Sohn Conference: Pediatric Cancer in a Post-genomic World

Journal Articles

Scott Armstrong

Chen CW, Armstrong SA. Targeting DOT1L and HOX gene expression in MLL-rearranged leukemia and beyond. Exp Hematol. 2015;43(8):673-84.

Chen CW, Koche RP, Sinha AU, et al. DOT1L inhibits SIRT1-mediated epigenetic silencing to maintain leukemic gene expression in MLL-rearranged leukemia. Nat Med. 2015;21(4):335-43.

Danis E, Yamauchi T, Echanique K, et al. Ezh2 controls an early hematopoietic program and growth and survival signaling in early T cell precursor acute lymphoblastic leukemia. Cell Rep. 2016;14(8):1953-65.

Kuhn MW, Armstrong SA. Designed to kill: novel menin-MLL inhibitors target MLL-rearranged leukemia. Cancer Cell. 2015;27(4):431-3.

Riedel SS, Haladyna JN, Bezzant M, et al. MLL1 and DOT1L cooperate with meningioma-1 to induce acute myeloid leukemia. J Clin Invest. 2016;126(4):1438-50.

Wang X, Chen CW, Armstrong SA. The role of DOT1L in the maintenance of leukemia gene expression. Curr Opin Genet Dev. 2016;36:68-72.

Brittany Campbell

Bouffet E, Larouche V, Campbell BB, et al. Immune checkpoint inhibition for hypermutant glioblastoma multiforme resulting from germline biallelic mismatch repair deficiency. J Clin Oncol. 2016. [Epub ahead of print]

Durno CA, Sherman PM, Aronson M, et al. Phenotypic and genotypic characterisation of biallelic mismatch repair deficiency (BMMR-D) syndrome. Eur J Cancer. 2015;51(8):977-83.

Shlien A, Campbell BB, de Borja R, et al. Combined hereditary and somatic mutations of replication error repair genes result in rapid onset of ultra-hypermutated cancers. Nat Genet. 2015;47(3):257-62.

Michael Dyer

Benavente CA, Dyer MA. Genetically engineered mouse and orthotopic human tumor xenograft models of retinoblastoma. Methods Mol Biol. 2015;1267:307-17.

Stewart E, Federico S, Karlstrom A, et al. The Childhood Solid Tumor Network: a new resource for the developmental biology and oncology research communities. Dev Biol. 2016;411(2):287-93.

Stewart E, Shelat A, Bradley C, et al. Development and characterization of a human orthotopic neuroblastoma xenograft. Dev Biol. 2015;407(2):344-55.

Richard Gilbertson

Gajjar A, Pfister SM, Taylor MD, Gilbertson RJ. Molecular insights into pediatric brain tumors have the potential to transform therapy. Clin Cancer Res. 2014;20(22):5630-40.

Gilbertson RJ. Driving glioblastoma to drink. Cell. 2014;157(2):289-90.

Huether R, Dong L, Chen X, et al. The landscape of somatic mutations in epigenetic regulators across 1,000 paediatric cancer genomes. Nat Commun. 2014;5:3630.

Nailor A, Walker DA, Jacques TS, et al. Highlights of Children with Cancer UK's Workshop on Drug Delivery in Paediatric Brain Tumours. Ecancermedicalscience. 2016;10:630.

Phoenix TN, Patmore DM, Boop S, et al. Medulloblastoma genotype dictates blood brain barrier phenotype. Cancer Cell. 2016;29(4):508-22.

Shuning He

Coustan-Smith E, Mullighan CG, Onciu M, et al. Early T-cell precursor leukaemia: a subtype of very high-risk acute lymphoblastic leukaemia. Lancet Oncol. 2009;10(2):147-56.

Langenau DM, Traver D, Ferrando AA, et al. Myc-induced T cell leukemia in transgenic zebrafish. Science. 2003;299(5608):887-90.

Lee J. Helman

Khan J, Helman LJ. Precision therapy for pediatric cancers. JAMA Oncol. 2016. [Epub ahead of print]

Pappo AS, Furman WL, Schultz KA, et al. Rare tumors in children: progress through collaboration. J Clin Oncol. 2015;33(27):3047-54. Review.

Thiele CJ, Helman LJ. Preface: the status of pediatric solid tumors in 2015. Crit Rev Oncog. 2015;20(3-4):v-vi.

Nada Jabado

Fontebasso AM, Jabado N. Pediatric brain tumors: genomics and epigenomics pave the way. Crit Rev Oncog. 2015;20(3-4):271-99.

Fontebasso AM, Shirinian M, Khuong-Quang DA, et al. Non-random aneuploidy specifies subgroups of pilocytic astrocytoma and correlates with older age. Oncotarget. 2015;6(31):31844-56.

Jones C, Karajannis MA, Jones DT, et al. Pediatric high-grade glioma: biologically and clinically in need of new thinking. Neuro Oncol. 2016. [Epub ahead of print]

Lu C, Jain SU, Hoelper D, et al. Histone H3K36 mutations promote sarcomagenesis through altered histone methylation landscape. Science. 2016;352(6287):844-9.

Nikbakht H, Panditharatna E, Mikael LG, et al. Spatial and temporal homogeneity of driver mutations in diffuse intrinsic pontine glioma. Nat Commun. 2016;7:11185.

Katherine A. Janeway

Harris MH, DuBois SG, Glade Bender JL, et al. Multicenter feasibility study of tumor molecular profiling to inform therapeutic decisions in advanced pediatric solid tumors: The Individualized Cancer Therapy (iCat) Study. JAMA Oncol. 2016. [Epub ahead of print]

Janeway KA, Place AE, Kieran MW, Harris MH. Future of clinical genomics in pediatric oncology. J Clin Oncol. 2013;31(15):1893-903.

Kris MG, Johnson BE, Berry LD, et al. Using multiplexed assays of oncogenic drivers in lung cancers to select targeted drugs. JAMA. 2014;311(19):1998-2006.

Mody RJ, Wu YM, Lonigro RJ, et al. Integrative clinical sequencing in the management of refractory or relapsed cancer in youth. JAMA. 2015;314(9):913-25.

Michael C. Jensen

Jensen MC, Riddell SR. Design and implementation of adoptive therapy with chimeric antigen receptor-modified T cells. Immunol Rev. 2014;257(1):127-44.

Jensen MC, Riddell SR. Designing chimeric antigen receptors to effectively and safely target tumors. Curr Opin Immunol. 2015;33:9-15.

Turtle CJ, Hanafi LA, Berger C, et al. CD19 CAR-T cells of defined CD4+:CD8+ composition in adult B cell ALL patients. J Clin Invest. 2016;126(6):2123-38.

Wang X, Popplewell LL, Wagner JR, et al. Phase I studies of central-memory-derived CD19 CAR T cell therapy following autologous HSCT in patients with B-cell NHL. Blood. 2016. [Epub]

Zah E, Lin MY, Silva-Benedict A, et al. T cells expressing CD19/CD20 bispecific chimeric antigen receptors prevent antigen escape by malignant B cells. Cancer Immunol Res. 2016;4(6):498-508.

Javed Khan

Chang W, Brohl AS, Patidar R, et al. MultiDimensional ClinOmics for precision therapy of children and adolescent young adults with relapsed and refractory cancer: a report from the Center for Cancer Research. Clin Cancer Res. 2016. [Epub ahead of print]

Hettmer S, Li Z, Billin AN, et al. Rhabdomyosarcoma: current challenges and their implications for developing therapies. Cold Spring Harb Perspect Med. 2014;4(11):a025650.

Li SQ, Cheuk AT, Shern JF, et al. Targeting wild-type and mutationally activated FGFR4 in rhabdomyosarcoma with the inhibitor ponatinib (AP24534). PLoS One. 2013;8(10):e76551.

Shern JF, Chen L, Chmielecki J, et al. Comprehensive genomic analysis of rhabdomyosarcoma reveals a landscape of alterations affecting a common genetic axis in fusion-positive and fusion-negative tumors. Cancer Discov. 2014;4(2):216-31.

Shern JF, Yohe ME, Khan J. Pediatric rhabdomyosarcoma. Crit Rev Oncog. 2015;20(3-4):227-43.

Andrew Kung

Aparicio S, Hidalgo M, Kung AL. Examining the utility of patient-derived xenograft mouse models. Nat Rev Cancer. 2015;15(5):311-6.

Oberg JA, Glade Bender JL, Cohn EG, et al. Overcoming challenges to meaningful informed consent for whole genome sequencing in pediatric cancer research. Pediatr Blood Cancer. 2015;62(8):1374-80.

Oberg JA, Sireci AN, Mansukhani MM, et al. Clinical implementation of genomic sequencing in pediatric oncology: identification and valuation of resources and costs associated with next-generation sequencing. Value Health. 2014;17(7):A645.

Yu C, Mannan AM, Yvone GM, et al. High-throughput identification of genotype-specific cancer vulnerabilities in mixtures of barcoded tumor cell lines. Nat Biotechnol. 2016;34(4):419-23.

