Predicting survival in patients with glioblastoma using MRI radiomic features extracted from radiation planning volumes.

BACKGROUND: Quantitative image analysis using pre-operative magnetic resonance imaging (MRI) has been able to predict survival in patients with glioblastoma (GBM). The study explored the role of postoperative radiation (RT) planning MRI-based radiomics to predict the outcomes, with features extracted from the gross tumor volume (GTV) and clinical target volume (CTV). METHODS: Patients with IDH-wildtype GBM treated with adjuvant RT having MRI as a part of RT planning process were included in the study. 546 features were extracted from each GTV and CTV. A LASSO Cox model was applied, and internal validation was performed using leave-one-out cross-validation with overall survival as endpoint. Cross-validated time-dependent area under curve (AUC) was constructed to test the efficacy of the radiomics model, and clinical features were used to generate a combined model. Analysis was done for the entire group and in individual surgical groups-gross total excision (GTR), subtotal resection (STR), and biopsy. RESULTS: 235 patients were included in the study with 57, 118, and 60 in the GTR, STR, and biopsy subgroup, respectively. Using the radiomics model, binary risk groups were feasible in the entire cohort (p < 0.01) and biopsy group (p = 0.04), but not in the other two surgical groups individually. The integrated AUC (iAUC) was 0.613 for radiomics-based classification in the biopsy subgroup, which improved to 0.632 with the inclusion of clinical features. CONCLUSION: Imaging features extracted from the GTV and CTV regions can lead to risk-stratification of GBM undergoing biopsy, while the utility in other individual subgroups needs to be further explored.

Quantitative mapping of individual voxels in the peritumoral region of IDH-wildtype glioblastoma to distinguish between tumor infiltration and edema.

PURPOSE: The peritumoral region (PTR) in glioblastoma (GBM) represents a combination of infiltrative tumor and vasogenic edema, which are indistinguishable on magnetic resonance imaging (MRI). We developed a radiomic signature by using imaging data from low grade glioma (LGG) (marker of tumor) and PTR of brain metastasis (BM) (marker of edema) and applied it on the GBM PTR to generate probabilistic maps. METHODS: 270 features were extracted from T1-weighted, T2-weighted, and apparent diffusion coefficient maps in over 3.5 million voxels of LGG (36 segments) and BM (45 segments) scanned in a 1.5T MRI. A support vector machine classifier was used to develop the radiomics model from approximately 50% voxels (downsampled to 10%) and validated with the remaining. The model was applied to over 575,000 voxels of the PTR of 10 patients with GBM to generate a quantitative map using Platt scaling (infiltrative tumor vs. edema). RESULTS: The radiomics model had an accuracy of 0.92 and 0.79 in the training and test set, respectively (LGG vs. BM). When extrapolated on the GBM PTR, 9 of 10 patients had a higher percentage of voxels with a tumor-like signature over radiological recurrence areas. In 7 of 10 patients, the areas under curves (AUC) were > 0.50 confirming a positive correlation. Including all the voxels from the GBM patients, the infiltration signature had an AUC of 0.61 to predict recurrence. CONCLUSION: A radiomic signature can demarcate areas of microscopic tumors from edema in the PTR of GBM, which correlates with areas of future recurrence.

MRI radiomics to differentiate between low grade glioma and glioblastoma peritumoral region.

