Loss Of 5q Research Articles

Abstract 2396: Heterogeneity of colorectal cancers and multiple liver metastases by mutation and copy number profiling

Abstract To investigate genetic and genomic heterogeneity in liver metastases of colorectal cancer, multiple tumor foci in the liver per patient were identified on the basis of radiological images and samples from these systematically biobanked. For a total of 67 patients, 258 metastatic tumor samples were collected (minimum of 2 and maximum of 9 samples per patient, median of 4 samples per patient). Additionally, for n = 14 patients, tumor samples from the corresponding primary colorectal cancer were also available. DNA isolated from these samples was subjected to mutation profiling by Sanger sequencing for the genes BRAF, KRAS, PIK3CA and TP53, as well as allele-specific copy number profiling by CytoscanHD SNP array. For the four genes investigated for mutations, 39/67 (58%) of the patients were homogeneous in their mutation profile, while 21/67 (32%) were heterogeneous (one or more samples had mutations not detected in all samples). For 7/67 patients (10%), no mutations were detected in any of these four genes. For 9/14 patients with matching primary tumors, there was complete concordance between the primary tumor and all liver metastases profiled for mutations, while the remaining 5/14 patients showed discordance in mutation profile between the primary tumor and at least one of the liver metastases profiled. High resolution copy number profiles of this cohort detected many of the aberrations previously described in primary colorectal cancer, specifically gain of chromosomes 7, 8q, 13q and 20q, and loss of chromosomes 5q, 8p, 15q, 17p and 18q. The extent of heterogeneity observed in patient-wise mutation profiles was largely mirrored in the copy number profiles of the same tumors. 16/27 (60%) of the patients for which both mutation and copy number profiles were available showed concordance in the degree of heterogeneity per patient. In summary, intra-individual heterogeneity varied from patient to patient, with results from the genetic and genomic data types largely concurring. Citation Format: Sharmini Alagaratnam, Tuva Høst Brunsell, Stine Aske Danielsen, Anita Sveen, Mette Eknæs, Merete Hektoen, Bård Røsok, Bjørn Atle Bjørnbeth, Arild Nesbakken, Ragnhild Lothe. Heterogeneity of colorectal cancers and multiple liver metastases by mutation and copy number profiling. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 2396.

32 - The Clinical Utility of a Combined Approach of CMA and NGS in Melanoma

Melanoma is a dangerous form of skin cancer arising from abnormal melanocytes. It accounts for only a small percentage of skin cancers, yet is responsible for the majority of deaths. The application of chromosomal microarray (CMA) with next-generation sequencing (NGS) could be a novel method for diagnosis, prognosis, and treatment of melanoma. Melanoma has non-random copy number changes and mutations. Copy number changes common in the diagnosis and prognosis of melanoma are loss of 1p, 6p, 6q, 7q, 9p, 10q and gain of 1q, 5p, 6, 6p, 6q, 7, 7p, 7q, 8, 8q, 11q, 12q. In order to determine the clinical utility of combining CMA and NGS, 30 cases sent to PathGroup were reviewed. All cases had at least one common copy number change. 28 cases (93.33%) had at least one targetable copy number change. The most common change was a loss of CDKN2A (9p21.3) in 26 cases (86.87%), which can be targeted by CDK4/6 inhibitors such as palbociclib. 28 cases were analyzed with NGS with 25 (89%) having a detectable mutation. The most common mutation was BRAF (predominantly V600E) mutation seen in 8 cases (32%), which can be targeted by RAS/RAF inhibitors such as trametinib and dabrafenib. 3 cases (10.71%) showed targetable genes by CMA when no mutation was seen. These data demonstrate the clinical utility of combining CMA and NGS in detecting copy number changes and mutations that can assist in diagnosis, prognosis, and treatment of melanoma.

AT-10ATYPICAL TERATOID RHABDOID TUMORS AND POORLY DIFFERENTIATED CHORDOMAS: DISTINCT MOLECULAR ENTITIES WITH SMARCB1/INI1 LOSS AND DISMAL PROGNOSIS

AT-10. ATYPICAL TERATOID RHABDOID TUMORS AND POORLY DIFFERENTIATED CHORDOMAS: DISTINCT MOLECULAR ENTITIES WITH SMARCB1/INI1 LOSS AND DISMAL PROGNOSIS Christian Thomas1, Volker Hovestadt2,3, Daniel Schrimpf2,3, Pascal Johann2,4, Susanne Bens5, Florian Oyen6, Hannes Vogel7, Felice Giangaspero8, Manila Antonelli8, Markus Riemenschneider9, Marie Christine Bernardo10, Caterina Giannini11, Nasir Ud Din12, Arie Perry13, Kathy Keyvani14, Frank van Landeghem15, David Sumerauer16, Peter Hauser17, David Capper3,4, Andrey Korshunov3,4, David T.W. Jones2,3, Stefan Pfister2,3, Reinhard Schneppenheim6, Reiner Siebert5, Michael C. Fruhwald18, Marcel Kool2,3, and Martin Hasselblatt1; University Hospital Munster, Munster, Germany; German Cancer Research Center (DKFZ), Heidelberg, Germany; German Cancer Consortium (DKTK), Heidelberg, Germany; University Hospital Heidelberg, Heidelberg, Germany; Christian-Albrechts-University Kiel, Kiel, Germany; University Medical CenterHamburg-Eppendorf,Hamburg,Germany;StanfordUniversitySchool of Medicine, Palo Alto, CA, USA; Sapienza University, Rome, Italy; Regensburg University Hospital, Regensburg, Germany; St. Lukes Medical Center, Quezon City, The Philippines; Mayo Clinic, Rochester, MN, USA; Aga Khan University Hospital, Karachi, Pakistan; UCSF Medical Center, San Francisco, CA, USA; University of Duisburg-Essen, Essen, Germany; University of Alberta, Edmonton, Canada; Charles University, Prague, Czech Republic; Semmelweis University, Budapest, Hungary; Childrens’ Hospital Augsburg and EU-RHAB Registry, Augsburg, Germany Chordomasare tumorsof the skull base and spine thought to arise fromremnants of the notochord. Pediatric chordomas showing atypia, increased mitotic activity, unstructured growth pattern and loss of SMARCB1/INI1 expression have been designated as “poorly differentiated chordomas”. It remains uncertain, however, if poorly differentiated chordoma represents a distinct entity or part of the AT/RT spectrum. Therefore, seven poorly differentiated chordomas, 14 conventional chordomas as well as 10 AT/RT of each of three molecular subgroups (i.e. TYR, MYC and SHH) were molecularly characterized including Illumina Infinium Human Methylation 450k Bead Chip profiling. Median age of the four boys and three girls harboring poorly differentiated chordomas was 7 years (range 1-11 years); median overall survival accounted for only 9 months (95% confidence interval 6-12 months). On histopathological examination, all poorly differentiated chordomas lacked SMARCB1/INI1 expression but showed nuclear brachyury expression. Brachyury expression was also observed in 2/30 AT/RT. Unsupervised clusteranalysis identified five methylation groups, including two distinct chordoma clusters, representing the poorly differentiated chordomas and the conventional chordomas, respectively, both clustering apart from the AT/RT subgroups. Losses of 22q affecting the SMARCB1 region were the only recurrentalteration inpoorlydifferentiatedchordomas andthevastmajorityofAT/ RT. Heterozygous or homozygous deletions affecting the SMARCB1 region could be confirmed on FISH and/or MLPA, while SMARCB1 point mutations were absent on sequencing. In conclusion, poorly differentiated chordoma represents a molecularly distinct entity with dismal prognosis, which can be reliably separated from conventional chordoma and AT/RT by a distinct methylation profile. Neuro-Oncology 18:iii1–iii6, 2016. doi:10.1093/neuonc/now065.9 #The Author(s) 2016. Published by Oxford University Press on behalf of the Society for Neuro-Oncology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Whole-genome profiling helps to classify phyllodes tumours of the breast

