PLoS ONE (Jan 2018)

Establishment of inflammation biomarkers-based nomograms to predict prognosis of advanced colorectal cancer patients based on real world data.

  • Guifang Guo,
  • Xiuxing Chen,
  • Wenzhuo He,
  • Haohua Wang,
  • Yixing Wang,
  • Pili Hu,
  • Yuming Rong,
  • Lei Fan,
  • Liangping Xia

DOI
https://doi.org/10.1371/journal.pone.0208547
Journal volume & issue
Vol. 13, no. 12
p. e0208547

Abstract

Read online

PurposeTo establish three novel prognostic nomograms with inflammatory factors for advanced colorectal cancer (ACRC), right-sided colon cancer (RSCC) and left-sided colorectal cancer (LSCRC) according to real world data.Materials and methodsACRC patients receiving medicine therapy from January 1st, 2005 to September 31th, 2015 in Sun Yat-sen University Cancer Center were enrolled. Inflammatory indicators such as the neutrophil-to-lymphocyte ratio (NLR), the platelet-lymphocyte ratio (PLR), carcinoembryonic antigen (CEA), carbohydrate antigen 19-9 (CA 19-9), lactate dehydrogenase (LDH) and C-reactive protein (CRP) were analyzed for establishing nomograms predicting overall survival (OS). Concordance index (C-index) determined predictive accuracy and discriminative ability.ResultsOur study selected 807 ACRC patients, 29.6% RSCC and 70.4% LSCRC. Median OS was 23.36 months. Patients at lower level of NLR, PLR, CEA, CA 19-9, LDH and CRP showed longer OS (P ConclusionsInflammatory factors enrolled in the proposed nomograms showed accurately individualized prognostic prediction, and prognostic factors for RSCC and LSCRC were different.