Materiały źródłowe

• #1 Predicting survival of glioblastoma from automatic whole-brain and tumor segmentation of MR images

  https://pmc.ncbi.nlm.nih.gov/articles/PMC9671967/

  Survival prediction models can potentially be used to guide treatment of glioblastoma patients. […] Prognosis is generally very poor, with a median overall survival (OS) of less than 15 months, and a 5-year OS rate of only 10%, even when aggressively treated. […] Following standard therapy, OS and progression-free survival (PFS) have been shown to correlate with several patient-specific features such as age, performance status and expression of O6-methylguanine-DNA-methyltransferase (MGMT). […] The literature on imaging biomarkers for glioblastoma survival prediction is currently dominated by radiomics, an approach in which hundreds or even thousands of features are extracted from delineated tumor regions of MR images, each quantifying some shape, texture, wavelet or histogram property.

• #2 Predicting survival of glioblastoma from automatic whole-brain and tumor segmentation of MR images | Scientific Reports

  https://www.nature.com/articles/s41598-022-19223-3

  Glioblastoma (GBM) is the most common malignant primary brain tumor in adults. Prognosis is generally very poor, with a median overall survival (OS) of less than 15 months, and a 5-year OS rate of only 10%, even when aggressively treated. Following standard therapy, OS and progression-free survival (PFS) have been shown to correlate with several patient-specific features such as age, performance status and expression of O6-methylguanine-DNA-methyltransferase (MGMT). However, the prognostic value of these features is still too low to guide treatment choices in individual patients. […] The proposed Hd95 features contain some information about where the tumor is located in the brain and its size, both of which have been studied before and shown to carry prognostic value. […] The best model for OS was achieved by combining the proposed features with the previously known prognostic clinical features: further addition of CoM, TCV, and CEV did not provide significant improvement.

• #3 About Glioblastoma

  https://braintumor.org/events/glioblastoma-awareness-day/about-glioblastoma/

  Glioblastoma (GBM) is one of the most complex, deadly, and treatment-resistant cancers. […] The five-year survival rate for glioblastoma patients is only 6.9 percent, and the average length of survival for glioblastoma patients is estimated to be only 8 months. […] Survival rates and mortality statistics for GBM have been virtually unchanged for decades. […] None of these treatments have succeeded in significantly extending patient lives beyond a few extra months. […] Despite these daunting facts and figures, there is hope. Science is advancing rapidly and there are promising research strategies moving forward.

• #4

  https://www.the-scientist.com/machine-learning-for-predicting-glioblastoma-prognosis-71402

  Researchers integrate scRNA-seq, spatial transcriptomics, and histology imaging data to show that spatial cellular architecture predicts glioblastoma prognosis. […] Scientists combine scRNA-seq, spatial transcriptomics, and histology imaging data to develop a machine learning program for glioblastoma prognosis. […] Glioblastoma (GBM) is the most common and aggressive malignant tumor in the central nervous system, with a five-year survival rate as low as 6.8 percent. […] Not only do these tumor cells have different expression profiles, but their cellular subtypes and spatial organization can also vary among patients, according to studies using single-cell RNA sequencing (scRNA-seq). […] Retaining spatial information while assessing GBM prognosis is important because cellular interactions and tumor architecture play critical roles in driving clonal evolution, tumor progression, and therapeutic resistance.

• #5 Glioblastoma (GBM): What It Is, Symptoms & Prognosis

  https://my.clevelandclinic.org/health/diseases/17032-glioblastoma

  Glioblastoma, formerly known as glioblastoma multiforme, is a devastating type of cancer that can result in death in fewer than six months without treatment. […] Glioblastoma may result in early death shortly after a diagnosis without treatment. But treatments are available. They may help you ease symptoms and stay comfortable or prolong your life. […] Most people live an average of 12 to 18 months after diagnosis. The five-year survival rate for glioblastoma is only about 5%. That means about 5% of people with GBM are still alive five years after their diagnosis.

• #6 What is the average glioblastoma survival rate? — Glioblastoma Research Organization

  https://www.gbmresearch.org/blog/glioblastoma-survival-rate

  GBM’s median survival rate for adults is 14.6 months, which can be devastating for patients and their loved ones. […] According to the National Brain Tumor Society, the five-year glioblastoma survival rate for patients is only 6.8 percent. […] The average length of survival for glioblastoma patients is estimated to be only 12 to 18 months. […] The glioblastoma survival rate constantly changes as the treatment strategy used changes. […] Despite these numbers, there is hope. Many new research techniques are in development, some of which have proved successful in changing data on glioblastoma treatment outcomes.

• #7 Survival Outcome Prediction in Glioblastoma: Insights from MRI Radiomics

  https://www.mdpi.com/1718-7729/31/4/165

  Survival Outcome Prediction in Glioblastoma: Insights from MRI Radiomics […] Despite significant advances in surgery and chemoradiotherapy over the last few decades, the prognosis for patients with GBM remains poor. The median survival remains low at only 15 months, with approximately 40% survival in the first year post diagnosis and 17% in the second year. […] The results of the present study indicate that predictive indices of survival in patients with GBM include both perfusion and diffusion indices, highlighting the significant role of microstructural or hemodynamic characteristics in determining patient outcomes. […] In our study, shape and first and second order radiomic features contributed to the prediction model from T1W, diffusion, and perfusion maps. […] In newly diagnosed wild-type GBM, MRI radiomic features derived from various components of the lesion on diffusion and perfusion parametric maps can predict survival in a non-invasive manner. This finding emphasizes the potential clinical value of radiomics in prognostication and risk stratification of GBM patients.