AimsThe aim of this study was to analyse a series of borderline and malignant phyllodes tumours (PTs) of the breast by whole-genome profiling to identify genomic markers that could help...

The chronological sequence of somatic mutations in early gastric carcinogenesis inferred from multiregion sequencing of gastric adenomas.

Mutation profiles and intratumoral heterogeneity are not well understood for benign gastric adenomas, some of which progress into malignant gastric adenocarcinomas. In this study, we performed whole-exome sequencing of three microsatellite stable (MSS) and two microsatellite instability-high (MSI-H) gastric adenomas with three regional tumor biopsies per case. We observed that the mutation abundance of benign gastric adenomas was comparable to those of gastric adenocarcinomas, suggesting that the mutational makeup for gastric carcinogenesis may already be achieved in benign adenomas. The extent of intratumoral heterogeneity was more substantial for MSS genomes in that only 1% - 14% of somatic mutations were common across the regional biopsies or ‘public’, while 50% - 94% of mutations were public in MSI-H gastric adenomas. We observed biallelic, loss-of-functional events of APC with truncating mutations and/or 5q losses for all cases, mostly as public events. All MSS gastric adenomas also harbored ARID2 truncating mutations, often as multiple, region-specific ones indicative of convergent evolution. Hotspot missense mutations on known cancer genes such as ERBB2 and KRAS were largely observed as region-specific aberrations. These findings suggest that biallelic functional APC inactivation initiates the gastric carcinogenesis and is followed by mutations of histone modifiers and then activation of known cancer-related genes. As the first exome-wide multi-region mutational profiling of gastric adenomas, our study provides clues on the chronological sequence of somatic mutations and their clonal architectures in early gastric carcinogenesis.

MiRNA expression profile of chronic lymphocytic leukemia patients with 13q deletion

Deletion 13q (13q-) is the most common cytogenetic aberration in chronic lymphocytic leukemia (CLL) and is associated with the most favorable prognosis as the sole cytogenetic abnormality. However, it is heterogeneous whereby CLL patients with higher percentages of 13q- cells (13q-H) have a more aggressive clinical course and a distinct gene expression profile. The microRNA (miRNA) expression profile of CLL gives additional biological and prognostic information, but its expression in 13q- CLL has not been examined in detail. The miRNA expression of clonal B cell lymphocytes (CD19+ cells) of 38 CLL patients and normal B cells of six healthy donors was analyzed. CLL patients with higher percentages of 13q- cells (≥80%) showed a different level of miRNA expression from patients with lower percentages (<80%). Interestingly, miR-143 was downregulated and miR-155 was overexpressed in 13q-H. This deregulation affected important validated target genes involved in apoptosis (BCL2, MDM2, TP53INP1) and proliferation (KRAS, PI3K-AKT signaling), that could lead to decreased apoptosis and increased proliferation in 13q-H patients. This study provides new evidence about the heterogeneity of the 13q deletion in CLL patients, showing that miRNA regulation could be involved in several significant pathways deregulated in CLL patients with a high number of losses in 13q.

MP49-01 NEXT GENERATION SEQUENCING OF CELL FREE DNA REVEALS GENOMIC ABERRATIONS IN METASTATIC UROTHELIAL CARCINOMA

You have accessJournal of UrologyBladder Cancer: Invasive III1 Apr 2016MP49-01 NEXT GENERATION SEQUENCING OF CELL FREE DNA REVEALS GENOMIC ABERRATIONS IN METASTATIC UROTHELIAL CARCINOMA Tilman Todenhöfer, Stanislav Volik, Bernie Eigl, Scott North, Sonal Brahmbhatt, Anne Haegert, Johannes Mischinger, Arnulf Stenzl, Stephane LeBihan, Alexander Wyatt, Colin Collins, and Peter Black Tilman TodenhöferTilman Todenhöfer More articles by this author , Stanislav VolikStanislav Volik More articles by this author , Bernie EiglBernie Eigl More articles by this author , Scott NorthScott North More articles by this author , Sonal BrahmbhattSonal Brahmbhatt More articles by this author , Anne HaegertAnne Haegert More articles by this author , Johannes MischingerJohannes Mischinger More articles by this author , Arnulf StenzlArnulf Stenzl More articles by this author , Stephane LeBihanStephane LeBihan More articles by this author , Alexander WyattAlexander Wyatt More articles by this author , Colin CollinsColin Collins More articles by this author , and Peter BlackPeter Black More articles by this author View All Author Informationhttps://doi.org/10.1016/j.juro.2016.02.414AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookTwitterLinked InEmail INTRODUCTION AND OBJECTIVES To date, only limited data is available on genomic alterations present in metastatic lesions of patients with metastatic urothelial carcinoma (mUC). The constant release of circulating cell free DNA (cfDNA) from tumor cells into the peripheral blood stream may provide a valuable source for detection of cancer associated somatic copy number variations and mutations. The value of next-generation sequencing (NGS) of cfDNA to identify unknown genomic mutations in patients with bladder cancer has not been addressed. The aim of the present pilot study was to evaluate the feasibility of analysis of somatic mutations and copy number variations in plasma cfDNA of patients with mUC. METHODS Peripheral blood (20ml) was collected from 8 patients with mUC before chemotherapy. Plasma cfDNA was isolated using the QIAAmp Circulating Nucleic Acid Kit. DNA was extracted from peripheral blood mononuclear cells (PBMCs) using the DNEasy Kit (Qiagen) for assessment of germ-line DNA alterations. Library construction was performed using the Ampliseq Comprehensive Cancer Panel (ThermoFisher). NGS was done on an ION Proton platform (ThermoFisher). Copy number variations and mutations were compared to publically available datasets from UC tissue. RESULTS The median amount of cfDNA extracted was 202.2 ng (range 42.3-530.4). We detected strong mutation candidates in 6/8 patients (Table 1) and copy number changes consistent with previously reported alterations in UC. Common copy number variations included gain of 1p (5/8 patients), 7p (5/8 patients), and 17q (3/8 patients including one with high-level focal amplification of ERBB2) and loss of 6q (6/8 patients). CONCLUSIONS Non-invasive characterization of genomic changes in peripheral blood of patients with mUC using NGS of plasma cfDNA is feasible and detects copy number changes and mutations in a high proportion of patients. Analysis of cfDNA might therefore provide an improved understanding of genomic changes in mUC that could enable the personalized use of targeted therapies. © 2016FiguresReferencesRelatedDetails Volume 195Issue 4SApril 2016Page: e664 Advertisement Copyright & Permissions© 2016MetricsAuthor Information Tilman Todenhöfer More articles by this author Stanislav Volik More articles by this author Bernie Eigl More articles by this author Scott North More articles by this author Sonal Brahmbhatt More articles by this author Anne Haegert More articles by this author Johannes Mischinger More articles by this author Arnulf Stenzl More articles by this author Stephane LeBihan More articles by this author Alexander Wyatt More articles by this author Colin Collins More articles by this author Peter Black More articles by this author Expand All Advertisement Advertisement PDF downloadLoading ...