BACKGROUND: The peritumoral region (PTR) of glioblastoma (GBM) appears as a T2W-hyperintensity and is composed of microscopic tumor and edema. Infiltrative low grade glioma (LGG) comprises tumor cells that seem similar to GBM PTR on MRI. The work here explored if a radiomics-based approach can distinguish between the two groups (tumor and edema versus tumor alone). METHODS: Patients with GBM and LGG imaged using a 1.5 T MRI were included in the study. Image data from cases of GBM PTR, and LGG were manually segmented guided by T2W hyperintensity. A set of 91 first-order and texture features were determined from each of T1W-contrast, and T2W-FLAIR, diffusion-weighted imaging sequences. Applying filtration techniques, a total of 3822 features were obtained. Different feature reduction techniques were employed, and a subsequent model was constructed using four machine learning classifiers. Leave-one-out cross-validation was used to assess classifier performance. RESULTS: The analysis included 42 GBM and 36 LGG. The best performance was obtained using AdaBoost classifier using all the features with a sensitivity, specificity, accuracy, and area of curve (AUC) of 91%, 86%, 89%, and 0.96, respectively. Amongst the feature selection techniques, the recursive feature elimination technique had the best results, with an AUC ranging from 0.87 to 0.92. Evaluation with the F-test resulted in the most consistent feature selection with 3 T1W-contrast texture features chosen in over 90% of instances. CONCLUSIONS: Quantitative analysis of conventional MRI sequences can effectively demarcate GBM PTR from LGG, which is otherwise indistinguishable on visual estimation.

Local control and patterns of failure for "Radioresistant" spinal metastases following stereotactic body radiotherapy compared to a "Radiosensitive" reference.

PURPOSE: The concept of a radioresistant (RR) phenotype has been challenged with use of stereotactic body radiotherapy (SBRT). We compared outcomes following SBRT to RR spinal metastases to a radiosensitive cohort. METHODS: Renal cell, melanoma, sarcoma, gastro-intestinal, and thyroid spinal metastases were identified as RR and prostate cancer (PCA) as radiosensitive. The primary endpoint was MRI-based local failure (LF). Secondary endpoints included overall survival (OS) and vertebral compression fracture (VCF). RESULTS: From a prospectively maintained database of 1394 spinal segments in 605 patients treated with spine SBRT, 173 patients/395 RR spinal segments were compared to 94 patients/185 PCA segments. Most received 24-28 Gy in 2 fractions (68.9%) and median follow-up was 15.5 months (range, 1.4-84.2 months). 1- and 2-year LF rates were 19.2% and 22.4% for RR metastases, respectively, which were significantly greater (p < 0.001) than PCA (3.2% and 8.4%, respectively). Epidural disease (HR: 2.47, 95% CI 1.65-3.71, p < 0.001) and RR histology (HR: 2.41, 95% CI 1.45-3.99, p < 0.001) predicted for greater LF. Median OS was 17.4 and 61.0 months for RR and PCA cohorts, respectively. Lung/liver metastases, polymetastatic disease and epidural disease predicted for worse OS. 2-year VCF rates were ~ 13% in both cohorts. Coverage of the CTV V90 (clinical target volume receiving 90% of prescription dose) by ≥ 87% (HR: 2.32, 95% CI 1.29-4.18, p = 0.005), no prior spine radiotherapy (HR: 1.96, 95% CI 1.09-3.55, p = 0.025), and a greater Spinal Instability Neoplasia Score (p = 0.013) predicted for VCF. CONCLUSIONS: Higher rates of LF were observed after spine SBRT in RR metastases. Optimization strategies include dose escalation and aggressive management of epidural disease.

Quantitative CEST and MT at 1.5T for monitoring treatment response in glioblastoma: early and late tumor progression during chemoradiation.

PURPOSE: Quantitative MRI (qMRI) was performed using a 1.5T protocol that includes a novel chemical exchange saturation transfer/magnetization transfer (CEST/MT) approach. The purpose of this prospective study was to determine if qMRI metrics at baseline, at the 10th and 20th fraction during a 30 fraction/6 week standard chemoradiation (CRT) schedule, and at 1 month following treatment could be an early indicator of response for glioblastoma (GBM). METHODS: The study included 51 newly diagnosed GBM patients. Four regions-of-interest (ROI) were analyzed: (i) the radiation defined clinical target volume (CTV), (ii) radiation defined gross tumor volume (GTV), (iii) enhancing-tumor regions, and (iv) FLAIR-hyperintense regions. Quantitative CEST, MT, T1 and T2 parameters were compared between those patients progressing within 6.9 months (early), and those progressing after CRT (late), using mixed modelling. Exploratory predictive modelling was performed to identify significant predictors of early progression using a multivariable LASSO model. RESULTS: Results were dependent on the specific tumor ROI analyzed and the imaging time point. The baseline CEST asymmetry within the CTV was significantly higher in the early progression cohort. Other significant predictors included the T2 of the MT pools (for semi-solid at fraction 20 and water at 1 month after CRT), the exchange rate (at fraction 20) and the MGMT methylation status. CONCLUSIONS: We observe the potential for multiparametric qMRI, including a novel pulsed CEST/MT approach, to show potential in distinguishing early from late progression GBM cohorts. Ultimately, the goal is to personalize therapeutic decisions and treatment adaptation based on non-invasive imaging-based biomarkers.