• #8 About Glioblastoma

  https://braintumor.org/events/glioblastoma-awareness-day/about-glioblastoma/

  Glioblastoma (GBM) is one of the most complex, deadly, and treatment-resistant cancers. […] The five-year survival rate for glioblastoma patients is only 6.9 percent, and the average length of survival for glioblastoma patients is estimated to be only 8 months. […] Survival rates and mortality statistics for GBM have been virtually unchanged for decades. […] None of these treatments have succeeded in significantly extending patient lives beyond a few extra months. […] Despite these daunting facts and figures, there is hope. Science is advancing rapidly and there are promising research strategies moving forward.

• #9 Predicting survival of glioblastoma from automatic whole-brain and tumor segmentation of MR images

  https://pmc.ncbi.nlm.nih.gov/articles/PMC9671967/

  Survival prediction models can potentially be used to guide treatment of glioblastoma patients. […] Prognosis is generally very poor, with a median overall survival (OS) of less than 15 months, and a 5-year OS rate of only 10%, even when aggressively treated. […] Following standard therapy, OS and progression-free survival (PFS) have been shown to correlate with several patient-specific features such as age, performance status and expression of O6-methylguanine-DNA-methyltransferase (MGMT). […] The literature on imaging biomarkers for glioblastoma survival prediction is currently dominated by radiomics, an approach in which hundreds or even thousands of features are extracted from delineated tumor regions of MR images, each quantifying some shape, texture, wavelet or histogram property.

• #10 Predicting survival of glioblastoma from automatic whole-brain and tumor segmentation of MR images | Scientific Reports

  https://www.nature.com/articles/s41598-022-19223-3

  Glioblastoma (GBM) is the most common malignant primary brain tumor in adults. Prognosis is generally very poor, with a median overall survival (OS) of less than 15 months, and a 5-year OS rate of only 10%, even when aggressively treated. Following standard therapy, OS and progression-free survival (PFS) have been shown to correlate with several patient-specific features such as age, performance status and expression of O6-methylguanine-DNA-methyltransferase (MGMT). However, the prognostic value of these features is still too low to guide treatment choices in individual patients. […] The proposed Hd95 features contain some information about where the tumor is located in the brain and its size, both of which have been studied before and shown to carry prognostic value. […] The best model for OS was achieved by combining the proposed features with the previously known prognostic clinical features: further addition of CoM, TCV, and CEV did not provide significant improvement.

• #11 What is the Prognosis of Glioblastoma? – Glioblastoma Foundation

  https://glioblastomafoundation.org/news/what-is-the-prognosis-of-glioblastoma

  Glioblastoma, or Glioblastoma Multiforme (GBM), is an aggressive stage 4 brain cancer. The prognosis for glioblastoma is generally very poor and current treatments often fall short. The median survival for a glioblastoma prognosis is just one year after diagnosis. Only 5% of glioblastoma cancer patients survive 5 years after diagnosis. […] Glioblastoma prognosis is closely related to extent of surgical resection. Improved prognosis is also seen in younger patients, females, those with IDH mutations, those undergoing gross total surgical resection, those with MGMT methylation, and those treated with combination chemotherapy plus radiation. […] Younger age has been associated with a better prognosis for glioblastoma. Patients under age 55 have an improved prognosis compared to patients over the age of 55. Prognosis is also improved for women in comparison to men.

• #12

  https://www.eortc.be/tools/recgbmcalculator/calculator.aspx

  These models were developed for patients who were previously treated by chemoradiation with temozolomide according to the results of Stupp at al.: Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. […] A patient with a performance status greater than 0 and more than one target lesion is predicted to have a median PFS of 1.6 months (1.4-1.8) and poor chance to be alive and free of progression at 6 months (PFS6= 2.2 % (0.2-10.2)). […] Patients with good performance status (PS=0) and one target lesion have an estimated median PFS of 3.3 months (1.9-5.5) months and PFS6 of 32.7% (19.3-44.8). […] For overall survival, patients predicted with a good prognosis have performance status 0, no baseline steroids, one target lesion and largest lesion has a maximum diameter equal to or lower than 42 mm. Their median OS is 18.2 month (11.4-38.9) months and OS12is 60.6% (46.0-72.3). […] Patients with poor performance status (PS2), steroids administered, more than one target lesion, and a maximum diameter of the largest lesion greater than 42 mm are predicted with a median OS of 2.6 months (1.9-3.9) and OS12 equal to 0% (0-0.3).

• #13 Predicting overall survival in glioblastoma patients using machine learning: an analysis of treatment efficacy and patient prognosis

  https://pmc.ncbi.nlm.nih.gov/articles/PMC12014569/

  The use of radiomics, a field focused on extracting detailed imaging features, has significantly enhanced the prognostic capabilities of ML. By analyzing high-dimensional data from MRI scans, radiomics enables models to detect subtle patterns related to tumor behavior and patient outcomes. […] Beyond imaging, molecular profiling has emerged as a critical component in GBM prognosis. Genomic and transcriptomic data have proven essential for identifying key survival markers, such as MGMT promoter methylation and IDH mutation status. […] Multimodal frameworks that combine radiomic, molecular, and clinical data have demonstrated exceptional potential for survival prediction. […] Despite the advancements, several obstacles remain in applying ML to GBM prognosis. One of the most pressing issues is class imbalance, where long-term survival categories are underrepresented in datasets.

• #14 Functional prediction of response to therapy prior to therapeutic intervention is associated with improved survival in patients with high-grade glioma | Scientific Reports

  https://www.nature.com/articles/s41598-024-68801-0

  High grade gliomas (HGG), including astrocytoma, IDH-mutant and glioblastoma, IDH-wildtype, are a class of aggressive brain cancers with extremely poor prognosis, and minimally effective treatment options. […] The adoption of this protocol improved median survival in HGG by 2.5 months over radiation treatment alone. Still, almost all patients undergo recurrence, and the 5 year survival rate remains less than 10%, as it has for the past 30 years. […] Methylation of the MGMT promoter is associated with better prognosis and survival with RT/TMZ compared to unmethylated patients. […] In general, all patients would benefit from a more direct knowledge of their predicted response to standard of care (SOC) prior to treatment to make informed decisions, potentially open the path to clinical trial enrollment, and maximize their time to recurrence.