Abstract IA11: Meningioma-1 cooperates with MLL and DOT1L to induce leukemia

Abstract Meningioma-1 (MN1) is frequently overexpressed in AML, and associated with a poor prognosis. In addition, MN1-TEL fusions are found in AML, underscoring the importance and possible driver function of MN1 in AML. Forced expression of MN1 in murine hematopoietic progenitors induces a highly aggressive leukemia as a single hit. The mechanism by which MN1 induces AML is unclear. MN1 is a transcriptional co-activator with almost no sequence or structural similarity to any other protein, and no targeted approaches to MN1-high AML are currently available. We sought to understand the mechanism by which MN1 induces AML with the goal to identify targetable downstream effectors. We found that the gene expression program induced by forced expression of MN1 in hematopoietic progenitors substantially overlaps with a set of genes that is downregulated in response of loss of the histone methyltransferase Dot1l in normal hematopoietic progenitors. This led us to hypothesize that the MN1-induced leukemogenic gene expression program might be dependent on Dot1l. We established MN1 leukemias on a Dot1l conditional background and found that loss of Dot1l indeed induced cell cycle arrest, differentiation and apoptosis, and prolonged the survival of transplanted mice in vivo. This was associated with the downregulation of the MN1-induced gene expression program. We next sought to investigate a possible mechanism for this observation. MN1 has been reported to be recruited to its target genes by ChIP-seq, but it does not itself possess sequence specific DNA binding capacity. The mediator of this recruitment is thus unclear. Since Dot1l has been shown to be required for the high level expression of MLL-fusion target genes in MLL-rearranged leukemias, we asked whether MN1 might cooperate with wild type MLL1, explaining the dependence on Dot1L. In order to test whether wild-type MLL1 is required for MN1 leukemias, we established AML on a Mll1-conditional background. Results largely phenocopied the loss of Dot1l, suggesting that MN1 cooperates with both MLL1 and Dot1L to induce leukemia. We are currently investigating whether MN1 binds MLL1 and/or DOT1L directly to exert this function using protein pull downs and ChIP-Seq. Finally, we asked whether our findings had relevance for human AML. MN1 overexpression is found over a wide range of different molecular subgroups but is relatively under-represented in MLL-rearranged AML, suggesting redundant pathways. A subgroup that frequently displays very high levels of MN1 expression are AML with a complex karyotype with loss of 5q or 7 sequences, and high expression of HOXA cluster genes. We analyzed the response of a human cell line and two patient samples with high MN1/high HOXA9 expression to inhibition of DOT1L, and found induction of differentiation and apoptosis similar to our mouse model. In summary, our data suggest that MN1 cooperates with wild type MLL1 to induce a leukemogenic gene expression program that results in AML, and that this program may be targetable by inhibiting DOT1L. Citation Format: Simone Riedel, Jessica Haladyna, Brett Stevens, Daniel Pollyea, Qi Wei, Craig Jordan, Patricia Ernst, Tobias Neff, Kathrin Bernt. Meningioma-1 cooperates with MLL and DOT1L to induce leukemia. [abstract]. In: Proceedings of the AACR Special Conference on Advances in Pediatric Cancer Research: From Mechanisms and Models to Treatment and Survivorship; 2015 Nov 9-12; Fort Lauderdale, FL. Philadelphia (PA): AACR; Cancer Res 2016;76(5 Suppl):Abstract nr IA11.

PRAME as an Independent Biomarker for Metastasis in Uveal Melanoma.

Uveal melanoma (UM) can be classified by gene expression profiling (GEP) into Class 1 (low metastatic risk) and Class 2 (high metastatic risk), the latter being strongly associated with mutational inactivation of the tumor suppressor BAP1. Nevertheless, a small percentage of Class 1 tumors give rise to metastatic disease. The purpose of this study was to identify biomarkers of metastasis in Class 1 tumors. A total of 389 consecutive patients with UM were assigned to Class 1 or Class 2 using a prospectively validated 12-gene prognostic classifier. Selected tumors were further analyzed using global GEP and single nucleotide polymorphism microarrays. PRAME (preferentially expressed antigen in melanoma) mRNA expression was analyzed in 64 Class 1 tumors by qPCR. Among Class 1 UMs, the most significant predictor of metastasis was PRAME mRNA expression (P = 0.0006). The 5-year actuarial rate of metastasis was 0% for Class1(PRAME-), 38% for Class1(PRAME+), and 71% for Class 2 tumors. Median metastasis-free survival for Class1(PRAME+) patients was 88 months, compared to 32 months for Class 2 patients. Findings were validated using three independent datasets, including one using disomy 3 to identify low-risk UM. Chromosome copy number changes associated with Class1(PRAME+) tumors included gain of 1q, 6p, 8q, and 9q and loss of 6q and 11q. PRAME expression was associated with larger tumor diameter (P = 0.05) and SF3B1 mutations (P = 0.003). PRAME is an independent prognostic biomarker in UM, which identifies increased metastatic risk in patients with Class 1 or disomy 3 tumors. This finding may further enhance the accuracy of prognostic testing and precision medicine for UM.

