Dataset Information

Integrative radiogenomics analysis for predicting molecular features and survival in clear cell renal cell carcinoma.

ABSTRACT:

Objectives

To assess the feasibility of predicting molecular characteristics by computed tomography (CT) radiomics features, and predicting overall survival (OS) using combination of omics data in clear cell renal cell carcinoma (ccRCC).

Methods

Genetic data of 207 ccRCC patients was retrieved from The Cancer Genome Atlas (TCGA) and matched contrast-enhanced CT images were obtained from The Cancer Imaging Archive (TCIA). Another cohort of 175 ccRCC patients from West China Hospital was used as external validation. We first applied radiomics features and machine learning algorithms to predict genetic mutations and mRNA-based molecular subtypes. Next, we established predictive models for OS based on single omics, combined omics (radiomics+genomics, radiomics+transcriptomics, radiomics+proteomics) and all features (multi-omics).

Results

Using radiomics features, random forest algorithm showed good capacity to identify the mutations VHL (AUC=0.971), BAP1 (AUC=0.955), PBRM1 (AUC=0.972), SETD2 (AUC=0.949), and molecular subtypes m1 (AUC=0.973), m2 (AUC=0.968), m3 (AUC=0.961), m4 (AUC=0.953). The TCGA cohort was divided into training (n=104) and validation (n=103) sets. The radiomics model had promising prognostic value for OS in validation set (5-year AUC=0.775) and external validation set (5-year AUC=0.755). In the validation set, the radiomics+omics models enhanced predictive accuracy than single-omics models, and the multi-omics model made further improvement (5-year AUC=0.846). High-risk group of validation set predicted by multi-omics model showed significantly poorer OS (HR=6.20, 95%CI: 3.19-8.44, p<0.0001).

Conclusions

CT radiomics might be a feasible approach to predict genetic mutations, molecular subtypes and OS in ccRCC patients. Integrative analysis of radiogenomics may improve the survival prediction of ccRCC patients.

SUBMITTER: Zeng H

PROVIDER: S-EPMC8064160 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:PURPOSE:To investigate associations between computed tomographic (CT) features of clear cell renal cell carcinoma (RCC) and mutations in VHL, PBRM1, SETD2, KDM5C, or BAP1 genes. MATERIALS AND METHODS:The institutional review board approved this retrospective, hypothesis-generating study of 233 patients with clear cell RCC and waived the informed consent requirement. The study was HIPAA compliant. Three radiologists independently reviewed pretreatment CT images of all clear cell RCCs without knowledge of their genomic profile. One radiologist measured largest diameter and enhancement parameters of each clear cell RCC. Associations between CT features and mutations in VHL, PBRM1, SETD2, KDM5C, and BAP1 genes were tested by using the Fisher exact test. Associations between mutations and size and enhancement were assessed by using the independent t test. Interreader agreement was calculated by using the Fleiss ?. RESULTS:Mutation frequencies among clear cell RCCs were as follows: VHL, 53.2% (124 of 233); PBRM1, 28.8% (67 of 233); SETD2, 7.3% (17 of 233); KDM5C, 6.9% (16 of 233); and BAP1, 6.0% (14 of 233). Mutations of VHL were significantly associated with well-defined tumor margins (P = .013), nodular tumor enhancement (P = .021), and gross appearance of intratumoral vascularity (P = .018). Mutations of KDM5C and BAP1 were significantly associated with evidence of renal vein invasion (P = .022 and .046, respectively). The genotype of solid clear cell RCC differed significantly from the genotype of multicystic clear cell RCC. While mutations of SETD2, KDM5C, and BAP1 were absent in multicystic clear cell RCC, mutations of VHL (P = .016) and PBRM1 (P = .017) were significantly more common among solid clear cell RCC. Interreader agreement for CT feature assessments ranged from substantial to excellent (? = 0.791-0.912). CONCLUSION:This preliminary radiogenomics analysis of clear cell RCC revealed associations between CT features and underlying mutations that warrant further investigation and validation.

Project description:PurposeTo construct a novel radiogenomics biomarker based on hypoxic-gene subset for the accurate prognostic prediction of clear cell renal cell carcinoma (ccRCC).Materials and methodsInitially, we screened for the desired hypoxic-gene subset by analysis using the GSEA database. Through univariate and multivariate cox regression hazard ratio analysis, survival-related hypoxia genes were identified, and a genomics signature was constructed in the TCGA database. Building on this, a hypoxia-gene related radiogenomics biomarker (prediction of hypoxia-genes signature by contrast-enhanced CT radiomics) was constructed in the TCIA-KIRC database by extracting features in the venous phase of contrast-enhanced CT images, selecting features using the mRMR and LASSO algorithms, and building logistic regression models. Finally, we validated the prognostic capability of the new biomarker for patients with ccRCC in an independent validation cohort at Huashan Hospital of Fudan University, Shanghai, China.ResultsThe hypoxia-related genomics signature consisting of five genes (IFT57, PABPN1, RNF10, RNF19B and UBE2T) was shown to be significantly associated with survival for patients with ccRCC in the TCGA database, delineated by grouping of the signature expression as either low- or high-risk. In the TCIA database, we constructed a radiogenomics biomarker consisting of 13 radiomics features that were optimal predictors of hypoxia-gene signature expression levels (low- or high-risk) in patients at each institution, that demonstrated AUC values of 0.91 and 0.91 in the training and validation groups, respectively. In the independent validation cohort at Huashan Hospital, our radiogenomics biomarker was significantly associated with prognosis in patients with ccRCC (p=0.0059).ConclusionsThe novel prognostic radiogenomics biomarker that was constructed achieved excellent correlation with prognosis in both the cohort of TCGA/TCIA-KIRC database and the independent validation cohort of Huashan hospital patients with ccRCC. It is anticipated that this work may assist in clinical preferential treatment decisions and promote the process of precision theranostics in the future.