• #15 The Effect of Re-operation on Survival in Patients with Recurrent Glioblastoma | Anticancer Research

  https://ar.iiarjournals.org/content/35/3/1743

  Glioblastoma (GBM) is the most frequent malignant tumor of the central nervous system (CNS) with an incidence of 4.8/100,000 cases per year. Surgical treatment followed by temozolomide (TMZ) concomitant with and adjuvant to radiotherapy (RT) is the cornerstone of treatment for newly-diagnosed GBM that allows for improvement in progression-free survival (PFS) and overall survival (OS). Nevertheless, despite optimal treatments, patients experience disease progression and median survival does not exceed 12-14 months with a 5-year survival rate of 10%. […] Re-surgery is often used both for confirmation of recurrent disease and for debulking to provide symptoms relief; however, there is no evidence that re-surgery might increase survival. […] In multivariate analysis, no significant effect of re-surgery was found, with age (p=0.001), MGMT methylation (p=0.002) and PFS at 6 months (p=0.0001) being significant prognostic factors.

• #16 COVPRIG robustly predicts the overall survival of IDH wild-type glioblastoma and highlights METTL1+ neural-progenitor-like tumor cell in driving unfavorable outcome | Journal of Translational Medicine | Full Text

  https://translational-medicine.biomedcentral.com/articles/10.1186/s12967-023-04382-2

  Accurately predicting the outcome of isocitrate dehydrogenase (IDH) wild-type glioblastoma (GBM) remains hitherto challenging. This study aims to Construct and Validate a Robust Prognostic Model for IDH wild-type GBM (COVPRIG) for the prediction of overall survival using a novel metric, genegene (GG) interaction, and explore molecular and cellular underpinnings. COVPRIG was designed for RNA-seq and microarray, respectively, and effectively identified patients at high risk of mortality. The predictive performance of COVPRIG was satisfactory, with area under the curve (AUC) ranging from 0.56 (CGGA693, RNA-seq, 6-month survival) to 0.79 (TCGA RNAseq, 18-month survival), which can be further validated by decision curves. The prognostic significance of COVPRIG was also validated in GBM including the IDH mutant samples. Notably, COVPRIG was comprehensively evaluated and externally validated, and a systemic review disclosed that COVPRIG outperformed current validated models with an integrated discrimination improvement (IDI) of 616%. Moreover, integrative bioinformatics analysis predicted an essential role of METTL1+ neural-progenitor-like (NPC-like) malignant cell in driving unfavorable outcome. This study provided a powerful tool for the outcome prediction for IDH wild-type GBM, and preliminary molecular underpinnings for future research. The outcome of IDH wild-type GBM can be heterogeneous. IDH wild-type GBM is classified as proneural, classical, and mesenchymal subtypes based on the transcriptome, with the latter having the worst prognosis. Compelling evidence illuminates that stem-like tumor cell, which is at the interface of neural and glioma biology, is essential in tumor progression and treatment resistance. In this study, we developed a robust prognostic model for IDH wild-type GBM through incorporating a novel parameter, GG interaction, to effectively identify patients at high mortality risk. In addition, comprehensive bioinformatics analysis pinpointed a subset of NPC-like cells as an essential player in driving unfavorable outcome. The prognostic significance of COVPRIG was demonstrated, with samples split into high- and low-risk groups by the median value. KM analysis found that an increased COVPRIG score was suggestive of a significantly decreased overall survival. The hazard ratio (HR) showed a dose-response association with groups. For example, the HRgroup3vs.1 was 3.14 in the TCGA RNA-seq cohort, higher than HRgroup2vs.1 (1.63). In sum, nomograms were constructed based on RNA-seq and microarray datasets for individualized prognostic prediction. Collectively, this study provided a reference for accurately determining the prognosis of IDH wild-type GBM.

• #17 What is the Prognosis of Glioblastoma? – Glioblastoma Foundation

  https://glioblastomafoundation.org/news/what-is-the-prognosis-of-glioblastoma

  Glioblastoma, or Glioblastoma Multiforme (GBM), is an aggressive stage 4 brain cancer. The prognosis for glioblastoma is generally very poor and current treatments often fall short. The median survival for a glioblastoma prognosis is just one year after diagnosis. Only 5% of glioblastoma cancer patients survive 5 years after diagnosis. […] Glioblastoma prognosis is closely related to extent of surgical resection. Improved prognosis is also seen in younger patients, females, those with IDH mutations, those undergoing gross total surgical resection, those with MGMT methylation, and those treated with combination chemotherapy plus radiation. […] Younger age has been associated with a better prognosis for glioblastoma. Patients under age 55 have an improved prognosis compared to patients over the age of 55. Prognosis is also improved for women in comparison to men.

• #18 What is the Prognosis of Glioblastoma? – Glioblastoma Foundation

  https://glioblastomafoundation.org/news/what-is-the-prognosis-of-glioblastoma

  Timing of diagnosis and treatment also influence glioblastoma prognosis. Early diagnosis and early surgical intervention by a neurosurgeon have been associated with a better prognosis. Glioblastoma prognosis is improved when the time between surgery and diagnosis is minimal. […] The overall prognosis for glioblastoma has changed little since the 1980s, and new treatments are desperately needed to improve outcomes for patients. […] While glioblastoma prognosis remains poor, we are making progress against the disease.

• #19 Survival prediction of glioblastoma patients using machine learning and deep learning: a systematic review | BMC Cancer | Full Text

  https://bmccancer.biomedcentral.com/articles/10.1186/s12885-024-13320-4

  Survival prediction is usually done from clinical features using statistical methods. However, with the advancement of artificial intelligence methods that include machine learning and deep learning, researchers have been eager to use these methods for survival prediction, which can combine pathology, histology, molecular, imaging, and clinical features. […] The current treatment involves removing the tumor surgically if possible, then giving radiotherapy along with temozolomide (TMZ) regimen; however, the tumor often comes back and becomes resistant/adapted to the treatment. The bad prognosis is related to many factors such as tumor diversity, weak immune response, invasion of the tumor into „normal” tissue, production of cancer stem cells, and the fast adaptation of the tumor to harsh treatment.

