Lingyu Qi1, Yan Yao2, Tingting Zhang1, Fubin Feng3, Chao Zhou3, Xia Xu4, Changgang Sun5. 1. College of First Clinical Medicine, Shandong University of Traditional Chinese Medicine, Jinan 250014, Shandong Province, China. 2. College of First Clinical Medicine, Weifang Medical University, Weifang 261041, Shandong Province, China. 3. Department of Oncology, Weifang Traditional Chinese Hospital, Weifang 261041, Shandong Province, China. 4. Pharmacy Department, Shandong Chest Hospital, Jinan 250014, Shandong Province, China. 5. College of First Clinical Medicine, Shandong University of Traditional Chinese Medicine, Jinan 250014, Shandong Province, China; Department of Oncology, Shandong University of Traditional Chinese Medicine, Jinan 250014, Shandong Province, China.. Electronic address: scgdoctor@126.com.
Abstract
BACKGROUND: Breast cancer (BRCA) is the most prevalent cancer that threatens female health. A growing body of evidence has demonstrated the non-negligible effects of messenger RNAs (mRNAs) on biological processes involved in cancers; however, there is no definite conclusion regarding the role of mRNAs in predicting the prognosis of BRCA patients. MATERIALS AND METHODS: We systematically screened the mRNA expression landscape and clinical data of samples from the Cancer Genome Atlas (TCGA). Univariate Cox analysis and robust likelihood-based survival analysis were conducted to identify key mRNAs associated with BRCA. Furthermore, risk scores based on multivariate Cox analysis divided the training set into high-risk and low-risk groups. ROC analysis determined the optimal cut-off point for patient classification of risk levels. The prognostic model was additionally validated in the testing set and complete dataset. Finally, we plotted the survival curves for the mRNAs used in our model. RESULTS: We obtained the original expression data of 13,617 mRNAs from a total of 1088 samples. After comprehensive survival analysis, the four-mRNA (ACSL1, OTUD3, PKD1L2, and WISP1) prognosis risk assessment model was constructed. Furthermore, the area under cure (AUC) was 0.834, indicating that the model was meaningful and reasonable. In each dataset, analysis based on the four-mRNA signature risk score indicated that the survival status of the group with high risk score was worse than that of the group with low risk scores. Patients with strong mRNA expression of OTUD3, PKD1L2, and WISP1 tended to have good prognosis, whereas patients with high ACSL1 expression tended to have poor prognosis. CONCLUSION: In summary, we constructed a four-mRNA prognosis risk assessment model for BRCA. The newly developed model offers more possibilities for assessing prognosis and guiding the selection of better treatment strategies for BRCA.
BACKGROUND:Breast cancer (BRCA) is the most prevalent cancer that threatens female health. A growing body of evidence has demonstrated the non-negligible effects of messenger RNAs (mRNAs) on biological processes involved in cancers; however, there is no definite conclusion regarding the role of mRNAs in predicting the prognosis of BRCApatients. MATERIALS AND METHODS: We systematically screened the mRNA expression landscape and clinical data of samples from the Cancer Genome Atlas (TCGA). Univariate Cox analysis and robust likelihood-based survival analysis were conducted to identify key mRNAs associated with BRCA. Furthermore, risk scores based on multivariate Cox analysis divided the training set into high-risk and low-risk groups. ROC analysis determined the optimal cut-off point for patient classification of risk levels. The prognostic model was additionally validated in the testing set and complete dataset. Finally, we plotted the survival curves for the mRNAs used in our model. RESULTS: We obtained the original expression data of 13,617 mRNAs from a total of 1088 samples. After comprehensive survival analysis, the four-mRNA (ACSL1, OTUD3, PKD1L2, and WISP1) prognosis risk assessment model was constructed. Furthermore, the area under cure (AUC) was 0.834, indicating that the model was meaningful and reasonable. In each dataset, analysis based on the four-mRNA signature risk score indicated that the survival status of the group with high risk score was worse than that of the group with low risk scores. Patients with strong mRNA expression of OTUD3, PKD1L2, and WISP1 tended to have good prognosis, whereas patients with high ACSL1 expression tended to have poor prognosis. CONCLUSION: In summary, we constructed a four-mRNA prognosis risk assessment model for BRCA. The newly developed model offers more possibilities for assessing prognosis and guiding the selection of better treatment strategies for BRCA.