Artificial intelligence (AI) applied to brain magnetic resonance imaging (MRI) has the potential to improve disease diagnosis and management but requires algorithms with generalizable knowledge that can perform well in a variety of clinical scenarios. The field has been constrained, thus far, by limited training data and task-specific models that do not generalize well across patient populations and medical tasks. Foundation models, by leveraging self-supervised learning, pretraining, and targeted adaptation, present a promising paradigm to overcome these limitations.
View Article and Find Full Text PDFCentral nervous system neuroblastoma with FOXR2 activation (NB-FOXR2) is a high-grade tumor of the brain hemispheres and a newly identified molecular entity. Tumors express dual neuronal and glial markers, leading to frequent misdiagnoses, and limited information exists on the role of FOXR2 in their genesis. To identify their cellular origins, we profiled the transcriptomes of NB-FOXR2 tumors at the bulk and single-cell levels and integrated these profiles with large single-cell references of the normal brain.
View Article and Find Full Text PDFBackground: Postoperative recurrence risk for pediatric low-grade gliomas (pLGGs) is challenging to predict by conventional clinical, radiographic, and genomic factors. We investigated if deep learning of MRI tumor features could improve postoperative pLGG risk stratification.
Methods: We used pre-trained deep learning (DL) tool designed for pLGG segmentation to extract pLGG imaging features from preoperative T2-weighted MRI from patients who underwent surgery (DL-MRI features).
Pediatric glioma recurrence can cause morbidity and mortality; however, recurrence pattern and severity are heterogeneous and challenging to predict with established clinical and genomic markers. Resultingly, almost all children undergo frequent, long-term, magnetic resonance (MR) brain surveillance regardless of individual recurrence risk. Deep learning analysis of longitudinal MR may be an effective approach for improving individualized recurrence prediction in gliomas and other cancers but has thus far been infeasible with current frameworks.
View Article and Find Full Text PDFPediatric low-grade glioma (pLGG) is the most common childhood brain tumor group. The natural history, when curative resection is not possible, is one of a chronic disease with periods of tumor stability and episodes of tumor progression. While there is a high overall survival rate, many patients experience significant and potentially lifelong morbidities.
View Article and Find Full Text PDFPurpose To develop and externally test a scan-to-prediction deep learning pipeline for noninvasive, MRI-based 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, = 214 [113 (52.8%) male; 104 (48.
View Article and Find Full Text PDFBackground: Cellular senescence can have positive and negative effects on the body, including aiding in damage repair and facilitating tumor growth. Adamantinomatous craniopharyngioma (ACP), the most common pediatric sellar/suprasellar brain tumor, poses significant treatment challenges. Recent studies suggest that senescent cells in ACP tumors may contribute to tumor growth and invasion by releasing a senesecence-associated secretory phenotype.
View Article and Find Full Text PDFA hallmark of high-risk childhood medulloblastoma is the dysregulation of RNA translation. Currently, it is unknown whether medulloblastoma dysregulates the translation of putatively oncogenic non-canonical open reading frames (ORFs). To address this question, we performed ribosome profiling of 32 medulloblastoma tissues and cell lines and observed widespread non-canonical ORF translation.
View Article and Find Full Text PDFWithin the last few decades, we have witnessed tremendous advancements in the study of pediatric low-grade gliomas (pLGG), leading to a much-improved understanding of their molecular underpinnings. Consequently, we have achieved successful milestones in developing and implementing targeted therapeutic agents for treating these tumors. However, the community continues to face many unknowns when it comes to the most effective clinical implementation of these novel targeted inhibitors or combinations thereof.
View Article and Find Full Text PDFBackground: Recurrent brain tumors are the leading cause of cancer death in children. Indoleamine 2,3-dioxygenase (IDO) is a targetable metabolic checkpoint that, in preclinical models, inhibits anti-tumor immunity following chemotherapy.
Methods: We conducted a phase I trial (NCT02502708) of the oral IDO-pathway inhibitor indoximod in children with recurrent brain tumors or newly diagnosed diffuse intrinsic pontine glioma (DIPG).
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.
Purpose: Treatment of wingless (WNT)-activated medulloblastoma (WNT+MB) with surgery, irradiation (XRT), and chemotherapy results in excellent outcomes. We studied the efficacy of therapy de-intensification by omitting XRT entirely in children with WNT+MB.
Patients And Methods: Tumors were molecularly screened to confirm the diagnosis of WNT+MB.
Spheroid culture systems have allowed in vitro propagation of cells unable to grow in canonical cell culturing conditions, and may capture cellular contexts that model tumor growth better than current model systems. The insights gleaned from genome-wide clustered regularly interspaced short palindromic repeat (CRISPR) screening of thousands of cancer cell lines grown in conventional culture conditions illustrate the value of such CRISPR pooled screens. It is clear that similar genome-wide CRISPR screens of three-dimensional spheroid cultures will be important for future biological discovery.
View Article and Find Full Text PDFA hallmark of high-risk childhood medulloblastoma is the dysregulation of RNA translation. Currently, it is unknown whether medulloblastoma dysregulates the translation of putatively oncogenic non-canonical open reading frames. To address this question, we performed ribosome profiling of 32 medulloblastoma tissues and cell lines and observed widespread non-canonical ORF translation.
View Article and Find Full Text PDFBackground: Pediatric low-grade gliomas (pLGGs) are the most common central nervous system tumor in children, characterized by RAS/MAPK pathway driver alterations. Genomic advances have facilitated the use of molecular targeted therapies, however, their long-term impact on tumor behavior remains critically unanswered.
Methods: We performed an IRB-approved, retrospective chart and imaging review of pLGGs treated with off-label targeted therapy at Dana-Farber/Boston Children's from 2010 to 2020.
Purpose: Anaplastic lymphoma kinase (ALK) aberrations have been identified in pediatric-type infant gliomas, but their occurrence across age groups, functional effects, and treatment response has not been broadly established.
Experimental Design: We performed a comprehensive analysis of ALK expression and genomic aberrations in both newly generated and retrospective data from 371 glioblastomas (156 adult, 205 infant/pediatric, and 10 congenital) with in vitro and in vivo validation of aberrations.
Results: ALK aberrations at the protein or genomic level were detected in 12% of gliomas (45/371) in a wide age range (0-80 years).