Pseudoprogression in GBM versus true progression in patients with glioblastoma: A multiapproach analysis

Deep Learning-Based Techniques in Glioma Brain Tumor Segmentation Using Multi-Parametric MRI: A Review on Clinical Applications and Future Outlooks

This comprehensive review explores the role of deep learning (DL) in glioma segmentation using multiparametric magnetic resonance imaging (MRI) data. The study surveys advanced techniques such as multiparametric MRI for capturing the complex nature of gliomas. It delves into the integration of DL with MRI, focusing on convolutional neural networks (CNNs) and their remarkable capabilities in tumor segmentation. Clinical applications of DL-based segmentation are highlighted, including treatment planning, monitoring treatment response, and distinguishing between tumor progression and pseudo-progression. Furthermore, the review examines the evolution of DL-based segmentation studies, from early CNN models to recent advancements such as attention mechanisms and transformer models. Challenges in data quality, gradient vanishing, and model interpretability are discussed. The review concludes with insights into future research directions, emphasizing the importance of addressing tumor heterogeneity, integrating genomic data, and ensuring responsible deployment of DL-driven healthcare technologies. EVIDENCE LEVEL: N/A TECHNICAL EFFICACY: Stage 2.

Perioperative imaging predictors of tumor progression and pseudoprogression: A systematic review

In high-grade gliomas, pseudoprogression after radiation treatment might dramatically impact patient's management. We searched for perioperative imaging predictors of pseudoprogression in high-grade gliomas according to PRISMA guidelines, using MEDLINE/Pubmed and Embase (until January 2024). Study design, sample size, setting, diagnostic gold standard, imaging modalities and contrasts, and differences among variables or measures of diagnostic accuracy were recorded. Study quality was assessed through the QUADAS-2 tool. Twelve studies (11 with MRI, one with PET; 1058 patients) were reviewed. Most studies used a retrospective design (9/12), and structural MRI (7/12). Studies were heterogeneous in metrics and diagnostic reference standards; patient selection bias was a frequent concern. Pseudoprogression and progression showed some significant group differences in perioperative imaging metrics, although often with substantial overlap. Radiomics showed moderate accuracy but requires further validation. Current literature is scarce and limited by methodological concerns, highlighting the need of new predictors and multiparametric approaches.

Early pseudoprogression and progression lesions in glioblastoma patients are both metabolically heterogeneous

The standard treatment in glioblastoma includes maximal safe resection followed by concomitant radiotherapy plus chemotherapy and adjuvant temozolomide. The first follow-up study to evaluate treatment response is performed 1 month after concomitant treatment, when contrast-enhancing regions may appear that can correspond to true progression or pseudoprogression. We retrospectively evaluated 31 consecutive patients at the first follow-up after concomitant treatment to check whether the metabolic pattern assessed with multivoxel MRS was predictive of treatment response 2 months later. We extracted the underlying metabolic patterns of the contrast-enhancing regions with a blind-source separation method and mapped them over the reference images. Pattern heterogeneity was calculated using entropy, and association between patterns and outcomes was measured with Cramér's V. We identified three distinct metabolic patterns-proliferative, necrotic, and responsive, which were associated with status 2 months later. Individually, 70% of the patients showed metabolically heterogeneous patterns in the contrast-enhancing regions. Metabolic heterogeneity was not related to the regions' size and only stable patients were less heterogeneous than the rest. Contrast-enhancing regions are also metabolically heterogeneous 1 month after concomitant treatment. This could explain the reported difficulty in finding robust pseudoprogression biomarkers.

MRI Treatment Response Assessment Maps (TRAMs) for differentiating recurrent glioblastoma from radiation necrosis

BACKGROUND

MRI treatment response assessment maps (TRAMs) were introduced to distinguish recurrent malignant glioma from therapy related changes. TRAMs are calculated with two contrast-enhanced T1-weighted sequences and reflect the "late" wash-out (or contrast clearance) and wash-in of gadolinium. Vital tumor cells are assumed to produce a wash-out because of their high turnover rate and the associated hypervascularization, whereas contrast medium slowly accumulates in scar tissue. To examine the real value of this method, we compared TRAMs with the pathology findings obtained after a second biopsy or surgery when recurrence was suspected.

METHODS

We retrospectively evaluated TRAMs in adult patients with histologically demonstrated glioblastoma, contrast-enhancing tissue and a pre-operative MRI between January 1, 2017, and December 31, 2022. Only patients with a second biopsy or surgery were evaluated. Volumes of the residual tumor, contrast clearance and contrast accumulation before the second surgery were analyzed.

RESULTS

Among 339 patients with mGBM who underwent MRI, we identified 29 repeated surgeries/biopsies in 27 patients 59 ± 12 (mean ± standard deviation) years of age. Twenty-eight biopsies were from patients with recurrent glioblastoma histology, and only one was from a patient with radiation necrosis. We volumetrically evaluated the 29 pre-surgery TRAMs. In recurrent glioblastoma, the ratio of wash-out volume to tumor volume was 36 ± 17% (range 1-73%), and the ratio of the wash-out volume to the sum of wash-out and wash-in volumes was 48 ± 21% (range 22-92%). For the one biopsy with radiation necrosis, the ratios were 42% and 54%, respectively.

CONCLUSIONS

Typical recurrent glioblastoma shows a > 20%ratio of the wash-out volume to the sum of wash-out and wash-in volumes. The one biopsy with radiation necrosis indicated that such necrosis can also produce high wash-out in individual cases. Nevertheless, the additional information provided by TRAMs increases the reliability of diagnosis.

