Faculty Bibliography

Medulloblastoma (MB) is the most common childhood malignant brain tumor and is a leading cause of cancer-related death in children. DNA methylation profiling has rapidly advanced our understanding of MB pathogenesis at the molecular level, but assessments in Saudi Arabian (SA)-MB cases are sparse. MBs can be sub-grouped according to methylation patterns from FPPE samples into Wingless (WNT-MB), Sonic Hedgehog (SHH-MB), Group 3 (G3), and Group 4 (G4) tumors. The WNT-MB and SHH-MB subgroups are characterized by gain-of function mutations that activate oncogenic cell signaling, whilst G3/G4 tumors show recurrent chromosomal alterations. Given that each subgroup has distinct clinical outcomes, the ability to subgroup SA-FPPE samples holds significant prognostic and therapeutic value. Here, we performed the first assessment of MB-DNA methylation patterns in an SA cohort using archival biopsy material (FPPE n = 49). Of the 41 materials available for methylation assessments, 39 could be classified into the major DNA methylation subgroups (SHH, WNT, G3, and G4). Furthermore, methylation analysis was able to reclassify tumors that could not be sub-grouped through next-generation sequencing, highlighting its superior accuracy for MB molecular classifications. Independent assessments demonstrated known clinical relationships of the subgroups, exemplified by the high survival rates observed for WNT tumors. Surprisingly, the G4 subgroup did not conform to previously identified phenotypes, with a high prevalence in females, high metastatic rates, and a large number of tumor-associated deaths. Taking our results together, we demonstrate that DNA methylation profiling enables the robust sub-classification of four disease sub-groups in archival FFPE biopsy material from SA-MB patients. Moreover, we show that the incorporation of DNA methylation biomarkers can significantly improve current disease-risk stratification schemes, particularly concerning the identification of aggressive G4 tumors. These findings have important implications for future clinical disease management in MB cases across the Arab world.

Infant high grade gliomas appear clinically distinct from their counterparts in older children, indicating that histopathologic grading may not accurately reflect the biology of these tumors. We have collected 241 cases under 4 years of age, and carried out histological review, methylation profiling, custom panel and genome/exome sequencing. After excluding tumors representing other established entities or subgroups, we identified 130 cases to be part of an 'intrinsic' spectrum of disease specific to the infant population. These included those with targetable MAP-kinase alterations, and a large proportion of remaining cases harboring gene fusions targeting ALK (n=31), NTRK1/2/3 (n=21), ROS1 (n=9) and MET (n=4) as their driving alterations, with evidence of efficacy of targeted agents in the clinic. These data strongly supports the concept that infant gliomas require a change in diagnostic practice and management.

BACKGROUND:Both genetic and methylation analysis have been shown to provide insight into the diagnosis and prognosis of many brain tumors. However, the implication of methylation profiling and its interaction with genetic alterations in pediatric low-grade gliomas (PLGGs) are unclear. METHODS:We performed a comprehensive analysis of PLGG with long term clinical follow up. In total 152 PLGGs were analyzed from a range of pathological subtypes including 40 gangliogliomas. Complete molecular analysis was compared to genome-wide methylation data and outcome in all patients. For further analysis of specific PLGG groups, including BRAF p.V600E mutant gliomas, we compiled an additional cohort of clinically and genetically defined tumors from 3 large centers. RESULTS:Unsupervised hierarchical clustering revealed 5 novel subgroups of PLGG. These were dominated by non-neoplastic factors such as tumor location and lymphocytic infiltration. Midline PLGG clustered while deep hemispheric lesions differed from lesions in the periphery. Mutations were distributed throughout these location-driven clusters of PLGG. A novel methylation cluster suggesting high lymphocyte infiltration was confirmed pathologically and exhibited worse progression-free survival compared to PLGG harboring similar molecular alterations (p = 0.008, multivariate analysis: p = 0.035). Although the current methylation classifier revealed low confidence in 44% of cases and failed to add information in most PLGG it was helpful in reclassifying rare cases. The addition of histopathological and molecular information to specific methylation subgroups such as pleomorphic xanthoastrocytoma (PXA)-like tumors could stratify these tumors into low and high risk (p = 0.0014). CONCLUSION/CONCLUSIONS:The PLGG methylome is affected by multiple non-neoplastic factors. Combined molecular and pathological analysis is key to provide additional information when methylation classification is used for PLGG in the clinical setting.