Diagnostic Neuroradiology Subspecialty Training: 1 Versus 2 Years; the Canadian Perspective.

BACKGROUND: Canada began a national reform of its post-graduate medical education training programs to a Competence By Design (CBD) model. Trends from accredited neuroradiology programs from the past 10 years were investigated to inform educators and stakeholders for this process. METHODS: A 13-question electronic survey was sent to program directors of all 8 accredited neuroradiology training programs in Canada. Data was requested for each year on the 2008-2019 graduating classes. Questions pertained to program enrolment; program completion; post-training employment; and the sufficiency of 1-year training programs. RESULTS: Response rate was 100%. Over the timeframe studied, the 2-year programs increased in size (P = 0.007), while the 1-year programs remained steady (P = 0.27). 12.2% of trainees enrolled in the 2-year program dropped out after 1 year, and were considered 1-year trainees thereafter. A higher proportion of 2-year trainees obtain positions within academic institutions (89.5 vs 67.2%, P = 0.0007), whereas a higher proportion of 1-year trainees obtain positions within non-academic institutions (29.3 vs 8.1%, P = 0.0007). A higher proportion of those with Canadian board certification in diagnostic radiology who completed a 2-year program obtained a position within a Canadian academic institution compared to non-certified 2-year trainees (P < 0.001). 71.4% of program directors agreed that a 1-year program was sufficient for non-academic staff positions. CONCLUSION: The length of the training program has significant impact on employment in academic vs non-academic institutions. This information can be used to guide the upcoming CBD initiative for neuroradiology programs.

Congenital Herniation of the Gyrus Rectus Resulting in Compressive Optic Neuropathy.

We report a 34-year-old male with a previously uninvestigated lifelong blindness of the right eye from compressive optic neuropathy secondary to congenital herniation of the gyrus rectus (HGR). His past medical history was otherwise unremarkable, with no history of prior head or ocular trauma. On examination, he had no light perception in the right eye, right relative afferent pupillary defect (RAPD), and primary optic atrophy. His left eye had normal visual acuity, color vision, and a healthy optic disc. There was a sensory exotropia in the right eye; however, extraocular movements were intact and the remainder of his neurological exam was normal. MRI revealed compression of the prechiasmatic right optic nerve from HGR and atrophy of the right optic nerve and optic chiasm (Figures 1 and 2), without any parenchymal mass lesions. There were no signal abnormalities in the optic nerves or the chiasm.

Updated Clinical Practice Guideline on Use of Gadolinium-Based Contrast Agents in Kidney Disease Issued by the Canadian Association of Radiologists.