Alex Kentsis

Henssen AG, Henaff E, Jiang E, et al. Genomic DNA transposition induced by human PGBD5. Elife. 2015;4.

Ortiz MV, Kobos R, Walsh M, et al. Integrating genomics into clinical pediatric oncology using the molecular tumor board at the Memorial Sloan Kettering Cancer Center. Pediatr Blood Cancer. 2016. [Epub ahead of print]

Gabriel Leprevier

Leprevier G, Sorensen PH. How does oncogene transformation render tumor cells hypersensitive to nutrient deprivation? Bioessays. 2014;36(11):1082-90.

Leprivier G, Remke M, Rotblat B, et al. The eEF2 kinase confers resistance to nutrient deprivation by blocking translation elongation. Cell. 2013;153(5):1064-79.

Leprivier G, Rotblat B, Khan D, et al. Stress-mediated translational control in cancer cells. Biochim Biophys Acta. 2015;1849(7):845-60.

A. Thomas Look

He S, Mansour MR, Zimmerman MW, et al. Synergy between loss of NF1 and overexpression of MYCN in neuroblastoma is mediated by the GAP-related domain. Elife. 2016;5.

Shin J, Padmanabhan A, de Groh ED, et al. Zebrafish neurofibromatosis type 1 genes have redundant functions in tumorigenesis and embryonic development. Dis Model Mech. 2012;5(6):881-94.

Zhu S, Lee JS, Guo F, et al. Activated ALK collaborates with MYCN in neuroblastoma pathogenesis. Cancer Cell. 2012;21(3):362-73.

Chao Lu

Bechet D, Gielen GG, Korshunov A, et al. Specific detection of methionine 27 mutation in histone 3 variants (H3K27M) in fixed tissue from high-grade astrocytomas. Acta Neuropathol. 2014;128(5):733-41.

Lu C, Jain SU, Hoelper D, et al. Histone H3K36 mutations promote sarcomagenesis through altered histone methylation landscape. Science. 2016;352(6287):844-9.

Lu C, Thompson CB. Metabolic regulation of epigenetics. Cell Metab. 2012;16(1):9-17.

Lu C, Venneti S, Akalin A, et al. Induction of sarcomas by mutant IDH2. Genes Dev. 2013;27(18):1986-98.

Venneti S, Felicella MM, Coyne T, et al. Histone 3 lysine 9 trimethylation is differentially associated with isocitrate dehydrogenase mutations in oligodendrogliomas and high-grade astrocytomas. J Neuropathol Exp Neurol. 2013;72(4):298-306.

David C. Lyden

André Mdo R, Pedro A, Lyden D. Cancer exosomes as mediators of drug resistance. Methods Mol Biol. 2016;1395:229-39.

Barcellos-Hoff MH, Lyden D, Wang TC. The evolution of the cancer niche during multistage carcinogenesis. Nat Rev Cancer. 2013;13(7):511-8.

Costa-Silva B, Aiello NM, Ocean AJ, et al. Pancreatic cancer exosomes initiate pre-metastatic niche formation in the liver. Nat Cell Biol. 2015;17(6):816-26.

Hoshino A, Costa-Silva B, Shen TL, et al. Tumour exosome integrins determine organotropic metastasis. Nature. 2015;527(7578):329-35.

Thakur BK, Zhang H, Becker A, et al. Double-stranded DNA in exosomes: a novel biomarker in cancer detection. Cell Res. 2014;24(6):766-9.

Elizabeth Maher

Choi C, Ganji SK, Madan A, et al. In vivo detection of citrate in brain tumors by 1H magnetic resonance spectroscopy at 3T. Magn Reson Med. 2014;72(2):316-23.

Choi C, Ganji SK, DeBerardinis RJ, et al. 2-hydroxyglutarate detection by magnetic resonance spectroscopy in IDH-mutated patients with gliomas. Nat Med. 2012;18(4):624-9.

De la Fuente MI, Young RJ, Rubel J, et al. Integration of 2-hydroxyglutarate-proton magnetic resonance spectroscopy into clinical practice for disease monitoring in isocitrate dehydrogenase-mutant glioma. Neuro Oncol. 2016;18(2):283-90.

Mashimo T, Pichumani K, Vemireddy V, et al. Acetate is a bioenergetic substrate for human glioblastoma and brain metastases. Cell. 2014;159(7):1603-14.

David Malkin

Alderfer MA, Zelley K, Lindell RB, et al. Parent decision-making around the genetic testing of children for germline TP53 mutations. Cancer. 2015;121(2):286-93.

Id Siad B, Malkin D. A functional variant in miR-605 modifies the age of onset in Li-Fraumeni syndrome. Cancer Genet. 2015;208(1-2):47-51.

Malkin D, Garber JE, Strong LC, Friend SH. CANCER. The cancer predisposition revolution. Science. 2016;352(6289):1052-3.

Morrissy AS, Garzia L, Shih DJ, et al. Divergent clonal selection dominates medulloblastoma at recurrence. Nature. 2016;529(7586):351-7.

Said BI, Kim H, Tran J, et al. Super-transactivation TP53 variant in the germline of a family with Li-Fraumeni syndrome. Hum Mutat. 2016. [Epub ahead of print]

John M. Maris

Bosse KR, Maris M. Advances in the translational genomics of neuroblastoma: From improving risk stratification and revealing novel biology to identifying actionable genomic alterations. Cancer. 2016;122(1):20-33.

Eleveld TF, Oldridge DA, Bernard V, et al. Relapsed neuroblastomas show frequent RAS-MAPK pathway mutations. Nat Genet. 2015;47(8):864-71.

Maris JM. Defining why cancer develops in children. N Engl J Med. 2015;373(24):2373-5.

Schnepp RW, Bosse KR, Maris JM. Improving patient outcomes with cancer genomics: unique opportunities and challenges in pediatric oncology. JAMA. 2015;314(9):881-3.

Schnepp RW, Khurana P, Attiyeh EF, et al. A LIN28B-RAN-AURKA signaling network promotes neuroblastoma tumorigenesis. Cancer Cell. 2015;28(5):599-609.

Yael Mossé

Bresler SC, Weiser DA, Huwe PJ, et al. ALK mutations confer differential oncogenic activation and sensitivity to ALK inhibition therapy in neuroblastoma. Cancer Cell. 2014;26(5):682-94.

Carpenter EL, Mossé YP. Targeting ALK in neuroblastoma—preclinical and clinical advancements. Nat Rev Clin Oncol. 2012;9(7):391-9.

DuBois SG, Marachelian A, Fox E, et al. Phase I Study of the aurora a kinase inhibitor Alisertib in combination with irinotecan and temozolomide for patients with relapsed or refractory neuroblastoma: A NANT (New Approaches to Neuroblastoma Therapy) Trial. J Clin Oncol. 2016;34(12):1368-75.

Mossé YP. Anaplastic lymphoma kinase as a cancer target in pediatric malignancies. Clin Cancer Res. 2016;22(3):546-52.

Norris RE, Shusterman S, Gore L, et al. Time to disease progression in children with relapsed or refractory neuroblastoma treated with ABT-751: a report from the Children's Oncology Group (ANBL0621). Pediatr Blood Cancer. 2014;61(6):990-6.

Kathy Pritchard-Jones

Charlton J, Pavasovic V, Pritchard-Jones K. Biomarkers to detect Wilms tumors in pediatric patients: where are we now? Future Oncol. 2015;11(15):2221-34.

Charlton J, Williams RD, Sebire NJ, et al. Comparative methylome analysis identifies new tumour subtypes and biomarkers for transformation of nephrogenic rests into Wilms tumour. Genome Med. 2015;7(1):11.

Dome JS, Graf N, Geller JI, et al. Advances in Wilms tumor treatment and biology: progress through international collaboration. J Clin Oncol. 2015;33(27):2999-3007.

Pritchard-Jones K, Graf N, van Tinteren H, Craft A. Evidence for a delay in diagnosis of Wilms' tumour in the UK compared with Germany: implications for primary care for children. Arch Dis Child. 2016;101(5):417-20.

Water AM, Pritchard-Jones K. Paediatrics: long-term effects of Wilms tumour therapy on renal function. Nat Rev Urol. 2015;12(8):423-4.

Zulekha A. Qadeer

Cheung NK, Dyer MA. Neuroblastoma: developmental biology, cancer genomics and immunotherapy. Nat Rev Cancer. 2013;13(6):397-411.

Wang C, Liu Z, Woo CW, et al. EZH2 Mediates epigenetic silencing of neuroblastoma suppressor genes CASZ1, CLU, RUNX3, and NGFR. Cancer Res. 2012;72(1):315-24.

Charles W. M. Roberts

Helming KC, Wang X, Roberts CW. Vulnerabilities of mutant SWI/SNF complexes in cancer. Cancer Cell. 2014;26(3):309-17.

Kim KH, Kim W, Howard TP, et al. SWI/SNF-mutant cancers depend on catalytic and non-catalytic activity of EZH2. Nat Med. 2015;21(12):1491-6.

Wang X, Haswell JR, Roberts CW. Molecular pathways: SWI/SNF (BAF) complexes are frequently mutated in cancer—mechanisms and potential therapeutic insights. Clin Cancer Res. 2014;20(1):21-7.