Of the four primary subgroups of medulloblastoma, the most frequent cytogenetic abnormality, i17q, distinguishes Groups 3 and 4 which carry the highest mortality; haploinsufficiency of 17p13.3 is a marker for particularly poor prognosis. At the terminal end of this locus lies miR-1253, a brain-enriched microRNA that regulates bone morphogenic proteins during cerebellar development. We hypothesized miR-1253 confers novel tumor-suppressive properties in medulloblastoma. Using two different cohorts of medulloblastoma samples, we first studied the expression and methylation profiles of miR-1253. We then explored the anti-tumorigenic properties of miR-1253 in parallel with a biochemical analysis of apoptosis and proliferation, and isolation of oncogenic targets using high-throughput screening. Deregulation of miR-1253 expression was noted, both in medulloblastoma clinical samples and cell lines, by epigenetic silencing via hypermethylation; specific de-methylation of miR-1253 not only resulted in rapid recovery of expression but also a sharp decline in tumor cell proliferation and target gene expression. Expression restoration also led to a reduction in tumor cell virulence, concomitant with activation of apoptotic pathways, cell cycle arrest, and reduction of markers of proliferation. We identified two oncogenic targets of miR-1253, CDK6 and CD276, whose silencing replicated the negative trophic effects of miR-1253. These data reveal novel tumor-suppressive properties for miR-1253, i.e. 1) loss of expression via epigenetic silencing, 2) negative trophic effects on tumor aggressiveness, and 3) downregulation of oncogenic targets.

Gene fusions are caused by chromosomal rearrangements and encode fusion proteins that can act as oncogenic drivers in cancers. Traditional methods for detecting oncogenic fusion transcripts include fluorescence in situ hybridization (FISH), reverse transcription polymerase chain reaction (RT-PCR) and immunohistochemistry (IHC). However, these methods are limited in scalability and pose significant technical and interpretational challenges. Next-generation sequencing (NGS) is a high-throughput method for detecting genetic abnormalities and providing prognostic and therapeutic information for cancer patients. We present our experience with the validation of a custom-designed Archer Anchored Multiplex PCR (AMP™) technology-based NGS technology, "NYU FUSION-SEQer" using RNA sequencing. We examine both analytical performance and clinical utility of the panel using 75 retrospective validation samples and 84 prospective clinical samples of solid tumors. Our panel showed robust sequencing performance with strong enrichment for target regions. The lower limit of detection was 12.5% tumor fraction at 125 ng of RNA input. The panel demonstrated excellent analytic accuracy, with 100% sensitivity, 100% specificity and 100% reproducibility on validation samples. Finally, in the prospective cohort, the panel detected fusions in 61% cases (n = 51), out of which 41% (n = 21) enabling diagnosis and 59% (n = 30) enabling treatment and prognosis. We demonstrate that the fusion panel can accurately, efficiently and cost-effectively detect the majority of known fusion genes, novel clinically relevant fusions and provides an excellent tool for discovery of new fusion genes in solid tumors.

