Evaluation of tumor shape features for overall survival prognosis in glioblastoma multiforme patients

Shape matters: unsupervised exploration of IDH-wildtype glioma imaging survival predictors

OBJECTIVES

This study examines clustering based on shape radiomic features and tumor volume to identify IDH-wildtype glioma phenotypes and assess their impact on overall survival (OS).

MATERIALS AND METHODS

This retrospective study included 436 consecutive patients diagnosed with IDH-wt glioma who underwent preoperative MR imaging. Alongside the total tumor volume, nine distinct shape radiomic features were extracted using the PyRadiomics framework. Different imaging phenotypes were identified using partition around medoids (PAM) clustering on the training dataset (348/436). The prognostic efficacy of these phenotypes in predicting OS was evaluated on the test dataset (88/436). External validation was performed using the public UCSF glioma dataset (n = 397). A decision-tree algorithm was employed to determine the relevance of features associated with cluster affiliation.

RESULTS

PAM clustering identified two clusters in the training dataset: Cluster 1 (n = 233) had a higher proportion of patients with higher sphericity and elongation, while Cluster 2 (n = 115) had a higher proportion of patients with higher maximum 3D diameter, surface area, axis lengths, and tumor volume (p < 0.001 for each). OS differed significantly between clusters: Cluster 1 showed a median OS of 23.8 compared to 11.4 months of Cluster 2 in the holdout test dataset (p = 0.002). Multivariate Cox regression showed improved performance with cluster affiliation over clinical data alone (C index 0.67 vs 0.59, p = 0.003). Cluster-based models outperformed the models with tumor volume alone (evidence ratio: 5.16-5.37).

CONCLUSION

Data-driven clustering reveals imaging phenotypes, highlighting the improved prognostic power of combining shape-radiomics with tumor volume, thereby outperforming predictions based on tumor volume alone in high-grade glioma survival outcomes.

CLINICAL RELEVANCE STATEMENT

Shape-radiomics and volume-based cluster analyses of preoperative MRI scans can reveal imaging phenotypes that improve the prediction of OS in patients with IDH-wild type gliomas, outperforming currently known models based on tumor size alone or clinical parameters.

KEY POINTS

Shape radiomics and tumor volume clustering in IDH-wildtype gliomas are investigated for enhanced prognostic accuracy. Two distinct phenotypic clusters were identified with different median OSs. Integrating shape radiomics and volume-based clustering enhances OS prediction in IDH-wildtype glioma patients.

Novel dual targeting cubosomes modified with angiopep-2 for co-delivery GNA and PLHSpT to brain glioma

3Glioblastoma multiforme is the most aggressive malignant brain tumor. However, the treatment of glioblastoma multiforme faces great challenges owing to difficult penetration of the blood-brain barrier. Therefore, more effective treatment strategies are desired quite urgently. In our study, a dual-targeting drug delivery system for co-loading with hydrophobic Gambogenic acid and hydrophilic PLHSpT was developed by cubosomes with angiopep-2 decorating. The Ang-cubs-(GNA + PLHSpT) was prepared by high-temperature emulsification-low-temperature solidification demonstrating excellent physical properties.Transmission electron microscopy revealed that Ang-cubs-(GNA + PLHSpT) was nearly spherical with a "core-shell" double-layer structure. Differential scanning calorimetry suggested that a new phase was formed. Small-angle X-ray scattering also verified that Ang-cubs-(GNA + PLHSpT) retains the Pn3m cubic. Moreover, laser confocal indicated that Ang-cubs-(GNA + PLHSpT) was capable of crossing BBB via binding to lipoprotein receptor-related protein-1, likely suggesting the potential tumor-specific targeting characteristic. Compared to free drug and cubs-(GNA + PLHSpT), Ang-cubs-(GNA + PLHSpT) was easily taken up by C6 cell and exhibited better anti-glioma effects in vitro. Importantly, GNA and PLHSpT co-loaded Ang-cubs could suppress tumor growth and significantly prolong survival in vivo. In conclusion, Ang-cubs-(GNA + PLHSpT) acts as a new dual-targeting drug delivery system for the treatment of GBM.

Association of quantitative radiomic shape features with functional outcome after surgery for primary sporadic dorsal spinal meningiomas

Objective

Spinal meningiomas (SM) account for 25%-46% of all primary spinal tumors and show an excellent long-term disease control in case of complete resection. Therefore, the postoperative functional outcome is of high importance. To date, reports on dorsally located SM are scarce. Moreover, the impact of radiomics shape features on the functional outcome after surgery for primary dorsal SMs has not been analyzed yet.

