Vorinostat, temozolomide or bevacizumab with irradiation and maintenance BEV/TMZ in pediatric high-grade glioma: A Children's Oncology Group Study.

Background: Outcomes for children with high-grade gliomas (HGG) remain poor. This multicenter phase II trial evaluated whether concurrent use of vorinostat or bevacizumab with focal radiotherapy (RT) improved 1-year event-free survival (EFS) compared to temozolomide in children with newly diagnosed HGG who received maintenance temozolomide and bevacizumab. Methods: Patients ≥ 3 and < 22 years with localized, non-brainstem HGG were randomized to receive RT (dose 54-59.4Gy) with vorinostat, temozolomide, or bevacizumab followed by 12 cycles of bevacizumab and temozolomide maintenance therapy. Results: Among 90 patients randomized, the 1-year EFS for concurrent bevacizumab, vorinostat, or temozolomide with RT was 43.8% (±8.8%), 41.4% (±9.2%), and 59.3% (±9.5%), respectively, with no significant difference among treatment arms. Three- and five-year EFS for the entire cohort was 14.8% and 13.4%, respectively, with no significant EFS difference among the chemoradiotherapy arms. IDH mutations were associated with more favorable EFS (P = .03), whereas H3.3 K27M mutations (P = .0045) and alterations in PIK3CA or PTEN (P = .025) were associated with worse outcomes. Patients with telomerase- and alternative lengthening of telomeres (ALT)-negative tumors (n = 4) had an EFS of 100%, significantly greater than those with ALT or telomerase, or both (P = .002). While there was no difference in outcomes based on TERT expression, high TERC expression was associated with inferior survival independent of the telomere maintenance mechanism (P = .0012). Conclusions: Chemoradiotherapy with vorinostat or bevacizumab is not superior to temozolomide in children with newly diagnosed HGG. Patients with telomerase- and ALT-negative tumors had higher EFS suggesting that, if reproduced, mechanism of telomere maintenance should be considered in molecular-risk stratification in future studies.

A Multi-Institutional Survey of Radiation Oncology Professionals' Knowledge, Attitudes, and Practice Behaviors Toward Sexual and Gender Minority Patients With Cancer.

Purpose: Sexual and gender minority (SGM) individuals have an increased risk of poor health outcomes, in part due to knowledge and training gaps in health care education. This study sought to evaluate the knowledge, attitudes, and practice behaviors of various health care role groups within radiation oncology toward SGM patients. Methods and Materials: A 38-item web-based survey was emailed to 1045 staff across 2 large radiation oncology departments. The survey assessed demographics, attitudes, knowledge, and practice behaviors. χ2 tests were performed to explore differences in survey responses by age, political affiliation, religious identity, year since graduation, and role groups. One-way analysis of variance tests were conducted to determine differences between respondents' confidence in knowledge and performance on the knowledge section of the survey. Thematic analysis was applied to the open discussion section. Results: Of the 223 respondents, 103 clinicians (physicians/advanced practice providers/nurses) and 120 nonclinicians (administrative staff, medical assistants, and other nonmedical staff) participated in the survey (21.3% response rate): 72.6% answered the knowledge questions; 93.5% stated they were comfortable treating sexual minorities, or lesbian, gay, bisexual, and queer + patients; 88% indicated comfort in treating transgender patients; 36.6% stated they were confident in their knowledge of the health needs of transgender patients; and 50.3% expressed confidence in treating lesbian, gay, bisexual, and queer + patients. Fewer nonclinicians than clinicians thought that gender identity, sexual orientation, and sex assigned at birth were important to provide the best care (P < .05). The open comments section identified key themes, including the belief that current educational tools are not helpful, desire for more educational formats (lectures, case-based learning, seminars), and an overall interest in SGM health education. Conclusions: Most staff feel comfortable in treating SGM patients but are less confident in the distinct needs of this population. Knowledge gaps persist for both clinicians and nonclinicians, indicating a need for further training specific to oncology care.

Noninvasive Molecular Subtyping of Pediatric Low-Grade Glioma with Self-Supervised Transfer Learning.