Wang X, Roberts CW. CARMA: CARM1 methylation of SWI/SNF in breast cancer. Cancer Cell. 2014;25(1):3-4.

Wilson BG, Roberts CW. SWI/SNF nucleosome remodellers and cancer. Nat Rev Cancer. 2011;11(7):481-92.

Michel Sadelain

Davila ML, Sadelain M. Biology and clinical application of CAR T cells for B cell malignancies. Int J Hematol. 2016. [Epub ahead of print]

Sadelain M. CAR therapy: the CD19 paradigm. J Clin Invest. 2015;125(9):3392-400.

Sadelain M. Tales of antigen evasion from CAR therapy. Cancer Immunol Res. 2016;4(6):473.

Sun J, Sadelain M. The quest for spatio-temporal control of CAR T cells. Cell Res. 2015;25(12):1281-2.

Zhao Z, Condomines M, van der Stegen SJ, et al. Structural design of engineered costimulation determines tumor rejection kinetics and persistence of CAR T cells. Cancer Cell. 2015;28(4):415-28.

Paul M. Sondel

Goldberg JL, Sondel P. Enhancing cancer immunotherapy via activation of innate immunity. Semin Oncol. 2015;42(4):562-72.

McDowell KA, Hank JA, DeSantes KB, et al. NK cell-based immunotherapies in pediatric oncology. J Pediatr Hematol Oncol. 2015;37(2):79-93.

Morris ZS, Guy EI, Francis DM, et al. In situ tumor vaccination by combining local radiation and tumor-specific antibody or immunocytokine treatments. Cancer Res. 2016. [Epub ahead of print]

Neri D, Sondel PM. Immunocytokines for cancer treatment: past, present and future. Curr Opin Immunol. 2016;40:96-102.

Wang W, Erbe AK, Hank JA, et al. NK cell-mediated antibody-dependent cellular cytotoxicity in cancer immunotherapy. Front Immunol. 2015;6:368.

Poul H. Sorensen

El-Naggar AM, Veinotte CJ, Cheng H, et al. Translational activation of HIF1α by YB-1 promotes sarcoma metastasis. Cancer Cell. 2015;27(5):682-97.

Hingorani P, Janeway K, Crompton BD, et al. Current state of pediatric sarcoma biology and opportunities for future discovery: a report from the sarcoma translational research workshop. Cancer Genet. 2016;209(5):182-94.

Qadir MA, Zhan SH, Kwok B, et al. ChildSeq-RNA: A next-generation sequencing-based diagnostic assay to identify known fusion transcripts in childhood sarcomas. J Mol Diagn. 2014;16(3):361-70.

Somasekharan SP, El-Naggar A, Leprivier G, et al. YB-1 regulates stress granule formation and tumor progression by translationally activating G3BP1. J Cell Biol. 2015;208(7):913-29.

Kimberley Stegmaier

Liu S, Yin L, Stroopinsky D, et al. MUC1-C oncoprotein promotes FLT3 receptor activation in acute myeloid leukemia cells. Blood. 2014;123(5):734-42.

Neumann T, Benajiba L, Göring S, et al. Evaluation of improved glycogen synthase kinase-3α inhibitors in models of acute myeloid leukemia. J Med Chem. 2015;58(22):8907-19.

Roti G, Carlton A, Ross KN, et al. Complementary genomic screens identify SERCA as a therapeutic target in NOTCH1 mutated cancer. Cancer Cell. 2013;23(3):390-405.

Roti G, Stegmaier K. New approaches to target T-ALL. Front Oncol. 2014;4:170.

Roti G, Stegmaier K. Targeting NOTCH1 in hematopoietic malignancy. Crit Rev Oncog. 2011;16(1-2):103-15.

Michael Taylor

Morrissy AS, Garzia L, Shih DJ, et al. Divergent clonal selection dominates medulloblastoma at recurrence. Nature. 2016;529(7586):351-7.

Northcott PA, Shih DJ, Peacock J, et al. Subgroup-specific structural variation across 1,000 medulloblastoma genomes. Nature. 2012;488(7409):49-56.

Pei Y, Liu KW, Wang J, et al. HDAC and PI3K antagonists cooperate to inhibit growth of MYC-driven medulloblastoma. Cancer Cell. 2016;29(3):311-23.

Ramaswamy V, Remke M, Bouffet E, et al. Risk stratification of childhood medulloblastoma in the molecular era: the current consensus. Acta Neuropathol. 2016;131(6):821-31.

Thompson EM, Hielscher T, Bouffet E, et al. Prognostic value of medulloblastoma extent of resection after accounting for molecular subgroup: a retrospective integrated clinical and molecular analysis. Lancet Oncol. 2016;17(4):484-95.

Craig B. Thompson

Carey BW, Finley LW, Cross JR, et al. Intracellular α-ketoglutarate maintains the pluripotency of embryonic stem cells. Nature. 2015;518(7539):413-6.

Finley LW, Zhang J, Ye J, et al. SnapShot: cancer metabolism pathways. Cell Metab. 2013;17(3):466-466.e2.

Intlekofer AM, Dematteo RG, Venneti S, et al. Hypoxia induces production of L-2-hydroxyglutarate. Cell Metab. 2015;22(2):304-11.

Lu C, Jain SU, Hoelper D, et al. Histone H3K36 mutations promote sarcomagenesis through altered histone methylation landscape. Science. 2016;352(6287):844-9.

Pavlova NN, Thompson CB. The emerging hallmarks of cancer metabolism. Cell Metab. 2016;23(1):27-47.

Matthew Vander Heiden

Davidson SM, Papagiannakopoulos T, Olenchock BA, et al. Environment impacts the metabolic dependencies of Ras-driven non-small cell lung cancer. Cell Metab. 2016;23(3):517-28.

Hirschey MD, DeBerardinis RJ, Diehl AM, et al. Dysregulated metabolism contributes to oncogenesis. Semin Cancer Biol. 2015;35 Suppl:S129-50.

Son S, Stevens MM, Chao HX, et al. Cooperative nutrient accumulation sustains growth of mammalian cells. Sci Rep. 2015;5:17401.

Yuan C, Clish CB, Wu C, et al. Circulating metabolites and survival among patients with pancreatic cancer. J Natl Cancer Inst. 2016;108(6):djv409.

Robert Wechsler-Reya

Becher OJ, Wechsler-Reya R. Cancer. For pediatric glioma, leave no histone unturned. Science. 2014;346(6216):1458-9.

Langenau DM, Sweet-Cordero A, Wechsler-Reya RJ, Dyer MA. Preclinical models provide scientific justification and translational relevance for moving novel therapeutics into clinical trials for pediatric cancer. Cancer Res. 2015;75(24):5176-86.

Pei Y, Liu KW, Wang J, et al. HDAC and PI3K antagonists cooperate to inhibit growth of MYC-driven medulloblastoma. Cancer Cell. 2016;29(3):311-23.

Pei Y, Moore CE, Wang J, et al. An animal model of MYC-driven medulloblastoma. Cancer Cell. 2012;21(2):155-67.

Rusert JM, Wu X, Eberhart CG, et al. Snapshot: medulloblastoma. Cancer Cell. 2014;26(6):940-940.e1.

Sponsors

Presented by

Highlights

Transition points in the patient journey, such as from asymptomatic disease to presentation of symptoms, from initial tests to working diagnosis, from untreated to starting a treatment plan, and from remission to relapse, offer opportunities to improve patient outcomes.

Next-generation sequencing can be deployed in the clinic, with sequencing data informing decision-making at the predictive, diagnostic, and prognostic levels and offering treatment targets.

Single-agent targeted therapies are unlikely to achieve cure, so precision therapies with combinations of agents against multiple targets are being investigated, using multi-dimensional "ClinOmics" approaches.

Successes and future directions for research and therapy

The improved diagnoses and outcomes that have been achieved in adult cancers can be attributed to advances in genetics and genomic technology, but such success has not yet been achieved in pediatric cancers. This field, according to Richard Gilbertson of the University of Cambridge, has fallen behind in integrating genomics and cell biology into the search for new treatments and the practice of clinical decision-making.

According to Gilbertson, success in pediatric cancers has followed from four developments: the realization that (1) medulloblastoma, leukemia, and other childhood cancers are not single diseases but rather (2) distinct diseases, with various cellular and molecular origins, (3) each with unique, clinical characteristics, and varied prognoses, (4) leading to the development of tailored and targeted therapies that have improved treatment outcomes over the long term.

The long-standing but long-dormant idea that molecular pathways gone awry in development are at the root of childhood cancers is now an active area of research. Gilbertson cited several studies from his and other groups that investigated the relationship between the capacity of stem cells to make an entire organ (stemness) and the capacity of these cells to give rise to tumors. These data, which hint at the intrinsic resistance of pediatric stem cells to cancer, might reveal mechanisms that could be used to prevent adult cancers.