In 2017, the Canadian Association of Radiologists issued a clinical practice guideline (CPG) regarding the use of gadolinium-based contrast agents (GBCAs) in patients with acute kidney injury (AKI), chronic kidney disease (CKD), or on dialysis due to mounting evidence indicating that nephrogenic systemic fibrosis (NSF) occurs with extreme rarity or not at all when using Group II GBCAs or the Group III GBCA gadoxetic acid (compared to first generation Group I linear GBCAs). One of the goals of the work group was to re-evaluate the CPG after 24 months to determine the effect of more liberal use of GBCA on reported cases of NSF in patients with AKI, CKD Stage 4 or 5 (estimated glomerular filtration rate [eGFR] < 30 mL/min/1.73 m2), or those that are dialysis-dependent. A comprehensive review of the literature was conducted by a subcommittee of the initial CPG panel between the dates of January 1, 2017-December 31, 2018 to identify new unconfounded cases of NSF linked to Group II or Group III GBCAs and an updated CPG developed. To our knowledge, when using a Group II or Group III GBCA between 2017-2018, only a single unconfounded case report of a fibrosing dermopathy has been reported in a patient who received gadobenate dimeglumine with Stage 2 CKD. No other unconfounded cases of NSF have been reported with Group II or III agents in during this timeframe. The subcommittee concluded that the main recommendations from the 2017 CPG should remain unaltered, but agreed that screening for renal disease in the outpatient setting is no longer justifiable, cost-effective or recommended. Patients on hemodialysis (HD) should, however, be identified prior to GBCA administration to arrange timely HD to optimize gadolinium clearance, although there remains no evidence that HD reduces the risk of NSF. When administering Group II or III GBCAs to patients with AKI, on dialysis or with severe CKD, informed consent relating to NSF is also no longer explicitly recommended.

Hypoxia Detection in Infiltrative Astrocytoma: Ferumoxytol-based Quantitative BOLD MRI with Intraoperative and Histologic Validation.

Purpose To validate ferumoxytol-based quantitative blood oxygenation level-dependent (BOLD) MRI for mapping oxygenation of human infiltrative astrocytomas by using intraoperative measurement of tissue oxygen tension and histologic staining. Materials and Methods Fifteen patients with infiltrative astrocytomas were recruited into this prospective multicenter study between July 2014 and December 2016. Prior to treatment, participants underwent preoperative quantitative BOLD MRI with ferumoxytol to generate tissue oxygen saturation (StO2) maps. Two intratumoral sites were identified, one with low StO2 and one with high StO2. Neuronavigation was used to locate sites intraoperatively for insertion of oxygen-sensing probes to measure local tissue oxygen tension (PtO2). Biopsies from both sites were taken and stained for markers of hypoxia (hypoxia-inducible factor 1α, carbonic anhydrase IX) and neoangiogenesis (vascular endothelial growth factor, endoglin [CD105]). Spearman correlation and nonparametric sign-rank tests were used to analyze data. Results Ten patients with median age of 58.5 years (interquartile range, 25 years; four men and six women) completed the study. Because there is no linear relationship between StO2 and PtO2, the ratios of low to high StO2 versus low to high PtO2 in each patient were compared and a significant correlation was found (r = 0.73; P = .01). Pathologic analyses revealed differences between carbonic anhydrase IX (P = .03) for sites of low StO2 versus high StO2. CD105 displayed a similar trend but was not significant (P = .09). Conclusion Ferumoxytol-based quantitative blood oxygenation level-dependent MRI can potentially be used as a noninvasive surrogate for oxygenation mapping in infiltrative astrocytomas. This technique can potentially be integrated in treatment planning for aggressive targeting of hypoxic areas in tumors.

Gadolinium Deposition in the Brain: A Systematic Review of Existing Guidelines and Policy Statement Issued by the Canadian Association of Radiologists.

Emerging evidence has confirmed that, following administration of a gadolinium-based contrast agent (GBCA), very small amounts of gadolinium will deposit in the brain of humans with intact blood-brain barriers. The literature is evolving rapidly and the degree to which gadolinium will deposit for a particular GBCA or class of GBCAs remains undetermined. Several studies suggest that linear GBCAs deposit more gadolinium in the brain compared with macrocyclic GBCAs; however, our understanding of the molecular composition of deposited gadolinium is preliminary, and the clinical significance of gadolinium deposition remains unknown. To date, there is no conclusive evidence linking gadolinium deposition in the brain with any adverse patient outcome. A panel of radiologists representing the Canadian Association of Radiologists was assembled to assist the Canadian medical imaging community in making informed decisions regarding the issue of gadolinium deposition in the brain. The objectives of the working group were: 1) to review the evidence from animal and human studies; 2) to systematically review existing guidelines and position statements issued by other organizations and health agencies; and 3) to formulate an evidence-based position statement on behalf of the Canadian Association of Radiologists. Based on our appraisal of the evidence and systematic review of 9 guidelines issued by other organizations, the working group established the following consensus statement. GBCA administration should be considered carefully with respect to potential risks and benefits, and only used when required. Standard dosing should be used and repeat administrations should be avoided unless necessary. Gadolinium deposition is one of several issues to consider when prescribing a particular GBCA. Currently there is insufficient evidence to recommend one class of GBCA over another. The panel considered it inappropriate to withhold a linear GBCA if a macrocyclic agent is unavailable, if hepatobiliary phase imaging is required, or if there is a history of severe allergic reaction to a macrocyclic GBCA. Further study in this area is required, and the evidence should be monitored regularly with policy statements updated accordingly.