Purpose To develop and externally test a scan-to-prediction deep learning pipeline for noninvasive, MRI-based BRAF mutational status classification for pediatric low-grade glioma. Materials and Methods This retrospective study included two pediatric low-grade glioma datasets with linked genomic and diagnostic T2-weighted MRI data of patients: Dana-Farber/Boston Children's Hospital (development dataset, n = 214 [113 (52.8%) male; 104 (48.6%) BRAF wild type, 60 (28.0%) BRAF fusion, and 50 (23.4%) BRAF V600E]) and the Children's Brain Tumor Network (external testing, n = 112 [55 (49.1%) male; 35 (31.2%) BRAF wild type, 60 (53.6%) BRAF fusion, and 17 (15.2%) BRAF V600E]). A deep learning pipeline was developed to classify BRAF mutational status (BRAF wild type vs BRAF fusion vs BRAF V600E) via a two-stage process: (a) three-dimensional tumor segmentation and extraction of axial tumor images and (b) section-wise, deep learning-based classification of mutational status. Knowledge-transfer and self-supervised approaches were investigated to prevent model overfitting, with a primary end point of the area under the receiver operating characteristic curve (AUC). To enhance model interpretability, a novel metric, center of mass distance, was developed to quantify the model attention around the tumor. Results A combination of transfer learning from a pretrained medical imaging-specific network and self-supervised label cross-training (TransferX) coupled with consensus logic yielded the highest classification performance with an AUC of 0.82 (95% CI: 0.72, 0.91), 0.87 (95% CI: 0.61, 0.97), and 0.85 (95% CI: 0.66, 0.95) for BRAF wild type, BRAF fusion, and BRAF V600E, respectively, on internal testing. On external testing, the pipeline yielded an AUC of 0.72 (95% CI: 0.64, 0.86), 0.78 (95% CI: 0.61, 0.89), and 0.72 (95% CI: 0.64, 0.88) for BRAF wild type, BRAF fusion, and BRAF V600E, respectively. Conclusion Transfer learning and self-supervised cross-training improved classification performance and generalizability for noninvasive pediatric low-grade glioma mutational status prediction in a limited data scenario. Keywords: Pediatrics, MRI, CNS, Brain/Brain Stem, Oncology, Feature Detection, Diagnosis, Supervised Learning, Transfer Learning, Convolutional Neural Network (CNN) Supplemental material is available for this article. © RSNA, 2024.

Aurora Kinase A inhibition enhances DNA damage and tumor cell death with 131I-MIBG therapy in high-risk neuroblastoma.

Background: Neuroblastoma is the most common extra-cranial pediatric solid tumor. 131I-metaiodobenzylguanidine (MIBG) is a targeted radiopharmaceutical highly specific for neuroblastoma tumors, providing potent radiotherapy to widely metastatic disease. Aurora kinase A (AURKA) plays a role in mitosis and stabilization of the MYCN protein in neuroblastoma. Here we explore whether AURKA inhibition potentiates a response to MIBG therapy. Results: Using an in vivo model of high-risk neuroblastoma, we demonstrated a marked combinatorial effect of 131I-MIBG and alisertib on tumor growth. In MYCN amplified cell lines, the combination of radiation and an AURKA A inhibitor increased DNA damage and apoptosis and decreased MYCN protein levels. Conclusion: The combination of AURKA inhibition with 131I-MIBG treatment is active in resistant neuroblastoma models and is a promising clinical approach in high-risk neuroblastoma.

Comparison of volumetric and 2D-based response methods in the PNOC-001 pediatric low-grade glioma clinical trial.

Background: Although response in pediatric low-grade glioma (pLGG) includes volumetric assessment, more simplified 2D-based methods are often used in clinical trials. The study's purpose was to compare volumetric to 2D methods. Methods: An expert neuroradiologist performed solid and whole tumor (including cyst and edema) volumetric measurements on MR images using a PACS-based manual segmentation tool in 43 pLGG participants (213 total follow-up images) from the Pacific Pediatric Neuro-Oncology Consortium (PNOC-001) trial. Classification based on changes in volumetric and 2D measurements of solid tumor were compared to neuroradiologist visual response assessment using the Brain Tumor Reporting and Data System (BT-RADS) criteria for a subset of 65 images using receiver operating characteristic (ROC) analysis. Longitudinal modeling of solid tumor volume was used to predict BT-RADS classification in 54 of the 65 images. Results: There was a significant difference in ROC area under the curve between 3D solid tumor volume and 2D area (0.96 vs 0.78, P = .005) and between 3D solid and 3D whole volume (0.96 vs 0.84, P = .006) when classifying BT-RADS progressive disease (PD). Thresholds of 15-25% increase in 3D solid tumor volume had an 80% sensitivity in classifying BT-RADS PD included in their 95% confidence intervals. The longitudinal model of solid volume response had a sensitivity of 82% and a positive predictive value of 67% for detecting BT-RADS PD. Conclusions: Volumetric analysis of solid tumor was significantly better than 2D measurements in classifying tumor progression as determined by BT-RADS criteria and will enable more comprehensive clinical management.