• #20 Functional prediction of response to therapy prior to therapeutic intervention is associated with improved survival in patients with high-grade glioma | Scientific Reports

  https://www.nature.com/articles/s41598-024-68801-0

  High grade gliomas (HGG), including astrocytoma, IDH-mutant and glioblastoma, IDH-wildtype, are a class of aggressive brain cancers with extremely poor prognosis, and minimally effective treatment options. […] The adoption of this protocol improved median survival in HGG by 2.5 months over radiation treatment alone. Still, almost all patients undergo recurrence, and the 5 year survival rate remains less than 10%, as it has for the past 30 years. […] Methylation of the MGMT promoter is associated with better prognosis and survival with RT/TMZ compared to unmethylated patients. […] In general, all patients would benefit from a more direct knowledge of their predicted response to standard of care (SOC) prior to treatment to make informed decisions, potentially open the path to clinical trial enrollment, and maximize their time to recurrence.

• #21 Predicting Survival in Glioblastoma Patients Using Diffusion MR Imaging Metrics—A Systematic Review

  https://www.mdpi.com/2072-6694/12/10/2858

  In this scenario, imaging represents an attractive option compared to more invasive approaches based on tissue-derived biomarkers. […] There is an urgent need to discover imaging biomarkers that can aid in the selection of patients who will likely derive the most benefit from surgical and/or chemotherapy and/or radiotherapy treatment in terms of overall survival (OS) and progression-free survival (PFS), which are the most commonly evaluated clinical endpoints. […] The most examined metrics were associated with the standard Diffusion Weighted Imaging (DWI) (34 studies) and Diffusion Tensor Imaging (DTI) models (17 studies). […] Our findings showed that quantitative diffusion MRI metrics provide useful information for predicting survival outcomes in GBM patients, mainly in combination with other clinical and multimodality imaging parameters.

• #22 Predicting Survival in Glioblastoma Patients Using Diffusion MR Imaging Metrics—A Systematic Review

  https://www.mdpi.com/2072-6694/12/10/2858

  In this scenario, imaging represents an attractive option compared to more invasive approaches based on tissue-derived biomarkers. […] There is an urgent need to discover imaging biomarkers that can aid in the selection of patients who will likely derive the most benefit from surgical and/or chemotherapy and/or radiotherapy treatment in terms of overall survival (OS) and progression-free survival (PFS), which are the most commonly evaluated clinical endpoints. […] The most examined metrics were associated with the standard Diffusion Weighted Imaging (DWI) (34 studies) and Diffusion Tensor Imaging (DTI) models (17 studies). […] Our findings showed that quantitative diffusion MRI metrics provide useful information for predicting survival outcomes in GBM patients, mainly in combination with other clinical and multimodality imaging parameters.

• #23 Predicting survival of glioblastoma from automatic whole-brain and tumor segmentation of MR images

  https://pmc.ncbi.nlm.nih.gov/articles/PMC9671967/

  Survival prediction models can potentially be used to guide treatment of glioblastoma patients. […] Prognosis is generally very poor, with a median overall survival (OS) of less than 15 months, and a 5-year OS rate of only 10%, even when aggressively treated. […] Following standard therapy, OS and progression-free survival (PFS) have been shown to correlate with several patient-specific features such as age, performance status and expression of O6-methylguanine-DNA-methyltransferase (MGMT). […] The literature on imaging biomarkers for glioblastoma survival prediction is currently dominated by radiomics, an approach in which hundreds or even thousands of features are extracted from delineated tumor regions of MR images, each quantifying some shape, texture, wavelet or histogram property.

• #24 Predicting overall survival in glioblastoma patients using machine learning: an analysis of treatment efficacy and patient prognosis

  https://pmc.ncbi.nlm.nih.gov/articles/PMC12014569/

  The use of radiomics, a field focused on extracting detailed imaging features, has significantly enhanced the prognostic capabilities of ML. By analyzing high-dimensional data from MRI scans, radiomics enables models to detect subtle patterns related to tumor behavior and patient outcomes. […] Beyond imaging, molecular profiling has emerged as a critical component in GBM prognosis. Genomic and transcriptomic data have proven essential for identifying key survival markers, such as MGMT promoter methylation and IDH mutation status. […] Multimodal frameworks that combine radiomic, molecular, and clinical data have demonstrated exceptional potential for survival prediction. […] Despite the advancements, several obstacles remain in applying ML to GBM prognosis. One of the most pressing issues is class imbalance, where long-term survival categories are underrepresented in datasets.

• #25 Survival Outcome Prediction in Glioblastoma: Insights from MRI Radiomics

  https://www.mdpi.com/1718-7729/31/4/165

  Survival Outcome Prediction in Glioblastoma: Insights from MRI Radiomics […] Despite significant advances in surgery and chemoradiotherapy over the last few decades, the prognosis for patients with GBM remains poor. The median survival remains low at only 15 months, with approximately 40% survival in the first year post diagnosis and 17% in the second year. […] The results of the present study indicate that predictive indices of survival in patients with GBM include both perfusion and diffusion indices, highlighting the significant role of microstructural or hemodynamic characteristics in determining patient outcomes. […] In our study, shape and first and second order radiomic features contributed to the prediction model from T1W, diffusion, and perfusion maps. […] In newly diagnosed wild-type GBM, MRI radiomic features derived from various components of the lesion on diffusion and perfusion parametric maps can predict survival in a non-invasive manner. This finding emphasizes the potential clinical value of radiomics in prognostication and risk stratification of GBM patients.