Abstract P6-07-04: Distinct repertoires of somatic mutations affecting driver genes in mucinous and neuroendocrine carcinomas of the breast

Abstract Introduction: Mucinous carcinoma of the breast (MCB) is a rare histologic type, which accounts for approximately 2% of all invasive breast cancers (IBCs) and is characterized by clusters of tumor cells floating in large amounts of extracellular mucin. MCB comprises two main subtypes based on architectural and cytologic features: mucinous A (paucicellular) and mucinous B (hypercellular). Although MCBs are low-grade ER-positive/HER2-negative tumors of luminal molecular subtype, these cancers lack concurrent losses of 16q and gains of 1q, the hallmark genetic features of low-grade ER+/HER2- breast cancers, and have low levels of genetic instability. Neuroendocrine carcinoma of the breast (NCB) accounts for 2% - 5% of IBCs and displays morphologic features similar to those of neuroendocrine tumors of other organs. Previous transcriptomic analyses have revealed that NCBs and mucinous B, but not mucinous A, breast cancers display similar gene expression profiles. The aims of this study were to determine whether MCBs and NCBs share a similar repertoire of somatic genetic alterations and if these aberrations are distinct from those reported in common forms of ER+/HER2- IBCs. Material and methods: DNA extracted from microdissected MCBs (n=7 mucinous A, n=6 mucinous B), NCBs (n=14) and adjacent normal tissues were subjected to massively parallel sequencing targeting all exons of 254 genes most frequently mutated in IBC or related to DNA repair. Somatic point mutations were identified using MuTect and somatic insertions and deletions (indels) were defined using Strelka and Varscan2. We retrieved mutations in the same 254 genes in common forms of ER+/HER2- IBC (n=252) from The Cancer Genome Atlas (TCGA). Results: The most frequently mutated genes in MCBs were GATA3 (31% of cases, 4/13, all frame-shift indels), followed by KMT2C (MLL3) and MAP3K1 (both 23%). GATA3 and KMT2C (29%) were the most frequently mutated genes in mucinous A cancers, whereas MAP3K1 (33%) was the most frequently mutated gene in mucinous B cancers. ARID1A mutations were found in three of 14 (21%) NCBs, of which 2 were truncating mutations. A comparative analysis of the repertoire of somatic mutations found in mucinous A, mucinous B and NCBs did not reveal any statistically significant differences. As compared to common forms of ER+/HER2- IBCs, MCBs were found to have a significantly lower frequency of PIK3CA (8% vs 42%, p=0.02) mutations, which was particularly evident in mucinous A cancers (0% vs 42%, p=0.04). NCBs displayed significantly higher frequencies of somatic mutations affecting ARID1A (21% vs 2%, respectively, p=0.006), FOXA1 (14% vs 2%, respectively, p&amp;lt;0.05) and a lower frequency of PIK3CA somatic mutations (14% vs 42%, respectively, p&amp;lt;0.05) than common forms of ER+/HER2- IBCs. Conclusion: The frequency of mutations affecting bona fide breast cancer genes differed among mucinous A, mucinous B and NCBs. The repertoire of somatic mutations found in MCBs and NCBs differed from that of common forms of ER+/HER2- IBCs, in particular by the low frequency of somatic mutations affecting PIK3CA. Citation Format: Piscuoglio S, Ng CKY, Marchio C, Eberle CA, Guerini-Rocco E, Mariani O, Vincent-Salomon A, Reis-Filho JS, Weigelt B. Distinct repertoires of somatic mutations affecting driver genes in mucinous and neuroendocrine carcinomas of the breast. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P6-07-04.

Pulmonary metastasis of a meningioma presenting as a solitary pulmonary nodule: 2 case reports

Distant metastases of meningioma are rare, especially in grade 1 meningiomas. In a recent literature review, only 115 cases were found. In almost all published cases, the meningioma was treated several years before the metastasis was diagnosed. The lungs are the most frequent site of metastasis. We describe two patients treated for meningioma (one case grade 1, the other grade 3) who were referred to the Respiratory Oncology Unit because of the incidental finding of a pulmonary nodule on routine chest radiography. Both had undergone several neurosurgical procedures but the last operation was more than 7 years before in both cases. Positron emission tomography scan was suggestive of a malignant lung tumour. The lesions were surgically removed. Pathology confirmed meningioma in both cases with the same WHO grade, immunohistochemical and genetic profiles as the original meningioma. Both patients recovered well from thoracic surgery. The patient with grade 3 meningioma died three years later from intracranial recurrence. When a patient previously treated for meningioma develops a nodular lung lesion, metastasis of the meningioma should be in the differential diagnosis list. Because of the occurrence of distant metastasis even in grade I meningiomas, we suggest that the grading system should take into account genetic changes in the meningioma. Chromosome 1p and 14q losses possibly explain the aggressive behaviour of the grade 1 meningioma.

Assessing intratumor heterogeneity and evolutionary distance in exome sequencing of colon cancer.

590 Background: Carcinogenesis is considered to be an evolutionary process by natural selection of cell clones which have acquired advantageous heritable characteristics. This adaption process, which resembles Darwin’s evolutionary theory, has also been suggested as a potential mechanism promoting resistance to anti-cancer treatment. Methods: We performed whole exome sequencing of four colon cancers at five morphologically different loci of the primary tumor tissue. To investigate intratumor heterogeneity we conducted structural variant analysis, copy number analysis and ploidy profiling. We propose an evolutionary distance of the tumor samples based on a distance function in copy number space. Results: Copy number and phylogenetic analysis demonstrated substantial intratumor heterogeneity, with all 20 samples (5 loci for 4 tumors) possessing a unique copy number profile. Our analysis further revealed cancer-specific, tumor-specific and locus-specific copy number variants. We found that at least one copy of 18q, including the genes DCC and SMAD4, as well as 8p, carrying the CSMD1 gene, was missing in all tumors. Additionally, gains of chromosome 7,8q, 13q, 20q and losses of 4q and 21q were frequent. Using our distance measure in copy number space, phylogenetic analysis grouped together the five loci of one tumor and showed their evolutionary branching. We found that our distance measure in copy number space is superior to a distance measure in SNV space for colon cancer. Conclusions: Our analysis reveals that the landscape of the tumor genome cannot be captured by single tumor-biopsy, which is one of the challenges to personalized-medicine as heterogeneity can lead to therapeutic failures. However, there can be characteristics of the tumor genomes that are stable across samples which can potentially be used as biomarkers. Nonetheless variability of copy number profiles of colorectal cancers might suggest a personalized therapy.

