An independently validated nomogram for isocitrate dehydrogenase-wild-type glioblastoma patient survival

Prognostic nomogram models for predicting survival probability in elderly glioblastoma patients

PURPOSE

To investigate the prognostic factors of survival and develop a predictive nomogram model for elderly GBM patients.

METHODS

Elderly patients (> = 65 years) with histologically diagnosed GBM were extracted from the SEER database. Survival analysis of overall survival (OS) was performed by Kaplan-Meier analysis. Univariate and multivariate Cox regression analyses were used to determine independent prognostic factors and these factors were used to further construct the nomogram model.

RESULTS

A total of 9068 elderly GBM patients (5122 males and 3946 females) were included, with a median age of 72 years (65-96 years). All patients were divided randomly into the training group (n = 6044) and the validation group (n = 3024) by a ratio of 2:1. Cox regression analyses on OS showed eight independent prognostic factors (race, age, tumor side, tumor size, metastasis, surgery, radiotherapy, and chemotherapy) in the training cohort. Also, seven variables (except for race) were identified on CSS in the training group. By comprising these variables, the nomogram models on OS and CSS for predicting the 6-month, 1-year, and 2-year survival probability were constructed and exhibited moderate consistency, respectively. Then, they could be validated well in the validation cohort and by C-index, time-dependent ROC curve, calibration plot, and DCA curve.

CONCLUSIONS

Nomogram models on OS and CSS could provide an applicable tool to predict the survival probability and provide clinical references regarding treatment strategies and prognosis.

Prognostic analysis and nomogram construction for older patients with IDH-wild-type glioblastoma

As many countries face an ageing population, the number of older patients with glioblastoma (GB) is increasing. Thus, there is an urgent need for prognostic models to aid in treatment decision-making and life planning. A total of 98 patients with isocitrate dehydrogenase (IDH)-wild-type GB aged ≥65 years were analysed from January 2012 to January 2020. Independent prognostic factors were identified by prognostic analysis. Using the independent prognostic factors for overall survival (OS), a nomogram was constructed by R software to predict the prognosis of older patients with IDH-wild-type GB. The concordance index (C-index) and receiver operating characteristic (ROC) curve were used to assess model discrimination, and the calibration curve was used to assess model calibration. Prognostic analysis showed that the extent of resection (EOR), adjusted Charlson comorbidity index (ACCI), O6-methylguanine-DNA methyltransferase (MGMT) methylation status, postoperative radiotherapy, and postoperative temozolomide (TMZ) chemotherapy were independent prognostic factors for OS. MGMT methylation status and subventricular zone (SVZ) involvement were independent prognostic factors for progression-free survival (PFS). A nomogram was constructed based on EOR, ACCI, MGMT methylation status, postoperative radiotherapy and postoperative TMZ chemotherapy to predict the 6-month, 12-month and 18-month OS of older patients with IDH-wild-type GB. The C-index of the nomogram was 0.72, and the ROC curves showed that the areas under the curve (AUCs) at 6, 12 and 18 months were 0.874, 0.739 and 0.779, respectively. The calibration plots showed that the nomogram was in good agreement with the actual observations in predicting the OS of older patients with IDH-wild-type GB. Older patients with IDH-wild-type GB can benefit from gross total resection (GTR), postoperative radiotherapy and postoperative TMZ chemotherapy. A high ACCI score and MGMT nonmethylation are poor prognostic factors. We constructed a nomogram including the ACCI to facilitate clinical decision-making and follow-up interval selection.

A validated prognostic nomogram for patients with H3 K27M-mutant diffuse midline glioma

H3 K27M-mutant diffuse midline glioma (H3 K27M-mt DMG) is a rare, highly invasive tumor with a poor prognosis. The prognostic factors of H3 K27M-mt DMG have not been fully identified, and there is no clinical prediction model for it. This study aimed to develop and validate a prognostic model for predicting the probability of survival in patients with H3 K27M-mt DMG. Patients diagnosed with H3 K27M-mt DMG in the West China Hospital from January 2016 to August 2021 were included. Cox proportional hazard regression was used for survival assessment, with adjustment for known prognostic factors. The final model was established using the patient data of our center as the training cohort and data from other centers for external independent verification. One hundred and five patients were ultimately included in the training cohort, and 43 cases from another institution were used as the validation cohort. The factors influencing survival probability in the prediction model included age, preoperative KPS score, radiotherapy and Ki-67 expression level. The adjusted consistency indices of the Cox regression model in internal bootstrap validation at 6, 12, and 18 months were 0.776, 0.766, and 0.764, respectively. The calibration chart showed high consistency between the predicted and observed results. The discrimination in external verification was 0.785, and the calibration curve showed good calibration ability. We identified the risk factors that affect the prognosis of H3 K27M-mt DMG patients and then established and validated a diagnostic model for predicting the survival probability of these patients.

