Advanced MRI, Radiomics and Radiogenomics in Unravelling Incidental Glioma Grading and Genetic Status: Where Are We?

Abstract

The 2021 WHO classification of brain tumours revolutionised the oncological field by emphasising the role of molecular, genetic and pathogenetic advances in classifying brain tumours. In this context, incidental gliomas have been increasingly identified due to the widespread performance of standard and advanced MRI sequences and represent a diagnostic and therapeutic challenge. The impactful decision to perform a surgical procedure deeply relies on the non-invasive identification of features or parameters that may correlate with brain tumour genetic profile and grading. Therefore, it is paramount to reach an early and proper diagnosis through neuroradiological techniques, such as MRI. Standard MRI sequences are the cornerstone of diagnosis, while consolidated and emerging roles have been awarded to advanced sequences such as Diffusion-Weighted Imaging/Apparent Diffusion Coefficient (DWI/ADC), Perfusion-Weighted Imaging (PWI), Magnetic Resonance Spectroscopy (MRS), Diffusion Tensor Imaging (DTI) and functional MRI (fMRI). The current novelty relies on the application of AI in brain neuro-oncology, mainly based on radiomics and radiogenomics models, which enhance standard and advanced MRI sequences in predicting glioma genetic status by identifying the mutation of multiple key biomarkers deeply impacting patients' diagnosis, prognosis and treatment, such as IDH, EGFR, TERT, MGMT promoter, p53, H3-K27M, ATRX, Ki67 and 1p19. AI-driven models demonstrated high accuracy in glioma detection, grading, prognostication, and pre-surgical planning and appear to be a promising frontier in the neuroradiological field. On the other hand, standardisation challenges in image acquisition, segmentation and feature extraction variability, data scarcity and single-omics analysis, model reproducibility and generalizability, the black box nature and interpretability concerns, as well as ethical and privacy challenges remain key issues to address. Future directions, rooted in enhanced standardisation and multi-institutional validation, advancements in multi-omics integration, and explainable AI and federated learning, may effectively overcome these challenges and promote efficient AI-based models in glioma management. The aims of our multidisciplinary review are to: (1) extensively present the role of standard and advanced MRI sequences in the differential diagnosis of iLGGs as compared to HGGs (High-Grade Gliomas); (2) give an overview of the current and main applications of AI tools in the differential diagnosis of iLGGs as compared to HGGs (High-Grade Gliomas); (3) show the role of MRI, radiomics and radiogenomics in unravelling glioma genetic profiles. Standard and advanced MRI, radiomics and radiogenomics are key to unveiling the grading and genetic profile of gliomas and supporting the pre-operative planning, with significant impact on patients' differential diagnosis, prognosis prediction and treatment strategies. Today, neuroradiologists are called to efficiently use AI tools for the in vivo, non-invasive, and comprehensive assessment of gliomas in the path towards patients' personalised medicine.

Keywords: DTI; EGFR; IDH; MGMT; PWI; high-grade glioma; incidental gliomas; low-grade glioma; personalised medicine; radiogenomics; radiomics.

Figure 1

Shows the brain MRIs of two patients affected by LGGs. A 32-year-old man affected by an incidental grade 2, IDH-mutant diffuse astrocytoma involving the inferior and lateral (fronto-opercular) portion of the left frontal lobe. Axial T2WI (A) demonstrate a well-circumscribed hyperintense lesion, with the typical T2WI/FLAIR (A,B) mismatch. A 31-year-old woman affected by an incidental grade 2, IDH-mutant diffuse astrocytoma involving the fronto-opercular portion of the right frontal lobe (C,D). Axial T2WI (C) and FLAIR (D) sequences show a round lesion with the typical T2/FLAIR mismatch sign. LGG: low-grade glioma; WI: weighted image; FLAIR: fluid-attenuated inversion recovery.

Figure 2

Compares the post-contrast T1WI sequences of patients affected by a grade 2, IDH-mutant diffuse astrocytoma involving the fronto-opercular portion of the right frontal lobe (A), a grade 4, IDH wild-type primary glioblastoma of the premotor area of the right frontal lobe (B), and a grade 4, IDH wild-type primary glioblastoma of the right frontal lobe. The LGG (A) appears hypointense with no contrast enhancement, while the HGGs (B,C) appear as inhomogeneous lesions with strong and irregular “ring-enhancement” surrounding the necrotic core (B,C). LGG: low-grade glioma; HGG: high-grade glioma; WI: weighted image.

Figure 3

A 48-year-old woman affected by histologically proven grade 2, IDH-mutant, 1p/19q codeleted oligodendroglioma centred in the right superior frontal gyrus and appearing heterogeneously hyperintense on axial T2WI and FLAIR sequences (A,B) and hypointense on T1WI (C) with poor/absent enhancement on post-contrast axial (D) and coronal fat-saturated T1WI (H). On DSC-PWI (G), the lesion is characterised by high rCBV, while there is an intermediate signal on ADC (E) and some “blooming” foci on SWI (F). WI: weighted image; FLAIR: fluid-attenuated inversion recovery; DSC: dynamic susceptibility contrast; PWI: perfusion WI; rCBV: relative cerebral blood volume; ADC: apparent diffusion coefficient; SWI: susceptibility weighted image.