Sudden unexplained death in childhood (SUDC) affects children >1-year-old whose cause of death remains unexplained following comprehensive case investigation and is often associated with hippocampal abnormalities. We prospectively performed systematic neuropathologic investigation in 20 SUDC cases, including (i) autopsy data and comprehensive ancillary testing, including molecular studies, (ii) ex vivo 3T MRI and extensive histologic brain samples, and (iii) blinded neuropathology review by 2 board-certified neuropathologists. There were 12 girls and 8 boys; median age at death was 33.3 months. Twelve had a history of febrile seizures, 85% died during apparent sleep and 80% in prone position. Molecular testing possibly explained 3 deaths and identified genetic mutations in TNNI3, RYR2, and multiple chromosomal aberrations. Hippocampal abnormalities most often affected the dentate gyrus (altered thickness, irregular configuration, and focal lack of granule cells), and had highest concordance between reviewers. Findings were identified with similar frequencies in cases with and without molecular findings. Number of seizures did not correlate with hippocampal findings. Hippocampal alterations were the most common finding on histological review but were also found in possibly explained deaths. The significance and specificity of hippocampal findings is unclear as they may result from seizures, contribute to seizure pathogenesis, or be an unrelated phenomenon.

AIMS/OBJECTIVE:DNA methylation-based central nervous system (CNS) tumour classification has identified numerous molecularly-distinct tumour types, and clinically relevant subgroups among known CNS tumour entities that were previously thought to represent homogeneous diseases. Our study aimed at characterising a novel, molecularly-defined variant of glioneuronal CNS tumour. PATIENTS AND METHODS/METHODS:DNA methylation profiling was performed using the Infinium MethylationEPIC or 450k BeadChip arrays (Illumina) and analysed using the 'conumee' package in R computing environment. Additional gene panel sequencing was also performed. Tumour samples were collected at the German Cancer Research Centre (DKFZ) and provided by multi-national collaborators. Histological sections were also collected and independently reviewed. RESULTS:Genome-wide DNA methylation data from >25,000 CNS tumours were screened for clusters separated from established DNA methylation classes, revealing a novel group comprising 31 tumours, mainly found in paediatric patients. This DNA methylation-defined variant of low-grade CNS tumours with glioneuronal differentiation displays recurrent monosomy 14, nuclear clusters within a morphology that is otherwise reminiscent of oligodendroglioma and other established entities with clear-cell histology, and a lack of genetic alterations commonly observed in other (paediatric) glioneuronal entities. CONCLUSIONS:DNA methylation-based tumour classification is an objective method of assessing tumour origins, which may aid in diagnosis, especially for atypical cases. With increasing sample size, methylation analysis allows for the identification of rare, putative new tumour entities, which are currently not recognized by the WHO classification. Our study revealed the existence of a DNA methylation-defined class of low-grade glioneuronal tumours with recurrent monosomy 14, oligodendroglioma-like features and nuclear clusters.

Intraoperative diagnosis is essential for providing safe and effective care during cancer surgery¹. The existing workflow for intraoperative diagnosis based on hematoxylin and eosin staining of processed tissue is time, resource and labor intensive^2,3. Moreover, interpretation of intraoperative histologic images is dependent on a contracting, unevenly distributed, pathology workforce⁴. In the present study, we report a parallel workflow that combines stimulated Raman histology (SRH)^5-7, a label-free optical imaging method and deep convolutional neural networks (CNNs) to predict diagnosis at the bedside in near real-time in an automated fashion. Specifically, our CNNs, trained on over 2.5 million SRH images, predict brain tumor diagnosis in the operating room in under 150 s, an order of magnitude faster than conventional techniques (for example, 20-30 min)². In a multicenter, prospective clinical trial (n = 278), we demonstrated that CNN-based diagnosis of SRH images was noninferior to pathologist-based interpretation of conventional histologic images (overall accuracy, 94.6% versus 93.9%). Our CNNs learned a hierarchy of recognizable histologic feature representations to classify the major histopathologic classes of brain tumors. In addition, we implemented a semantic segmentation method to identify tumor-infiltrated diagnostic regions within SRH images. These results demonstrate how intraoperative cancer diagnosis can be streamlined, creating a complementary pathway for tissue diagnosis that is independent of a traditional pathology laboratory.