Prediction of Response to Radiotherapy by Characterizing the Transcriptomic Features in Clinical Tumor Samples across 15 Cancer Types

Abstract

Purpose: Radiotherapy (RT) is one of the major cancer treatments. However, the responses to RT vary among individual patients, partly due to the differences of the status of gene expression and mutation in tumors of patients. Identification of patients who will benefit from RT will improve the efficacy of RT. However, only a few clinical biomarkers were currently used to predict RT response. Our aim is to obtain gene signatures that can be used to predict RT response by analyzing the transcriptome differences between RT responder and nonresponder groups.

Materials and methods: We obtained transcriptome data of 1664 patients treated with RT from the TCGA database across 15 cancer types. First, the genes with a significant difference between RT responder (R group) and nonresponder groups (PD group) were identified, and the top 100 genes were used to build the gene signatures. Then, we developed the predictive model based on binary logistic regression to predict patient response to RT.

Results: We identified a series of differentially expressed genes between the two groups, which are involved in cell proliferation, migration, invasion, EMT, and DNA damage repair pathway. Among them, MDC1, UCP2, and RBM45 have been demonstrated to be involved in DNA damage repair and radiosensitivity. Our analysis revealed that the predictive model was highly specific for distinguishing the R and PD patients in different cancer types with an area under the curve (AUC) ranging from 0.772 to 0.972. It also provided a more accurate prediction than that from a single-gene signature for the overall survival (OS) of patients.

Conclusion: The predictive model has a potential clinical application as a biomarker to help physicians create optimal treatment plans. Furthermore, some of the genes identified here may be directly involved in radioresistance, providing clues for further studies on the mechanism of radioresistance.

Figure 1

Statistical histogram of cancer types of samples and grouping.

Figure 2

Heatmap of the top 100 differentially genes for all cancer species analyzed. Both samples and genes were clustered with average linkage.

Figure 3

Histogram of pathway enrichment of top 100 differential genes.

Figure 4

mRNA levels of differentially expressed genes between response and progressive disease tumors. The distribution of gene expression values in R or PD samples was drawn through the boxplot, and the p value is marked (t-test).

Figure 5

ROC curve of differentially expressed genes between response tumors and progressive disease tumors.

Figure 6

Overall survival curve of differential expression between reactive tumor and progressive disease tumor. The abscissa of the survival curve is the observation time and the ordinate is the survival rate. The median expression level of each gene was taken as the threshold, the high expression level group was higher than the median, and the low expression level group was lower than the median. The log-rank test was used to test the statistical significance of the two groups of data.

Figure 7

ROC curve, predictor chart, and overall survival rate chart of reactive tumor and progressive disease tumor after logistic regression. (a) ROC curve. (b) The ordinate is sorted from small to large according to the prediction index, and the threshold is 0.758. (c) Overall survival curve.

Figure 8

Heatmap of differentially expressed genes in different tumor types.

Figure 9

Boxplot of differentially expressed genes in different tumor types.

Figure 10

ROC map of differentially expressed genes in different tumor types.

Figure 11

Overall survival curve of differentially expressed genes in different tumor types.

Figure 12

Logistic regression prediction ROC diagram of different tumor types.

Figure 13

Logistic regression predictors of different tumor types.

Figure 14

Survival curves of different tumor types.