- Patients in the low-risk group were more likely to have higher tumor mutational burden, stem cell characteristics, and higher PD-L1 expression levels. - Although most studies of alterations in cancer me- tabolism have focused on glucose metabolism (i.e., the Warburg effect), the role of abnormal lipid metabolism in cancer cells has been gradually recognized over the past few years [14, 15]. - the search used the key- words “colon cancer”, “rectum cancer”, “adenocarcin- oma”, “Homo sapiens”, and “Expression profiling.” The gene expression microarray dataset GSE39582 was se- lected and downloaded because it had the largest sample size for CRC [29]. - A total of 1044 genes were found to be involved in the lipid metabolism process after removal of overlapping genes.. - The risk score of each sample was calculated by the following formula: (expGene: the expression level of LMRG in the TCGA or GEO cohort. - Coef: the coefficient of LMRG in the LASSO Cox regression model in the training set).. - Finally, 57 LMRG were shown to be related to prognosis in the training set. - The AUCs of the signature were 0.6901 at 1 year, 0.6776 at 3 years, and 0.5945 at 5 years in the training set (Fig. - 2 Identification of the differentially expressed and prognosis-related genes in the TCGA cohort. - K-M survival curves showed that high-risk pa- tients had significantly shorter OS than low-risk cases in the training set (Fig. - 3E, more patients died in the high-risk group, whereas the majority survived in the low-risk group. - In the training set, principal components analysis (PCA) was used to obtain expression patterns in the low- and high-risk groups. - In the validation set, the high-risk patients again had significantly shorter OS than those in the low-risk group (Fig. - Independent prognostic value of the risk signature The results showed that the risk score based on the four-LMRG signature was an independent prognostic factor in the training set, with hazard ratio (HR. - Similar results were obtained in the validation set, showing the independence of the risk signature with HR CI . - There were also more stem cell characteristics in the low-risk group, including significantly higher. - The OS of high-risk patients was significantly shorter than that of low-risk patients in the different clinicopatho- logical characteristic subgroups.. - 0.05, 487 samples of the total set were included in the subsequent ana- lysis. - As shown in the cor- relation heatmap, the results revealed that the risk score was negatively correlated with most of the im- mune cells (Supplementary Fig. - In this study, the risk score was negatively correlated with PD-L1 mRNA in the training set (R. - 3 Development and validation of the risk signature. - (C) Time-dependent ROC curves for 1-, 3-, and 5-year OS in the training group. - (H) The survival curves showed significant differences between high- and low-risk patients in the validation set ( P <. - (I) Time-dependent ROC curves for 1-, 3-, and 5-year OS in the validation set. - (J) Risk score distribution (above) and survival status (below) of CRC patients in the validation set. - 4 Univariate and multivariate Cox regression analysis showed that the risk score was an independent prognostic factor both in the training and validation sets. - (A) Results of univariate Cox regression analysis in the training set. - (B) Results of multivariate Cox regression analysis in the training set.. - (C) Results of univariate Cox regression analysis in the validation set. - (D) Results of multivariate Cox regression analysis in the validation set. - The top 20 genes with the highest mutation frequency in CRC are shown in the waterfall plots in Fig. - The results of the analysis showed that the TMB score was higher in the low-risk group than in the high-risk group (P = 0.0002, Fig. - 5 Stratified analysis of the risk signature. - This could be one of the reasons for the better prognosis of patients in the low-risk group.. - Functional enrichment in the low- risk group was focused on energy-metabolism-related functions, including fatty acid metabolism and oxidative phosphorylation (Fig. - Each variable in the nomogram was assigned a weighted score based on the multivariate Cox regression coefficient. - (B) Prognostic value of the nomogram for predicting 1-, 3-, and 5-year overall survival rate in the training set. - (C) Prognostic value of the nomogram for predicting 1-, 3-, and 5-year OS rates in the validation set. - (D-F) Calibration plots suggest that the nomogram ’ s predictions of 1-year (D), 3-year survival (E), and 5-year survival (F) match well with the actual observed probabilities in the training set. - The more the blue lines and dashed lines in the graph coincide, the better the predictive performance of the. - (G-I) The calibration plots showed that the actual observed probabilities were in agreement with the predictive values from the nomogram for 1-year (G), 3-year (H), and 5-year survival (I) in the validation set. - Therefore, the lipid-related metabolism risk signature represents alter- ations in the TIME of CRC. - Gene names and mutation frequency are shown in the bar chart on the left. - org/10.1186/s . - Immune landscape of the risk signature.. - Mutation landscape of the risk signature.. - 9 Enriched gene sets in the Hallmark collection for high- and low-risk patients. - The lines above the X axis indicate gene sets enriched in the high-risk group, and the lines under the X axis indicate the gene sets enriched in the low-risk group. - https://doi.org/10.1038/s . - https://doi.. - https://doi.org/10.1016/S . - https://doi.org/10.6004/jnccn.2020.0048.. - https://doi.org . - https://doi.org/10.1016/j.ajhg . - https://doi.org/10.1245/s . - https://doi.org/10.1016/j.canlet . - https://doi.org CCR-19-1543.. - https://doi.org CAN-19-1027.. - https://doi.org/1 0.1016/j.cell . - https://doi.org/10.1016/j.semca ncer . - https://doi.org/1 0.1186/s . - https://doi.org/10.3390/ijms16010924.. - https://doi.org CAN-16-0651.. - https://doi.org/10.1016/j.cmet . - https://doi.org/10.1016/j.critrevonc . - https://doi.org/10.1002/ijc.30709.. - https://doi.org/10.1007/s . - https://doi.org/10.1016/j.bbalip . - https://doi.org/10.1016/j.molcel . - https://doi.org/10.1172/JCI127201.. - https://doi.org/10.1210/jc . - https://doi.org/10.1111/jcmm.14647.. - https://doi.org/10.1038/. - https://doi.org/10.1371/journal.pmed.1001453.. - doi.org/10.1093/bioinformatics/btg405.. - https://doi.org/10.1073/pnas . - https://doi.org/10.1093/bioinformatics/bts034.. - doi.org/10.1089/omi.2011.0118.. - https://doi.org/10.1093/nar/gkaa970.. - Penalized cox regression analysis in the high-dimensional and low-sample size settings, with applications to microarray gene expression data. - https://doi.org/10.1093/bioinforma tics/bti422.. - https://doi.org/10.1186/s . - https://doi.org/10.1002/stem.1837.. - https://doi.org/10.101 6/j.cell . - https://doi.org/10.1038/nmeth.3337.. - https://doi.org/10.1093/bioinformatics/btw325.. - https://doi.org/10.1101/gr . - doi.org CIR-18-0927.. - https://doi.org CCR-19-1105.. - https://doi.org/10.1002/oby.22591.. - https://doi.org/10.1038/s x.. - https://doi.org/10.1186/s x.. - https://doi.org/10.1 053/j.gastro . - https://doi.org/10.1016/j.celrep . - https://doi.org/10.1038/s y.. - org/10.1016/j.bbcan . - https://doi.org/10.7554/eLife.10250.. - org/10.1016/j.immuni . - https://doi.org/10.15252/emmm.201910698.. - https://doi.org/10.101 6/j.canlet
Xem thử không khả dụng, vui lòng xem tại trang nguồn hoặc xem
Tóm tắt