Molecular Characterization of Philadelphia Chromosome Positive Acute Myeloid Leukemia - New Provisional Entity?

IntroductionPhiladelphia chromosome positive (Ph+) acute myeloid leukemia (Ph+AML) is discussed to be a new provisional entity for the upcoming WHO classification. Whether Ph+AML represents a distinct entity or rather embodies chronic myeloid leukemia in myeloid blast crisis (CML-BC) without preceding clinical manifestation is under debate mostly due to lack of robust criteria to reliably differentiate these two diseases. Further, while Ph+AML is clearly distinguishable from Ph+ acute lymphoblastic leukemia (Ph+ALL) based on immunophenotyping, recent studies demonstrated that Ph+AML retain typical characteristics of lymphoid disease.AimPh+AML, CML-BC and Ph+ALL were analyzed by a panel of 24 genes and array CGH to get more insights into these Ph+ leukemias and to potentially define delimiting genetic features.Patients and MethodsWe examined 24 pts with Ph+AML (11 females/13 males, median age: 58 (24-83)), 11 CML-BC (7 females/4 males, median age: 60 (32-79)) and 11 Ph+ALL (7 females/4 males, median age: 67 (44-77)). AML and ALL were diagnosed according to WHO classification by morphology, MPO and flow cytometry. CML-BC all were diagnosed as CML before and treated accordingly. All cases revealed the BCR-ABL1 fusion gene. Next generation sequencing was performed for ASXL1, BCOR, CBL, CSF3R, DNMT3A, ETV6, FLT3 tyrosine kinase domain(FLT3 -TKD), IDH1/2, JAK1/2/3, KRAS, NPM1, NRAS, PTPN11, RUNX1, TET2, TP53, WT1 and ZRSR2 using the MiSeq Instrument (Illumina, San Diego, CA). Partial tandem duplications in MLL (MLL- PTD), internal tandem duplications in FLT3 (FLT3- ITD)and deletions in IKZF1 were analyzed by quantitative real-time PCR or genescan analysis. 45 cases were investigated by array CGH (Agilent, Waldbronn, Germany).ResultsWith respect to cytogenetic abnormalities besides Philadelphia chromosome, several unbalanced abnormalities were found in Ph+ALL (mean: 9, range: 2-31), while less aberrations were found in Ph+AML and CML-BC (mean: 4, range: 0-28 and mean: 4, range: 0-16, respectively; p=0.03). Of these the most prominent aberrations which were present in all three groups included loss of 7p encompassing IKZF1 (Ph+AML: 7/23, 30%; Ph+ALL: 8/11, 73%; CML-BC: 2/10, 20%), loss of 9p encoding CDKN2A/B (Ph+AML: 2/23, 9%; Ph+ALL: 6/11, 55%; CML-BC: 2/10, 20%) as well as gain of 8q (Ph+AML: 6/23, 26%; Ph+ALL: 3/11, 27%; CML-BC:4/11, 36%). Loss of 5q (5/23, 22%), gain of 13q (4/23, 17%) and loss of 21q (3/23, 9%) was exclusively present in Ph+AML. While in Ph+ALL loss of 10q (3/11, 27%), 2p (3/11, 18%), 11q (3/11, 18%) and gain of 4q (3/11, 18%) was exclusively found. Regarding recurrent balanced aberrations no rearrangements were found in Ph+AML and Ph+ALL, while 3/11 (27%) CML-BC pts harbored balanced 3q26-rearrangements.With respect to molecular genetics, alterations were found in 15/24 (63%) Ph+AML, 8/11 (73%) CML-BC and 8/11 (73%) Ph+ALL pts. Commonly shared molecular aberrations were deletions in IKZF1 (Ph+AML: 3/24, 13%; Ph+ALL: 8/11, 73%; CML-BC: 2/10, 20%) as well as mutations (mut) in RUNX1 (Ph+AML: 5/19, 26%; Ph+ALL: 1/7, 14%; CML-BC: 5/10, 50%). Further, mut found in Ph+AML and CML-BC affected ASXL1 (Ph+AML: 2/22, 9%; CML-BC: 2/11, 18%) and IDH1 (Ph+AML: 2/22, 9%; CML-BC: 1/11, 9%). Additionally, Ph+AML harbored alterations in TP53 (3/21, 14%), TET2 (2/21, 10%) and DNMT3A (1/20, 5%). For CML-BC, additional mut were found in WT1 (2/9, 22%), ETV6 (1/9, 11%) and KRAS (1/9, 11%). Regarding FLT3 -ITD, NPM1 and the remaining genes no alterations were found. Overall, Ph+ALL differed from the combined cohort of Ph+AML and CML-BC in that mut in ASXL1, DNMT3A, ETV6, IDH1, KRAS, TET2 and WT1 as well as MLL -PTD occurred not in the former but only in the latter (12/31 cases with at least one gene mutated, p=0.07). Intriguingly, the mean±SD number of mut in these genes did not significantly differ between Ph+AML and CML-BC cases (0.36±0.58 vs. 0.67±0.71, p=0.23).ConclusionComparing cytogenetic alterations Ph+AML could be clearly distinguished from CML-BC or Ph+ALL by harboring loss of 5q and gain of 13q, which are typically found in myeloid diseases. Beside, Ph+AML and CML-BC showed a high frequency of molecular mutations which were hardly found in Ph+ALL. This data supports the concept discussed by the WHO that Ph+AML is a specific entity and can be distinguished from CML-BC and Ph+ALL. However, further studies are warranted to define the most appropriate parameters to distinguish Ph+AML from CML-BC. DisclosuresWeber:MLL Munich Leukemia Laboratory: Employment. Meggendorfer:MLL Munich Leukemia Laboratory: Employment. Nadarajah:MLL Munich Leukemia Laboratory: Employment. Perglerová:MLL2 s.r.o.: Employment. Schnittger:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Kern:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership.