Transition points in the patient journey, such as from asymptomatic disease to presentation of symptoms, could be central to genomics-driven advances in treatment outcomes. (Image courtesy of Richard Gilbertson)

Gilbertson also identified four challenges in pediatric cancer. He argued that opportunities for advances lie in decisions at transition points in the disease, such as the transition from asymptomatic to symptomatic disease, a time of opportunity for (1) early diagnosis, which is not currently available in pediatric cancers. In the eight adult cancers with the highest mortality rates, early diagnosis at stage 1 or 2 improves 10-year survival compared to late diagnosis at stage 3 or 4 by 4-fold (81% vs. 22% survival), highlighting the benefit of early diagnosis even without new treatments.

(2) Knowledge of disease biology is seldom applied in the diagnostic process, which is outdated and, Gilbertson argued, (3) does not incorporate preclinical data from humanized mouse models of disease. Those models can recapitulate tumor responses to conventional treatments and be used to assess the effectiveness of new treatments.

(4) Post-treatment disease management is another area for improvement. Clinicians currently treat relapsed or recurring disease after tumors have become bulky and so molecularly heterogeneous that targeted therapy is no longer useful. Gilbertson stressed the need to routinely screen patients for circulating tumor DNA, tumor cells, and other tumor macromolecules, which can detect relapsing breast cancer 9 months before patients manifest clinical symptoms. These methods, along with advanced imaging such as hyperpolarized magnetic resonance imaging (MRI), could be used to proactively and noninvasively monitor for drug resistance, detect subclinical disease, and track metastases, allowing clinicians to re-treat pediatric patients before the disease becomes unmanageable.

Gilbertson closed by calling for the pathology community to embrace genomics, routine methylation profiling, and transcriptomics in clinical practice. He urged the adult and pediatric research communities to abandon silos, and particularly to share knowledge in genomics.

Precision medicine: the PIPseq experience

Cancer research is in a post-genomic era. Andrew Kung, who directs the Precision in Pediatric Sequencing (PIPseq) program at Columbia University, described how the field has moved beyond large-scale genome-wide association studies to employ individual genome-wide causation studies of direct benefit to one patient.

The PIPseq program began in 2014, aiming to bring next-generation sequencing to the clinic, with regulatory compliance and state approval, for routine use in treatment decision-making for all cancer patients. It is a cross-disciplinary effort of several medical departments, including pediatric oncology, molecular pathology, surgery, bioinformatics, medical genetics, and bioethics. Whole-exome (all expressed genes) and transcriptome (RNA transcripts) sequencing of cancerous and normal tissue is available for all high-risk and relapsed patients.

The Precision in Pediatric Sequencing (PIPseq) cancer medicine program. (Image courtesy of Andrew Kung)

Kung's case studies show that the results of this testing influenced clinical decisions in about 65% of patients. Forty percent of the cancer-related genetic alterations identified in those patients were actionable, meaning that clinicians could use the data to choose specific, targeted drugs for patients who otherwise would have had no treatment options. Genome sequencing data also pointed to non-actionable alterations in about 3% of patients, showing which treatments would be ineffective.

In a quarter of patients, sequencing data resolved a questionable diagnosis and provided information to add to the biological and mechanistic understanding of pediatric cancers. In 13% of cases, sequencing data identified germline mutations that revealed cancer susceptibility, providing an opportunity to monitor family members for cancer.

Cancer genomics, novel biomarkers and drivers, and precision therapeutics

The National Cancer Institute's Javed Khan, an expert on rhabdomyosarcoma (RMS) and other pediatric cancers, began by recapping how first-generation genomics helped identify FGFR4, a gene for a cell-surface receptor tyrosine kinase, as a cancer-causing driver gene in RMS. Khan now uses next-generation genomics to uncover other driver genes and to understand how the proteins they encode interact in this cancer's signaling networks. A single therapeutic agent is unlikely to work against a cancer with mutations that, working through multiple pathways, create built-in redundancies; network analysis generates data that helps pinpoint a combination of precision therapies against the cancer. Khan's group used an integrated, 'omics approach—which accounted for the timing of new somatic and germline alterations, for the proteomic landscape of RMS, and for the clonal heterogeneity of the tumors—to chart the evolutionary history of cancer in individual patients.

A multidimensional "ClinOmics" approach combines various sequencing methods with multidisciplinary treatment in real-time. It is being implemented in precision therapy that targets multiple actionable mutations in tumors. (Image courtesy of Javed Khan)

Researchers recently discovered that pediatric cancers have a low mutational burden at diagnosis, with few actionable alterations, but a higher mutational burden at recurrence, when up to 50% of mutations are actionable in somatic cells and up to 10% are actionable in germline cells. These findings, along with data from studies like Khan's, spurred the NCI to develop a multidimensional "ClinOmics" platform. Researchers will conduct multiple types of sequencing on pediatric cancer biopsies, generating data on exome and transcriptome alterations and on single nucleotide polymorphisms (SNPs), which are useful for research and for treatment decisions. The platform is being tested in a so-called umbrella trial for precision therapy, in patients with 20 different cancers.

Highlights

Mutations in one subunit of the multimodular SWI/SNF chromatin remodeling complex may create vulnerabilities in other components of the complex, which could be targeted therapeutically.

Understanding how modifications of histone complexes control gene expression in leukemias has helped researchers identify new mechanisms to target.

In some types of aggressive pediatric brain tumors, cancer-driving histone mutations arise first and occur along with specific oncogenic mutations that arise later, suggesting a need for new therapeutic strategies to target these oncohistone partnerships.

Identifying and attacking vulnerabilities in SWI/SNF-mutated cancers

Chromatin regulatory complexes—macromolecules that package DNA in cell nucleus—control genes expression, chromosome integrity, and developmental fate decisions. Changes in these epigenetic mediators can lead to cancer even if DNA itself is not mutated or damaged. Epigenetic mechanisms such as chromatin remodeling and histone modifications are involved in various pediatric cancers.

The SWI/SNF complex is a chromatin remodeler that controls gene expression by modifying the architecture of tightly condensed genomic DNA (heterochromatin) to allow access to transcription machinery. The complex is also a tumor suppressor, and researchers have found mutations that dismantle this activity in about 20% of adult and pediatric cancers.

Charles Roberts of St. Jude Children's Research Hospital is studying how inactivation of the complex's SNF5 subunit leads to a highly aggressive type of pediatric cancer called rhabdoid tumor. SNF5 loss causes tumors to become dependent on other chromatin regulators, such as EZH2, an enzyme that normally methylates DNA and silences gene expression but whose mutation or overexpression can fuel cancer growth. Small molecule inhibitors of EZH2 are in phase I clinical trials targeting SNF5-mutant pediatric and adult cancers.

The EZH2 inhibitor tazemetostat demonstrated clinical activity in a phase I trial in patients with relapsed or refractory solid tumors. (Image courtesy of Charles Roberts)

Zulekha Qadeer, a graduate student at the Icahn School of Medicine at Mount Sinai, gave a hot topic talk on the epigenetic consequences of mutations in an SNF5-like chromatin remodeler called ATRX in pediatric neuroblastoma. ATRX normally recruits the EZH2 enzyme to deposit methylation marks and thereby silence genes. Qadeer's work suggests that ATRX mutations prevent EZH2 recruitment and promote a deregulated chromatin state that can promote sensitivity to EZH2 inhibitors.

Aberrant histone modification, a new therapeutic target

The main protein components of chromatin are histones, which act as spools around which DNA is wound. Histone tails are susceptible to methylation, phosphorylation, and other chemical modifications by various enzymes; these modifications affect histones' ability to regulate genes.

Scott Armstrong of Memorial Sloan Kettering Cancer Center studies histone modifications that control gene expression in MLL-fusion leukemia, an acute form of the disease that occurs in 70% of mixed-lineage leukemia (MLL) cases. H3K79 methylation (methylation of lysine at position 79 of histone H3) is carried out by the enzyme DOT1L, and researchers are testing DOT1L small molecule inhibitors as therapeutic agents in MLL-fusion leukemias. Patients in one small study had complete remission after treatment with a DOT1L small molecule inhibitor. Armstrong's team is also working on boosting the anti-leukemic efficacy of these inhibitors by combining them with other inhibitors that block the activity of an MLL-fusion partner protein called menin.

Diffuse intrinsic pontine gliomas (DIPG) are aggressive pediatric brain tumors with poor survival outcomes. Nada Jabado of McGill University described the work of her group and others to identify two age-dependent mutations, in the lysine tails of histones H3.1 and H3.3, driving these tumors. Spatial and temporal analysis of autopsy samples via deep sequencing shows that these mutations arise first, in particular locations in the brain, and always partner with later mutations in growth factor, tumor suppressor, or other genes. Stopping aggressive DIPG tumors might thus require targeting these oncohistone partnerships, rather than targeting single growth factor or tumor suppressor genes.

