Alvaro J Huamani
Conference 2023 Live Talk
Talk title
Transcriptome-based survival analysis identifies ASXL1 and MAP4K4 as prognostic markers in microsatellite instability gastric cancer
Authors and Affiliations
Alvaro Huamani1,2,10, Anthony Campos3,10, Kevin Magano4,10, Obert Marin5,10, Jenny Bonifacio5,6, Alexis Murillo7,10, Cesar Ortiz8,9,10
1. Professional School of Medical Technology, National University of San Marcos, Lima, Peru
2. Scientific Society of Medical Technology Students in the area of Clinical Laboratory and Pathological Anatomy, Universidad Nacional Mayor de San Marcos, Lima, Peru
3. Professional School of Biology, National University Federico Villarreal Lima, Peru
4. Professional School of Genetics and Biotechnology, Universidad Nacional Mayor de San Marcos, Lima, Peru
5. Department of Medical Microbiology, Faculty of Medicine of the Universidad Nacional Mayor de San Marcos, Lima, Peru
6.Tumor Tissue Bank, National Institute of Neoplastic Diseases, Lima, Peru
7. Center for Translational Research in Oncology, Cancer Institute of the State of Sao Paulo, Sao Paulo, Brazil
8. Cell-Based Therapy Center, University of Sao Paulo, Ribeirao Preto, Brazil
9. Division of Hematology, University of Sao Paulo School of Medicine, Sao Paulo, Brazil
10. Immunology and Cancer Research Group (INMUCA), Lima, Peru
Abstract
Background
Microsatellite instability (MSI) tumors are identified by alteration in mismatch repair genes. In gastric cancer (GC), MSI tumors characterize patients with a high probability to respond well to the treatment; nevertheless, little is known about the genetic causes associated with an aggressive disease present in some of these patients. Here, we associated clinical and transcriptome data of patients with MSI-GC in order to evaluate genes associated with prognosis.
Methods
We included patients from two independent studies, The Cancer Genome Atlas (TCGA, n=68) and the Asian Cancer Research Group (ACRG, n=68) datasets. Clinical and transcriptome data were retrieved from the Firebrowse database and the Gene Expression Omnibus (GEO), GSE66229, respectively. To identify genes associated with prognosis, the patients were dichotomized in groups of high and low gene expression, for each gene of the transcriptome (> 20 000 genes). Then, genes associated with prognosis were identified after Cox regression analysis and area under the curve calculation (AUC>0.5). Only genes associated with overall survival and disease-free survival in the two cohorts, were considered. Next, we predicted the cellular pathways associated with the identified genes by Gene Set Enrichment Analysis. Finally, we predicted non-tumor cell fractions in the tumor microenvironment by using the TIMER 2.0 resource. All calculations were performed with R 4.1.1 software.
Results
We identified two genes, ASXL1 and MAP4K4, which were associated with short OS and DFS rates. For ASXL1high, we found a HRtcga=5.36 (p=0.0013) and HRacrg=4.39 (p=0.0005), while for MAP4K4high, were HRtcga=2.89 (p=0.0123) and HRacrg=3.86 (p=0.0036). Our multivariate Cox regression demonstrated that the both genes have a prognosis significance that is not influenced by relevant clinical variables such as age and tumor stage. In addition, AUC values for both genes were more than 0.6. Interestingly, we found that tumors with high expression of MAP4K4 have a microenvironment enriched with tumor-associated fibroblasts and monocytes. Additionally, we found that MAP4K4high tumors would have an energetic metabolic status characterized by high mitochondrial respiration. Finally, we did not find a specific microenvironment composition or cellular pathways activation in ASXL1high tumors.
Conclusions
The high expression of ASXL1 and MAP4K4 is associated with lower survival rates. Furthermore, MAP4K4 contributes to the specific microenvironment composition and alters the tumor metabolic status.