• #26 Survival Outcome Prediction in Glioblastoma: Insights from MRI Radiomics

  https://www.mdpi.com/1718-7729/31/4/165

  Survival Outcome Prediction in Glioblastoma: Insights from MRI Radiomics […] Despite significant advances in surgery and chemoradiotherapy over the last few decades, the prognosis for patients with GBM remains poor. The median survival remains low at only 15 months, with approximately 40% survival in the first year post diagnosis and 17% in the second year. […] The results of the present study indicate that predictive indices of survival in patients with GBM include both perfusion and diffusion indices, highlighting the significant role of microstructural or hemodynamic characteristics in determining patient outcomes. […] In our study, shape and first and second order radiomic features contributed to the prediction model from T1W, diffusion, and perfusion maps. […] In newly diagnosed wild-type GBM, MRI radiomic features derived from various components of the lesion on diffusion and perfusion parametric maps can predict survival in a non-invasive manner. This finding emphasizes the potential clinical value of radiomics in prognostication and risk stratification of GBM patients.

• #27 Survival prediction of glioblastoma patients using machine learning and deep learning: a systematic review | BMC Cancer | Full Text

  https://bmccancer.biomedcentral.com/articles/10.1186/s12885-024-13320-4

  Survival prediction is usually done from clinical features using statistical methods. However, with the advancement of artificial intelligence methods that include machine learning and deep learning, researchers have been eager to use these methods for survival prediction, which can combine pathology, histology, molecular, imaging, and clinical features. […] The current treatment involves removing the tumor surgically if possible, then giving radiotherapy along with temozolomide (TMZ) regimen; however, the tumor often comes back and becomes resistant/adapted to the treatment. The bad prognosis is related to many factors such as tumor diversity, weak immune response, invasion of the tumor into „normal” tissue, production of cancer stem cells, and the fast adaptation of the tumor to harsh treatment.

• #28 Predicting overall survival in glioblastoma patients using machine learning: an analysis of treatment efficacy and patient prognosis

  https://pmc.ncbi.nlm.nih.gov/articles/PMC12014569/

  Glioblastoma (GBM), the most aggressive primary brain tumor, poses a significant challenge in predicting patient survival due to its heterogeneity and resistance to treatment. Accurate survival prediction is essential for optimizing treatment strategies and improving clinical outcomes. […] The application of machine learning to GBM metadata offers a robust approach to predicting patient survival. The study highlights the potential of ML models to enhance clinical decision-making and contribute to personalized treatment strategies, with a focus on accuracy, reliability, and interpretability. […] GBM is a highly aggressive brain tumor with poor outcomes, presenting unique challenges for accurate prognosis and treatment planning. Traditional approaches often struggle to account for the complex biological and clinical variability inherent to GBM, necessitating advanced methodologies to improve survival predictions.

• #29 Predicting overall survival in glioblastoma patients using machine learning: an analysis of treatment efficacy and patient prognosis

  https://pmc.ncbi.nlm.nih.gov/articles/PMC12014569/

  Glioblastoma (GBM), the most aggressive primary brain tumor, poses a significant challenge in predicting patient survival due to its heterogeneity and resistance to treatment. Accurate survival prediction is essential for optimizing treatment strategies and improving clinical outcomes. […] The application of machine learning to GBM metadata offers a robust approach to predicting patient survival. The study highlights the potential of ML models to enhance clinical decision-making and contribute to personalized treatment strategies, with a focus on accuracy, reliability, and interpretability. […] GBM is a highly aggressive brain tumor with poor outcomes, presenting unique challenges for accurate prognosis and treatment planning. Traditional approaches often struggle to account for the complex biological and clinical variability inherent to GBM, necessitating advanced methodologies to improve survival predictions.

• #30 Predicting overall survival in glioblastoma patients using machine learning: an analysis of treatment efficacy and patient prognosis

  https://pmc.ncbi.nlm.nih.gov/articles/PMC12014569/

  This study aims to build on these advancements by integrating molecular, and clinical data into an advanced ML framework to improve GBM survival predictions. […] The best results were achieved using an XGBoost algorithm, which attained an average ROC-AUC of 0.90 with a standard deviation of 0.07 and an accuracy of 0.78 on the test data. […] These findings highlight the efficacy of XGB and ET classifiers in handling complex predictive tasks, with XGB slightly outperforming others in terms of stability and overall performance. […] The findings of this study highlight the transformative potential of ML in GBM prognosis. Accurate survival predictions have profound implications for patient care, from guiding individualized treatment strategies to identifying candidates for experimental therapies and optimizing resource allocation.

• #31 Predicting overall survival in glioblastoma patients using machine learning: an analysis of treatment efficacy and patient prognosis

  https://pmc.ncbi.nlm.nih.gov/articles/PMC12014569/

  This study aims to build on these advancements by integrating molecular, and clinical data into an advanced ML framework to improve GBM survival predictions. […] The best results were achieved using an XGBoost algorithm, which attained an average ROC-AUC of 0.90 with a standard deviation of 0.07 and an accuracy of 0.78 on the test data. […] These findings highlight the efficacy of XGB and ET classifiers in handling complex predictive tasks, with XGB slightly outperforming others in terms of stability and overall performance. […] The findings of this study highlight the transformative potential of ML in GBM prognosis. Accurate survival predictions have profound implications for patient care, from guiding individualized treatment strategies to identifying candidates for experimental therapies and optimizing resource allocation.

• #32 Survival prediction of glioblastoma patients using machine learning and deep learning: a systematic review | BMC Cancer | Full Text

  https://bmccancer.biomedcentral.com/articles/10.1186/s12885-024-13320-4

  Recent studies aim to provide better survival estimates based on multiple types of data such as clinical, molecular, and imaging data. A combination of multiple types of data may give a more detailed description of the underlying characteristics that affect survival and their relationships than single methods. […] The results of this systematic review show that machine learning and deep learning methods have been widely applied to survival prediction of glioblastoma patients, using different types of input data, such as clinical features, molecular markers, imaging features, radiomics features, omics data, or a combination of them. […] Predicting survival with high accuracy can assist personalized clinical decision-making in glioblastoma patients.

