An immune infiltration signature to predict the overall survival of patients with colon cancer

Immune infiltration of tumors has been increasingly accepted as a prognostic factor in colon cancer. Here, we aim to develop a novel immune signature, based on estimated immune landscape from tumor transcriptomes, to predict the overall survival of patients with colon cancer. The compositions of 22...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IUBMB life 2019-11, Vol.71 (11), p.1760-1770
Hauptverfasser: Peng, Dengfa, Wang, Lianjun, Li, Huqing, Cai, Changsong, Tan, Yong, Xu, Bin, Le, Hao
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Immune infiltration of tumors has been increasingly accepted as a prognostic factor in colon cancer. Here, we aim to develop a novel immune signature, based on estimated immune landscape from tumor transcriptomes, to predict the overall survival of patients with colon cancer. The compositions of 22 immune cell subtypes from three microarray datasets were characterized with the CIBERSORT deconvolution algorithm. A prognostic immunoscore (PIS) model for overall survival prediction was established by using least absolute shrinkage and selection operator (LASSO) penalized regression analysis. A total of 17 immune cell markers were screened out in the LASSO model and were then aggregated to generate the PIS. In the training cohort (n = 490), patients with high PIS exhibited a remarkably poorer overall survival than those with low PIS. Similar results were obtained in patients with different TNM stages and in patients receiving adjunctive chemotherapy or not. Multivariate Cox regression indicated that the PIS was an independent predictor for overall survival in colon cancer (hazard ratio: 2.734, 95% confidence interval: 2.052–3.643, p
ISSN:1521-6543
1521-6551
DOI:10.1002/iub.2124