Everolimus for Children With Recurrent or Progressive Low-Grade Glioma: Results From the Phase II PNOC001 Trial.

PURPOSE: The PNOC001 phase II single-arm trial sought to estimate progression-free survival (PFS) associated with everolimus therapy for progressive/recurrent pediatric low-grade glioma (pLGG) on the basis of phosphatidylinositol 3-kinase (PI3K)/AKT/mammalian target of rapamycin (mTOR) pathway activation as measured by phosphorylated-ribosomal protein S6 and to identify prognostic and predictive biomarkers. PATIENTS AND METHODS: Patients, age 3-21 years, with progressive/recurrent pLGG received everolimus orally, 5 mg/m2 once daily. Frequency of driver gene alterations was compared among independent pLGG cohorts of newly diagnosed and progressive/recurrent patients. PFS at 6 months (primary end point) and median PFS (secondary end point) were estimated for association with everolimus therapy. RESULTS: Between 2012 and 2019, 65 subjects with progressive/recurrent pLGG (median age, 9.6 years; range, 3.0-19.9; 46% female) were enrolled, with a median follow-up of 57.5 months. The 6-month PFS was 67.4% (95% CI, 60.0 to 80.0) and median PFS was 11.1 months (95% CI, 7.6 to 19.8). Hypertriglyceridemia was the most common grade ≥3 adverse event. PI3K/AKT/mTOR pathway activation did not correlate with clinical outcomes (6-month PFS, active 68.4% v nonactive 63.3%; median PFS, active 11.2 months v nonactive 11.1 months; P = .80). Rare/novel KIAA1549::BRAF fusion breakpoints were most frequent in supratentorial midline pilocytic astrocytomas, in patients with progressive/recurrent disease, and correlated with poor clinical outcomes (median PFS, rare/novel KIAA1549::BRAF fusion breakpoints 6.1 months v common KIAA1549::BRAF fusion breakpoints 16.7 months; P < .05). Multivariate analysis confirmed their independent risk factor status for disease progression in PNOC001 and other, independent cohorts. Additionally, rare pathogenic germline variants in homologous recombination genes were identified in 6.8% of PNOC001 patients. CONCLUSION: Everolimus is a well-tolerated therapy for progressive/recurrent pLGGs. Rare/novel KIAA1549::BRAF fusion breakpoints may define biomarkers for progressive disease and should be assessed in future clinical trials.

Re-irradiation of recurrent IDH-wildtype glioblastoma in the bevacizumab and immunotherapy era: Target delineation, outcomes and patterns of recurrence.

Introduction and background: While recurrent glioblastoma patients are often treated with re-irradiation, there is limited data on the use of re-irradiation in the setting of bevacizumab (BEV), temozolomide (TMZ) re-challenge, or immune checkpoint inhibition (ICI). We describe target delineation in patients with prior anti-angiogenic therapy, assess safety and efficacy of re-irradiation, and evaluate patterns of recurrence. Materials and methods: Patients with a histologically confirmed diagnosis of glioblastoma treated at a single institution between 2013 and 2021 with re-irradiation were included. Tumor, treatment and clinical data were collected. Logistic and Cox regression analysis were used for statistical analysis. Results: One hundred and seventeen recurrent glioblastoma patients were identified, receiving 129 courses of re-irradiation. In 66 % (85/129) of cases, patients had prior BEV. In the 80 patients (62 %) with available re-irradiation plans, 20 (25 %) had all T2/FLAIR abnormality included in the gross tumor volume (GTV). Median overall survival (OS) for the cohort was 7.3 months, and median progression-free survival (PFS) was 3.6 months. Acute CTCAE grade ≥ 3 toxicity occurred in 8 % of cases. Concurrent use of TMZ or ICI was not associated with improved OS nor PFS. On multivariable analysis, higher KPS was significantly associated with longer OS (p < 0.01). On subgroup analysis, patients with prior BEV had significantly more marginal recurrences than those without (26 % vs. 13 %, p < 0.01). Conclusion: Re-irradiation can be safely employed in recurrent glioblastoma patients. Marginal recurrence was more frequent in patients with prior BEV, suggesting a need to consider more inclusive treatment volumes incorporating T2/FLAIR abnormality.

