Faculty Bibliography

PURPOSE/OBJECTIVE:Isocitrate dehydrogenase (IDH) mutant gliomas are a distinct glioma molecular subtype for which no effective molecularly-directed therapy exists. Low-grade gliomas, which are 80-90% IDH mutant, have high RNA levels of the cell surface Notch ligand DLL3. We sought to determine DLL3 expression by immunohistochemistry in glioma molecular subtypes and the potential efficacy of an anti-DLL3 antibody drug conjugate (ADC), rovalpituzumab tesirine (Rova-T), in IDH mutant glioma. EXPERIMENTAL DESIGN/METHODS:We evaluated DLL3 expression by RNA using TCGA data and by immunohistochemistry in a discovery set of 63 gliomas and 20 non-tumor brain tissues and a validation set of 62 known IDH wildtype and mutant gliomas using a monoclonal anti-DLL3 antibody. Genotype was determined using a DNA methylation array classifier or by sequencing. The effect of Rova-T on patient-derived endogenous IDH mutant glioma tumorspheres was determined by cell viability assay. RESULTS:Compared to IDH wildtype glioblastoma, IDH mutant gliomas have significantly higher DLL3 RNA (P<1x10-15) and protein by immunohistochemistry (P=0.0014 and P<4.3x10-6 in the discovery and validation set, respectively). DLL3 immunostaining was intense and homogeneous in IDH mutant gliomas, retained in all recurrent tumors, and detected in only 1 of 20 non-tumor brains. Patient-derived IDH mutant glioma tumorspheres overexpressed DLL3 and were potently sensitive to Rova-T in an antigen-dependent manner. CONCLUSIONS:DLL3 is selectively and homogeneously expressed in IDH mutant gliomas and can be targeted with Rova-T in patient-derived IDH mutant glioma tumorspheres. Our findings are potentially immediately translatable and have implications for therapeutic strategies that exploit cell surface tumor-associated antigens.

Superficial/cutaneous malignant peripheral nerve sheath tumor is a rare soft tissue neoplasm that shares morphological, immunohistochemical, and molecular features with spindle/desmoplastic melanoma. We aimed to identify a methylome signature to distinguish these two entities. We analyzed 15 cases of spindle/desmoplastic melanoma and 15 cases of cutaneous malignant peripheral nerve sheath tumor in 23 men and 7 women. DNA from formalin-fixed, paraffin-embedded tissues was extracted and processed using the Illumina Infinium Methylation EPIC array interrogating 866,562 CpG sites. Using a home-grown informatics pipeline, we identified differentially methylated positions between the two entities. Functional network analysis for enrichment signatures was performed using DAVID tools. Identified differentially methylated positions were compared with the Cancer Genome Atlas's cutaneous melanoma dataset and a recently published malignant peripheral nerve sheath tumor dataset to assess the specificity of the identified signature. Unsupervised hierarchical clustering showed different patterns of methylation in cutaneous malignant peripheral nerve sheath tumor and spindle/desmoplastic melanoma. Two probes, cg20783223 and cg13332552, colocalized in the promoter region of BCAT1 and miR-2504. Pathway analysis highlighted enrichment in a subset of genes involved in breast and gastric cancer centered on BCAT1 and downstream activated genes in the mTOR pathway. Our study identifies BCAT1 as a novel methylome signature distinguishing spindle/desmoplastic melanoma from cutaneous malignant peripheral nerve sheath tumor.

Visual inspection of histopathology slides is one of the main methods used by pathologists to assess the stage, type and subtype of lung tumors. Adenocarcinoma (LUAD) and squamous cell carcinoma (LUSC) are the most prevalent subtypes of lung cancer, and their distinction requires visual inspection by an experienced pathologist. In this study, we trained a deep convolutional neural network (inception v3) on whole-slide images obtained from The Cancer Genome Atlas to accurately and automatically classify them into LUAD, LUSC or normal lung tissue. The performance of our method is comparable to that of pathologists, with an average area under the curve (AUC) of 0.97. Our model was validated on independent datasets of frozen tissues, formalin-fixed paraffin-embedded tissues and biopsies. Furthermore, we trained the network to predict the ten most commonly mutated genes in LUAD. We found that six of them-STK11, EGFR, FAT1, SETBP1, KRAS and TP53-can be predicted from pathology images, with AUCs from 0.733 to 0.856 as measured on a held-out population. These findings suggest that deep-learning models can assist pathologists in the detection of cancer subtype or gene mutations. Our approach can be applied to any cancer type, and the code is available at https://github.com/ncoudray/DeepPATH .