Meningeoma-1 Cooperates with MLL and DOT1L to Induce Leukemia

Meningioma-1 (MN1) is frequently overexpressed in AML, and associated with a poor prognosis. In addition, MN1-TEL fusions are found in AML, underscoring the importance and possible driver function of MN1 in AML. Forced expression of MN1 in murine hematopoietic progenitors induces a highly aggressive leukemia as a single hit. The mechanism by which MN1 induces AML is unclear. MN1 is a transcriptional co-activator with almost no sequence or structural similarity to any other protein, and no targeted approaches to MN1-high AML are currently available.We sought to understand the mechanism by which MN1 induces AML with the goal to identify targetable downstream effectors. We found that the gene expression program induced by forced expression of MN1 in hematopoietic progenitors substantially overlaps with a set of genes that is downregulated in response to loss of the histone methyltransferase Dot1l in normal hematopoietic progenitors. This led us to hypothesize that the MN1-induced leukemogenic gene expression program might be dependent on Dot1l. We established MN1 leukemias on a Dot1l conditional knockout background and found that of Dot1l indeed induced cell cycle arrest, differentiation and apoptosis, and prolonged the survival of transplanted mice in vivo. This was associated with the downregulation of the MN1-induced gene expression program. We next sought to investigate a possible mechanism for this observation. MN1 has been reported to be recruited to its target genes based on ChIP-seq, but it does not itself possess sequence specific DNA binding capacity. The mediator of this recruitment is thus unclear. Since Dot1l has been shown to be required for the high level expression of MLL-fusion target genes in MLL-rearranged leukemias, we asked whether MN1 might cooperate with wild type MLL, explaining the dependence on Dot1l. In order to test whether wild-type MLL is required for MN1 leukemias, we established AML on a MLL-conditional background. Results phenocopied the loss of Dot1l, suggesting that MN1 cooperates with both MLL and Dot1l to induce leukemia. We are currently investigating whether MN1 binds MLL and/or DOT1L directly to exert this function using biochemical and genomic approaches. Finally, we asked whether our findings had relevance for human AML. MN1 overexpression is found over a wide range of different molecular subgroups but is relatively under-represented in MLL-rearranged AML, suggesting redundant pathways. A subgroup that frequently displays very high levels of MN1 expression are AML with a complex karyotype with loss of 5q or 7 sequences, and high expression of HOXA cluster genes. We analyzed the response of a human cell line and two patient samples with high MN1/high HOXA9 expression to inhibition of DOT1L, and found induction of differentiation and apoptosis similar to our mouse model.In summary, our data suggest that MN1 cooperates with wild type MLL to induce a leukemogenic gene expression program that results in AML, and that this program may be targetable by inhibiting DOT1L. DisclosuresPollyea:GlycoMimetics: Other: Member of data safety monitoring board; Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Ariad: Consultancy; Karyopharm: Consultancy; Pfizer: Consultancy; Agios: Consultancy, Membership on an entity's Board of Directors or advisory committees. Armstrong:Epizyme, Inc: Consultancy. Ernst:Amgen: Other: 528 shares of stock. Neff:Epizyme: Patents & Royalties: US Patent 62/026583 Dot1l Inhibition in Patients with MN1 High AML filed 7/2014 Coinventor: Tobias Neff.

Genomic and Targeted Mutational Analysis of T/NK-Cell Post-Transplant Lymphoproliferative Disorders Provides Insight into Disease Biology

BackgroundPost-transplant lymphoproliferative disorders (PTLDs) encompass heterogeneous entities with different histomorphology and behavior. T/NK-cell PTLDs (T/NK-PTLDs) are rare, accounting for 2-15% of all PTLD, which are often EBV-negative and associated with a poor prognosis. The pathogenesis of T/NK-PTLDs and the underlying molecular alterations are presently unknown. In order to better understand their pathogenesis, we performed high resolution DNA profiling for copy number alterations and targeted mutational analyses of 18T/NK-PTLDs.MethodsCases of T/NK-PTLD were selected from the archives of two institutions. Morphological features were assessed and immunohistochemistry (IHC) was performed to classify lymphomas according to the WHO 2008 criteria. DNA was extracted from formalin-fixed paraffin-embedded tissue. Copy number analysis (n=16; Affymetrix Oncoscan FFPE Assay) and targeted next-generation sequencing for 467 cancer-associated genes, after capture of all exons or select whole genes (n=17; Illumina HiSeq 2500), were performed. Variants with allele frequency <10%, allele prevalence >1% in 1000 genomes project, read depth <10x, or quality score <10 were excluded and cross referenced with COSMIC, PROVEAN, and SIFT databases. Mismatch repair deficiency was assessed using IHC and PCR analysis for a panel of mononucleotide repeat markers (Promega). Cases exhibiting microsatellite instability (MSI) at ≥2 loci were classified as MSI-high.ResultsThe T/NK-PTLDs represented virtually the entire spectrum of currently recognized T- and NK-cell lymphomas. Peripheral T-cell lymphoma (PTCL), NOS were the most frequent (n=6), extranodal location was common (n=14, 78%) and only a minority were EBV+ (n=2, 11%). Copy number alterations were observed in 13 cases (81%; complex =11, simple=2). In contrast to B-cell PTLDs, EBV+ T/NK-PTLDs showed more complex copy number changes than EBV- cases. Somatic mutations were detected in 17/18 cases (94%). Overall, 343 variants were identified: 277 missense (80%), 40 indels (12%) and 14 nonsense (4%). An average of 20.1 variants/case was observed with an average 594.8-fold coverage.Mutations of epigenetic mediators and chromatin remodeling complex genes, including TET2 (n=5), MLL2 (n=4), MLL3 (n=4), DNMT3a (n=3), ARID1B (n=3) and ARID2 (n=2), were the most frequent alterations, seen in 11/17 (65%) cases.Disruption of TP53 was identified in 6/18 (33%) cases; biallelic inactivation in 3 cases with co-occurrence of missense (n=2) or nonsense (n=1) mutation and 17p LOH, monoallelic missense mutations in two cases, and loss of 17p encompassing the p53 locus in one case.Mutually exclusive activating STAT3 and STAT5B mutations were identified in 5/17 (29%) cases. Mutations targeting the STAT3 SH2 domain were seen in one EBV- cutaneous anaplastic large cell lymphoma and two EBV+ PTLDs (PTCL, NOS and extranodal NK/T cell lymphoma (ENKTCL)). Both EBV+ cases also showed LOH of the STAT3 locus on 17q and 6q loss encompassing PTPRK, the phosphorylase responsible for downregulation of STAT3 signaling. A JAK3 variant (S568P) adjacent a known hotspot was also identified in the ENKTCL.Missense mutations in RHOA and FYN were detected in one angioimmunoblastic T-cell lymphoma (AITL). Mutations were also observed in genes not previously implicated in T/NK-cell lymphomas, including potentially damaging missense variants in MED12 (n=3) and TBX3 (n=3), and a novel missense MTOR mutation (W1456L) involving the FAT domain.Recurrent copy number alterations comprised gains of 7q (n=4, including 2 hepatosplenic T-cell lymphomas [HSTCL]), 8q (n=3; MYC) and 9q (n=2; ABL1, SYK), and losses of 7p (n=4) and 6q (n=3; FOXO3, PRDM1).Contrary to prior reports, MSI was uncommon in our series despite use of azathioprine therapy (n=9), with only 1 of 3 cases harboring mutations in mismatch repair genes was confirmed to be MSI-H.ConclusionOur analysis of a large series of T/NK-PTLDs represents the first survey of genomic and genetic changes in these lymphomas, and adds to the growing list of genes and pathways altered in T/NK-cell neoplasms. Our findings suggest shared genomic and molecular aberrations with T-cell lymphomas occurring in "immunocompetent" individuals. Functional studies are needed to assess if some of the observed genetic alterations could serve as therapeutic targets for these aggressive neoplasms. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