In diffuse intrinsic pontine gliomas, driver mutations in two oncohistones occur in partnership with specific mutations in other cancer-driving pathways. (Image courtesy of Nada Jabado)

Chao Lu, a postdoctoral fellow in David Allis's laboratory at the Rockefeller University, presented a hot topic talk on H3K36M histone mutations (lysine-to-methionine at position 36 in histone H3.3), which impair chondrocyte (cartilage cell) differentiation and can lead to mesenchymal sarcoma in children. Mesenchymal stem cells (MSCs) are the progenitors of chondrocytes, and in normal MSCs, PRC1—a chromatin remodeling protein complex that normally represses genes during development—assists development by masking or silencing certain genes. The H3K36M mutations disrupt normal PRC1 binding, restructuring chromatin by opening new domains that compete for PRC1 binding. The mutations thereby "unmask" formerly repressed genes to prevent chrondrocyte differentiation and initiate sarcomagenesis.

Highlights

Efforts to understand the genetic basis of neuroblastoma have identified several predisposing genes and a tumor-initiating signaling network that is clinically relevant.

Knowledge of the genomic landscape and cancer-predisposing mechanisms in Wilms tumor has opened the possibility of developing early diagnosis and prevention strategies.

Clinical monitoring of individuals carrying cancer-predisposing mutations combined with next-generation sequencing could help clinicians detect cancers before clinical manifestation.

Tumors with a high mutational burden, such as those driven by biallelic mismatch repair deficiency (bMMRD), have a higher neoantigen load and are more susceptible to immunotherapy with checkpoint inhibitors such as anti-PD-1 agents.

Unraveling cancer-predisposing gene networks

Unlike adult cancers, childhood cancers, particularly brain tumors, often arise alongside cancer predisposition syndromes. The genetics of cancer predisposition elucidates how cancer develops in children and guides not only treatment but also familial pre-cancer monitoring and counseling. Family history is not a reliable indicator of cancer predisposition: in the most recent proof, a next-generation sequencing study of predisposing germline mutations in 1120 pediatric patients found no family history of cancer for more than half of children with the mutations.

John Maris of the Children's Hospital of Philadelphia and the University of Pennsylvania explained his group's work to understand how predisposing and subsequent mutations together initiate and establish pediatric cancers such as neuroblastoma. Germline mutations in ALK and PHOX2B genes were the only known predisposing mutations for neuroblastoma until genome-wide association study (GWAS) and zebrafish modeling identified several new predisposing genes and a model for how the LIN28-RAN-AURKA signaling network promotes neuroblastoma tumorigenesis.

Predisposing mutations to neuroblastoma such as ALK mutations are rare but have a large effect on tumorigenesis. (Image courtesy of John Maris)

Kathy Pritchard-Jones of the Great Ormond Street Hospital for Children NHS Foundation Trust and University College London studies genetic predisposition to cancer in Wilms tumor, which can occur in one or both (bilateral) lobes of the kidneys. More than 80% of Wilms tumor cases are "sporadic," without a known family history. Up to 20% of cases are associated with a WT1 gene mutation, and 70%, with a chromosomal alteration, 11p15, combined with overexpression of IGF2 protein. In a major advance in understanding the development of these tumors, four studies published in 2015 found mutations in DROSHA and DICER1 genes that control messenger RNA processing in 20%–30% of cases. Yet little is known about the genomic landscape of bilateral Wilms tumor, although evidence points to an association with cancer-predisposing somatic (non-heritable) mutations.

Applying these findings in the clinic, researchers conducting a pilot study called IMPORT are collecting blood, tissue, and urine samples from over 300 patients from the time of diagnosis to the end of treatment and analyzing the molecular, genetic, and epigenetic evolution of the tumors to look for biomarkers of risk stratification and early diagnosis and to identify prevention strategies.

Genomic analysis of Wilms tumors, from diagnosis to the end of treatment, could help pinpoint strategies to improve patient outcomes. (Image courtesy of Kathy Pritchard-Jones)

Defining new strategies for cancers with high mutational burden

Li-Fraumeni syndrome (LFS), a cancer predisposition disorder, is characterized by early onset of disease and multiple types of cancer during a patient's life. It is associated with germline mutations in the TP53 gene, which encodes the tumor suppressor p53, the "guardian" of the genome that is functionally inactivated in most cancers.

David Malkin of the Hospital for Sick Children explained how clinical surveillance strategies, such as his group's Toronto Protocol, combine clinical monitoring with next-generation DNA sequencing to detect presymptomatic tumors in patients carrying TP53 mutations. The approach is associated with improved patient survival. Analysis of genomic rearrangements, repetitive sequence elements, gene fusions, and novel transcripts in medulloblastoma, an aggressive brain tumor, suggests that TP53 mutations initiate a genomic catastrophe. Germline TP53 mutations may make the genome vulnerable to chromothripsis, thousands of clustered chromosomal rearrangements in a single event within small, localized regions in one or more chromosomes.

The researchers are now investigating how these mutations lead to LFS phenotypes and whether circulating tumor DNA (as a biomarker) and novel MRI-based imaging can detect early tumors in mutation carriers. They are also using genetically modified zebrafish and mouse models to screen for candidate drugs and p53-targeting peptides that could be used therapeutically.

Biallelic mismatch repair deficiency (bMMRD), another childhood cancer predisposition syndrome, is associated with pediatric glioblastomas, highly aggressive brain tumors with poor survival outcomes. Caused by homozygous germline mutations in one of four mismatch repair (MMR) genes—which correct mistakes in DNA base pair matching during replication—bMMRD is one of the most penetrant syndromes, with 100% of mutation carriers developing cancer within two decades.

In her hot topic talk, Brittany Campbell, a graduate student at the International bMMRD Consortium and the Sick Kids Cancer Sequencing Program, showed evidence that bMMRD tumors carry a very high mutational burden and, as a result, are susceptible to immune checkpoint inhibitors such as anti-PD-1 drugs. Tumors with a high mutational burden have more novel tumor proteins, antigens for which are presented on the cell surface. Drugs that block the PD-1 pathway, which tumors subvert to dampen the immune response, free T cells to destroy neoantigen-bearing tumor cells. This approach has so far been successful in 7 pediatric patients with bMMRD glioblastoma, a brain cancer with 10-times more mutations and 7- to 16-times more neoantigens than adult immunoresponsive tumors, such as melanoma.

Alex Kentsis of Memorial Sloan Kettering Cancer Center gave a hot topic talk on a newly discovered DNA transposase called PGBD5. Transposases are enzymes that enable movement of mobile sequences called transposons within the genome, activity that can create genomic instability. His team has shown that PGBD5 is overexpressed in childhood tumors and that it contributes to cellular transformation in rhabdoid tumors by driving a distinct class of regulatory and structural genomic rearrangements that are required for tumorigenesis.

Highlights

Pediatric tumors are distinct from adult tumors in having more mutations in genes involved in metabolic and chromatin programming.

New therapeutic strategies against pediatric ALL include targeting mutant NOTCH-1 signaling specifically in ALL cells by conjugating an inhibitor to folate, which is exclusively taken up by cancer cells, and by metabolically reprogramming cells to undergo normal differentiation by suppressing a mitochondrial enzyme that regulates ATP production.

The metabolite 2-hydroxyglutarate, which blocks differentiation through histone modifications, could be used as a biomarker and surveillance tool to predict the conversion of low-grade, indolent brain tumors into aggressive, malignant disease.

Metabolic networks vary between cell lineages, and tumors adapt the existing networks to survive; thus, lineage may affect tumor sensitivity to drugs that target metabolic changes.

Levels of the substrate α-ketoglutarate have a yin-and-yang effect on methylation and demethylation of genes that maintain pluripotency and that induce cell differentiation.

Blocking cancer metabolic pathways to reroute cell differentiation

Normal cells respond to nutrient starvation with metabolic reprogramming that reduces anabolic processes of protein and fatty acid synthesis and increases catabolic processes of glycolysis and fatty acid oxidation. Tumor cells need to adapt to nutrient starvation as they grow, and they hijack cell metabolism and reprogramming pathways to become more aggressive. Research is underway to understand cells' altered metabolism and reprogramming to devise therapies targeting cancer metabolism.

Interrupting folic acid metabolism has been a mainstay of therapy for acute lymphoid leukemia (ALL) since Sidney Farber's discovery, in the 1940s, of ALL's dependency on folate metabolism. But with ALL pediatric cure rates of 55%, new approaches that target this process are needed.

Kimberly Stegmaier of the Dana-Farber / Boston Children's Cancer and Blood Disorders Center is pursuing two options. The first approach targets mutations in NOTCH1, which are present in up to 60% of pediatric T-cell ALL (T-ALL) cases. A high-throughput screen identified a calcium-channel inhibitor called thapsigargin with anti-leukemic effects. This natural molecule blocks posttranslational maturation of mutant NOTCH1, preventing it from reaching the cell surface and signaling aberrantly. Thapsigargin is extremely toxic, so Stegmaier developed a conjugate drug that targets it to ALL cells, taking advantage of the high expression of folate receptors on ALL cells by attaching thapsigargin to folate. The conjugate drug increases the therapeutic window of thapsigargin and has shown anti-leukemic activity in vitro and ex vivo in patient cells but has yet to be tested in patients.