Genetic architecture of the acute and persistent immune cell response after radiation exposure.

Hematologic toxicity is a common side effect of multimodal cancer therapy. Nearly all animal studies investigating the causes of radiotherapy-induced hematologic toxicity use inbred strains with limited genetic diversity and do not reflect the diverse responses observed in humans. We used the population-based Collaborative Cross (CC) mouse resource to investigate the genetic architecture of the acute and persistent immune response after radiation exposure by measuring 22 immune parameters in 1,720 CC mice representing 35 strains. We determined relative acute and persistent radiation resistance scores at the individual strain level considering contributions from all immune parameters. Genome-wide association analysis identified quantitative trait loci associated with baseline and radiation responses. A cross-species radiation resistance score predicted recurrence-free survival in medulloblastoma patients. We present a community resource of immune parameters and genome-wide association analyses before and after radiation exposure for future investigations of the contributions of host genetics on radiosensitivity.

Automated temporalis muscle quantification and growth charts for children through adulthood.

Lean muscle mass (LMM) is an important aspect of human health. Temporalis muscle thickness is a promising LMM marker but has had limited utility due to its unknown normal growth trajectory and reference ranges and lack of standardized measurement. Here, we develop an automated deep learning pipeline to accurately measure temporalis muscle thickness (iTMT) from routine brain magnetic resonance imaging (MRI). We apply iTMT to 23,876 MRIs of healthy subjects, ages 4 through 35, and generate sex-specific iTMT normal growth charts with percentiles. We find that iTMT was associated with specific physiologic traits, including caloric intake, physical activity, sex hormone levels, and presence of malignancy. We validate iTMT across multiple demographic groups and in children with brain tumors and demonstrate feasibility for individualized longitudinal monitoring. The iTMT pipeline provides unprecedented insights into temporalis muscle growth during human development and enables the use of LMM tracking to inform clinical decision-making.

Clinical outcomes of radiation therapy for transgender and gender-expansive people with cancer.

Introduction: Approximately 1.6 million people in the US identify as transgender, many of whom undergo gender-affirming medical or surgical therapies. While transgender individuals are diagnosed with cancer at similar rates as those who are cisgender, the impacts of radiation therapy on outcomes of gender-affirming care in transgender, nonbinary, and gender-expansive people with cancer are understudied. We report on the experiences and outcomes of transgender and gender-expansive patients receiving radiation therapy for cancer treatment. Methods: This study is a multi-institutional retrospective review of patients evaluated from 2005-2019 identified as transgender or gender-expansive in the medical record and treated with radiation therapy. Results: We identified 23 patients who received radiation to 32 sites, including 12 (38%) to the brain, head, or neck, 8 (25%) to the thorax, and 7 (22%) to the pelvis. Seventeen patients (74%) received gender-affirming hormone therapy and 13 patients (57%) underwent gender-affirming surgery. Four patients had pelvic radiation before or after gender-affirming pelvic surgery, including two trans women who had pelvic radiation after vaginoplasty. Four patients had radiation to the chest or thorax and gender-affirming chest or breast surgery, including two trans men with breast cancer. Two pediatric patients developed hypopituitarism and hypogonadism secondary to radiation therapy and, as adults, changed their hormone replacement therapy to affirm their transgender identities. Discussion: Transgender people with cancer undergo radiation therapy for a wide range of cancers. Understanding their prior gender-affirming medical or surgical treatments and future gender affirmation goals may identify important considerations for their oncologic care.

Inaugural Results of the Individualized Screening Trial of Innovative Glioblastoma Therapy: A Phase II Platform Trial for Newly Diagnosed Glioblastoma Using Bayesian Adaptive Randomization.