Comparison of Near-Infrared Imaging Agents Targeting the PTPmu Tumor Biomarker

PURPOSE

Maximal, safe resection of solid tumors is considered a critical first step in successful cancer treatment. The advent of fluorescence image-guided surgery (FIGS) using non-specific agents has improved patient outcomes, particularly in the case of glioblastoma. Molecularly targeted agents that recognize specific tumor biomarkers have the potential to augment these gains. Identification of the optimal combination of targeting moiety and fluorophore is needed prior to initiating clinical trials.

PROCEDURES

A 20-amino acid peptide (SBK2) recognizing the receptor protein-tyrosine phosphatase mu (PTPmu)-derived tumor-specific biomarker, with or without a linker, was conjugated to three different near-infrared fluorophores: indocyanine green (ICG), IRDye® 800CW, and Tide Fluor™ 8WS. The in vivo specificity, time course, and biodistribution were evaluated for each using mice with heterotopic human glioma tumors that express the PTPmu biomarker to identify component combinations with optimal properties for FIGS.

RESULTS

SBK2 conjugated to ICG demonstrated excellent specificity for gliomas in heterotopic tumors. SBK2-ICG showed significantly higher in vivo tumor labeling compared to the Scram-ICG control from 10 min to 24 h, p < 0.01 at all timepoints, following injection, as well as a significantly higher ex vivo tumor signal at 24 h, p < 0.001. Inserting a six-amino acid linker between the targeting peptide and ICG increased the clearance rate and resulted in significantly higher in vivo tumor signal relative to its linker-containing Scrambled control from 10 min to 8 h, p < 0.05 at all timepoints, after dosing. Agents made with the more hydrophilic IRDye® 800CW and Tide Fluor™ 8WS showed no specific tumor labeling relative to the controls. The IRDye 800CW-conjugated agents cleared within 1 h, while the non-specific fluorescent tumor signal generated by the Tide Fluor 8WS-conjugated agents persists beyond 24 h.

CONCLUSIONS

The SBK2 PTPmu-targeting peptide conjugated to ICG specifically labels heterotopic human gliomas grown in mice between 10 min and 24 h following injection. Similar molecules constructed with more hydrophilic dyes demonstrated no specificity. These studies present a promising candidate for use in FIGS of PTPmu biomarker-expressing tumors.

Perfusion MR prior to radiotherapy is a strong predictor of survival in high-grade gliomas after proton and carbon ion radiotherapy

Background

To assess the survival predictability of perfusion magnetic resonance imaging (MRI) by the normalized cerebral blood volume (nCBV) prior to particle beam radiotherapy (PBRT) in high-grade glioma (HGG) patients underwent particle therapy.

Methods

The study retrieved dynamic susceptibility contrast MRI acquired prior to PBRT between 6/2015 and 3/2019 in 45 patients with HGG. Maximum nCBV (nCBVmax) within or adjacent to surgical/tumor bed was measured using 'hot-spot' method. The predictive values of nCBVmax for progression-free survival (PFS) and overall survival (OS) were assessed in univariate Kaplan-Meier curve and multivariate Cox proportional hazards (CPH) models. Nomograms based on CPH results were constructed to individualize the predicted probability of OS and PFS.

Results

The Kaplan-Meier curves and all CPH models based on nCBVmax as continuous variable (nCBVmax-C), group by cut-off derived from median value and Youden-index method showed that nCBVmax prior to radiotherapy was a strong predictor for both PFS and OS in HGG patients who underwent PBRT. Nomograms built on CPH models showed similar excellent performance in both discrimination and calibration.

Conclusions

Perfusion imaging prior to PBRT is a strong predictor of survival in HGG. Novel perfusion MR-based nomogram with prospective validation could potentially be formally used in future clinical practice to individualize survival probability.

Glioma survival prediction from whole-brain MRI without tumor segmentation using deep attention network: a multicenter study

OBJECTIVES

To develop and validate a deep learning model for predicting overall survival from whole-brain MRI without tumor segmentation in patients with diffuse gliomas.