The second approach targets aberrant differentiation of acute myeloid leukemia (AML) cells, focusing on differentiation pathways and related genes perturbed in AML. The researchers identified MTHFD2, a mitochondrial enzyme whose very high expression in cancer compared to very low expression in normal tissue makes it the most differentially expressed protein in cancer. Stegmaier showed that MTHFD2 suppression leads cancer cells to differentiate into normal-looking cells and impairs leukemic progression in vivo. Her work has also connected aberrant differentiation of leukemia cells to perturbations in cell metabolism, showing that MTHFD2 suppression affects the tricarboxylic (TCA) cycle and the production of related metabolites and reverses the metabolic signature of pluripotent stem cells. The next step is to test the metabolic consequences of suppressing MTHFD2 with small molecule inhibitors in AML and other leukemia cells.

Early thymocyte progenitor–ALL (ETP-ALL) is a T-ALL subtype that carries a distinct gene-expression signature and a high risk of treatment failure. In her hot topic talk topic, Shuning He, a research fellow in Thomas Look's laboratory at Dana-Farber Cancer Institute, explained her work on the JDP2 gene, which encodes a transcription factor that regulates pro-survival signaling, initiates T-ALL when overexpressed in zebrafish models, and is highly expressed in ETP-ALL (compared to non-ETP-ALL). Indeed, the level of the protein predicts the risk of treatment failure for ETP-ALL patients. Screening for drugs that kill thymocytes overexpressing JDP2 identified BET bromodomain inhibitors, which are known to work against some forms of acute myeloid leukemia, and CRM inhibitors, which block nuclear protein export.

Reengineering the machinery of cancer metabolism

Low-grade brain tumors, such as grade 2 astrocytoma, can remain stable and indolent for years before abruptly transforming into high-grade aggressive tumors, such as grade 4 glioblastoma, which has a 16-month median overall survival. Elizabeth Maher of the University of Texas Southwestern Medical Center focuses on tumor transition from low-grade to high-grade types, and on interventions to prevent this transformation.

Targeting tumors' metabolic machinery—"emptying the fuel tank" of these tumors—could be a therapeutic strategy. Using a protocol she developed to trace radiolabeled glucose during tumor resecting surgery, Mayer found that rapidly proliferating tumors use both glucose and acetate, and possibly other substrates, to fuel their growth. To understand the metabolic requirements of slow-growing tumors, she examined the metabolic changes in pediatric grade 2 gliomas, 35% of which have gain-of-function mutations in isocitrate dehydrogenase 1 and 2 (IDH 1/2), resulting in the production of a metabolite called 2-hydroxyglutarate (2-HG) that is known to deregulate DNA and histone methylation and thereby block differentiation and promote growth.

Maher found that 2-HG levels stay remarkably stable throughout the years of stable or indolent disease in low-grade gliomas and abruptly increase during transformation to high-grade disease. Thus 2-HG could be used as a diagnostic and prognostic biomarker and as a surveillance tool. To be able to prevent the metabolic reprogramming that drives tumor progression, researchers need to further define the additional genetic changes that occur during the stable period as the cells are "revving up their engines" and accumulating 2-HG.

Levels of 2-hydroxyglutarate stay low and stable during indolent disease but rise abruptly after transformation to aggressive disease, suggesting that this substrate can be used as a surveillance tool and a predictive biomarker of tumor transformation. (Image courtesy of Elizabeth Maher)

Gabriel Leprivier of Ben-Gurion University of the Negev gave a hot topic talk on his work in Poul Sorensen's laboratory at the University of British Columbia. He focused on how the eEF2 kinase (eEF2K), which is hijacked by tumors to adapt to nutrient stress, supports cell survival under starvation conditions. Transcriptomics (mRNA sequencing) of nutrient-deprived cultured cells with and without eEF2K activity along with gene set enrichment analysis identified fatty acid metabolism as a downstream target of eEF2K activity. By increasing fatty acid oxidation, eEF2K helps maintain acetyl-CoA levels under starvation to keep the TCA cycle running to produce adenosine triphosphate (ATP).

Different tumors solve metabolic problems in different ways that are influenced by tumor genetics. Yet it is not clear whether and how variations in tumor metabolic phenotypes are affected by the tumor environment or cell lineage. Matthew Vander Heiden of the Koch Institute at Massachusetts Institute of Technology studies glucose metabolism in tumor cells from K-RAS/p53-driven lung cancers in mice. He found that these cells convert glucose into either lactate or TCA-cycle metabolites while inside the tumor environment in mice, but mostly convert glucose into lactate when cultured in vitro, reverting to the alternate process when reimplanted into the mouse tumor environment.

Experiments comparing K-RAS/p53-driven lung tumors to pancreatic tumors from mice revealed that each tumor type derives amino acids from different sources using different metabolic processes, despite having the same genetic driver. These results suggest that cell lineages have unique metabolic networks and that tumors arising within these lineages adapt the existing network to sustain themselves. Such a model implies that, like metabolic dependency, drug sensitivity could also be affected by tumor lineage and environment.

Yin and yang: metabolic substrates, methylation and demethylation, and stemness

In his keynote address, Craig Thompson of Memorial Sloan Kettering Cancer Center also discussed the link between metabolism regulation and cell differentiation, focusing on how metabolic and epigenetic mutations fuel the stem cell phenotype that is a hallmark of pediatric cancers. Mutations in metabolism-regulating IDH1/2 genes are the most common in human cancers, occurring in adult cancers and, with particularly high frequency, in pediatric cancers; about 60% of childhood bone cancers and 80% of all intermediate grade gliomas (brain tumors) and secondary glioblastomas carry IDH mutations.

In the energy-producing TCA cycle, IDH1/2 normally converts isocitrate to α-ketoglutarate (α-KG), which, along with the ten-eleven translocation (TET) enzymes, demethylates both histones and DNA and thereby un-silences genes required for cell differentiation. When mutated, heterodimers of wildtype and mutant IDH1 instead produce 2-HG, which antagonizes TET enzymes and blocks histone and DNA demethylation. In the model Thompson has proposed, stem cell genes and lineage-specific genes necessary for differentiation are activated or repressed in a mutually antagonistic way depending on the level of α-KG, the fuel for the reactions underlying methylation and demethylation.

Thompson's group found that gliomas with IDH mutations are rife with both hypo- and hypermethylation, the former resulting in the expression of stem cell genes that would normally be silenced and the latter resulting in the silencing of lineage-specific genes that would normally be active. The result is maintenance of a pro-cancer pluripotent stem cell phenotype instead of a differentiated phenotype. The group further showed that α-KG directly promotes pluripotency through demethylation of stemness genes in embryonic stem cells. These data validate the proposed model that explains how the yin-and-yang effect of metabolic substrate α-KG on methylation and demethylation controls both stem cell and lineage genes.

Interestingly, Thompson's group also found that ERK inhibitors maintain the stemness of embryonic stem cells in much the same way that α-KG levels trigger the shift from hyper- to hypomethylation of stem cell genes. He therefore cautioned that ERK inhibitors, which are being studied as anticancer agents in many types of cancer, might not be a good option for progenitor-cell-derived childhood tumors such as gliomas.

Highlights

A new therapeutic approach for aggressive pediatric sarcomas could involve blocking YB-1, an RNA-binding helicase that facilitates metastasis by activating mechanisms that help cancer cells adapt to stress.

Understanding of the expression patterns of cell surface integrin receptors on tumor exosomes, which create metastatic niches, has driven the design of decoy peptides that block exosome adhesion via the integrins, thereby preventing metastatic tumor formation.

Differences in gene expression profiles between primary and metastatic tumors suggest that therapies targeting mutations found in the primary tumor may be ineffective against metastatic tumors.

Preventing stress granule formation to block metastasis

Metastasis from the original tumor site to distant vital organs such as the liver, lung, and brain is a devastating step in cancer progression that is responsible for more than 90% of all cancer-related deaths. It is the most powerful predictor of poor outcomes in high-risk childhood sarcomas, but much of the biology of this phenomenon is still a black box.

Poul Sorensen of the University of British Columbia focuses on YB-1, an RNA-binding helicase that is highly expressed in most high-risk childhood sarcomas and is associated with poor outcomes. His group found that YB-1 facilitates metastasis in two ways. First, it enhances the translation of HIF-1A mRNA under hypoxic conditions, which facilitates tumor cell adaptation to stress and triggers sarcoma invasion and metastasis. Second, it translationally activates another protein, G3BP1, which is critical for the formation of stress granules—sites of mRNA storage where HIF-1A and G3BP1 are either silenced until needed or degraded. Small molecule inhibitors of G3BP1 prevent stress granule formation and block sarcoma metastasis in vivo. YB-1 thus appears to be a cell plasticity factor that controls selective mRNA translation and stress granule formation. Sorensen is exploring the possibility that stress granules bestow metastatic capacity by reducing oxidative stress.

Decoding the molecular basis of exosome-mediated metastasis

David Lyden of Weill Cornell Medical College studies the role of tumor-derived exosomes—microvesicles that are shed by tumors into blood and urine—in metastasis. His group previously found that tumor exosomes, which carry genetic material, proteins, and lipids, create pre-metastatic niches—environments that are favorable to tumor growth at future metastatic sites. Exosomes have long been known to be organotrophic, meaning that they do not randomly invade new tissue but prefer certain tissues and cell types, depending on the cell type they are derived from. Researchers previously thought they functioned as storage compartments or as pathogen-spreading vehicles in infectious diseases.