PURPOSE: The Individualized Screening Trial of Innovative Glioblastoma Therapy (INSIGhT) is a phase II platform trial that uses response adaptive randomization and genomic profiling to efficiently identify novel therapies for phase III testing. Three initial experimental arms (abemaciclib [a cyclin-dependent kinase [CDK]4/6 inhibitor], neratinib [an epidermal growth factor receptor [EGFR]/human epidermal growth factor receptor 2 inhibitor], and CC-115 [a deoxyribonucleic acid-dependent protein kinase/mammalian target of rapamycin inhibitor]) were simultaneously evaluated against a common control arm. We report the results for each arm and examine the feasibility and conduct of the adaptive platform design. PATIENTS AND METHODS: Patients with newly diagnosed O6-methylguanine-DNA methyltransferase-unmethylated glioblastoma were eligible if they had tumor genotyping to identify prespecified biomarker subpopulations of dominant glioblastoma signaling pathways (EGFR, phosphatidylinositol 3-kinase, and CDK). Initial random assignment was 1:1:1:1 between control (radiation therapy and temozolomide) and the experimental arms. Subsequent Bayesian adaptive randomization was incorporated on the basis of biomarker-specific progression-free survival (PFS) data. The primary end point was overall survival (OS), and one-sided P values are reported. The trial is registered with ClinicalTrials.gov (identifier: NCT02977780). RESULTS: Two hundred thirty-seven patients were treated (71 control; 73 abemaciclib; 81 neratinib; 12 CC-115) in years 2017-2021. Abemaciclib and neratinib were well tolerated, but CC-115 was associated with ≥ grade 3 treatment-related toxicity in 58% of patients. PFS was significantly longer with abemaciclib (hazard ratio [HR], 0.72; 95% CI, 0.49 to 1.06; one-sided P = .046) and neratinib (HR, 0.72; 95% CI, 0.50 to 1.02; one-sided P = .033) relative to the control arm but there was no PFS benefit with CC-115 (one-sided P = .523). None of the experimental therapies demonstrated a significant OS benefit (P > .05). CONCLUSION: The INSIGhT design enabled efficient simultaneous testing of three experimental agents using a shared control arm and adaptive randomization. Two investigational arms had superior PFS compared with the control arm, but none demonstrated an OS benefit. The INSIGhT design may promote improved and more efficient therapeutic discovery in glioblastoma. New arms have been added to the trial.

Radiotherapy for Primary Pediatric Central Nervous System Malignancies: Current Treatment Paradigms and Future Directions.

BACKGROUND: Central nervous system tumors are the most common solid tumors in childhood. Treatment paradigms for pediatric central nervous system malignancies depend on elements including tumor histology, age of patient, and stage of disease. Radiotherapy is an important modality of treatment for many pediatric central nervous system malignancies. SUMMARY: While radiation contributes to excellent overall survival rates for many patients, radiation also carries significant risks of long-term side effects including neurocognitive decline, hearing loss, growth impairment, neuroendocrine dysfunction, strokes, and secondary malignancies. In recent decades, clinical trials have demonstrated that with better imaging and staging along with more sophisticated radiation planning and treatment set-up verification, smaller treatment volumes can be utilized without decrement in survival. Furthermore, the development of intensity-modulated radiotherapy and proton-beam radiotherapy has greatly improved conformality of radiation. KEY MESSAGES: Recent changes in radiation treatment paradigms have decreased risks of short- and long-term toxicity for common histologies and in different age groups. Future studies will continue to develop novel radiation regimens to improve outcomes in aggressive central nervous system tumors, integrate molecular subtypes to tailor radiation treatment, and decrease radiation-associated toxicity for long-term survivors.

Noninvasive molecular subtyping of pediatric low-grade glioma with self-supervised transfer learning.

