogram achieved AUC values of 0.855, 0.759, and 0.717 for 1, 3, and 5 year overall survival prediction, respectively, in the training set, and 0.952, 0.682, and 0.658 for the corresponding time points in the validation set.ConclusionThis model significantly improves existing prognostic methods and provides clinicians with a more comprehensive and clinically applicable tool for predicting outcomes in patients with RC.