Methods

We retrospectively performed an analysis of shape-based radiomic features in 3D slicer software and quantified the tumor volume, surface area, sphericity, surface area to volume ratio and tumor canal ratio. Subsequently, we evaluated the correlation between the radinomic parameters and the postoperative outcome according to Modified Japanese Orthopedic Association (mJOA) score.

Results

Between 2010 and 2022, we identified 24 Females and 2 Males operated on dorsal SMs in our institutional database. The most common SM localization was thoracic spine (n = 20), followed by cervical (n = 4), and lumbar (n = 2). The univariate analysis and the receiver operating characteristic (ROC) analysis showed a strong diagnostic performance of sphericity in the prediction of postoperative functional outcome based on mJOA score (AUC of 0.79, sphericity cut-of value 0.738; p = 0.01). Subsequently, the patients were divided into two groups (mJOA improved vs. mJOA stable/worsened). Patients with improved mJOA score showed significantly higher sphericity (0.79 ± 0.1 vs. 0.70 ± 1.0; p = 0.03). Finally, we divided the cohort based on sphericity (<0.738 and ≥0.738). The group with higher sphericity exhibited a significantly higher positive mJOA difference 3 months postoperatively (16.6 ± 1.4 vs. 14.8 ± 3.7; p = 0.03).

Conclusion

In our study investigating primary sporadic dorsal SMs, we demonstrated that a higher degree of sphericity may be a positive predictor of postoperative improvement, as indicated by the mJOA score.

Time-to-event overall survival prediction in glioblastoma multiforme patients using magnetic resonance imaging radiomics

PURPOSE

Glioblastoma Multiforme (GBM) represents the predominant aggressive primary tumor of the brain with short overall survival (OS) time. We aim to assess the potential of radiomic features in predicting the time-to-event OS of patients with GBM using machine learning (ML) algorithms.

MATERIALS AND METHODS

One hundred nineteen patients with GBM, who had T1-weighted contrast-enhanced and T2-FLAIR MRI sequences, along with clinical data and survival time, were enrolled. Image preprocessing methods included 64 bin discretization, Laplacian of Gaussian (LOG) filters with three Sigma values and eight variations of Wavelet Transform. Images were then segmented, followed by the extraction of 1212 radiomic features. Seven feature selection (FS) methods and six time-to-event ML algorithms were utilized. The combination of preprocessing, FS, and ML algorithms (12 × 7 × 6 = 504 models) was evaluated by multivariate analysis.

RESULTS

Our multivariate analysis showed that the best prognostic FS/ML combinations are the Mutual Information (MI)/Cox Boost, MI/Generalized Linear Model Boosting (GLMB) and MI/Generalized Linear Model Network (GLMN), all of which were done via the LOG (Sigma = 1 mm) preprocessing method (C-index = 0.77). The LOG filter with Sigma = 1 mm preprocessing method, MI, GLMB and GLMN achieved significantly higher C-indices than other preprocessing, FS, and ML methods (all p values < 0.05, mean C-indices of 0.65, 0.70, and 0.64, respectively).

CONCLUSION

ML algorithms are capable of predicting the time-to-event OS of patients using MRI-based radiomic and clinical features. MRI-based radiomics analysis in combination with clinical variables might appear promising in assisting clinicians in the survival prediction of patients with GBM. Further research is needed to establish the applicability of radiomics in the management of GBM in the clinic.

Impact of Shape Irregularity in Medial Sphenoid Wing Meningiomas on Postoperative Cranial Nerve Functioning, Proliferation, and Progression-Free Survival

Medial sphenoid wing meningiomas (MSWM) are surgically challenging skull base tumors. Irregular tumor shapes are thought to be linked to histopathology. The present study aims to investigate the impact of tumor shape on postoperative functioning, progression-free survival, and neuropathology. This monocentric study included 74 patients who underwent surgery for primary sporadic MSWM (WHO grades 1 and 2) between 2010 and 2021. Furthermore, a systematic review of the literature regarding meningioma shape and the MIB-1 index was performed. Irregular MSWM shapes were identified in 31 patients (41.9%). Multivariable analysis revealed that irregular shape was associated with postoperative cranial nerve deficits (OR: 5.75, 95% CI: 1.15-28.63, p = 0.033). In multivariable Cox regression analysis, irregular MSWM shape was independently associated with tumor progression (HR:8.0, 95% CI: 1.04-62.10, p = 0.046). Multivariable regression analysis showed that irregular shape is independently associated with an increased MIB-1 index (OR: 7.59, 95% CI: 2.04-28.25, p = 0.003). A systematic review of the literature and pooled data analysis, including the present study, showed that irregularly shaped meningiomas had an increase of 1.98 (95% CI: 1.38-2.59, p < 0.001) in the MIB-1 index. Irregular MSWM shape is independently associated with an increased risk of postoperative cranial nerve deficits and a shortened time to tumor progression. Irregular MSWM shapes might be caused by highly proliferative tumors.