Purpose: To develop and externally validate a scan-to-prediction deep-learning pipeline for noninvasive, MRI-based BRAF mutational status classification for pLGG. Materials and Methods: We conducted a retrospective study of two pLGG datasets with linked genomic and diagnostic T2-weighted MRI of patients: BCH (development dataset, n=214 [60 (28%) BRAF fusion, 50 (23%) BRAF V600E, 104 (49%) wild-type), and Child Brain Tumor Network (CBTN) (external validation, n=112 [60 (53%) BRAF-Fusion, 17 (15%) BRAF-V600E, 35 (32%) wild-type]). We developed a deep learning pipeline to classify BRAF mutational status (V600E vs. fusion vs. wildtype) via a two-stage process: 1) 3D tumor segmentation and extraction of axial tumor images, and 2) slice-wise, deep learning-based classification of mutational status. We investigated knowledge-transfer and self-supervised approaches to prevent model overfitting with a primary endpoint of the area under the receiver operating characteristic curve (AUC). To enhance model interpretability, we developed a novel metric, COMDist, that quantifies the accuracy of model attention around the tumor. Results: A combination of transfer learning from a pretrained medical imaging-specific network and self-supervised label cross-training (TransferX) coupled with consensus logic yielded the highest macro-average AUC (0.82 [95% CI: 0.70-0.90]) and accuracy (77%) on internal validation, with an AUC improvement of +17.7% and a COMDist improvement of +6.4% versus training from scratch. On external validation, the TransferX model yielded AUC (0.73 [95% CI 0.68-0.88]) and accuracy (75%). Conclusion: Transfer learning and self-supervised cross-training improved classification performance and generalizability for noninvasive pLGG mutational status prediction in a limited data scenario.

Children's Oncology Group's 2023 blueprint for research: Central nervous system tumors.

Tumors of the central nervous system (CNS) are a leading cause of morbidity and mortality in the pediatric population. Molecular characterization in the last decade has redefined CNS tumor diagnoses and risk stratification; confirmed the unique biology of pediatric tumors as distinct entities from tumors that occur in adulthood; and led to the first novel targeted therapies receiving Food and Drug Administration (FDA) approval for children with CNS tumors. There remain significant challenges to overcome: children with unresectable low-grade glioma may require multiple prolonged courses of therapy affecting quality of life; children with high-grade glioma have a dismal long-term prognosis; children with medulloblastoma may suffer significant short- and long-term morbidity from multimodal cytotoxic therapy, and approaches to improve survival in ependymoma remain elusive. The Children's Oncology Group (COG) is uniquely positioned to conduct the next generation of practice-changing clinical trials through rapid prospective molecular characterization and therapy evaluation in well-defined clinical and molecular groups.

Multi-institutional Characterization of Outcomes for Pediatric and Young Adult Patients With High-Risk Myxopapillary Ependymoma After Radiation Therapy.

PURPOSE: Myxopapillary ependymoma (MPE) is a rare, typically slow-growing subtype of spinal ependymomas. There are no standard guidelines for radiotherapy and long-term outcomes after radiation, particularly patterns of relapse, for pediatric and young adult (YA) patients with MPE remain under-characterized. METHODS AND MATERIALS: This is an Institutional Review Board-approved multi-institutional retrospective cohort study of 60 pediatric and YA patients diagnosed with MPE and received radiotherapy between 2000-2020. Clinical and treatment characteristics, and long-term outcomes were recorded. Site(s) of progression was compared to radiation fields. Survival outcomes were analyzed using Kaplan-Meier method. Cumulative incidence of local in-field progression (CILP) after initial radiotherapy was analyzed using Gray's method with out-of-field-only progression as a competing risk. Univariate analyses were performed using Cox proportional hazard's model. RESULTS: The median age at radiation was 14.8 years (range: 7.1-26.5). At time of radiotherapy, 45 (75.0%) and 35 (58.3%) patients had gross residual and multifocal disease, respectively. Forty-eight (80.0%), seven (11.7%) and five (8.3%) patients received involved field radiotherapy, craniospinal irradiation, and whole spine radiation, respectively. Median follow-up from end of radiotherapy was 6.2 years (range: 0.6-21.0). Five-year overall survival, progression-free survival, and CILP were 100%, 60.8% and 4.1%, respectively. Both local recurrences were at sites of gross residual disease. Of the eighteen out-of-field first recurrences after radiotherapy, all were superior to the initial treatment field and nine had intracranial relapse. On univariate analyses, distant-only recurrence before radiation (HR: 4.00, 95% CI: 1.54-10.43, p = 0.005) was significantly associated with shorter time to progression. CONCLUSIONS: While the risk of recurrence within the radiation field is low, pediatric and YA patients with high-risk MPE remain at risk for recurrences in the spine above the radiation field and intracranially after radiotherapy. Future prospective studies are needed to investigate the appropriate radiation field and dose based on the extent of metastases.