Figure 4

Comparison between a large grade 2 IDH-mutant glioma involving the left insular region (A–C) and a grade 4 IDH wild-type glioma involving the left frontal lobe (D–F). Both lesions appear heterogeneously hyperintense on T2WI (A,D) with no significant contrast enhancement on T1WI (B,E). Nevertheless, PWI-DSC rCBV suggests the differential diagnosis based on the increased perfusion identifiable in the grade 4 IDH wild-type glioma (F) as compared to the absence of increased perfusion of the grade 2 IDH-mutant glioma (C). WI: weighted image; DSC: dynamic susceptibility contrast; PWI: perfusion WI; rCBV: relative cerebral blood volume.

Figure 5

Shows a grade 4, IDH wild-type primary glioblastoma of the right frontal lobe (A–C) and a grade 4, IDH wild-type primary glioblastoma of the premotor area of the right frontal lobe (D–F). Both lesions appear inhomogeneously hyperintense on T2WI (B,D) and FLAIR (C,F), and inhomogeneously hypointense on T1WI (A,E). Peripheral vasogenic oedema appears hypointense on T1WI (A,E) and hyperintense on T2WI/FLAIR (B–D,F). WI: weighted image; FLAIR: fluid-attenuated inversion recovery.

Figure 6

Compares the PWI-DSC-derived rCBV maps of patients affected by a grade 2, IDH-mutant diffuse astrocytoma involving the fronto-opercular portion of the right frontal lobe (A), a grade 4, IDH wild-type primary glioblastoma of the right frontal lobe (B), and a grade 4, IDH wild-type primary glioblastoma of the premotor area of the right frontal lobe. The LGG (A) shows no increased signal on the rCBV map, while the HGGs (B,C) present inhomogenously increased signal on the rCBV maps in the non-necrotic peripheral tumoral areas (B,C). DSC: dynamic susceptibility contrast; PWI: perfusion WI; rCBV: relative cerebral blood volume; LGG (low-grade glioma).

Figure 7

Compares the multi-voxel MRS with intermediate TE of an incidental grade 2, IDH-mutant diffuse astrocytoma involving the fronto-opercular portion of the right frontal lobe, which demonstrates a reduction in NAA and an increase in Cho peaks with inversion of the Cho/NAA ratio (A), and a grade 4, IDH wild-type primary glioblastoma of the right frontal lobe, which demonstrates a very high peak of Cho and a decreased NAA peak, resulting in the inversion of the Cho/NAA ratio (B). Cho: choline; NAA: N-Acetyl-Aspartate; MRS: magnetic resonance spectroscopy; TE: echo time.

Figure 8

Compares the SWI sequences of patients affected by a grade 2, IDH-mutant diffuse astrocytoma involving the fronto-opercular portion of the right frontal lobe (A), a grade 4, IDH wild-type primary glioblastoma of the right frontal lobe (B), and a grade 4, IDH wild-type primary glioblastoma of the premotor area of the right frontal lobe (C). No “blooming” foci are present in the SWI sequence of the LGG (A), while HGGs show inhomogeneously decreased signal intensity in the tumoral vital areas, suggestive of neovascularisation. SWI: susceptibility weighted image; LGG: low-grade glioma; HGG: high-grade glioma.

Figure 9

Compares the DWI/ADC sequences of patients affected by an incidental grade 2, IDH-mutant diffuse astrocytoma involving the inferior and lateral (fronto-opercular) portion of the left frontal lobe (A,B), a grade 2, IDH-mutant diffuse astrocytoma involving the fronto-opercular portion of the right frontal lobe (C), a grade 4, IDH wild-type primary glioblastoma of the premotor area of the right frontal lobe (D), and a grade 4, IDH wild-type primary glioblastoma of the right frontal lobe (E). LGGs (A–C) do not show diffusion restriction, while HGGs (D,E) demonstrate significant diffusion restriction of the peripheral vital areas. DWI: diffusion-weighted image; ADC: apparent diffusion coefficient; LGG: low-grade glioma; HGG: high-grade glioma.

Figure 10

Shows the DTI-derived FA map of a grade 2, IDH-mutant diffuse astrocytoma involving the fronto-opercular portion of the right frontal lobe, which demonstrates no disruption of the right cortico-spinal tract (A), and the DTI-derived FA maps of a grade 4, IDH wild-type primary glioblastoma of the right frontal lobe, which demonstrates the disruption of the white matter tracts involved and surrounding the HGG (B). (C) Shows the DTI-derived FA map of a grade 4, IDH wild-type primary glioblastoma of the premotor areas of the right frontal lobe, which shows disrupted white matter tracts caused by the lesion, which is anterior to the right cortico-spinal tract. These findings are easily detectable when the DTI-derived FA map is superimposed on the post-contrast T1WI (D). DTI: diffusion tensor imaging; FA: fractional anisotropy; WI: weighted image; HGG: high-grade glioma.

Figure 11

Compares the fMRIs of: an incidental grade 2, IDH-mutant diffuse astrocytoma involving the inferior and lateral (fronto-opercular) portion of the left frontal lobe, whose task-based fMRI study shows a cluster of mouth activation close to the posterior margin of the lesion (A); a grade 4, IDH wild-type primary glioblastoma of the right frontal lobe, task-based fMRI study shows a bilateral cluster of mouth activation (B); and a grade 4, IDH wild-type primary glioblastoma of the premotor area of the right frontal lobe, task-based fMRI study shows a cluster of activation corresponding to the left hand, whose brain area appears close but not infiltrated by the lesion (C). fMRI (functional MRI).