Radiologic Differences between Human Papillomavirus-Related and Human Papillomavirus-Unrelated Oropharyngeal Carcinoma on Diffusion-Weighted Imaging.

BACKGROUND AND PURPOSE: Human papilloma virus-related oropharyngeal carcinoma (HPV+ OPC) is a unique entity compared to HPV-unrelated (HPV-) OPC. Previous studies were inconclusive regarding the differences between HPV+ and HPV- OPCs on diffusion-weighted imaging (DWI). This study sought to determine if there is an association between HPV status and apparent diffusion coefficient (ADC) values as an imaging biomarker in OPCs. MATERIALS AND METHODS: OPC patients with pretreatment MRI including DWI were retrospectively reviewed and analyzed as a blinded, controlled cohort. HPV status was biopsy-ascertained with p16 staining. ADC values were determined by placing the largest possible circular region of interest in solid portions of primary tumors and/or metastatic lymph nodes. Necrotic and cystic portions were excluded. RESULTS: Twenty-eight HPV+ and 12 HPV- patients were included. Adjusted for age and sex, ADC values were significantly lower in HPV+ OPC primary tumors (p = 0.013) and lymph node metastases (p = 0.013). The area under the curve (AUC) was 0.85 and 0.90, respectively. A model with a linear combination of the 2 variables yielded an AUC of 0.92. CONCLUSION: ADC values were significantly lower in both the primary tumors and lymph node metastases in HPV+ OPCs compared to HPV- OPCs. These results confirm the results of prior studies.

Safety of Magnetic Resonance Imaging in Pregnancy.

Magnetic resonance imaging is being increasingly used to diagnose and follow up a variety of medical conditions in pregnancy, both for maternal and fetal indications. However, limited data regarding its safe use in pregnancy may be a source of anxiety and avoidance for both patients and their healthcare providers. In this review, we critically discuss the main safety concerns of Magnetic Resonance Imaging (MRI) in pregnancy including energy deposition, acoustic noise, and use of contrast agents, supported by data from animal and human studies. Use of maternal sedatives and concerns related to occupational exposure in pregnant personnel are also addressed. Exposure to gadolinium-based contrast agents and sedation for MRI during pregnancy should be avoided whenever feasible.

Synthesis of diffusion-weighted MRI scalar maps from FLAIR volumes using generative adversarial networks.

Introduction: Acquisition and pre-processing pipelines for diffusion-weighted imaging (DWI) volumes are resource- and time-consuming. Generating synthetic DWI scalar maps from commonly acquired brain MRI sequences such as fluid-attenuated inversion recovery (FLAIR) could be useful for supplementing datasets. In this work we design and compare GAN-based image translation models for generating DWI scalar maps from FLAIR MRI for the first time. Methods: We evaluate a pix2pix model, two modified CycleGANs using paired and unpaired data, and a convolutional autoencoder in synthesizing DWI fractional anisotropy (FA) and mean diffusivity (MD) from whole FLAIR volumes. In total, 420 FLAIR and DWI volumes (11,957 images) from multi-center dementia and vascular disease cohorts were used for training/testing. Generated images were evaluated using two groups of metrics: (1) human perception metrics including peak signal-to-noise ratio (PSNR) and structural similarity (SSIM), (2) structural metrics including a newly proposed histogram similarity (Hist-KL) metric and mean squared error (MSE). Results: Pix2pix demonstrated the best performance both quantitatively and qualitatively with mean PSNR, SSIM, and MSE metrics of 23.41 dB, 0.8, 0.004, respectively for MD generation, and 24.05 dB, 0.78, 0.004, respectively for FA generation. The new histogram similarity metric demonstrated sensitivity to differences in fine details between generated and real images with mean pix2pix MD and FA Hist-KL metrics of 11.73 and 3.74, respectively. Detailed analysis of clinically relevant regions of white matter (WM) and gray matter (GM) in the pix2pix images also showed strong significant (p < 0.001) correlations between real and synthetic FA values in both tissue types (R = 0.714 for GM, R = 0.877 for WM). Discussion/conclusion: Our results show that pix2pix's FA and MD models had significantly better structural similarity of tissue structures and fine details than other models, including WM tracts and CSF spaces, between real and generated images. Regional analysis of synthetic volumes showed that synthetic DWI images can not only be used to supplement clinical datasets, but demonstrates potential utility in bypassing or correcting registration in data pre-processing.