METHODS

In this multicenter retrospective study, two deep learning models were built for survival prediction from MRI, including a DeepRisk model built from whole-brain MRI, and an original ResNet model built from expert-segmented tumor images. Both models were developed using a training dataset (n = 935) and an internal tuning dataset (n = 156) and tested on two external test datasets (n = 194 and 150) and a TCIA dataset (n = 121). C-index, integrated Brier score (IBS), prediction error curves, and calibration curves were used to assess the model performance.

RESULTS

In total, 1556 patients were enrolled (age, 49.0 ± 13.1 years; 830 male). The DeepRisk score was an independent predictor and can stratify patients in each test dataset into three risk subgroups. The IBS and C-index for DeepRisk were 0.14 and 0.83 in external test dataset 1, 0.15 and 0.80 in external dataset 2, and 0.16 and 0.77 in TCIA dataset, respectively, which were comparable with those for original ResNet. The AUCs at 6, 12, 24, 26, and 48 months for DeepRisk ranged between 0.77 and 0.94. Combining DeepRisk score with clinicomolecular factors resulted in a nomogram with a better calibration and classification accuracy (net reclassification improvement 0.69, p < 0.001) than the clinical nomogram.

CONCLUSIONS

DeepRisk that obviated the need of tumor segmentation can predict glioma survival from whole-brain MRI and offers incremental prognostic value.

KEY POINTS

• DeepRisk can predict overall survival directly from whole-brain MRI without tumor segmentation. • DeepRisk achieves comparable accuracy in survival prediction with deep learning model built using expert-segmented tumor images. • DeepRisk has independent and incremental prognostic value over existing clinical parameters and IDH mutation status.

Optimizing management of the elderly patient with glioblastoma: Survival prediction online tool based on BC Cancer Registry real-world data

BACKGROUND

Glioblastoma (GBM) is associated with fatal outcomes and devastating neurological presentations especially impacting the elderly. Management remains controversial and representation in clinical trials poor. We generated 2 nomograms and a clinical decision making web tool using real-world data.

METHODS

Patients ≥60 years of age with histologically confirmed GBM (ICD-O-3 histology codes 9440/3, 9441/3, and 9442/3) diagnosed 2005-2015 were identified from the BC Cancer Registry (n = 822). Seven hundred and twenty-nine patients for which performance status was captured were included in the analysis. Age, performance and resection status, administration of radiation therapy (RT), and chemotherapy were reviewed. Nomograms predicting 6- and 12-month overall survival (OS) probability were developed using Cox proportional hazards regression internally validated by c-index. A web tool powered by JavaScript was developed to calculate the survival probability.

RESULTS

Median OS was 6.6 months (95% confidence interval [CI] 6-7.2 months). Management involved concurrent chemoradiation (34%), RT alone (42%), and chemo alone (2.3%). Twenty-one percent of patients did not receive treatment beyond surgical intervention. Age, performance status, extent of resection, chemotherapy, and RT administration were all significant independent predictors of OS. Patients <80 years old who received RT had a significant survival advantage, regardless of extent of resection (hazard ratio range from 0.22 to 0.60, CI 0.15-0.95). A nomogram was constructed from all 729 patients (Harrell's Concordance Index = 0.78 [CI 0.71-0.84]) with a second nomogram based on subgroup analysis of the 452 patients who underwent RT (Harrell's Concordance Index = 0.81 [CI 0.70-0.90]). An online calculator based on both nomograms was generated for clinical use.

CONCLUSIONS

Two nomograms and accompanying web tool incorporating commonly captured clinical features were generated based on real-world data to optimize decision making in the clinic.

Independently validated sex-specific nomograms for predicting survival in patients with newly diagnosed glioblastoma: NRG Oncology RTOG 0525 and 0825

Background/purpose

Glioblastoma (GBM) is the most common primary malignant brain tumor. Sex has been shown to be an important prognostic factor for GBM. The purpose of this study was to develop and independently validate sex-specific nomograms for estimation of individualized GBM survival probabilities using data from 2 independent NRG Oncology clinical trials.

Methods

This analysis included information on 752 (NRG/RTOG 0525) and 599 (NRG/RTOG 0825) patients with newly diagnosed GBM. The Cox proportional hazard models by sex were developed using NRG/RTOG 0525 and significant variables were identified using a backward selection procedure. The final selected models by sex were then independently validated using NRG/RTOG 0825.