Epigenetic patterns on the level of DNA methylation have already shown to separate clinically relevant subgroups of meningiomas. Based on a reference set (Sahm et al., Lancet Oncol 2017), an epigenetic meningioma classifier employing DNA methylation patterns is made available through molecularneurpathology.org. We now set out to identify prognostic implications of epigenetic modification on the proteome level, particularly modifications of histones. First focus was on H3K27 trimethylation (H3K27me3). H3K27me3 was assessed by immunohistochemistry on 232 meningiomas. In 194 cases, trimethylation was detected in tumor cells. In 25 cases, staining was limited to vessels while all tumor cells were negative. Finally, 13 cases yielded equivocal staining patterns. Reduced abundance of H3K27me3 in cases with staining limited to vessels was confirmed by mass spectrometry on a subset of cases. Lack of staining for H3K27me3 in all tumor cells was significantly associated with more rapid progression (p=0.009). In line, H3K27me3 negative cases were associated with a DNA methylation pattern of the more aggressive types among the recently introduced DNA methylation groups. Also, NF2 and SUFU mutations were enriched among cases with lack of H3K27me3 in tumor cells (p<0.0001 and p=0.029, respectively). H3K27me3 staining pattern added significant prognostic insight in WHO grade II cases and in the compound subset of WHO grade I and II cases (p=0.04 and p=0.007, respectively). However, it did not further stratify within WHO grade III cases. Collectively, this data indicate that epigenetic modifications beyond DNA methylation are involved in the aggressiveness of meningioma. It also suggests that H3K27me3 immunohistochemistry might be a useful adjunct in meningioma diagnostics, particularly for cases with WHO grade II histology or at the borderline between WHO grade I and II. Ongoing studies evaluate the role of histone marks other than H3K27me3 and consequences on the proteomic composition of meningioma cells by high-throughput mass spectrometry

In contrast to adulthood, meningiomas are rare among children and adolescents. However, the molecular relations between both groups have not been elucidated in detail. We have analyzed 41 tumor samples from 37 pediatric meningioma patients (female: 17, male: 20; age range: 1-21 years). Atypical meningioma WHO grade II was the most frequent histological subtype (N=14, 38%). Most tumors were located at the convexity (N=18) or the skull base (N=15). Lack of SMO, AKT, KLF4/TRAF7 mutation in Sanger sequencing (n=22) prompted whole genome sequencing of a subset (n=7). All cases exhibited bi-allelic mutation of NF2 (combined large deletion and germline (5/7) or somatic (2/7) base exchanges/frameshifts). Subsequently, representative samples of all 37 patients were subjected to 450K DNA methylation profiling and remaining DNA to sequencing of a brain tumor specific gene panel. Loss of chromosome 22 was frequently detected (N=28, 76%), followed by loss of chromosome 1 (N=12, 32%) and chromosome 18 (N=7, 19%). Moreover, a separation into three groups was evident: One group covering all clearcell meningiomas with enrichment for SMARCE1 mutations, a second group dominated by atypical meningiomas, and a third group covering benign WHO grade I meningiomas, as well as rare subtypes such as rhabdoid meningiomas. Compared to adult tumors, the majority of pediatric meningiomas clustered in a separate group both by unsupervised hierarchical and clustering and t-stochastic nearest neighbor embedding. Analysis of four tumor recurrences did not reveal changes compared to the primary tumor. These data suggest that pediatric meningiomas are fundamentally different from adult counterparts