Alzheimer's and vascular disease classification using regional texture biomarkers in FLAIR MRI.

Interactions between subcortical vascular disease and dementia due to Alzheimer's disease (AD) are unclear, and clinical overlap between the diseases makes diagnosis challenging. Existing studies have shown regional microstructural changes specific to each disease, and that textures in fluid-attenuated inversion recovery (FLAIR) MRI images may characterize abnormalities in tissue microstructure. This work aims to investigate regional FLAIR biomarkers that can differentiate dementia cohorts with and without subcortical vascular disease. FLAIR and diffusion MRI (dMRI) volumes were obtained in 65 mild cognitive impairment (MCI), 21 AD, 44 subcortical vascular MCI (scVMCI), 22 Mixed etiology, and 48 healthy elderly patients. FLAIR texture and intensity biomarkers were extracted from the normal appearing brain matter (NABM), WML penumbra, blood supply territory (BST), and white matter tract regions of each patient. All FLAIR biomarkers were correlated to dMRI metrics in each region and global WML load, and biomarker means between groups were compared using ANOVA. Binary classifications were performed using Random Forest classifiers to investigate the predictive nature of the regional biomarkers, and SHAP feature analysis was performed to further investigate optimal regions of interest for differentiating disease groups. The regional FLAIR biomarkers were strongly correlated to MD, while all biomarker regions but white matter tracts were strongly correlated to WML burden. Classification between Mixed disease and healthy, AD, and scVMCI patients yielded accuracies of 97%, 81%, and 72% respectively using WM tract biomarkers. Classification between scVMCI and healthy, MCI, and AD patients yielded accuracies of 89%, 84%, and 79% respectively using penumbra biomarkers. Only the classification between AD and healthy patients had optimal results using NABM biomarkers. This work presents novel regional FLAIR biomarkers that may quantify white matter degeneration related to subcortical vascular disease, and which indicate that investigating degeneration in specific regions may be more important than assessing global WML burden in vascular disease groups.

30 Gy in 4 Stereotactic Body Radiotherapy Fractions for Complex Spinal Metastases: Mature Outcomes Supporting This Novel Regimen.