• #33 Survival prediction of glioblastoma patients using machine learning and deep learning: a systematic review | BMC Cancer | Full Text

  https://bmccancer.biomedcentral.com/articles/10.1186/s12885-024-13320-4

  Recent studies aim to provide better survival estimates based on multiple types of data such as clinical, molecular, and imaging data. A combination of multiple types of data may give a more detailed description of the underlying characteristics that affect survival and their relationships than single methods. […] The results of this systematic review show that machine learning and deep learning methods have been widely applied to survival prediction of glioblastoma patients, using different types of input data, such as clinical features, molecular markers, imaging features, radiomics features, omics data, or a combination of them. […] Predicting survival with high accuracy can assist personalized clinical decision-making in glioblastoma patients.

• #34 Predicting survival of glioblastoma from automatic whole-brain and tumor segmentation of MR images | Scientific Reports

  https://www.nature.com/articles/s41598-022-19223-3

  The Hd95 features thus seem to bring prognostic value that is not contained in simple size and location based features. […] The proposed Hd95 features measure how much each brain structure is deformed compared to its expected shape in the absence of pathology, and therefore they contain information about the location of the tumor, which has been shown previously to be a prognostic factor for OS. Nevertheless, our results show that the proposed Hd95 features carry richer prognostic information for predicting OS than tumor location alone. […] The application of survival models exploiting advanced imaging features, such as the ones proposed here, could potentially help minimize bias in stratification in future clinical trials. High quality prognostic information could also potentially guide clinicians in adjusting the intensity of interventions, based on expected outcome and quality-of-life considerations.

• #35 Survival Outcome Prediction in Glioblastoma: Insights from MRI Radiomics

  https://www.mdpi.com/1718-7729/31/4/165

  Survival Outcome Prediction in Glioblastoma: Insights from MRI Radiomics […] Despite significant advances in surgery and chemoradiotherapy over the last few decades, the prognosis for patients with GBM remains poor. The median survival remains low at only 15 months, with approximately 40% survival in the first year post diagnosis and 17% in the second year. […] The results of the present study indicate that predictive indices of survival in patients with GBM include both perfusion and diffusion indices, highlighting the significant role of microstructural or hemodynamic characteristics in determining patient outcomes. […] In our study, shape and first and second order radiomic features contributed to the prediction model from T1W, diffusion, and perfusion maps. […] In newly diagnosed wild-type GBM, MRI radiomic features derived from various components of the lesion on diffusion and perfusion parametric maps can predict survival in a non-invasive manner. This finding emphasizes the potential clinical value of radiomics in prognostication and risk stratification of GBM patients.

• #36 Functional prediction of response to therapy prior to therapeutic intervention is associated with improved survival in patients with high-grade glioma | Scientific Reports

  https://www.nature.com/articles/s41598-024-68801-0

  Patients with high-grade glioma (HGG) have an extremely poor prognosis compounded by a lack of advancement in clinical care over the past few decades. […] Median survival between test-predicted temozolomide responders and test-predicted temozolomide non-responders revealed a statistically significant increase in progression-free survival when using the test to predict response across multiple subgroups including HGG (5.8 months), glioblastoma (4.7 months), and MGMT unmethylated glioblastoma (4.7 months). Overall survival was also positively separated across the subgroups at 7.6, 5.1, and 6.3 months respectively. […] The strong correlation of 3D Predict Glioma test results with clinical outcomes demonstrates that this functional test is prognostic in patients treated with RT/TMZ and supports aligning clinical treatment to test-predicted response across varying HGG subgroups.

• #37 Functional prediction of response to therapy prior to therapeutic intervention is associated with improved survival in patients with high-grade glioma | Scientific Reports

  https://www.nature.com/articles/s41598-024-68801-0

  Patients with high-grade glioma (HGG) have an extremely poor prognosis compounded by a lack of advancement in clinical care over the past few decades. […] Median survival between test-predicted temozolomide responders and test-predicted temozolomide non-responders revealed a statistically significant increase in progression-free survival when using the test to predict response across multiple subgroups including HGG (5.8 months), glioblastoma (4.7 months), and MGMT unmethylated glioblastoma (4.7 months). Overall survival was also positively separated across the subgroups at 7.6, 5.1, and 6.3 months respectively. […] The strong correlation of 3D Predict Glioma test results with clinical outcomes demonstrates that this functional test is prognostic in patients treated with RT/TMZ and supports aligning clinical treatment to test-predicted response across varying HGG subgroups.

• #38 Predicting survival of glioblastoma from automatic whole-brain and tumor segmentation of MR images | Scientific Reports

  https://www.nature.com/articles/s41598-022-19223-3

  Glioblastoma (GBM) is the most common malignant primary brain tumor in adults. Prognosis is generally very poor, with a median overall survival (OS) of less than 15 months, and a 5-year OS rate of only 10%, even when aggressively treated. Following standard therapy, OS and progression-free survival (PFS) have been shown to correlate with several patient-specific features such as age, performance status and expression of O6-methylguanine-DNA-methyltransferase (MGMT). However, the prognostic value of these features is still too low to guide treatment choices in individual patients. […] The proposed Hd95 features contain some information about where the tumor is located in the brain and its size, both of which have been studied before and shown to carry prognostic value. […] The best model for OS was achieved by combining the proposed features with the previously known prognostic clinical features: further addition of CoM, TCV, and CEV did not provide significant improvement.

