Identification of intrinsic imaging subtypes using clustering analysis based on dynamic contrast-enhanced magnetic resonance imaging radiomics features for gliomas: preliminary associations with gene expression profiles

Abstract

Background: There has been no research based on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) radiomics for the stratification diagnosis and prognostic evaluation of gliomas. The study aimed to identify multiple glioma subtypes and decipher the gene expression profiles linked with different subtypes.

Methods: Cross-sectional and retrospective data of 189 patients were collected. The static radiomics features were obtained at three time points (0, 90, and 300 s) corresponding to pre-contrast, arterial, and delayed phases, respectively. The dynamic radiomics features were retrieved by determining the temporal anisotropy of these three phases. Multi-omics clustering was used to identify intrinsic radiomics subtypes within the cohort. The association between the radiomics clusters and gene expression profiles was evaluated through the analysis of variance.

Results: The patients in cluster 3 were oldest. Cluster 3 and cluster 1 had higher frequency of grade 4, high Ki-67 level, glioblastoma isocitrate dehydrogenase (IDH) wild-type, and unmethylated O6-methylguanine-DNA methyltransferase (MGMT) promoter. Cluster 3 had the highest frequency of epidermal growth factor receptor (EGFR) amplification and cyclin-dependent kinase inhibitor (CDKN) 2A/B homozygous deletion. Cluster 1 had the highest frequency of EGFR non-mutant. Cluster 4 and cluster 2 had a higher frequency of astrocytoma IDH-mutant. Cluster 4 had a higher frequency of grade 3, oligodendroglioma IDH-mutant and 1p/19q codeleted, MGMT promoter methylation, and EGFR non-amplification. Cluster 2 had a higher frequency of grade 2, low Ki-67 level, and patients without CDKN 2A/B homozygous deletion. There were no associations for other molecular markers between clusters.

Conclusions: The intrinsic imaging subtypes obtained from DCE-MRI radiomics features provide a new insight into glioma classification, potentially guiding the diagnosis.

Keywords: Dynamic contrast-enhanced (DCE); diagnosis; glioma; radiomics.

Figure 1

Image processing and tumor delineation. (A) MR images of T₁CE and T₂WI; (B) MR images of T₁CE and T₂WI: 3D resampled; (C) T₁CE registered to T₂WI; (D) MR images of T₁CE and T₂WI: skull stripping; (E) MR images of T₁CE and T₂WI: TSOMT to 128³ tensor; (F) VOIs (MR images of T₁CE and T₂WI): CNN to predict VOIs on tensor. Red color indicates non-enhancing tumor; blue color indicates enhancing tumor; green color indicates peritumoral edema. (G) VOIs (MR images of T₁CE and T₂WI): inverse TSOMT (VOIs back to raw image). Red color indicates non-enhancing tumor; blue color indicates enhancing tumor; green color indicates peritumoral edema. (H) VOIs registered to DCE-MRI. Red color indicates non-enhancing tumor; blue color indicates enhancing tumor; green color indicates peritumoral edema. 3D, three-dimensional; CNN, convolutional neural network; DCE-MRI, dynamic contrast-enhanced magnetic resonance imaging; MR, magnetic resonance; T₁CE, T₁ contrast-enhanced; T₂WI, T₂-weighted imaging; TSOMT, two-stage optimal mass transport; VOIs, volumes of interest.

Figure 2

Consensus heatmap—the CM was derived from 10 advanced multi-omics clustering algorithms. Each element in the CM represents the frequency of paired patients in the same cluster, and from the CM, a hierarchical clustering was used to get a stable clustering result. CM, consensus matrix; CS, cluster subtype.

Figure 3

Flow chart of glioma patients. DCE, dynamic contrast-enhanced.

Figure 4

Using Clustering Prediction Index to determine the optimized number of cluster k.

Figure 5

The radiomics feature distribution heatmap in 4 subtypes. P is corresponding to timepoint 1, A is corresponding to timepoint 2, and V is corresponding to timepoint 3. TA, temporal anisotropy radiomics features.

Figure 6

Representative cases and DCE curves from each cluster. (A) Cluster 1: example for MR images (T₁CE) and DCE curve of a 55-year-old male with glioblastoma, IDH wild-type, grade 4. Ki-67 =30%, MGMT promoter unmethylation, EGFR non-amplification and non-mutant, CDKN 2A/B deletion. (B) Cluster 2: example for MR images (FLAIR) and DCE curve of a 53-year-old female with astrocytoma, IDH-mutant, grade 2. Ki-67 =5%, MGMT promoter methylation, TERT wild-type, EGFR non-amplification and non-mutant, without CDKN 2A/B deletion and chromosomes +7/−10. (C) Cluster 3: example for MR images (T₁CE) and DCE curve of a 65-year-old female with glioblastoma, IDH wild-type, grade 4. Ki-67 =50%, MGMT promoter unmethylation, EGFR amplification and mutant, CDKN 2A/B deletion and with chromosomes +7/−10. (D) Cluster 4: example for MR images (T₂WI) and DCE curve of a 44-year-old male with oligodendroglioma, IDH-mutant, 1p/19q codeleted, grade 2. Ki-67 =5%, MGMT promoter methylation. CDKN, cyclin-dependent kinase inhibitor; DCE, dynamic contrast-enhanced; EGFR, epidermal growth factor receptor; FLAIR, fluid-attenuated inversion recovery; IDH, isocitrate dehydrogenase; MGMT, O6-methylguanine-DNA methyltransferase; MR, magnetic resonance; T₁CE, T₁ contrast-enhanced; T₂WI, T₂-weighted imaging.