Relapse patterns and radiation dose exposure in IDH wild-type glioblastoma at first radiographic recurrence following chemoradiation

PURPOSE

To quantify the radiation dose distribution and lesion morphometry (shape) at baseline, prior to chemoradiation, and at the time of radiographic recurrence in patients with glioblastoma (GBM).

METHODS

The IMRT dose distribution, location of the center of mass, sphericity, and solidity of the contrast enhancing tumor at baseline and the time of tumor recurrence was quantified in 48 IDH wild-type GBM who underwent postoperative IMRT (2 Gy daily for total of 60 Gy) with concomitant and adjuvant temozolomide.

RESULTS

Average radiation dose within enhancing tumor at baseline and recurrence was ≥ 60 Gy. Centroid location of the enhancing tumor shifted an average of 11.3 mm at the time of recurrence with respect to pre-IMRT location. A positive correlation was observed between change in centroid location and PFS in MGMT methylated patients (P = 0.0007) and Cox multivariate regression confirmed centroid distance from baseline was associated with PFS when accounting for clinical factors (P = 0.0189). Lesion solidity was higher at recurrence compared to baseline (P = 0.0118). Tumors that progressed > 12 weeks after IMRT were significantly more spherical (P = 0.0094).

CONCLUSION

Most GBMs recur local within therapeutic IMRT doses; however, tumors with longer PFS occurred further from the original tumor location and were more solid and/or nodular.

Identifying risk factors of intracerebral hemorrhage stability using explainable attention model

Segmentation of intracerebral hemorrhage (ICH) helps improve the quality of diagnosis, draft the desired treatment methods, and clinically observe the variations with healthy patients. The clinical utilization of various ICH progression scoring systems has limitations due to the systems' modest predictive value. This paper proposes a single pipeline of a multi-task model for end-to-end hemorrhage segmentation and risk estimation. We introduce a 3D spatial attention unit and integrate it into the state-of-the-art segmentation architecture, UNet, to enhance the accuracy by bootstrapping the global spatial representation. We further extract the geometric features from the segmented hemorrhage volume and fuse them with clinical features such as CT angiography (CTA) spot, Glasgow Coma Scale (GCS), and age to predict the ICH stability. Several state-of-the-art machine learning techniques such as multilayer perceptron (MLP), support vector machine (SVM), gradient boosting, and random forests are applied to train stability estimation and to compare the performances. To align clinical intuition with model learning, we determine the shapely values (SHAP) and explain the most significant features for the ICH risk scoring system. A total of 79 patients are included, of which 20 are found in critical condition. Our proposed single pipeline model achieves a segmentation accuracy of 86.3%, stability prediction accuracy of 78.3%, and precision of 82.9%; the mean square error of exact expansion rate regression is observed to be 0.46. The SHAP analysis reveals that CTA spot sign, age, solidity, location, and length of the first axis of the ICH volume are the most critical characteristics that help define the stability of the stroke lesion. We also show that integrating significant geometric features with clinical features can improve the ICH progression scoring by predicting long-term outcomes. Graphical abstract Overview of our proposed method comprising of spatial attention and feature extraction mechanisms. The architecture is trained on the input CT images, and the first step output is the predicted segmentation of the hemorrhagic region. The output is fed into a geometric feature extractor and is fused with clinical features to estimate ICH stability using a multilayer perceptron (MLP).

Extent of resection and survival in patients with glioblastoma multiforme: Systematic review and meta-analysis

BACKGROUND

Glioblastoma multiforme (GBM) owes an ominous prognosis: its mean overall survival is 14 months. The extent of surgical resection (ESR) highlights among factors in which an association has been found to a somewhat better prognosis. However, the association between greater ESR and prolonged overall (OS) survival is not a constant finding nor a proven cause-and-effect phenomenon. To our objective is to establish the strength of association between ESR and OS in patients with GBM through a systematic review and meta-analysis.

METHODS

In accordance with PRISMA-P recommendations, we conducted a systematic literature search; we included studies with adult patients who had undergone craniotomy for GBM. Our primary outcome is overall postoperative survival at 12 and 24 months. We reviewed 180 studies, excluded 158, and eliminated 8; 14 studies that suited our requirements were analyzed.