Results

Final nomograms were built by sex. Age at diagnosis, KPS, MGMT promoter methylation and location of tumor were common significant predictors of survival for both sexes. For both sexes, tumors in the frontal lobes had significantly better survival than tumors of multiple sites. Extent of resection, and use of corticosteroids were significant predictors of survival for males.

Conclusions

A sex specific nomogram that assesses individualized survival probabilities (6-, 12- and 24-months) for patients with GBM could be more useful than estimation of overall survival as there are factors that differ between males and females. A user friendly online application can be found here—https://npatilshinyappcalculator.shinyapps.io/SexDifferencesInGBM/.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11060-021-03886-5.

Deep learning applications in neuro-oncology

Deep learning (DL) is a relatively newer subdomain of machine learning (ML) with incredible potential for certain applications in the medical field. Given recent advances in its use in neuro-oncology, its role in diagnosing, prognosticating, and managing the care of cancer patients has been the subject of many research studies. The gamut of studies has shown that the landscape of algorithmic methods is constantly improving with each iteration from its inception. With the increase in the availability of high-quality data, more training sets will allow for higher fidelity models. However, logistical and ethical concerns over a prospective trial comparing prognostic abilities of DL and physicians severely limit the ability of this technology to be widely adopted. One of the medical tenets is judgment, a facet of medical decision making in DL that is often missing because of its inherent nature as a "black box." A natural distrust for newer technology, combined with a lack of autonomy that is normally expected in our current medical practices, is just one of several important limitations in implementation. In our review, we will first define and outline the different types of artificial intelligence (AI) as well as the role of AI in the current advances of clinical medicine. We briefly highlight several of the salient studies using different methods of DL in the realm of neuroradiology and summarize the key findings and challenges faced when using this nascent technology, particularly ethical challenges that could be faced by users of DL.

Therapeutic Perspective of Temozolomide Resistance in Glioblastoma Treatment

Glioblastoma (GB) is the most lethal form of primary brain neoplasm. TMZ is the first-line standard treatment, but the strong resistance constrains the efficacy in clinical use. GB contains glioma stem cells (GSCs), which contribute to TMZ resistance, promote cell survival evolvement, and repopulate the tumor mass. This review summarizes the TMZ-resistance mechanisms and discusses several potential therapies from the conservative opinion of GSC-targeted therapy orientation to the current view of TMZ resistance-aimed efficacy, which will provide an understanding of the role of heterogeneity in drug resistance and improve therapeutic efficacy in general.

A nomogram for individualized prediction of overall survival in patients with newly diagnosed glioblastoma: a real-world retrospective cohort study

Background

This study aimed to identify the most valuable predictors of prognosis in glioblastoma (GBM) patients and develop and validate a nomogram to estimate individualized survival probability.

Methods

We conducted a real-world retrospective cohort study of 987 GBM patients diagnosed between September 2010 and December 2018. Computer generated random numbers were used to assign patients into a training cohort (694 patients) and internal validation cohort (293 patients). A least absolute shrinkage and selection operator (LASSO)-Cox model was used to select candidate variables for the prediction model. Cox proportional hazards regression was used to estimate overall survival. Models were internally validated using the bootstrap method and generated individualized predicted survival probabilities at 6, 12, and 24 months, which were compared with actual survival.

Results

The final nomogram was developed using the Cox proportional hazards model, which was the model with best fit and calibration. Gender, age at surgery, extent of tumor resection, radiotherapy, chemotherapy, and IDH1 mutation status were used as variables. The concordance indices for 6-, 12-, 18-, and 24-month survival probabilities were 0.776, 0.677, 0.643, and 0.629 in the training set, and 0.725, 0.695, 0.652, and 0.634 in the validation set, respectively.

Conclusions

Our nomogram that assesses individualized survival probabilities (6-, 12-, and 24-month) in newly diagnosed GBM patients can assist healthcare providers in optimizing treatment and counseling patients.

Trial registration: retrospectively registered.

A novel nomogram for predicting the risk of epilepsy occurrence after operative in gliomas patients without preoperative epilepsy history

OBJECTIVE

Epilepsy is a common complication in glioma patients after undergoing brain tumor surgery combined with chemotherapy and/or radiotherapy. Whether antiepileptic drug prophylaxis could be used in these patients remains an open question. The purpose of this study was to produce a model for predicting the risk of epilepsy occurrence in such patients.

