In This Issue

in this issuE

AACR Project gEnIE Facilitates sharing of genomic and Clinical Data

• Project GENIE launched with data • Project GENIE aims to promote • Data from Project GENIE may guide from 19,000 patients from 8 institu data sharing to enhance precision identification of drug targets and tions with a variety of tumor types . medicine research . biomarkers in patients with cancer .

Genomic profi ling of tumors accessible in the cBioPortal for Cancer Genomics. Despite the has become increasingly com- differences in genomic testing at the contributing centers, the mon across cancer types, but genomic data collected so far are largely concordant across the data are not regularly made the centers, and mutation rates are similar to those reported by available to the entire research The Cancer Genome Atlas. Initial results from Project GENIE community. The American Asso- suggest that more than 30% of tumors harbor potentially ciation for Cancer Research clinically actionable mutations. Advantages of this platform (AACR) launched the Genomics, include integration of clinical data from electronic health Evidence, Neoplasia, Informa- records and increased statistical power to potentially facilitate tion, Exchange (GENIE) project understanding of the clinical relevance of somatic mutations in partnership with eight academic institutions to facilitate and improve patient selection for targeted therapies. The large-scale sharing of genomic and clinical data. The AACR establishment of infrastructure for integrating genomic and Project GENIE Consortium released the fi rst of set of data clinical data has the potential to aid identifi cation of thera- from 19,000 patients in January 2017, and the number of sam- peutic targets and biomarkers of treatment and response in ples included is expected to grow to more than 100,000 within patients with cancer to enhance precision medicine research 5 years as more centers join the Consortium. Data from par- and improve patient outcomes. n ticipating centers include matched clinical and genomic data from a variety of tumor types that are harmonized and made See article, p. 818.

TBk1/Ikkd Is a Potential Target in MAPk Inhibitor–Resistant Melanoma

• A SOX10associated 5gene bio • SOX10independent, BRAF/MEK • TBK1/IKKε addiction defines an marker discriminates MAPK inhibitor– inhibitor–resistant melanoma is innate immune subtype with YAP sensitive and –resistant melanomas . sensitive to TBK1/IKKε inhibitors . activation and epigenetic remodeling .

The effi cacy of BRAF/MEK the noncanonical IκB kinases TANK-binding kinase 1 (TBK1) inhibitors in BRAF-mutant mela- and IKKε (encoded by IKBKE), independent of BRAF status. noma is limited by the emergence TBK1/IKKε–sensitive melanomas were enriched for innate of resistance, which often occurs immune signaling and exhibited TBK1/IKKε–mediated acti- via mechanisms that reactivate vation of YAP and AKT survival signaling, which may underlie MAPK signaling. To characterize the dependency of this subtype on TBK1/IKKε. In addition, potential alternative therapeutic TBK1/IKKε–sensitive melanomas were characterized by targets in BRAF/MEK inhibitor– increased expression of nicotinamide N-methyltransferase, resistant melanoma, Eskiocak which has been implicated in epigenetic remodeling and and colleagues performed an chromatin relaxation. Consistent with this fi nding, global integrative analysis of a functional genomic screen and copy- H3K27 trimethylation was reduced in TBK1/IKKε–sensitive number variation in melanoma cell lines and tumors. Among cells, and inhibition of the H3K27 methyltransferase EZH2 the identifi ed melanoma cell survival genes, addiction to was suffi cient to confer sensitivity to TBK1/IKKε inhibitors SOX10, assessed by expression of a 5-gene biomarker, dis- in previously resistant cells. Together, these fi ndings defi ne tinguished two mechanistic subtypes of melanoma: SOX10- potential molecular predictors of response to BRAF/MEK dependent, BRAF-mutant melanomas sensitive to BRAF/ inhibition and identify TBK1/IKKε as a candidate therapeu- MEK inhibitors and SOX10-independent, BRAF/MEK inhib- tic target for BRAF/MEK inhibitor–resistant melanoma. n itor–resistant melanomas. Targeted therapy–resistant melanomas were found to be selectively sensitive to inhibition of See article, p. 832.

August 2017 CANCER DISCOVERY | 783

00-CD-16-FM_Aug.indd 783 7/20/17 10:07 AM In ThIs IssUE

hDAC and BET Inhibitors synergize to Promote Antitumor Immune Responses

• The HDAC6 inhibitor ricolinostat in • Ricolinostat promotes Tcell activ • Dual HDAC6/BET targeting may poten combination with the BET inhibitor ity and JQ1 reduces Treg activity to tially enhance the antitumor immune JQ1 promotes Tcell function . suppress tumor growth . response in patients with NSCLC .