By injecting exosomes purified from various human cancers into mice, assessing their distribution in the lung, brain, bone, and liver, and performing proteomics, Lyden's team found specific expression patterns of transmembrane receptors called integrins. These receptors guide the organ-specific targeting of exosomes and enable metastasis by upregulating S100 proteins, which promote pro-inflammatory and pro-migratory activity. The ability of integrin-blocking decoy peptides to destroy tumor exosome adhesion suggests that peptide-derived drugs could prevent exosomes from creating niches for future sites of tumor growth.

The expression patterns of integrin receptors on tumor exosomes determine organ-specific metastasis. (Image courtesy of David Lyden)

Actionable mutations from primary tumors might not be targetable post-relapse

About 30% of pediatric patients with medulloblastoma have metastatic disease at diagnosis, and most children will have metastatic disease at relapse. Although metastatic medulloblastoma has a very poor prognosis, most research on this type of brain tumor has focused on primary tumors, which are widely available for analysis because they are easier to resect than metastatic tumors.

However, Michael Taylor of the Hospital for Sick Children and the University of Toronto has found that primary and metastatic tumors in medulloblastoma are not equivalent: the gene expression signatures of primary medulloblastoma tumors are vastly different from those of metastatic tumors that occur locally or at other locations, with less than 10% of common genetic events occurring in all three sites. This finding implies that a targeted therapy developed based on a mutation detected at diagnosis might no longer work in relapsed patients who are highly unlikely to still carry that mutation.

It has long been thought that primary medulloblastoma becomes metastatic by spreading to other locations exclusively through the cerebrospinal fluid. Taylor's group has disproved this idea by revealing the presence of circulating medulloblastoma cells and tumor DNA in blood, a finding that could have diagnostic and prognostic value. The circulating tumor cells seem to reach the brain, after they detach from the primary tumor, by expressing CCL2, a chemokine that guides cell migration. Whether interrupting this migration via anti-CCL2 antibodies could prevent metastasis is unknown.

Highlights

Next-generation CAR T cells can be engineered to graded potency levels, can induce responses against tumors that are refractory to chemotherapy, and can be used in synergy with other immune modulating agents.

Approaches to increase the efficacy and safety of CAR therapy include coadministration with T-APCs that work like a vaccine to boost CAR T cell activity, and preemptive anti-inflammatory treatment to decrease CAR T cell toxicity.

Response to immunotherapy against solid tumors can be enhanced by combining local radiation therapy with intratumoral injection of tumor-specific antibodies, in part via antibody-dependent cell-mediated cytotoxicity.

Optimizing CD19 CAR Therapy

Chimeric antigen receptors (CARs) are synthetically engineered T cell receptors that bestow T cells with specificity to any desired antigen and with enhanced potency to overcome immune escape mechanisms and then eradicate tumors. CAR T cells offer myriad advantages over natural T cells. T-cell engineering allows CAR T cells to recognize antigens that are not presented by HLA molecules—antigens other than peptides—and enables both CD4 and CD8 T cells to attack tumors while also increasing both the potency and the longevity of these cells.

Michel Sadelain's group at Memorial Sloan Kettering Cancer Center began working on CAR T cell-mediated cancer therapy in the early 2000s, engineering T cells specific for CD19, a surface receptor expressed on B-cell lineage lymphomas and leukemias. In phase I trials in patients with relapsed B-cell ALL that is unresponsive to chemotherapy, these CAR19 T cells achieve complete response in 82% of adults, regardless of age, chemotherapy history, and disease burden, and in 73% of children.

T-cell activation typically involves a sequence of three signals: the first from the antigen through the T cell receptor, the second through the costimulatory CD28 receptor, and the third through the 41BB costimulatory receptor following its engagement of the 41BB ligand on the tumor cell or antigen-presenting cell. Sadelain's group previously showed that engineering the 41BB ligand onto T cells could trigger autocostimulation and generate very potent antitumor T cells in a prostate cancer mouse model.

The team has applied this insight to generate human CAR19 T cells with graded potency and longevity, and the researchers plan to begin clinical trials in late 2016. Microarray studies also revealed that CAR T cells' potency depends on increased induction of IRF7, which transforms the tumor microenvironment to boost the immune response. Sadelain therefore postulates that CAR T cells could be useful against pediatric tumors, which are notoriously low in mutational burden and elicit poor immune responses against tumor antigens.

CAR T cells engineered for costimulation through CD28 as well as 41BB have potent antitumor activity. Mechanisms that can be exploited to further boost this potency are being explored. (Image courtesy of Michel Sadelain)

Engineering enhanced CAR T cells against B cell malignancies

Michael Jensen of Seattle Children's Research Institute and Fred Hutchinson Cancer Research Center spoke about his group's work to systematically optimize CAR19 T cells to treat B-cell leukemias. Patient-derived T cells used to engineer CAR T cells have variable compositions of naïve, effector, and memory cells, and different ratios of CD4 and CD8 subsets. These variations can reduce the success of receptor engraftment and render the engineered cells less effective and less safe when they are reinfused into the patient. Protocols are in place to purify and homogenize CAR T cell products, improving the cells' antitumor activity in vivo. A phase I pediatric leukemia trial in refractory/relapsed cases yielded response rates of 93%, with 73% of patients surviving a year.

The magnitude of the CAR19 T cell graft and the duration of its persistence depends on CD19 antigen levels in patient bone marrow. To prevent graft loss when antigen levels are low, Jensen's team now creates CD19-expressing T-antigen presenting cells (T-APCs) from patient cells at the same time as the engineered CAR19 T cells. The T-APCs work like a vaccine, boosting CD19 antigen levels and consequently CAR T cell numbers when infused in vivo. Having completed a proof-of-concept study of CAR20 T cells in monkey models, the researchers now plan to test the approach in patients.

Constitutively active CAR T cells can be toxic, producing copious amounts of cytokines and infiltrating the nervous system. Preemptive treatment with anti-IL-6 antibodies and the steroid dexamethasone dampens the immune response to remedy this problem. To prevent relapse caused by the emergence of CD19 escape variants that CAR19 T cells no longer recognize, Jensen's team produce bispecific CARs which recognize CD19 and a second tumor antigen and thereby retain antitumor potency for longer. Jensen plans to next tackle other tumor types, particularly solid tumors, using these strategies.

Harnessing both innate and adaptive immune mechanisms to boost antitumor response

Currently, antitumor immunotherapy in the clinic primarily involves T-cell checkpoint inhibitors, which mobilize the CD8 T-cell arm of the immune system. Paul Sondel of the University of Wisconsin School of Medicine and Public Health is pursuing a broader approach that harnesses innate and antibody responses against tumors. His group's discovery that natural killer (NK) cells activated by the cytokine IL-2 kill tumor cells coated with antitumor antibodies through antibody-dependent cell-mediated cytotoxicity contributed to a combined approach developed by the Children’s Oncology Group that is now an FDA-approved therapy against neuroblastoma.

Sondel's group is now developing an approach that combines radiation with an intratumorally injected immunocytokine—a fusion protein that links a tumor-specific antibody to IL-2. While this combination has been successful against microscopic tumors, targeting macroscopic or metastasized tumors is more difficult because tumor-specific responses at the irradiated primary tumor site may be suppressed by regulatory T (Treg) cells from a non-treated secondary tumor site. The team has overcome this challenge by adding Treg-depleting anti-CTLA-4 antibodies to this cocktail, effectively turning tumors into in situ vaccines that boost a systemic immune response.

In a mouse model of melanoma that has poor immunogenicity, a combination approach of local radiation, intratumorally injected tumor-specific antibodies conjugated to IL-2, and regulatory T cell-depleting anti-CTLA4 antibodies boosts both local and systemic antitumor responses. (Image courtesy of Paul Sondel)

Highlights

Neuroblastoma tumorigenesis driven by RAS/MAPK pathway mutations that are refractory to current treatments can be targeted by combining RAS/MAPK inhibition with other key pathways.

Patient-derived xenograft mouse models are valuable preclinical tools for pediatric tumor research; they have helped identify druggable targets, new combination approaches, and better treatment protocols.

Clinical trials are underway to test the combination of histone deacetylation inhibitors and PI3K inhibitors in medulloblastoma, a combination that blocked tumor growth in vitro and in vivo.

Identifying synergistic drug combinations in neuroblastoma

An embryonic tumor that arises in the peripheral nervous system, neuroblastoma accounts for 10% of pediatric cancer deaths. Using a transgenic zebrafish model of the disease that closely resembles the human version, A. Thomas Look's group at the Dana-Farber Cancer Institute has uncovered most of the genetic and signaling pathways underlying the cancer. Overexpression of the oncogene MYCN combined with loss-of-function mutations in the tumor suppressor gene NF1—associated with poor outcomes in human neuroblastoma—results in a rapid onset of highly penetrant and aggressive disease, with almost all of the fish developing disease by 3 weeks of age.