METHODS

The clinicopathologic data of glioma patients after tumor treatment were reviewed in this study. Univariate and multivariate logistic regression analyses were carried out to analyze the correlation between the clinicopathologic data and the risk of epilepsy occurrence. A nomogram was built according to the multivariate logistic regression model results.

RESULTS

A total of 219 patients with gliomas were reviewed. Univariate analyses revealed that age, WHO glioma classification, CD34, EGFR, Ki67, MGMT, P53 and VIM were significantly associated with the risk of epilepsy occurrence. Multivariate analyses revealed that age, WHO glioma classification, CD34, EGFR, MGMT, and VIM were predictors of risk of epilepsy occurrence. A nomogram of the risk of epilepsy occurrence was built based on statistically significant variables from the multivariate logistic regression analysis. The c-index of the nomogram was 0.755 (95 % confidence interval (CI), 0.742-0.769).

SIGNIFICANCE

This nomogram model provides reliable information about the risk of epilepsy occurrence for oncologists and neurological physicians.

The Influence of Gene Aberrations on Survival in Resected IDH Wildtype Glioblastoma Patients: A Single-Institution Study

This prospective population-based study on a group of 132 resected IDH-wildtype (IDH-wt) glioblastoma (GBM) patients assesses the prognostic and predictive value of selected genetic biomarkers and clinical factors for GBM as well as the dependence of these values on the applied therapeutic modalities. The patients were treated in our hospital between June 2006 and June 2015. Clinical data and tumor samples were analyzed to determine the frequencies of TP53, MDM2, EGFR, RB1, BCR, and CCND1 gene aberrations and the duplication/deletion statuses of the 9p21.3, 1p36.3, 19q13.32, and 10p11.1 chromosome regions. Cut-off values distinguishing low (LCN) and high (HCN) copy number status for each marker were defined. Additionally, MGMT promoter methylation and IDH1/2 mutation status were investigated retrospectively. Young age, female gender, Karnofsky scores (KS) above 80, chemoradiotherapy, TP53 HCN, and CCND1 HCN were identified as positive prognostic factors, and smoking was identified as a negative prognostic factor. Cox proportional regression models of the chemoradiotherapy patient group revealed TP53 HCN and CCND1 HCN to be positive prognostic factors for both progression-free survival and overall survival. These results confirmed the influence of key clinical factors (age, KS, adjuvant oncotherapy, and smoking) on survival in GBM IDH-wt patients and demonstrated the prognostic and/or predictive importance of CCND1, MDM2, and 22q12.2 aberrations.

GLIMPSE: a glioblastoma prognostication model using ensemble learning-a surveillance, epidemiology, and end results study

Purpose

Glioblastoma is one of the most common and aggressive brain tumors in the world with a poor prognosis. A glioblastoma prognostication model has the potential to improve the cancer's standard of care. No other paper has looked at using ensemble learning with a population database to predict multiple binary glioblastoma survival outcomes.

Methods

We utilized ensemble learning to design, build, and test a prognostication system for glioblastoma for short-, intermediate- and long-term survival, based on various clinical features. We used the population database SEER which covers 17 different registries. The most important prognostic features were identified and used as a clinical feature set. The statistical feature set was determined using Random Forests. The accuracy, sensitivity, specificity, area under the receiver operating characteristic (AUROC), positive predictive value (PPV), and negative predictive value (NPV) were reported.

Results

Statistically-determined feature sets had the best performance. All the top models for short, intermediate, and long-term survival were random forests. With regards to short-term survival, top model had metrics AUROC = 0.937, accuracy = 86%, specificity = 88%, sensitivity = 85%, NPV = 85%, and PPV = 87%. For long-term survival, the top model had AUROC = 0.893, accuracy = 81%, specificity = 79%, sensitivity = 83%, NPV = 82%, and PPV = 79%. The top intermediate-term survival prediction had AUROC ≥ 0.780 and the other metrics were at least 70%.

Conclusions

Our ensemble models were high-performing and achieved AUROCs as high as 0.94, highlighting the importance of balancing, using ensemble techniques and statistical feature selection. Our models can potentially be used by clinicians after external validation.

A Nomogram Predicts Individual Prognosis in Patients With Newly Diagnosed Glioblastoma by Integrating the Extent of Resection of Non-Enhancing Tumors

Background

The extent of resection of non-contrast enhancing tumors (EOR-NCEs) has been shown to be associated with prognosis in patients with newly diagnosed glioblastoma (nGBM). This study aimed to develop and independently validate a nomogram integrated with EOR-NCE to assess individual prognosis.

