Cargando…
Clustering analysis and prognostic model based on PI3K/AKT-related genes in pancreatic cancer
BACKGROUND: Pancreatic cancer is one of most aggressive malignancies with a dismal prognosis. Activation of PI3K/AKT signaling is instrumental in pancreatic cancer tumorigenesis. The aims of this study were to identify the molecular clustering, prognostic value, relationship with tumor immunity and...
Autores principales: | , , , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Frontiers Media S.A.
2023
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10140326/ https://www.ncbi.nlm.nih.gov/pubmed/37124502 http://dx.doi.org/10.3389/fonc.2023.1112104 |
_version_ | 1785033134111195136 |
---|---|
author | Deng, Xiangying He, Xu Yang, Zehua Huang, Jing Zhao, Lin Wen, Min Hu, Xiyuan Zou, Zizheng |
author_facet | Deng, Xiangying He, Xu Yang, Zehua Huang, Jing Zhao, Lin Wen, Min Hu, Xiyuan Zou, Zizheng |
author_sort | Deng, Xiangying |
collection | PubMed |
description | BACKGROUND: Pancreatic cancer is one of most aggressive malignancies with a dismal prognosis. Activation of PI3K/AKT signaling is instrumental in pancreatic cancer tumorigenesis. The aims of this study were to identify the molecular clustering, prognostic value, relationship with tumor immunity and targeting of PI3K/AKT-related genes (PARGs) in pancreatic cancer using bioinformatics. METHODS: The GSEA website was searched for PARGs, and pancreatic cancer-related mRNA data and clinical profiles were obtained through TCGA downloads. Prognosis-related genes were identified by univariate Cox regression analysis, and samples were further clustered by unsupervised methods to identify significant differences in survival, clinical information and immune infiltration between categories. Next, a prognostic model was constructed using Lasso regression analysis. The model was well validated by univariate and multivariate Cox regression analyses, Kaplan−Meier survival analysis and ROC curves, and correlations between risk scores and patient pathological characteristics were identified. Finally, GSEA, drug prediction and immune checkpoint protein analyses were performed. RESULTS: Pancreatic cancers were divided into Cluster 1 (C1) and Cluster 2 (C1) according to PARG mRNA expression. C1 exhibited longer overall survival (OS) and higher immune scores and CTLA4 expression, whereas C2 exhibited more abundant PD-L1. A 6-PARG-based prognostic model was constructed to divide pancreatic cancer patients into a high-risk score (HRS) group and a low-risk score (LRS) group, where the HRS group exhibited worse OS. The risk score was defined as an independent predictor of OS. The HRS group was significantly associated with pancreatic cancer metastasis, aggregation and immune score. Furthermore, the HRS group exhibited immunosuppression and was sensitive to radiotherapy and guitarbine chemotherapy. Multidrug sensitivity prediction analysis indicated that the HRS group may be sensitive to PI3K/AKT signaling inhibitors (PIK-93, GSK2126458, CAL-101 and rapamycin) and ATP concentration regulators (Thapsigargin). In addition, we confirmed the oncogenic effect of protein phosphatase 2 regulatory subunit B’’ subunit alpha (PPP2R3A) in pancreatic cancer in vitro and in vivo. CONCLUSIONS: PARGs predict prognosis, tumor immune profile, radiotherapy and chemotherapy drug sensitivity and are potential predictive markers for pancreatic cancer treatment that can help clinicians make decisions and personalize treatment. |
format | Online Article Text |
id | pubmed-10140326 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2023 |
publisher | Frontiers Media S.A. |
record_format | MEDLINE/PubMed |