Significance of a Critical Set of 11q Chromosome Aberrations for Diagnosis of MYC Negative Burkitt Lymphoma

Background: Burkitt lymphoma (BL) is characterized by a non-specific morphology and immunophenotype, a high proliferation rate, MYC rearrangements (MYC +), and by a simple karyotype. However, 5% of BL cases have no MYC rearrangements (MYC -) detectable by FISH. It is a matter of debate whether a true MYC (-) BL does exist.The WHO 2008 classification does not clearly define MYC (-) BL cases, and such cases are often misdiagnosed and treated as diffuse large B-cell lymphoma (DLBCL). We have previously described a provisional category of aggressive B-cell lymphoma unclassifiable (BCLU) with recurrent chromosome 11q aberrations, referred to as B-NHL(11q), with clinical, pathomorphological, and gene expression profile features typicalof BL,but MYC (-). B-NHLs(11q) carry proximal gains and telomeric losses of 11q. Karyotyping (CC) and FISH defined the gain region as dup(11)(q23q13) involving CCND1, ATM and KMT2A. As we have recently shown, BL and B-NHL(11q) express different levels of CD38 and CD16&CD56, and both have lower levels of miRNA-155, -21 and -26a than DLBCL. Here we describe a series of BL patients with a set of critical 11q aberrations and propose a diagnostic algorithm for a rapid work-up.Methods: Within a group of 82 BL cases diagnosed and treated with the BL protocol at our institution, we identified 15 cases of B-NHL(11q) with BL features and 11q aberrations: MYC (-) in 11(male/female 10/1, median age [range] 24 [18-62]) and MYC (+) in 4 cases (male/female 3/1, median age [range] 36.5 [20-82]). In MYC (-) pts, the disease was confined to a single site in 82%, was bulky (>7 cm) in 64% with diameter >20 cm in 45% of cases. BL, BCLU and DLBCL diagnosis according to WHO 2008 classification was based on histopathological/immunohistochemical examination (HP/IHC), CC, FISH, and clinical characteristics in all pts. For the final evaluation, the flow cytometry (FCM) immunophenotype, array comparative genomic hybridization (aCGH) data, and miRNA expression was assessed on samples obtained by the fine needle aspiration biopsy (FNAB). In the B-NHL(11q) cases we identified 11q duplication, dup(11q), with an inversion (inv) of the duplicated region and a deletion of its telomeric region, referred to as critical set of 11q aberrations, as opposed to non-critical aberration set that did not involve all three changes. B-NHL(11q) cells were evaluated with the panel of antibodies by IHC (CD20/CD10/BCL6/ BCL2/MUM1/MYC/Ki-67/CD43/CD44), and by FCM with CD (19, 20, 22, 23, 52, 79β, 81, 5, 25, 38, 43, 44, 45, 16&56, 56, 52, 62L, 71, 200), FMC7, HLADR, and BCL2. All B-NHL(11q) cases were evaluated by CC, FISH (MYC, BCL2, BCL6, CCND1, ATM, KMT2A and telomeric 11q) and aCGH. The relative positions of CCND1, ATM, and KMT2A within a duplicated region on the aberrant chromosome 11 were used to identify inversions.Results: A median follow-up of MYC (-) B-NHL(11q) pts treated with the BL regimen was 30 months, and 2-yr OS was 72% (95% CI: 45%, 99%). In 53% of B-NHL(11q) pts tingible body macrophages were less pronounced than in classic BL. All MYC (-) and MYC (+) B-NHLs(11q) cases presented the same phenotype and Ki-67 index of 100% and met the IHC criteria for BL. In MYC (-) and MYC (+) B-NHLs(11q) pts, all with a simple or less simple karyotype, we confirmed a critical or non-critical 11q aberrations in 10 and 5 pts, respectively. In 87% of pts we identified dup(11q), of two types: the larger part between 11q12.1 and 11q24.3 bands, and the smaller part between 11q22.3 and 11q24.1, with an additional multiplication of KMT2A inside the duplication region. In 13% of cases an inv without dup(11q) was detected. We found an inv of dup(11q) region in 73% of all cases, and no inv in dup(11q) in 2 MYC (+) cases only. Bulky tumors of >20 cm correlated with increased KMT2A copy number in B-NHLs(11q)cases.Conclusions: B-NHL(11q) cases are clinically homogenous while 11q aberrations are heterogeneous. We believe that BLs MYC (-) do exist. Combination of HP/IHC with FNAB/FCM/CC/FISH is a reliable method for credible diagnosis of BLMYC (-). BLMYC (-) should only be diagnosed in cases where critical 11q aberrations and a simple karyotype are identified. BCLU(11q) or DLBCL(11q) cases should be diagnosed if there are more complex karyotypes accompanied by 11q aberrations of any type. We hypothesize that in BLMYC (-) and other aggressive B-NHL(11q), 11q aberrations may determine clinical and pathomorphological features equivalent to those resulting from MYC rearrangements. DisclosuresWalewski:Gilead: Consultancy, Honoraria, Other: travel, accommodation; Seattle: Other: travel, accommodation; GSK/Novartis: Research Funding; Genetics: Other: travel, accommodation; Celgene: Honoraria, Other: travel, accommodation, Research Funding; Teva: Consultancy, Honoraria; Servier: Consultancy; Karyopharm: Consultancy; Boehringer Ingelheim: Consultancy; Ariad: Consultancy; Takeda: Consultancy, Honoraria, Other; Roche: Consultancy, Honoraria, Other: travel, accommodation, Research Funding; Sanofi: Honoraria, Other: travel, accommodation; Mundipharma: Consultancy, Honoraria, Research Funding; Janssen-Cilag: Consultancy.

Reprofiling Metastatic Samples for Chromosome 9p and 14q Aberrations as a Strategy to Overcome Tumor Heterogeneity in Clear-cell Renal Cell Carcinoma.