BACKGROUND AND OBJECTIVES: We designed a 30 Gy in 4 fractions stereotactic body radiotherapy protocol, as an alternative option to our standard 2-fraction approach, for primarily large volume, multilevel, or previously radiated spinal metastases. We report imaging-based outcomes of this novel fractionation scheme. METHODS: The institutional database was reviewed to identify all patients who underwent 30 Gy/4 fractions from 2010 to 2021. Primary outcomes were magnetic resonance-based vertebral compression fracture (VCF) and local failure per treated vertebral segment. RESULTS: We reviewed 245 treated segments in 116 patients. The median age was 64 years (range, 24-90). The median number of consecutive segments within the treatment volume was 2 (range, 1-6), and the clinical target volume (CTV) was 126.2 cc (range, 10.4-863.5). Fifty-four percent had received at least 1 previous course of radiotherapy, and 31% had previous spine surgery at the treated segment. The baseline Spinal Instability Neoplastic Score was stable, potentially unstable, and unstable for 41.6%, 51.8%, and 6.5% of segments, respectively. The cumulative incidence of local failure was 10.7% (95% CI 7.1-15.2) at 1 year and 16% (95% CI 11.5-21.2) at 2 years. The cumulative incidence of VCF was 7.3% (95% CI 4.4-11.2) at 1 year and 11.2% (95% CI 7.5-15.8) at 2 years. On multivariate analysis, age ≥68 years ( P = .038), CTV volume ≥72 cc ( P = .021), and no previous surgery ( P = .021) predicted an increased risk of VCF. The risk of VCF for CTV volumes <72 cc/≥72 cc was 1.8%/14.6% at 2 years. No case of radiation-induced myelopathy was observed. Five percent of patients developed plexopathy. CONCLUSION: 30 Gy in 4 fractions was safe and efficacious despite the population being at increased risk of toxicity. The lower risk of VCF in previously stabilized segments highlights the potential for a multimodal treatment approach for complex metastases, especially for those with a CTV volume of ≥72 cc.

Mature Local Control and Reirradiation Rates Comparing Spine Stereotactic Body Radiation Therapy With Conventional Palliative External Beam Radiation Therapy.

PURPOSE: Stereotactic body radiation therapy (SBRT) improves complete pain response for painful spinal metastases compared with conventional external beam radiation therapy (cEBRT). We report mature local control and reirradiation rates in a large cohort of patients treated with SBRT versus cEBRT enrolled previously in the Canadian Clinical Trials Group Symptom Control 24 phase 2/3 trial. METHODS AND MATERIALS: One hundred thirty-seven of 229 (60%) patients randomized to 24 Gy in 2 SBRT fractions or 20 Gy in 5 cEBRT fractions were retrospectively reviewed. By including all treated spinal segments, we report on 66 patients (119 spine segments) treated with SBRT and 71 patients (169 segments) treated with cEBRT. The primary outcomes were magnetic resonance-based local control and reirradiation rates for each treated spine segment. RESULTS: The median follow-up was 11.3 months (interquartile range, 5.3-27.7 months), and median overall survival in the SBRT and cEBRT cohorts were 21.6 (95% confidence interval [CI], 11.3, upper bound not reached) and 18.9 (95% CI, 12.2-29.1) months (P = .428), respectively. The cohorts were balanced with respect to radioresistant histology and presence of mass (paraspinal and/or epidural disease extension). Risk of local failure after SBRT versus cEBRT at 6, 12, and 24 months were 2.8% (95% CI, 0.8%-7.4%) versus 11.2% (95% CI, 6.9%-16.6%), 6.1% (95% CI, 2.5%-12.1%) versus 28.4% (95% CI, 21.3%-35.9%), and 14.8% (95% CI, 8.2-23.1%) versus 35.6% (95% CI, 27.8%-43.6%), respectively (P < .001). cEBRT (hazard ratio [HR], 3.48; 95% CI, 1.94-6.25; P < .001) and presence of mass (HR, 2.07; 95% CI, 1.29-3.31; P = .002) independently predicted local failure on multivariable analysis. The 1-year reirradiation rates and median times to reirradiation after SBRT versus cEBRT were 2.2% (95% CI, 0.4-7.0%) versus 15.8% (95% CI, 10.4%-22.3%) (P = .002) and 22.9 months versus 9.5 months, respectively. cEBRT (HR, 2.60; 95% CI, 1.27-5.30; P = .009) and radioresistant histology (HR, 2.00; 95% CI, 1.12-3.60; P = .020) independently predicted for reirradiation. Eight of 12 iatrogenic vertebral compression fractures were after SBRT and 4 of 12 after cEBRT; grade 3 adverse fracture effects were isolated to the SBRT cohort (5 of 12). CONCLUSIONS: Risk of local failure and reirradiation is lower with SBRT compared with cEBRT for spinal metastases. Although the iatrogenic vertebral compression fracture rates were within expectations, grade 3 vertebral compression fractures were isolated to the SBRT cohort.