Randomized phase III trial of metabolic imaging-guided dose escalation of radio-chemotherapy in patients with newly diagnosed glioblastoma (SPECTRO GLIO trial)

BACKGROUND

Glioblastoma (GBM) systematically recurs after a standard 60 Gy radio-chemotherapy regimen. Since magnetic resonance spectroscopic imaging (MRSI) has been shown to predict the site of relapse, we analyzed the effect of MRSI-guided dose escalation on overall survival (OS) of patients with newly diagnosed GBM.

METHODS

In this multicentric prospective phase III trial, patients who had undergone biopsy or surgery for a GBM were randomly assigned to a standard dose (SD) of 60 Gy or a high dose (HD) of 60 Gy with an additional simultaneous integrated boost totaling 72 Gy to MRSI metabolic abnormalities, the tumor bed and residual contrast enhancements. Temozolomide was administered concomitantly and maintained for 6 months thereafter.

RESULTS

One hundred and eighty patients were included in the study between March 2011 and March 2018. After a median follow-up of 43.9 months (95% CI [42.5; 45.5]), median OS was 22.6 months (95% CI [18.9; 25.4]) versus 22.2 months (95% CI [18.3; 27.8]) for HD, and median progression-free survival was 8.6 (95% CI [6.8; 10.8]) versus 7.8 months (95% CI [6.3; 8.6]), in SD versus HD, respectively. No increase in toxicity rate was observed in the study arm. The pseudoprogression rate was similar across the SD (14.4%) and HD (16.7%) groups. For O(6)-methylguanine-DNA methyltransferase (MGMT) methylated patients, the median OS was 38 months (95% CI [23.2; NR]) for HD patients versus 28.5 months (95% CI [21.1; 35.7]) for SD patients.

CONCLUSION

The additional MRSI-guided irradiation dose totaling 72 Gy was well tolerated but did not improve OS in newly diagnosed GBM.

TRIAL REGISTRATION

NCT01507506; registration date: December 20, 2011. https://clinicaltrials.gov/ct2/show/NCT01507506?cond=NCT01507506&rank=1.

The STEMRI trial: Magnetic resonance spectroscopy imaging can define tumor areas enriched in glioblastoma stem-like cells

Despite maximally safe resection of the magnetic resonance imaging (MRI)-defined contrast-enhanced (CE) central tumor area and chemoradiotherapy, most patients with glioblastoma (GBM) relapse within a year in peritumoral FLAIR regions. Magnetic resonance spectroscopy imaging (MRSI) can discriminate metabolic tumor areas with higher recurrence potential as CNI+ regions (choline/N-acetyl-aspartate index >2) can predict relapse sites. As relapses are mainly imputed to glioblastoma stem-like cells (GSCs), CNI+ areas might be GSC enriched. In this prospective trial, 16 patients with GBM underwent MRSI/MRI before surgery/chemoradiotherapy to investigate GSC content in CNI-/+ biopsies from CE/FLAIR. Biopsy and derived-GSC characterization revealed a FLAIR/CNI+ sample enrichment in GSC and in gene signatures related to stemness, DNA repair, adhesion/migration, and mitochondrial bioenergetics. FLAIR/CNI+ samples generate GSC-enriched neurospheres faster than FLAIR/CNI-. Parameters assessing biopsy GSC content and time-to-neurosphere formation in FLAIR/CNI+ were associated with worse patient outcome. Preoperative MRI/MRSI would certainly allow better resection and targeting of FLAIR/CNI+ areas, as their GSC enrichment can predict worse outcomes.

Disease Assessments in Patients with Glioblastoma

PURPOSE OF REVIEW

The neuro-oncology team faces a unique challenge when assessing treatment response in patients diagnosed with glioblastoma. Magnetic resonance imaging (MRI) remains the standard imaging modality for measuring therapeutic response in both clinical practice and clinical trials. However, even for the neuroradiologist, MRI interpretations are not straightforward because of tumor heterogeneity, as evidenced by varying degrees of enhancement, infiltrating tumor patterns, cellular densities, and vasogenic edema. The situation is even more perplexing following therapy since treatment-related changes can mimic viable tumor. Additionally, antiangiogenic therapies can dramatically decrease contrast enhancement giving the false impression of decreasing tumor burden. Over the past few decades, several approaches have emerged to augment and improve visual interpretation of glioblastoma response to therapeutics. Herein, we summarize the state of the art for evaluating the response of glioblastoma to standard therapies and investigational agents as well as challenges and future directions for assessing treatment response in neuro-oncology.

RECENT FINDINGS

Monitoring glioblastoma responses to standard therapy and novel agents has been fraught with many challenges and limitations over the past decade. Excitingly, new promising methods are emerging to help address these challenges. Recently, the Response Assessment in Neuro-Oncology (RANO) working group proposed an updated response criteria (RANO 2.0) for the evaluation of all grades of glial tumors regardless of IDH status or therapies being evaluated. In addition, advanced neuroimaging techniques, such as histogram analysis, parametric response maps, morphometric segmentation, radio pharmacodynamics approaches, and the integrating of amino acid radiotracers in the tumor evaluation algorithm may help resolve equivocal lesion interpretations without operative intervention. Moreover, the introduction of other techniques, such as liquid biopsy and artificial intelligence could complement conventional visual assessment of glioblastoma response to therapies. Neuro-oncology has evolved over the past decade and has achieved significant milestones, including the establishment of new standards of care, emerging therapeutic options, and novel clinical, translational, and basic research. More recently, the integration of histopathology with molecular features for tumor classification has marked an important paradigm shift in brain tumor diagnosis. In a similar manner, treatment response monitoring in neuro-oncology has made considerable progress. While most techniques are still in their inception, there is an emerging body of evidence for clinical application. Further research will be critically important for the development of impactful breakthroughs in this area of the field.