Losses of chromosomes 9p and 14q are associated with worse outcomes in patients affected by clear-cell renal cell carcinoma (RCC) and are helpful for prognostic risk stratification. Both chromosomal loci harbor several hot-spot molecular pathways suitable for targeted therapeutic interventions. Intratumor heterogeneity may foster tumor adaptation and therapeutic failure. We sought to investigate the presence of losses of the hot spots of chromosomal loci 9p and 14q in primary clear-cell RCC and matched metastatic tissues. CD10 and CD13 were performed on 7 cases of clear-cell RCC with hematogenous tissue metastases. Cytogenetic fluorescence in situ hybridization analysis was performed on primary and matched metastatic tissues using specific probes mapping the 9p and the 14q loci. The loss of chromosome 9p was observed in 85% of both primary clear-cell RCCs and in matched metastases; 14% showed discordance between primary and matched metastases showing gains. The loss of chromosome 14q was observed in 58% of both primary and matched metastases. Only 3/7 (42%) did show an equal status of loss of chromosome 14q. Heterogeneity of the cytogenetic status between metastatic and primary clear-cell RCCs is observed for the loss of chromosome 14q rather than chromosome 9p. The impact of chromosome 14q cytogenetic status, harboring the HIF1 gene, a major driver for the angiogenenic switch, may drive the efficacy of targeted inhibitors, whereas the loss of chromosome 9p, harboring other hot-spot genes, seems to be related to the metastatic behavior per se, without cytogenetic modulation. Reprofiling the metastatic tissue, as compared with the primary tumor, in patients affected by metastatic RCC could be a novel approach to overcome resistance to VEGF(Rs)-targeting agents.

Genomic Copy Number Alterations in Renal Cell Carcinoma with Sarcomatoid Features

Sarcomatoid changes in renal cell carcinoma are associated with a poorprognosis. The identification of genetic alterations that drive this aggressivephenotype could aid in the development of more effective targeted therapies.In this study we aimed to pinpoint unique copy number alterations in sarcomatoid renal cell carcinoma compared to classical renal cell carcinoma subtypes. Genomic copy number analysis was performed using single nucleotide polymorphism based microarrays on tissue extracted from the tumors of 81 patients who underwent renal mass excision, including 17 with sarcomatoid renal cell carcinoma. Sarcomatoid renal cell carcinoma showed a significantly higher number of copy number alterations than clear cell, papillary and chromophobe renal cell carcinoma (mean 18.0 vs 5.8, 6.5 and 7.2, respectively, p <0.0001). Copy number losses of chromosome arms 9q, 15q, 18p/q and 22q, and gains of 1q and 8q occurred in a significantly higher proportion of sarcomatoid renal cell carcinomas than in the other 3 histologies. Patients with sarcomatoid renal cell carcinoma demonstrated significantly worse overall survival compared to those without that condition on Kaplan-Meier analysis (p = 0.0001). Patients with 9 or more copy number alterations also demonstrated significantly worse overall survival than those with fewer than 9 copy number alterations (p = 0.004). Sarcomatoid changes in renal cell carcinoma are associated with ahigh rate of chromosomal imbalances with losses of 9q, 15q, 18p/q and 22q, andgains of 1q and 8q occurring at significantly higher frequencies in comparison to nonsarcomatoid renal cell carcinoma. Identifying candidate driver genes or tumor suppressor loci in these chromosomal regions may help identify targets for future therapies.

GENO-27GENOMIC CHARACTERIZATION OF PRIMARY CENTRAL NERVOUS SYSTEM LYMPHOMA REVEALS PREDOMINANT NONSYNONYMOUS SOMATIC MUTATIONS OF MYD88 AND PIM1 GENES

Primary central nervous system lymphoma (PCNSL) is a rare malignancy confined to CNS. Despite low tendency of systemic dissemination, its prognosis is poor with median overall survival of 4 years when treated with standard of care high-dose methotrexate-based chemoradiotherapy. Despite majority of PCNSL is diffuse large B-cell lymphoma (DLBCL), its response to therapy and biological behavior differ from the systemic disease, necessitating detailed investigation of its pathogenesis and genetic alterations. Here we performed whole-exome sequencing (WES) on 41 PCNSL and paired normal specimens as well as RNA sequencing. Seven cases were classified as germinal center B-cell-like (GCB) subtype (17%), while 34 cases were non-GCB. Progression-free survival was significantly longer in cases with GCB subtype as in systemic DLBCL (p = 0.027). WES yielded sequencing data with a mean coverage of 158Χ with 98% of the targeted bases covered by ≥20 independent reads. We identified a total of 17,385 somatic mutations, of which 62% were nonsynonymous single nucleotide variations. The predominant nucleotide substitution was a C > T transition (57.8%), especially in the context of the sequence GCG. WES revealed a substantial extent of aberrant somatic hypermutation (aSHM) with a high frequency of nonsynonymous somatic mutations especially in PIM1 (90.2%) and BTG2 (65.9%). Genes mutated other than aSHM include MYD88 (85.4%), most of which being L265P activating mutation. These mutations were not associated with GCB/non-GCB subtypes. Notably, mutations in a number of genes related to the NF-κB signaling pathway were also prominent. Copy number aberration analysis identified gains in chromosomes 1q, 7, 12, and 18q, and a loss of 6q. These results suggest that MYD88 and PIM1 mutations may play an important role in PCNSL development regardless of GCB subtypes, and NF-κB pathway could be the therapeutic targets of this intractable disease.

Next generation sequencing in synovial sarcoma reveals novel gene mutations.

Over 95% of all synovial sarcomas (SS) share a unique translocation, t(X;18), however, they show heterogeneous clinical behavior. We analyzed multiple SS to reveal additional genetic alterations besides the translocation. Twenty-six SS from 22 patients were sequenced for 409 cancer-related genes using the Comprehensive Cancer Panel (Life Technologies, USA) on an Ion Torrent platform. The detected variants were verified by Sanger sequencing and compared to matched normal DNAs. Copy number variation was assessed in six tumors using the Oncoscan array (Affymetrix, USA). In total, eight somatic mutations were detected in eight samples. These mutations have not been reported previously in SS. Two of these, in KRAS and CCND1, represent known oncogenic mutations in other malignancies. Additional mutations were detected in RNF213, SEPT9, KDR, CSMD3, MLH1 and ERBB4. DNA alterations occurred more often in adult tumors. A distinctive loss of 6q was found in a metastatic lesion progressing under pazopanib, but not in the responding lesion. Our results emphasize t(X;18) as a single initiating event in SS and as the main oncogenic driver. Our results also show the occurrence of additional genetic events, mutations or chromosomal aberrations, occurring more frequently in SS with an onset in adults.