Epigenetic drugs including ther, ricolinostat increased surface expression of MHC class histone deacetylase (HDAC) II molecules and CD86 on monocytes and tumor-associated inhibitors and bromodomain macrophages, suggesting that ricolinostat promotes pheno- and extraterminal domain (BET) typic changes that support improved antigen presentation and inhibitors have shown prom- costimulatory capabilities. These changes were recapitulated in ise in a variety of tumor types. vivo in a mouse model of NSCLC, where ricolinostat treatment Although immunomodulatory promoted a more mature tumor-infi ltrating macrophage phe- effects have been reported, the notype and increased tumor T-cell activation. The BET inhibi- effects of HDAC and BET inhibitor JQ1 acted on Treg cells within tumors, disrupting their tors on the tumor immune gene expression patterns and diminishing their suppressive microenvironment have not been fully elucidated, prompt- activity. Dual inhibition with ricolinostat and JQ1 enhanced ing Adeegbe and colleagues to investigate the immunoregula- the antitumor activity of tumor-infi ltrating T cells, suppressed tory properties of these drugs in non–small cell lung cancer tumor growth, and extended survival in mice with NSCLC. (NSCLC). In peripheral blood mononuclear cells (PBMC) and Collectively, these fi ndings uncover immunoregulatory effects tumors from patients with NSCLC, the selective HDAC6 of HDAC and BET inhibition, and suggest that dual targeting inhibitor ricolinostat reduced the proportion of suppressive may enhance the antitumor immune response in NSCLC. n CD4+FOXP3+ regulatory T cells (Treg) and increased expression of CD69 on T cells, suggesting T-cell activation. Fur- See article, p. 852.

Comprehensive Methylation Profiling Reveals Epigenetic subtypes in AML

• Differential methylation of nonpro • Comprehensive methylome sequenc • Comutation of IDH2 and DNMT3A moter regulatory elements defines ing is superior to promoterspecific results in epigenetic antagonism the epigenetic subtype in AML . analysis for capturing AML biology . that may sensitize to MEK inhibitors .

The effects of disrupted DNA cytosine methylation. AML with dominant hypermethyla- methylation patterns in regions tion was associated with epigenetic disruption of promoters. outside of promoters and CpG Conversely, AML with dominant hypomethylation exhibited islands are not well understood greater disruption of distal and intronic regions. Co-muta- in acute myeloid leukemia tion of IDH2 and DNMT3A, which have opposing effects on (AML). Glass and colleagues per- DNA methylation, resulted in an epigenetic antagonism, and formed enhanced reduced rep- this subset of AML had the least differential methylation. In resentation bisulfi te sequencing preleukemic hematopoietic stem cells, induction of Dnmt3A (ERRBS) on a panel of 119 pri- and Idh2 mutations promoted epigenetic antagonism prior to mary AMLs to evaluate cytosine malignant transformation, indicating it is not a consequence methylation changes in nonpromoter gene regulatory ele- of transformation. Transcriptome analysis in patients with ments. This comprehensive methylome sequencing approach IDH2/DNMT3A-mutant AML revealed upregulation of the more precisely linked genetic lesions to disrupted cytosine RAS signaling signature; accordingly, double-mutant cells methylation patterns than promoter-specifi c methylation. exhibited sensitivity to MEK inhibition in ex vivo experiments. CpG methylation in “gene neighborhoods” between 2kb Altogether, this comprehensive DNA methylation profi ling and 50kb from the transcription start or end site was best analysis shows that differential methylation of nonpromoter able to capture the epigenetic patterning observed in AML. regulatory elements drives AML epigenetic identity. n Active enhancers displayed strong focal changes in methylation whereas promoters exhibited less robust changes in See article, p. 868.

784 | CANCER DISCOVERY August 2017 www.aacrjournals.org

00-CD-16-FM_Aug.indd 784 7/20/17 10:07 AM In ThIs IssUE

PAX3–FOXO1-Positive Rhabdomyosarcomas Are Dependent on BRD4

• PAX3–FOXO1 promotes chromatin • PAX3–FOXO1 interacts with MYOD, • BET inhibitors may potentially be remodeling to establish myogenic su MYOG, and MYCN, and recruits beneficial in patients with fusion perenhancers in rhabdomyosarcoma . BRD4 to superenhancers . positive rhabdomyosarcoma .