Methods

Data for this nomogram were based on 301 patients hospitalized for nGBM from October 2011 to April 2019 at the Beijing Tiantan Hospital, Capital Medical University. These patients were randomly divided into derivation (n=181) and validation (n=120) cohorts at a ratio of 6:4. To evaluate predictive accuracy, discriminative ability, and clinical net benefit, concordance index (C-index), receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA) were calculated for the extent of resection of contrast enhancing tumor (EOR-CE) and EOR-NCE nomograms. Comparison between these two models was performed as well.

Results

The Cox proportional hazards model was used to establish nomograms for this study. Older age at diagnosis, Karnofsky performance status (KPS)<70, unmethylated O⁶-methylguanine-DNA methyltransferase (MGMT) status, wild-type isocitrate dehydrogenase enzyme (IDH), and lower EOR-CE and EOR-NCE were independent factors associated with shorter survival. The EOR-NCE nomogram had a higher C-index than the EOR-CE nomogram. Its calibration curve for the probability of survival exhibited good agreement between the identical and actual probabilities. The EOR-NCE nomogram showed superior net benefits and improved performance over the EOR-CE nomogram with respect to DCA and ROC for survival probability. These results were also confirmed in the validation cohort.

Conclusions

An EOR-NCE nomogram assessing individualized survival probabilities (12-, 18-, and 24-month) for patients with nGBM could be useful to provide patients and their relatives with health care consultations on optimizing therapeutic approaches and prognosis.

A validated integrated clinical and molecular glioblastoma long-term survival-predictive nomogram

Background

Glioblastoma (GBM) is the most common primary malignant brain tumor in adulthood. Despite multimodality treatments, including maximal safe resection followed by irradiation and chemotherapy, the median overall survival times range from 14 to 16 months. However, a small subset of GBM patients live beyond 5 years and are thus considered long-term survivors.

Methods

A retrospective analysis of the clinical, radiographic, and molecular features of patients with newly diagnosed primary GBM who underwent treatment at The University of Texas MD Anderson Cancer Center was conducted. Eighty patients had sufficient quantity and quality of tissue available for next-generation sequencing and immunohistochemical analysis. Factors associated with survival time were identified using proportional odds ordinal regression. We constructed a survival-predictive nomogram using a forward stepwise model that we subsequently validated using The Cancer Genome Atlas.

Results

Univariate analysis revealed 3 pivotal genetic alterations associated with GBM survival: both high tumor mutational burden (P = .0055) and PTEN mutations (P = .0235) negatively impacted survival, whereas IDH1 mutations positively impacted survival (P < .0001). Clinical factors significantly associated with GBM survival included age (P < .0001), preoperative Karnofsky Performance Scale score (P = .0001), sex (P = .0164), and clinical trial participation (P < .0001). Higher preoperative T1-enhancing volume (P = .0497) was associated with shorter survival. The ratio of TI-enhancing to nonenhancing disease (T1/T2 ratio) also significantly impacted survival (P = .0022).

Conclusions

Our newly devised long-term survival-predictive nomogram based on clinical and genomic data can be used to advise patients regarding their potential outcomes and account for confounding factors in nonrandomized clinical trials.

A Comparison Study of Machine Learning (Random Survival Forest) and Classic Statistic (Cox Proportional Hazards) for Predicting Progression in High-Grade Glioma after Proton and Carbon Ion Radiotherapy

Background

Machine learning (ML) algorithms are increasingly explored in glioma prognostication. Random survival forest (RSF) is a common ML approach in analyzing time-to-event survival data. However, it is controversial which method between RSF and traditional cornerstone method Cox proportional hazards (CPH) is better fitted. The purpose of this study was to compare RSF and CPH in predicting tumor progression of high-grade glioma (HGG) after particle beam radiotherapy (PBRT).

Methods

The study enrolled 82 consecutive HGG patients who were treated with PBRT at Shanghai Proton and Heavy Ion Center between 6/2015 and 11/2019. The entire cohort was split into the training and testing set in an 80/20 ratio. Ten variables from patient-related, tumor-related and treatment-related information were utilized for developing CPH and RSF for predicting progression-free survival (PFS). The model performance was compared in concordance index (C-index) for discrimination (accuracy), brier score (BS) for calibration (precision) and variable importance for interpretability.