Expert-level pediatric brain tumor segmentation in a limited data scenario with stepwise transfer learning.

Purpose: Artificial intelligence (AI)-automated tumor delineation for pediatric gliomas would enable real-time volumetric evaluation to support diagnosis, treatment response assessment, and clinical decision-making. Auto-segmentation algorithms for pediatric tumors are rare, due to limited data availability, and algorithms have yet to demonstrate clinical translation. Methods: We leveraged two datasets from a national brain tumor consortium (n=184) and a pediatric cancer center (n=100) to develop, externally validate, and clinically benchmark deep learning neural networks for pediatric low-grade glioma (pLGG) segmentation using a novel in-domain, stepwise transfer learning approach. The best model [via Dice similarity coefficient (DSC)] was externally validated and subject to randomized, blinded evaluation by three expert clinicians wherein clinicians assessed clinical acceptability of expert- and AI-generated segmentations via 10-point Likert scales and Turing tests. Results: The best AI model utilized in-domain, stepwise transfer learning (median DSC: 0.877 [IQR 0.715-0.914]) versus baseline model (median DSC 0.812 [IQR 0.559-0.888]; p<0.05). On external testing (n=60), the AI model yielded accuracy comparable to inter-expert agreement (median DSC: 0.834 [IQR 0.726-0.901] vs. 0.861 [IQR 0.795-0.905], p=0.13). On clinical benchmarking (n=100 scans, 300 segmentations from 3 experts), the experts rated the AI model higher on average compared to other experts (median Likert rating: 9 [IQR 7-9]) vs. 7 [IQR 7-9], p<0.05 for each). Additionally, the AI segmentations had significantly higher (p<0.05) overall acceptability compared to experts on average (80.2% vs. 65.4%). Experts correctly predicted the origins of AI segmentations in an average of 26.0% of cases. Conclusions: Stepwise transfer learning enabled expert-level, automated pediatric brain tumor auto-segmentation and volumetric measurement with a high level of clinical acceptability. This approach may enable development and translation of AI imaging segmentation algorithms in limited data scenarios.

Children's Oncology Group's 2023 blueprint for research: Radiation oncology.

Radiation oncology is an integral part of the multidisciplinary team caring for children with cancer. The primary goal of our committee is to enable the delivery of the safest dose of radiation therapy (RT) with the maximal potential for cure, and to minimize toxicity in children by delivering lower doses to normal tissues using advanced technologies like intensity-modulated RT (IMRT) and proton therapy. We provide mentorship for y ators and are actively involved in educating the global radiation oncology community. We are leaders in the effort to discover novel radiosensitizers, radioprotectors, and advanced RT technologies that could help improve outcomes of children with cancer.

Serum miRNA-based signature indicates radiation exposure and dose in humans: A multicenter diagnostic biomarker study.

PURPOSE: Mouse and non-human primate models showed that serum miRNAs may be used to predict the biological impact of radiation doses. We hypothesized that these results can be translated to humans treated with total body irradiation (TBI), and that miRNAs may be used as clinically feasible biodosimeters. METHODS: To test this hypothesis, serial serum samples were obtained from 25 patients (pediatric and adults) who underwent allogeneic stem-cell transplantation and profiled for miRNA expression using next-generation sequencing. miRNAs with diagnostic potential were quantified with qPCR and used to build logistic regression models with lasso penalty to reduce overfitting, identifying samples drawn from patients who underwent total body irradiation to a potentially lethal dose. RESULTS: Differential expression results were consistent with previous studies in mice and non-human primates. miRNAs with detectable expression in this and two prior animal sets allowed for distinction of the irradiated from non-irradiated samples in mice, macaques and humans, validating the miRNAs as radiation-responsive through evolutionarily conserved transcriptional regulation mechanisms. Finally, we created a model based on the expression of miR-150-5p, miR-30b-5p and miR-320c normalized to two references and adjusted for patient age with an AUC of 0.9 (95%CI:0.83-0.97) for identifying samples drawn after irradiation; a separate model differentiating between high and low radiation dose achieved AUC of 0.85 (95%CI: 0.74-0.96). CONCLUSIONS: We conclude that serum miRNAs reflect radiation exposure and dose for humans undergoing TBI and may be used as functional biodosimeters for precise identification of people exposed to clinically significant radiation doses.