• #39 About Glioblastoma

  https://braintumor.org/events/glioblastoma-awareness-day/about-glioblastoma/

  Glioblastoma (GBM) is one of the most complex, deadly, and treatment-resistant cancers. […] The five-year survival rate for glioblastoma patients is only 6.9 percent, and the average length of survival for glioblastoma patients is estimated to be only 8 months. […] Survival rates and mortality statistics for GBM have been virtually unchanged for decades. […] None of these treatments have succeeded in significantly extending patient lives beyond a few extra months. […] Despite these daunting facts and figures, there is hope. Science is advancing rapidly and there are promising research strategies moving forward.

• #40 Predicting overall survival in glioblastoma patients using machine learning: an analysis of treatment efficacy and patient prognosis

  https://pmc.ncbi.nlm.nih.gov/articles/PMC12014569/

  The use of radiomics, a field focused on extracting detailed imaging features, has significantly enhanced the prognostic capabilities of ML. By analyzing high-dimensional data from MRI scans, radiomics enables models to detect subtle patterns related to tumor behavior and patient outcomes. […] Beyond imaging, molecular profiling has emerged as a critical component in GBM prognosis. Genomic and transcriptomic data have proven essential for identifying key survival markers, such as MGMT promoter methylation and IDH mutation status. […] Multimodal frameworks that combine radiomic, molecular, and clinical data have demonstrated exceptional potential for survival prediction. […] Despite the advancements, several obstacles remain in applying ML to GBM prognosis. One of the most pressing issues is class imbalance, where long-term survival categories are underrepresented in datasets.

• #41 Predicting Survival in Glioblastoma Patients Using Diffusion MR Imaging Metrics—A Systematic Review

  https://www.mdpi.com/2072-6694/12/10/2858

  However, their role in prediction and evaluation of survival outcomes has not been fully addressed and results are often controversial or unsatisfactory. […] Therefore, the aim of this systematic review is to collect, summarize and discuss all studies evaluating the role of diffusion MRI metrics in predicting survival in GBM patients and raise awareness for future research in this field.

• #42 Predicting Survival in Glioblastoma Patients Using Diffusion MR Imaging Metrics—A Systematic Review

  https://www.mdpi.com/2072-6694/12/10/2858

  However, their role in prediction and evaluation of survival outcomes has not been fully addressed and results are often controversial or unsatisfactory. […] Therefore, the aim of this systematic review is to collect, summarize and discuss all studies evaluating the role of diffusion MRI metrics in predicting survival in GBM patients and raise awareness for future research in this field.

• #43 Survival prediction of glioblastoma patients using machine learning and deep learning: a systematic review | BMC Cancer | Full Text

  https://bmccancer.biomedcentral.com/articles/10.1186/s12885-024-13320-4

  Recent studies aim to provide better survival estimates based on multiple types of data such as clinical, molecular, and imaging data. A combination of multiple types of data may give a more detailed description of the underlying characteristics that affect survival and their relationships than single methods. […] The results of this systematic review show that machine learning and deep learning methods have been widely applied to survival prediction of glioblastoma patients, using different types of input data, such as clinical features, molecular markers, imaging features, radiomics features, omics data, or a combination of them. […] Predicting survival with high accuracy can assist personalized clinical decision-making in glioblastoma patients.

• #44

  https://www.the-scientist.com/machine-learning-for-predicting-glioblastoma-prognosis-71402

  Researchers integrate scRNA-seq, spatial transcriptomics, and histology imaging data to show that spatial cellular architecture predicts glioblastoma prognosis. […] Scientists combine scRNA-seq, spatial transcriptomics, and histology imaging data to develop a machine learning program for glioblastoma prognosis. […] Glioblastoma (GBM) is the most common and aggressive malignant tumor in the central nervous system, with a five-year survival rate as low as 6.8 percent. […] Not only do these tumor cells have different expression profiles, but their cellular subtypes and spatial organization can also vary among patients, according to studies using single-cell RNA sequencing (scRNA-seq). […] Retaining spatial information while assessing GBM prognosis is important because cellular interactions and tumor architecture play critical roles in driving clonal evolution, tumor progression, and therapeutic resistance.

• #45

  https://www.the-scientist.com/machine-learning-for-predicting-glioblastoma-prognosis-71402

  Researchers integrate scRNA-seq, spatial transcriptomics, and histology imaging data to show that spatial cellular architecture predicts glioblastoma prognosis. […] Scientists combine scRNA-seq, spatial transcriptomics, and histology imaging data to develop a machine learning program for glioblastoma prognosis. […] Glioblastoma (GBM) is the most common and aggressive malignant tumor in the central nervous system, with a five-year survival rate as low as 6.8 percent. […] Not only do these tumor cells have different expression profiles, but their cellular subtypes and spatial organization can also vary among patients, according to studies using single-cell RNA sequencing (scRNA-seq). […] Retaining spatial information while assessing GBM prognosis is important because cellular interactions and tumor architecture play critical roles in driving clonal evolution, tumor progression, and therapeutic resistance.

• #46

  https://link.springer.com/article/10.1007/s13246-024-01443-8

  To propose a style transfer model for multi-contrast magnetic resonance imaging (MRI) images with a cycle-consistent generative adversarial network (CycleGAN) and evaluate the image quality and prognosis prediction performance for glioblastoma (GBM) patients from the extracted radiomics features. […] The survival time had a significant difference between good and poor prognosis groups for both real and synthesized T2w (p0.05). However, the survival time had no significant difference between good and poor prognosis groups for both real and synthesized T1w. […] It was found that the synthesized image could be used for prognosis prediction. […] The proposed prognostic model using CycleGAN could reduce the cost and time of image scanning, leading to a promotion to build the patients outcome prediction with multi-contrast images.

• #47

  https://link.springer.com/article/10.1007/s11060-024-04715-1

  However, surgical resection of the tumor carries the risk of causing or exacerbating functional impairment, a significant concern for the clinician, as postsurgical functional preservation (e.g., quality of life) is known to correlate with overall patient outcomes. […] To that end, the capability to forecast postsurgical functional outcomes from the initial diagnosis could prove advantageous in surgical planning and for better-informing patients of their likely treatment outcomes. […] Our results indicate that these models can accurately predict postoperative functional outcomes at the time of initial diagnosis. […] The strongest predictors identified by the model included FC between somatomotor, visual, auditory, and reward networks. […] Age was also a strong predictor. However, tumor volume was only a moderate predictor.