RESULTS

The initial level of evidence of all studies is low, and it may be degraded to very low according to GRADE criteria because of design issues. The definition of different levels of the extent of resection is heterogeneous and poorly defined. We found a great amount of variation in the methodology of the operation and the adjuvant treatment protocol. The combined result for relative risk (RR) for OS for 12 months analysis is 1.25 [95% confidence interval (95% CI) 1.14-1.36, P < .01], absolute risk reduction (ARR) of 15.7% (95% CI 11.9-19.4), relative risk reduction (RRR) of 0.24 (95% CI 0.18-0.31), number needed to treat (NNT) 6; for 24-month analysis RR is 1.59 (95% CI 1.11-2.26, P < .01) ARR of 11.5% (95% CI 7.7-15.1), relative risk reduction (RRR) of 0.53 (95% CI 0.33-0.76), (NNT) 9. In each term analysis, the proportion of alive patients who underwent more extensive resection is significantly higher than those who underwent subtotal resection.

CONCLUSION

Our results sustain a weak but statistically significant association between the ESR and OS in patients with GBM obtained from observational studies with a very low level of evidence according to GRADE criteria. As a consequence, any estimate of effect is very uncertain. Current information cannot sustain a cause-and-effect relationship between these variables.

Introduction to radiomics and radiogenomics in neuro-oncology: implications and challenges

Neuro-oncology largely consists of malignancies of the brain and central nervous system including both primary as well as metastatic tumors. Currently, a significant clinical challenge in neuro-oncology is to tailor therapies for patients based on a priori knowledge of their survival outcome or treatment response to conventional or experimental therapies. Radiomics or the quantitative extraction of subvisual data from conventional radiographic imaging has recently emerged as a powerful data-driven approach to offer insights into clinically relevant questions related to diagnosis, prediction, prognosis, as well as assessing treatment response. Furthermore, radiogenomic approaches provide a mechanism to establish statistical correlations of radiomic features with point mutations and next-generation sequencing data to further leverage the potential of routine MRI scans to serve as "virtual biopsy" maps. In this review, we provide an introduction to radiomic and radiogenomic approaches in neuro-oncology, including a brief description of the workflow involving preprocessing, tumor segmentation, and extraction of "hand-crafted" features from the segmented region of interest, as well as identifying radiogenomic associations that could ultimately lead to the development of reliable prognostic and predictive models in neuro-oncology applications. Lastly, we discuss the promise of radiomics and radiogenomic approaches in personalizing treatment decisions in neuro-oncology, as well as the challenges with clinical adoption, which will rely heavily on their demonstrated resilience to nonstandardization in imaging protocols across sites and scanners, as well as in their ability to demonstrate reproducibility across large multi-institutional cohorts.

A Novel Four-Gene Signature Associated With Immune Checkpoint for Predicting Prognosis in Lower-Grade Glioma

The overall survival of patients with lower grade glioma (LGG) varies greatly, but the current histopathological classification has limitations in predicting patients’ prognosis. Therefore, this study aims to find potential therapeutic target genes and establish a gene signature for predicting the prognosis of LGG. CD44 is a marker of tumor stem cells and has prognostic value in various tumors, but its role in LGG is unclear. By analyzing three glioma datasets from Gene Expression Omnibus (GEO) database, CD44 was upregulated in LGG. We screened 10 CD44-related genes via protein–protein interaction (PPI) network; function enrichment analysis demonstrated that these genes were associated with biological processes and signaling pathways of the tumor; survival analysis showed that four genes (CD44, HYAL2, SPP1, MMP2) were associated with the overall survival (OS) and disease-free survival (DFS)of LGG; a novel four-gene signature was constructed. The prediction model showed good predictive value over 2-, 5-, 8-, and 10-year survival probability in both the development and validation sets. The risk score effectively divided patients into high- and low- risk groups with a distinct outcome. Multivariate analysis confirmed that the risk score and status of IDH were independent prognostic predictors of LGG. Among three LGG subgroups based on the presence of molecular parameters, IDH-mutant gliomas have a favorable OS, especially if combined with 1p/19q codeletion, which further confirmed the distinct biological pattern between three LGG subgroups, and the gene signature is able to divide LGG patients with the same IDH status into high- and low- risk groups. The high-risk group possessed a higher expression of immune checkpoints and was related to the activation of immunosuppressive pathways. Finally, this study provided a convenient tool for predicting patient survival. In summary, the four prognostic genes may be therapeutic targets and prognostic predictors for LGG; this four-gene signature has good prognostic prediction ability and can effectively distinguish high- and low-risk patients. High-risk patients are associated with higher immune checkpoint expression and activation of the immunosuppressive pathway, providing help for screening immunotherapy-sensitive patients.