Spine Stereotactic Body Radiotherapy for Prostate Cancer Metastases and the Impact of Hormone Sensitivity Status on Local Control.

BACKGROUND: Stereotactic body radiotherapy (SBRT) is used to deliver ablative dose of radiation to spinal metastases. OBJECTIVE: To report the first dedicated series of spine SBRT specific to prostate cancer (PCa) metastases with outcomes reported according to hormone sensitivity status. METHODS: A prospective database was reviewed identifying patients with PCa treated with spine SBRT. This included those with hormone-sensitive PCa (HSPC) and castrate-resistant PCa (CRPC). The primary end point was MRI-based local control (LC). RESULTS: A total of 183 spine segments in 93 patients were identified; 146 segments had no prior radiation and 37 had been previously radiated; 27 segments were postoperative. The median follow-up was 31 months. At the time of SBRT, 50 patients had HSPC and the remaining 43 had CRPC. The most common fractionation scheme was 24-28 Gy in 2 SBRT fractions (76%). LC rates at 1 and 2 years were 99% and 95% and 94% and 78% for the HSPC and CRPC cohorts, respectively. For patients treated with de novo SBRT, a higher risk of local failure was observed in patients with CRPC (P = .0425). The 1-year and 2-year overall survival rates were significantly longer at 98% and 95% in the HSPC cohort compared with 79% and 65% in the CRPC cohort (P = .0005). The cumulative risk of vertebral compression fracture at 2 years was 10%. CONCLUSION: Favorable LC rates were observed after spine SBRT for PCa metastases; strategies to improve long-term LC in patients with CRPC require further investigation.

Pattern of Recurrence of Glioblastoma Versus Grade 4 IDH-Mutant Astrocytoma Following Chemoradiation: A Retrospective Matched-Cohort Analysis.

Background and Purpose: To quantitatively compare the recurrence patterns of glioblastoma (isocitrate dehydrogenase-wild type) versus grade 4 isocitrate dehydrogenase-mutant astrocytoma (wild type isocitrate dehydrogenase and mutant isocitrate dehydrogenase, respectively) following primary chemoradiation. Materials and Methods: A retrospective matched cohort of 22 wild type isocitrate dehydrogenase and 22 mutant isocitrate dehydrogenase patients were matched by sex, extent of resection, and corpus callosum involvement. The recurrent gross tumor volume was compared to the original gross tumor volume and clinical target volume contours from radiotherapy planning. Failure patterns were quantified by the incidence and volume of the recurrent gross tumor volume outside the gross tumor volume and clinical target volume, and positional differences of the recurrent gross tumor volume centroid from the gross tumor volume and clinical target volume. Results: The gross tumor volume was smaller for wild type isocitrate dehydrogenase patients compared to the mutant isocitrate dehydrogenase cohort (mean ± SD: 46.5 ± 26.0 cm3 vs 72.2 ± 45.4 cm3, P = .026). The recurrent gross tumor volume was 10.7 ± 26.9 cm3 and 46.9 ± 55.0 cm3 smaller than the gross tumor volume for the same groups (P = .018). The recurrent gross tumor volume extended outside the gross tumor volume in 22 (100%) and 15 (68%) (P= .009) of wild type isocitrate dehydrogenase and mutant isocitrate dehydrogenase patients, respectively; however, the volume of recurrent gross tumor volume outside the gross tumor volume was not significantly different (12.4 ± 16.1 cm3 vs 8.4 ± 14.2 cm3, P = .443). The recurrent gross tumor volume centroid was within 5.7 mm of the closest gross tumor volume edge for 21 (95%) and 22 (100%) of wild type isocitrate dehydrogenase and mutant isocitrate dehydrogenase patients, respectively. Conclusion: The recurrent gross tumor volume extended beyond the gross tumor volume less often in mutant isocitrate dehydrogenase patients possibly implying a differential response to chemoradiotherapy and suggesting isocitrate dehydrogenase status might be used to personalize radiotherapy. The results require validation in prospective randomized trials.