Results

The CPH model demonstrated a better performance in terms of integrated C-index (62.9%) and BS (0.159) compared to RSF model (C-index = 61.1%, BS = 0.174). In the context of variable importance, CPH model indicated that age (P = 0.024), WHO grade (P = 0.020), IDH gene (P = 0.019), and MGMT promoter status (P = 0.040) were significantly correlated with PFS in the univariate analysis; multivariate analysis showed that age (P = 0.041), surgical completeness (P = 0.084), IDH gene (P = 0.057), and MGMT promoter (P = 0.092) had a significant or trend toward the relation with PFS. RSF showed that merely IDH and age were of positive importance for predicting PFS. A final nomogram was developed to predict tumor progression at the individual level based on CPH model.

Conclusions

In a relatively small dataset with HGG patients treated with PBRT, CPH outperformed RSF for predicting tumor progression. A comprehensive criterion with accuracy, precision, and interpretability is recommended in evaluating ML prognostication approaches for clinical deployment.

Prognosis Analysis and Validation of m6A Signature and Tumor Immune Microenvironment in Glioma

Glioma is one of the most typical intracranial tumors, comprising about 80% of all brain malignancies. Several key molecular signatures have emerged as prognostic biomarkers, which indicate room for improvement in the current approach to glioma classification. In order to construct a more veracious prediction model and identify the potential prognosis-biomarker, we explore the differential expressed m⁶A RNA methylation regulators in 665 gliomas from TCGA-GBM and TCGA-LGG. Consensus clustering was applied to the m6A RNA methylation regulators, and two glioma subgroups were identified with a poorer prognosis and a higher grade of WHO classification in cluster 1. The further chi-squared test indicated that the immune infiltration was significantly enriched in cluster 1, indicating a close relation between m⁶A regulators and immune infiltration. In order to explore the potential biomarkers, the weighted gene co-expression network analysis (WGCNA), along with Least absolute shrinkage and selection operator (LASSO), between high/low immune infiltration and m⁶A cluster 1/2 groups were utilized for the hub genes, and four genes (TAGLN2, PDPN, TIMP1, EMP3) were identified as prognostic biomarkers. Besides, a prognostic model was constructed based on the four genes with a good prediction and applicability for the overall survival (OS) of glioma patients (the area under the curve of ROC achieved 0.80 (0.76-0.83) and 0.72 (0.68-0.76) in TCGA and Chinese Glioma Genome Atlas (CGGA), respectively). Moreover, we also found PDPN and TIMP1 were highly expressed in high-grade glioma from The Human Protein Atlas database and both of them were correlated with m6A and immune cell marker in glioma tissue samples. In conclusion, we construct a novel prognostic model which provides new insights into glioma prognosis. The PDPN and TIMP1 may serve as potential biomarkers for prognosis of glioma.

Treatment and surgical factors associated with longer-term glioblastoma survival: a National Cancer Database study

Background

Insufficient data exist to characterize factors associated with longer-term survival of glioblastoma (GBM). A population-based analysis of GBM longer-term survivors (LTS) in the United States was conducted to investigate the association between treatment, demographic, surgical factors, and longer-term survival.

Methods

From the National Cancer Database, GBM patients were identified using ICD-O-3 histology codes 9440-9442/3, 2005-2015 and were divided into routine (≤3 years) and longer-term (>3 years) overall survival (OS) groups. Univariable and multivariable logistic regression analysis was used to assess factors associated with longer-term survival. A subset analysis was performed to further investigate the association of extent of resection and treatment combinations on OS outcomes.

Results

A total of 93 036 patients with GBM met study criteria. Among these patients, 8484 were LTS and 84 552 were routine survivors (RS). When comparing LTS (OS of >3 years) with RS (OS of ≤3 years), younger age, insured status, metro/urban residence, treatment at academic facility, and fewer comorbidities were associated with longer-term survival. In addition, trimodality therapy (chemotherapy + radiation + surgery) was associated with having best odds of longer-term survival (odds ratio = 4.89, 95% confidence interval [3.58, 6.68]); 74% of LTS received such therapy compared with 51% of RS. Subset analysis revealed that total resection is only associated with longer-term survival status for those receiving trimodality therapy or surgery only.

Conclusions

In a population-based analysis, standard of care surgery and chemo radiation connote a survival advantage in GBM. Among those receiving standard of care, having a total resection is most beneficial for longer-term survival status.