BACKGROUND: Up to 20% of meningiomas are aggressive tumors with high recurrence rates and poor prognosis. Biomarkers predicting the risk of an unfavorable clinical course are lacking although aberrations in NF2, increased copy number variations and a hypomethylated phenotype have been associated with more aggressive behavior. Mutational signatures (MS) are characteristic patterns of somatic mutations seen in cancer genomes associated with aging, exposure to certain mutagens, or defective DNA repair. We aimed to identify MS patterns in clinically aggressive meningiomas.
METHOD(S): We performed whole exome sequencing of 18 de novo meningiomas (locally invasive and recurrent WHO I, n=6; atypical WHO II, n=4; anaplastic WHO III, n=8). Median PFS was 18.9 months. Copy numbers and DNA methylation phenotype were assessed by DNA methylation array analysis. Mutational signatures were identified using published signature algorithms (COSMIC).
RESULT(S): MS1 and MS5 (aging) were found in 18 (100%) cases. MS associated with defective DNA MMR were highly prevalent: MS20 and MS26 were detected in 18 (100%) and MS6 in 2 (12%) cases. MS12 (unknown etiology) was present in 14 (82%) cases. Despite the association with defective DNA MMR, none (0%) of the MS6 cases harbored somatic mutations associated with DNA MMR while MS12 tumors were enriched for mutations in DNA MMR (43%), chromatin remodeling (36%) and other cancer-associated genes (7%). MS6 tumors had significantly lower indels compared to non-MS6 tumors (p=0.01). Tumors with mutations in chromatin remodeling genes had a significantly higher rate of single nucleotide variants (SNVs) compared to cases without such mutations (p=0.02).
CONCLUSION(S): MS associated with defective DNA MMR were highly prevalent in this set of aggressive meningiomas. However, despite the association with DNA MMR, MS6 meningiomas harbored no somatic mutations associated with DNA MMR while MS12 tumors were enriched for mutations in DNA MMR, chromatin remodeling and cancerassociated genes

BACKGROUND: Tumor heterogeneity presents a major challenge to cancer diagnosis and treatment. In addition to interpatient tumor variability, intratumoral heterogeneity characterized by distinct molecular and phenotypic profiles is increasingly recognized as a major cause of therapy resistance and cancer recurrence. Because DNA methylation patterns are largely responsible for determining cell-type-specific functioning, we hypothesized that distinct DNA methylation and proteomic alterations could be identified in histologically-defined invasive and proliferative tumor cell populations in human isocitrate dehydrogenase 1 (IDH1)- mutated and wild-type glioblastoma (GBM).
METHOD(S): Formalin-fixed paraffin-embedded tissue sections of human adult IDH1-mutated and wild-type GBM were laser-microdissected (LM) into perinecrotic pseudopalisading tumor cells (PPCs), non-pseudopalisading tumor core cells (NPPCs), invasive subpial spread (SPS) and perivascular satellitosis tumor cells and brain adjacent to tumor cells prior to analysis and compared to non-microdissected tumor (NMT) and/or germline DNA. Genomewide DNA methylation and chromosomal copy numbers were determined with Infinium MethylationEPIC 850K BeadChip and intratumoral DNA methylation patterns compared by unsupervised hierarchical clustering. Label-free quantitative liquid chromatography-mass spectrometry of proteins was performed and proteins differentially expressed across LM areas subjected to pathway enrichment analysis.
RESULT(S): Unsupervised hierarchical classification of DNA methylation patterns for each LM area and NMT demonstrated remarkable clustering for all patients, based on methylation probe and methylated gene patterns. Proteomics analysis showed upregulation of hypoxia-inducible factor-1 inducible proteins in hypoxic PPCs. Out of 1819 proteins quantified, 5 were overexpressed and 9 underexpressed more than 10-fold in SPS compared with NPPCs and associated with alterations in metabolism, transport, extracellular matrix and apoptosis. Correlation of protein expression and DNA methylation patterns was noted.
CONCLUSION(S): Compared to NPPCs, SPS cells migrating toward the invasive edge share a relatively consistent epigenetic and proteomic signature, suggesting potentially targetable common mechanism(s) of invasion shared among GBM