Development of a Transcription Factor-Based Prognostic Model for Predicting the Immune Status and Outcome in Pancreatic Adenocarcinoma

Pancreatic adenocarcinoma (PAAD) is the most common primary malignancy of the pancreas. Growing studies indicate that transcription factors (TFs) are abnormally expressed in PAAD. We, therefore, aimed to evaluate the effect of TFs in PAAD and develop a TF-based prognostic signature for the patients. The expression of the TFs and the clinical characteristics were obtained from TCGA datasets. The levels of the TFs were evaluated in PAAD tissues or nontumor tissues. Kyoto Encyclopedia of Genes and Genomes (KEGG) was used to determine the potential function of the dysregulated TFs. To create a prognostic signature, we used univariate and multivariate Cox regression. In addition, the relationship between risk score and tumor microenvironment was analyzed. In this study, we observed 19 increased and 10 decreased TFs in PAAD tissues. KEGG assays indicated that dysregulated TFs were involved in transcriptional misregulation in cancer. Multivariate Cox analysis identified two prognostic factors, Zinc finger protein 488 and BCL11A; and we developed a risk score model by these two factors. The Kaplan-Meier estimator suggested that patients with high risk exhibited a shorter overall survival than those with low risk. The receiver operating characteristic curve proved that the accuracy of this prognostic signature was 0.686 in predicting the 5-year survival. In addition, we observed that the high score was distinctly related to advanced tumor stage and immune infiltrates. Taken together, we developed a novel TF-related model which could be applied as a potential prognostic tool for PAAD and may guide the choice of immunotherapies.

Automated summary

This study aimed to evaluate the role of transcription factors (TFs) in pancreatic adenocarcinoma (PAAD) and develop a TF-based prognostic signature. The results showed that certain TFs were abnormally expressed in PAAD tissues and were associated with transcriptional misregulation in cancer. By conducting multivariate analysis, the researchers identified two TFs as prognostic factors and developed a risk score model for predicting patient survival. Additionally, high risk scores were associated with advanced tumor stage and immune infiltrates.