• #48

  https://link.springer.com/article/10.1007/s11060-024-04715-1

  However, surgical resection of the tumor carries the risk of causing or exacerbating functional impairment, a significant concern for the clinician, as postsurgical functional preservation (e.g., quality of life) is known to correlate with overall patient outcomes. […] To that end, the capability to forecast postsurgical functional outcomes from the initial diagnosis could prove advantageous in surgical planning and for better-informing patients of their likely treatment outcomes. […] Our results indicate that these models can accurately predict postoperative functional outcomes at the time of initial diagnosis. […] The strongest predictors identified by the model included FC between somatomotor, visual, auditory, and reward networks. […] Age was also a strong predictor. However, tumor volume was only a moderate predictor.

• #49 Predicting overall survival in glioblastoma patients using machine learning: an analysis of treatment efficacy and patient prognosis

  https://pmc.ncbi.nlm.nih.gov/articles/PMC12014569/

  This study aims to build on these advancements by integrating molecular, and clinical data into an advanced ML framework to improve GBM survival predictions. […] The best results were achieved using an XGBoost algorithm, which attained an average ROC-AUC of 0.90 with a standard deviation of 0.07 and an accuracy of 0.78 on the test data. […] These findings highlight the efficacy of XGB and ET classifiers in handling complex predictive tasks, with XGB slightly outperforming others in terms of stability and overall performance. […] The findings of this study highlight the transformative potential of ML in GBM prognosis. Accurate survival predictions have profound implications for patient care, from guiding individualized treatment strategies to identifying candidates for experimental therapies and optimizing resource allocation.

• #50 Predicting the survival of patients with glioblastoma using deep learning: a systematic review | Egyptian Journal of Neurosurgery | Full Text

  https://ejns.springeropen.com/articles/10.1186/s41984-025-00385-x

  This systematic review suggests that utilizing DL to analyze multiparametric imaging data for glioblastoma patients can greatly increase the ability to stratify patients according to survival risk. […] Before therapeutic integration can happen, however, external multicenter repeatability studies must show generalizability, and expected benefits must be measured across various clinical practice situations. […] The findings of this study highlight the need for additional research into the development of predictive models that incorporate larger patient cohorts, more robust imaging modalities, and supplementary multi-omics data. […] Calibration stability across variable acquisition methodologies must be addressed to use models for risk categorization effectively through methods like dataset harmonization.

• #51 Predicting the survival of patients with glioblastoma using deep learning: a systematic review | Egyptian Journal of Neurosurgery | Full Text

  https://ejns.springeropen.com/articles/10.1186/s41984-025-00385-x

  This systematic review suggests that utilizing DL to analyze multiparametric imaging data for glioblastoma patients can greatly increase the ability to stratify patients according to survival risk. […] Before therapeutic integration can happen, however, external multicenter repeatability studies must show generalizability, and expected benefits must be measured across various clinical practice situations. […] The findings of this study highlight the need for additional research into the development of predictive models that incorporate larger patient cohorts, more robust imaging modalities, and supplementary multi-omics data. […] Calibration stability across variable acquisition methodologies must be addressed to use models for risk categorization effectively through methods like dataset harmonization.

• #52 Predicting the survival of patients with glioblastoma using deep learning: a systematic review | Egyptian Journal of Neurosurgery | Full Text

  https://ejns.springeropen.com/articles/10.1186/s41984-025-00385-x

  This systematic review suggests that utilizing DL to analyze multiparametric imaging data for glioblastoma patients can greatly increase the ability to stratify patients according to survival risk. […] Before therapeutic integration can happen, however, external multicenter repeatability studies must show generalizability, and expected benefits must be measured across various clinical practice situations. […] The findings of this study highlight the need for additional research into the development of predictive models that incorporate larger patient cohorts, more robust imaging modalities, and supplementary multi-omics data. […] Calibration stability across variable acquisition methodologies must be addressed to use models for risk categorization effectively through methods like dataset harmonization.

• #53

  https://link.springer.com/article/10.1007/s11060-024-04715-1

  These findings indicate the feasibility of achieving classification accuracy exceeding 90% before surgical intervention, using basic demographics, tumor volume, and RS-fMRI measures. […] Our results indicate tumor volume was only a moderate predictor of KPS, which might initially appear counterintuitive. However, this finding underscores the critical point that tumor location in relation to functional networks could have a more significant influence on functional outcomes than the size of the tumor alone. […] Our findings show a distinct dichotomy between the anatomical location of the tumor and functional connectivity (FC) changes, specifically concerning their predictive ability for KPS. […] This apparent dissociation may be due to several potential mechanisms. […] Taken together, multiple parallel processes could potentially be responsible for FC disruption and associated functional impairment: damage to gray matter regions with specific functional network affiliations, the disruption of direct/indirect connections mediated by white matter damage, and overall health. […] By incorporating these models into clinical practice, we stand to enhance patient care, enabling personalized treatment plans that balance quality of life with survival.

• #54 What is the average glioblastoma survival rate? — Glioblastoma Research Organization

  https://www.gbmresearch.org/blog/glioblastoma-survival-rate

  GBM’s median survival rate for adults is 14.6 months, which can be devastating for patients and their loved ones. […] According to the National Brain Tumor Society, the five-year glioblastoma survival rate for patients is only 6.8 percent. […] The average length of survival for glioblastoma patients is estimated to be only 12 to 18 months. […] The glioblastoma survival rate constantly changes as the treatment strategy used changes. […] Despite these numbers, there is hope. Many new research techniques are in development, some of which have proved successful in changing data on glioblastoma treatment outcomes.

Zobacz więcej