Prediction of radiosensitivity in non-small cell lung cancer based on computed tomography and tumor genomics: a multiple real world cohort study

Abstract

Background: The varying degrees of radiotherapy sensitivity of tumors limit the efficacy of tumor radiotherapy. In this research, based on single cell sequence data we used radiomics to help identify and screen feature signatures to distinguish varying radiosensitivity in different regions of the target area of non-small cell lung cancer can provide a new pattern to assess sensitivity of radiotherapy and assist in clinical decision-making.

Methods: This retrospective study included CT radiology data from 454 patients diagnosed with non-small cell lung cancer in multiple real-world cohorts prior to radiotherapy. The tumor primary target area was delineated on a training set (n = 154) and segmented to obtain a radiogenomic single signature. The radiogenomic signature LCDigital-RT, which can predict radiosensitivity, was developed by combining transcriptome sequencing signature radiosensitivity index and validated on two independent external validation sets (n = 74) and (n = 160). Besides, we also described the single-cell landscape of non-small cell lung cancer with different radiosensitivity, attempting to explain the potential biological mechanism at the single-cell level.

Results: By constructing solely from the single radiomics feature signature, pre LCDigital-RT can effectively identify populations with differences in radiation sensitivity in non-small cell lung cancer, with AUCs of 0.759, 0.728 and 0.745 for the training and two external validation sets, respectively. However, LCDigital-RT has a greater advantage, with a training set AUC of 0.837, which has been well validated in the JXCH cohort (AUC = 0.789) and GDPH cohort (AUC = 0.791). With the help of LCDigital-RT, patients can be divided into radiation sensitive and radiation resistant groups, and there is a significant difference in the characteristics of primary tumor lesions between the two groups. We have also enriched the interpretability of our radiogenomic features in biology at the single-cell level, demonstrating their enormous value in clinical translational research.

Conclusions: We have developed an LCDigital RT prediction tool that will help predict populations at risk of radiation sensitivity differences. By visualizing the thermal map of the primary tumor area, we can assist in the development of radiotherapy plans, reduce the occurrence of radiation toxicity events, and improve radiotherapy efficacy. At the same time, it provides a reference basis for evaluating radiation sensitivity from imaging, genetics, and other aspects.

Keywords: Non-small cell lung cancer; Radiogenomics; Radiosensitivity; Radiotherapy target area; Single cell sequencing technology.

Fig. 1

The workflow for the development of the LCDigital-RT to predict radiotherapy sensitivity in NSCLC patients. GTVp Primary Gross Tumor Volume, RSI Radiosensitivity Index, TP True Positive, FN False Negative, FP False Positive, TN True Negative, LGB Light gradient boosting machine, LASSO Least absolute shrinkage and selection operator, LR Logistic regression, SVM Support vector machine, RSF Randomforest

Fig. 2

Patient eligibility for this multicenter cohort study. Retrospective collection of clinical data and corresponding imaging data from JXCH and GDPH patients for the development of radiomics signatures. NSCLC Non-small cell lung cancer, SCLC Small cell lung cancer

Fig. 3

Development and performance of pre-LCDigital RT. A A bar chart of the population composition of radiation sensitive and radiation resistant individuals in a multicenter cohort after radiological assessment. B A petal plot comparing AUC values of multiple machine learning models. C Mountain map showing the differences in radiological characteristics between radiation sensitive and radiation resistant groups. D Compare the ROC curves of the training set, test set, JXCH external validation set, and GDPH external validation set

Fig. 4

Development and performance of LCDigital RT. A A bar chart of the population composition of radiation sensitive and radiation resistant individuals in a multicenter cohort evaluated by RSI score. B VEEN plot of the intersection of radiological feature sets related to RSI scores in a multicenter cohort. C A petal plot comparing AUC values of multiple machine learning models. D Mantel test heatmap between RSI score and radiological features. E Compare the ROC curves of the training set, test set, JXCH external validation set, and GDPH external validation set

Fig. 5

GTVp Radiosensitivity Thermogram. Red and blue represent the values on the heatmap

Fig. 6

Exploring at the molecular level and biological mechanisms. A Single cell landscape of 11 cell types. B RSI score enrichment level at the single-cell level. C Tumor cell cycle landscape and principal component inference. D KEGG enrichment analysis. E GO enrichment analysis. F Metabolic pathway and metabolite activity analysis. G Ranking of genes related to cytoTRACE score and expression at the single-cell level. H Bar chart of cell communication quantity and intensity. I The performance of RS and RR groups in intercellular communication signals