A genomic score to predict local control among patients with brain metastases managed with radiation.

BACKGROUND: Clinical predictors of local recurrence following radiation among patients with brain metastases (BrM) provide limited explanatory power. We developed a DNA-based signature of radiotherapeutic efficacy among patients with BrM to better characterize recurrence risk. METHODS: We identified 570 patients with 1487 BrM managed with whole-brain (WBRT) or stereotactic radiation therapy at Brigham and Women's Hospital/Dana-Farber Cancer Institute (2013-2020) for whom next-generation sequencing panel data (OncoPanel) were available. Fine/Gray's competing risks regression was utilized to compare local recurrence on a per-metastasis level among patients with versus without somatic alterations of likely biological significance across 84 genes. Genes with a q-value ≤ 0.10 were utilized to develop a "Brain-Radiation Prediction Score" ("Brain-RPS"). RESULTS: Genomic alterations in 11 (ATM, MYCL, PALB2, FAS, PRDM1, PAX5, CDKN1B, EZH2, NBN, DIS3, and MDM4) and 2 genes (FBXW7 and AURKA) were associated with decreased or increased risk of local recurrence, respectively (q-value ≤ 0.10). Weighted scores corresponding to the strength of association with local failure for each gene were summed to calculate a patient-level RPS. On multivariable Fine/Gray's competing risks regression, RPS [1.66 (1.44-1.91, P < .001)], metastasis-associated edema [1.60 (1.16-2.21), P = .004], baseline size [1.02 (1.01-1.03), P < .001] and receipt of WBRT without local therapy [4.04 (2.49-6.58), P < .001] were independent predictors of local failure. CONCLUSIONS: We developed a genomic score to quantify local recurrence risk following brain-directed radiation. To the best of our knowledge, this represents the first study to systematically correlate DNA-based alterations with radiotherapeutic outcomes in BrM. If validated, Brain-RPS has potential to facilitate clinical trials aimed at genome-based personalization of radiation in BrM.

Feasibility of the Audio-Visual Assisted Therapeutic Ambience in Radiotherapy (AVATAR) System for Anesthesia Avoidance in Pediatric Patients: A Multicenter Trial.

PURPOSE: The Audio-Visual Assisted Therapeutic Ambience in Radiotherapy (AVATAR) system was the first published radiation therapy (RT)-compatible system to reduce the need for pediatric anesthesia through video-based distraction. We evaluated the feasibility of AVATAR implementation and effects on anesthesia use, quality of life, and anxiety in a multicenter pediatric trial. METHODS AND MATERIALS: Pediatric patients 3 to 10 years of age preparing to undergo RT at 10 institutions were prospectively enrolled. Children able to undergo at least 1 fraction of RT using AVATAR without anesthesia were considered successful (S). Patients requiring anesthesia for their entire treatment course were nonsuccessful (NS). The PedsQL3.0 Cancer Module (PedsQL) survey assessed quality of life and was administered to the patient and guardian at RT simulation, midway through RT, and at final treatment. The modified Yale Preoperative Anxiety Scale (mYPAS) assessed anxiety and was performed at the same 3 time points. Success was evaluated using the χ2 test. PedsQL and mYPAS scores were assessed using mixed effects models with time points evaluated as fixed effects and a random intercept on the subject. RESULTS: Eighty-one children were included; median age was 7 years. AVATAR was successful at all 10 institutions and with photon and proton RT. There were 63 (78%) S patients; anesthesia was avoided for a median of 20 fractions per patient. Success differed by age (P = .04) and private versus public insurance (P < .001). Both patient (P = .008) and parent (P = .006) PedsQL scores significantly improved over the course of RT for patients aged 5 to 7. Anxiety in the treatment room decreased for both S and NS patients over RT course (P < .001), by age (P < .001), and by S versus NS patients (P < .001). CONCLUSIONS: In this 10-center prospective trial, anesthesia avoidance with AVATAR was 78% in children aged 3 to 10 years, higher than among age-matched historical controls (49%; P < .001). AVATAR implementation is feasible across multiple institutions and should be further studied and made available to patients who may benefit from video-based distraction.