spelling | pubmed-101403262023-04-29 Clustering analysis and prognostic model based on PI3K/AKT-related genes in pancreatic cancer Deng, Xiangying He, Xu Yang, Zehua Huang, Jing Zhao, Lin Wen, Min Hu, Xiyuan Zou, Zizheng Front Oncol Oncology BACKGROUND: Pancreatic cancer is one of most aggressive malignancies with a dismal prognosis. Activation of PI3K/AKT signaling is instrumental in pancreatic cancer tumorigenesis. The aims of this study were to identify the molecular clustering, prognostic value, relationship with tumor immunity and targeting of PI3K/AKT-related genes (PARGs) in pancreatic cancer using bioinformatics. METHODS: The GSEA website was searched for PARGs, and pancreatic cancer-related mRNA data and clinical profiles were obtained through TCGA downloads. Prognosis-related genes were identified by univariate Cox regression analysis, and samples were further clustered by unsupervised methods to identify significant differences in survival, clinical information and immune infiltration between categories. Next, a prognostic model was constructed using Lasso regression analysis. The model was well validated by univariate and multivariate Cox regression analyses, Kaplan−Meier survival analysis and ROC curves, and correlations between risk scores and patient pathological characteristics were identified. Finally, GSEA, drug prediction and immune checkpoint protein analyses were performed. RESULTS: Pancreatic cancers were divided into Cluster 1 (C1) and Cluster 2 (C1) according to PARG mRNA expression. C1 exhibited longer overall survival (OS) and higher immune scores and CTLA4 expression, whereas C2 exhibited more abundant PD-L1. A 6-PARG-based prognostic model was constructed to divide pancreatic cancer patients into a high-risk score (HRS) group and a low-risk score (LRS) group, where the HRS group exhibited worse OS. The risk score was defined as an independent predictor of OS. The HRS group was significantly associated with pancreatic cancer metastasis, aggregation and immune score. Furthermore, the HRS group exhibited immunosuppression and was sensitive to radiotherapy and guitarbine chemotherapy. Multidrug sensitivity prediction analysis indicated that the HRS group may be sensitive to PI3K/AKT signaling inhibitors (PIK-93, GSK2126458, CAL-101 and rapamycin) and ATP concentration regulators (Thapsigargin). In addition, we confirmed the oncogenic effect of protein phosphatase 2 regulatory subunit B’’ subunit alpha (PPP2R3A) in pancreatic cancer in vitro and in vivo. CONCLUSIONS: PARGs predict prognosis, tumor immune profile, radiotherapy and chemotherapy drug sensitivity and are potential predictive markers for pancreatic cancer treatment that can help clinicians make decisions and personalize treatment. Frontiers Media S.A. 2023-04-14 /pmc/articles/PMC10140326/ /pubmed/37124502 http://dx.doi.org/10.3389/fonc.2023.1112104 Text en Copyright © 2023 Deng, He, Yang, Huang, Zhao, Wen, Hu and Zou https://creativecommons.org/licenses/by/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. |
spellingShingle | Oncology Deng, Xiangying He, Xu Yang, Zehua Huang, Jing Zhao, Lin Wen, Min Hu, Xiyuan Zou, Zizheng Clustering analysis and prognostic model based on PI3K/AKT-related genes in pancreatic cancer |
title | Clustering analysis and prognostic model based on PI3K/AKT-related genes in pancreatic cancer |
title_full | Clustering analysis and prognostic model based on PI3K/AKT-related genes in pancreatic cancer |
title_fullStr | Clustering analysis and prognostic model based on PI3K/AKT-related genes in pancreatic cancer |
title_full_unstemmed | Clustering analysis and prognostic model based on PI3K/AKT-related genes in pancreatic cancer |
title_short | Clustering analysis and prognostic model based on PI3K/AKT-related genes in pancreatic cancer |
title_sort | clustering analysis and prognostic model based on pi3k/akt-related genes in pancreatic cancer |
topic | Oncology |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10140326/ https://www.ncbi.nlm.nih.gov/pubmed/37124502 http://dx.doi.org/10.3389/fonc.2023.1112104 |
work_keys_str_mv | AT dengxiangying clusteringanalysisandprognosticmodelbasedonpi3kaktrelatedgenesinpancreaticcancer AT hexu clusteringanalysisandprognosticmodelbasedonpi3kaktrelatedgenesinpancreaticcancer AT yangzehua clusteringanalysisandprognosticmodelbasedonpi3kaktrelatedgenesinpancreaticcancer AT huangjing clusteringanalysisandprognosticmodelbasedonpi3kaktrelatedgenesinpancreaticcancer AT zhaolin clusteringanalysisandprognosticmodelbasedonpi3kaktrelatedgenesinpancreaticcancer AT wenmin clusteringanalysisandprognosticmodelbasedonpi3kaktrelatedgenesinpancreaticcancer AT huxiyuan clusteringanalysisandprognosticmodelbasedonpi3kaktrelatedgenesinpancreaticcancer AT zouzizheng clusteringanalysisandprognosticmodelbasedonpi3kaktrelatedgenesinpancreaticcancer |