In this model, tumor survival and proliferation are driven by aberrantly upregulated RAS/MAPK pathway signaling. Look's team has used the zebrafish model to screen for drugs against this pathway, identifying the therapeutic index and synergistic effects of the MEK inhibitor trametinib and the retinoid isotretinoin at several dosage combinations via isobologram analysis. These findings could identify new treatments for relapsed disease, given the frequency of mutations in RAS/MAPK pathway genes in relapsed neuroblastoma.

Medulloblastomas, the most common malignant brain tumors in children, include four genomically distinct diseases driven by mutations in different oncogenic pathways and with different treatment responses and prognoses. In contrast to tumors driven by Wnt signaling, tumors driven by Myc signaling, which account for up to a third of medulloblastoma cases, are very aggressive, highly metastatic, and usually fatal.

Identifying novel drug combinations using preclinical mouse models for medulloblastoma

Robert Wechsler-Reya's group at the Sanford Burnham Prebys Medical Discovery Institute created a mouse model of medulloblastoma driven by overexpression of the Myc oncogene. The mice—injected with cerebellar stem cells isolated from neonatal mice and infected with viruses encoding Myc and a mutant form of the tumor suppressor p53—develop tumors that resemble human medulloblastoma and show upregulation of PI3K signaling and downregulation of the tumor suppressor Foxo1, which is a known antagonist of Myc.

A high-throughput screen using a chemical library of thousands of compounds identified more than 20 drugs that kill murine medulloblastoma cells while sparing normal brain cells. Four were histone deacetylase (HDAC) inhibitors, which block HDACs from inactivating chromatin and thereby derepress tumor suppressor genes. Panobinostat, an HDAC inhibitor approved for use in multiple myeloma but untested in medulloblastoma, synergized with the PI3K inhibitor buparlisib to block Myc-driven tumor growth in the mouse model and in patient-derived xenograft (PDX) models—immunocompromised mice implanted with tumor cells isolated from a patient's primary tumor.

The group is now applying the high-throughput screening approach to PDX lines, which are then subjected to gene expression profiling, whole exome sequencing, methylation analysis, and drug response profile analysis. The researchers hope this approach will correlate drug responses to genes and pathways activated in tumors and will help predict the response of different tumor subgroups to different drugs.

Most human solid tumors do not metastasize when grown subcutaneously in immunocompromised mice, including PDX models, unless the tumor tissue is engrafted in the organ type of its origin. Michael Dyer's group at St. Jude Children's Research Hospital, which has created one of the largest collections of patient-derived orthotopic xenografts, uses orthotopic PDX mouse models to study a host of childhood solid tumors. The group aims to understand tumor biology and to establish a preclinical platform for testing drug candidates and interventions before testing them in pediatric patients.

These PDX models have already been useful in translational research on pediatric solid tumors. A PDX model for rhabdomyosarcoma, a highly aggressive cancer that develops in the muscle, revealed deregulation of CDK genes, which control the cell cycle, in addition to the well-known RAS pathway mutations seen in most patients. Combining a RAS pathway inhibitor with a CDK inhibitor synergistically improved treatment outcomes compared to either treatment individually. Using a PDX model for retinoblastoma, a solid tumor of the eye, Dyer also illustrated how researchers can use these models to improve treatment protocols or avoid interventions that are not therapeutic.

Highlights

Single-agent therapies are of limited use in many tumor types; there is an urgent need to understand how targeted therapies can be safely combined with each other or with chemotherapy and immunotherapy agents.

A validated next-generation sequencing (NGS)–derived, biomarker-based strategy in pediatric cancer trials will help expedite the identification of patient-specific drug combinations for difficult-to-treat diseases such as relapsed neuroblastoma.

A need for new rules in the post-genomics era war against pediatric cancer

Lee Helman of the National Cancer Institute proposed new guidelines for pediatric cancer trials for targeted therapy. He urged researchers to rethink the definition of a response in a clinical trial, arguing that current measures do not correlate to meaningful improvement for the patient. He also advocated that researchers stop testing single agents, arguing that perturbing a single component in a signaling network simply activates other components via feedback loops or redundancies in the network. This idea has already been borne out in several failed cancer trials and in studies that show that more than one agent is needed to achieve a prolonged and deep response to cancer.

Helman pointed out that, because tumors evolve, treatment decisions for recurring cancers should not be based on genetic or genomic data from the time of diagnosis. He also urged oncologists to learn how targeted agents alter the immune response, especially as immunotherapy becomes a treatment option for more cancers, so that both modes of therapy can be safely and effectively combined.

Because tumors evolve and genomic profiles change over time, treatment decisions for progressing cancers should not be based on data from the time of diagnosis. (Image courtesy of Lee Helman)

Using large-scale genomics to confront the challenges of relapsed neuroblastoma

Despite the identification of ALK, a cell surface receptor tyrosine kinase, as a druggable molecular target in neuroblastoma, several challenges remain in making ALK-targeted therapy a mainstay for this pediatric cancer. Relapsed disease remains largely incurable and difficult to study because it is difficult to collect samples, and aspects of disease biology that affect treatment response, such as clonal evolution of tumors and mechanisms of resistance to chemotherapy and targeted therapies, are only beginning to be defined.

Yael Mossé of the Children's Hospital of Philadelphia helped established ALK as the major familial neuroblastoma gene more than a decade ago. She recently showed that not all mutations of the gene are created equal. Some are non-oncogenic, others do not affect ALK function. Mutations that affect function occur in the ALK tyrosine kinase domain—which is critical for the protein's ability to phosphorylate targets—and are found in around 8% of patients. These mutations have different sensitivities to the inhibitor crizotinib, which is currently used to treat ALK-mutated lung cancer.

These findings suggest that ALK genomic status could be important in therapeutic decision-making. Mossé's group identified ALK mutations that confer primary resistance to crizotinib, which led to the development of second-generation ALK inhibitors to overcome this resistance.

Mossé also described work by her group and collaborators to study how neuroblastoma genomes evolve at relapse. Their studies show a 3-fold increase in ALK mutations at relapse versus diagnosis, with similar trends in other genes as well, further supporting the idea that combination therapies rather than single-agent treatments should be the preferred strategy for these tumors. Mossé's group is conducting a clinical trial for relapsed or primary refractory high-risk neuroblastoma, in which researchers use a biopsy at relapse combined with next-generation sequencing results to identify biomarkers with which to assign patients to combination therapy with an ALK inhibitor and a second drug specific to individual mutation profiles.

A clinical trial strategy to personalize therapy and identify the right targeted drug combination for relapsed neuroblastoma. (Image courtesy of Yael Mossé)

Implementing precision cancer medicine for pediatric tumors

Katherine Janeway of Dana-Farber / Boston Children's Cancer and Blood Disorders Center spoke about her group's involvement in the multicenter iCat1 study of individualized cancer therapy. Its objective was to assess the feasibility of identifying actionable mutations and to make recommendations for individualized therapy in pediatric patients or young adults with high-risk recurrent or refractory solid tumors. Contrary to expectations, the study found that there is a sufficiently high prevalence of potentially actionable mutations in pediatric solid tumors to justify using measures such as molecular profiling in treatment decisions.

Molecular profiling of tumors yielded results with potential clinical significance in 43% of patients, but only 3 of the 31 patients with a recommendation received a match-targeted therapy. Janeway speculated that the low numbers of patients receiving a matched treatment might be explained by their poor clinical status (these are patients with advanced diseases who might be ineligible to enroll in trials), the unavailability of the correct drugs, or the presence of unknown or non-actionable mutations.

The effects of match-targeted therapy on patient outcomes still need to be assessed. Janeway discussed several studies with different methodologies that are focused on understanding precision cancer therapy for pediatric tumors.

A multicenter clinical genomics study in pediatric oncology showed that a substantial number of relapsed or refractory tumors carry actionable mutations. (Image courtesy of Katherine Janeway)

How should the rapidly expanding body of molecular information about pediatric tumors be integrated into diagnostic procedures?

How should molecular subtype-specific treatment strategies be integrated into clinical protocols?

Histone mutations that affect gene expression in cancer have been identified, but what is the precise role of these mutations in tumorigenesis and will targeting them be an effective antitumor strategy?

What are the mechanisms of oncogenic transformation that make tumor cells hypersensitive to nutrient deprivation and other forms of metabolic stress, and can any of these mechanisms be therapeutically targeted?

How can the emerging genetic and epigenetic findings in tumors with high risk of relapse, such as Wilms tumor or medulloblastoma, be translated into reliable biomarker-based risk stratification strategies?

The molecular classification of medulloblastoma into distinct clinical diseases is a step forward in devising precision treatment strategies, but why are these diseases different in their prognoses, and what are the factors that make them segregate into different diseases or dictate which disease manifests?

Are the current criteria for defining responses to immunotherapy adequate and accurate, and how should they be updated to measure responses to immunotherapy combinations?

How do the various targeted treatments affect or alter the immune milieu in tumors, and how should the best combination strategies that include immunotherapy for different tumor types be selected?