Alveolar rhabdomyosarcoma genes. Superenhancer analysis revealed a distinct landscape is a myogenic cancer driven by in fusion-positive rhabdomyosarcoma compared with fusion- expression of the oncogenic negative, and identifi ed MYOD, MYOG, and MYCN as master chimeric transcription factors regulators controlled by superenhancers. PAX3–FOXO1 col- PAX3–FOXO1 or PAX7–FOXO1. laborated with these master regulators at superenhancers to Fusion-positive rhabdomyosar- promote gene expression, and acted as a pioneering factor, coma generally affects children recruiting p300, MED1, and BRD4 to enhancers to facilitate and young adults and is asso- looping and transcription. These fi ndings provide a ration- ciated with a poor survival. ale for the evaluation of BET inhibitors in fusion-positive PAX3–FOXO1-positive tumors rhabdomyosarcoma, and treatment with the BET inhibitor are associated with elevated expression of MYCN and the JQ1 reduced expression of PAX3–FOXO1 target genes. BRD4 myogenic transcription factors MYOD1 and MYOG, but was required for the function and stability of PAX3–FOXO1, the epigenetic mechanisms by which PAX3–FOXO1 dys- and JQ1 inhibited PAX3–FOXO1 activity and suppressed the regulates chromatin have not been described. Gryder and growth of fusion-positive rhabdomyosarcoma xenografts in colleagues used chromatin immunoprecipitation sequenc- vivo. Collectively, these fi ndings demonstrate that PAX3– ing to map the genome-wide landscape of histone modifi - FOXO1 promotes chromatin remodeling and superenhancer cations in fusion positive rhabdomyosarcoma cell lines to establishment, and support further investigation of BET determine how PAX fusions regulate the myogenic program. inhibitors in fusion-positive rhabdomyosarcoma. n The large majority of PAX3–FOXO1 was localized at active distal enhancers, where it promoted expression of target See article, p. 884.

Bap1 and Pbrm1 Loss Determine Renal Cell Carcinoma Tumor grade

• Genetically engineered mouse models • Loss of Vhl and Bap1 induces high • ccRCC may arise from Bowman cap of Bap1 and Pbrm1deficient ccRCC grade ccRCC, whereas loss of Vhl and sule cells, and loss of BAP1 and recapitulate the human disease . Pbrm1 induces lowgrade ccRCC . PBRM1 may determine tumor grade .

VHL is inactivated in the expressed in human BAP1- and PBRM1-defi cient tumors. large majority of clear cell renal Co-deletion of Vhl and Bap1 resulted in high-grade ccRCC cell carcinomas (ccRCC), and with immunohistochemical and morphologic features that is frequently accompanied by mimicked the human disease. In contrast, loss of Vhl and loss of BAP1 or PBRM1, which Pbrm1 induced low-grade ccRCC; however, a small subset also reside on chromosome 3p. of tumors were high grade and exhibited mTORC1 activa- PBRM1 is a SWI/SNF chroma- tion. Disruption of a single allele of the mTORC1 negative tin-remodeling complex compo- regulator Tsc1 was suffi cient to activate mTORC1 in some nent mutated in approximately tumors and induce high-grade ccRCC in Vhl−/−;Pbrm1−/− mice. half of ccRCCs, and BAP1 is a Using Cre-drivers expressed in proximal tubules instead of deubiquitinating enzyme mutated in approximately 10% to Pax8-Cre did not result in tumorigenesis in response to Vhl 15% of ccRCCs. BAP1 mutations are associated with aggres- and Bap1 or Pbrm1 loss, suggesting that ccRCC may not sive higher-grade tumors and shorter survival than PBRM1 arise from proximal tubules, but instead may arise from mutations, but it is not clear if PBRM1 and BAP1 mutations Bowman capsule cells. These mouse models suggest that drive these differences. Gu and colleagues developed mouse BAP1 and PBRM1 loss may drive ccRCC and determine models of Bap1- or Pbrm1-defi cient ccRCC using Pax8-Cre tumor grade. n to drive kidney cell specifi c deletion. PAX8 was selected because it is a nephric lineage transcription factor uniformly See article, p. 900.

In This Issue is written by Cancer Discovery editorial staff . Readers are encouraged to consult the original articles for full details .

August 2017 CANCER DISCOVERY | 785

00-CD-16-FM_Aug.indd 785 7/20/17 10:07 AM In This Issue

Cancer Discov 2017;7:783-785.

Updated version Access the most recent version of this article at: http://cancerdiscovery.aacrjournals.org/content/7/8/783

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerdiscovery.aacrjournals.org/content/7/8/783. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.