Search

Microarray pathway analysis

Standard therapies used for the treatment of acute myeloid leukemia (AML) are cytotoxic agents that target rapidly proliferating cells. Unfortunately, this therapeutic approach has limited efficacy and significant toxicity and the majority of AML patients still die of their disease. In contrast to the poor prognosis of most AML patients, most individuals with a rare subtype of AML, acute promyelocytic leukemia, can be cured by differentiation therapy using regimens containing all-trans retinoic acid. GSK3 has been previously identified as a therapeutic target in AML where its inhibition can lead to the differentiation and growth arrest of leukemic cells. Unfortunately, existing GSK3 inhibitors lead to suboptimal differentiation activity making them less useful as clinical AML differentiation agents. Here, we describe the discovery of a novel GSK3 inhibitor, GS87. GS87 was discovered in efforts to optimize GSK3 inhibition for AML differentiation activity. Despite GS87's dramatic ability to induce AML differentiation, kinase profiling reveals its high specificity in targeting GSK3 as compared with other kinases. GS87 demonstrates high efficacy in a mouse AML model system and unlike current AML therapeutics, exhibits little effect on normal bone marrow cells. GS87 induces potent differentiation by more effectively activating GSK3-dependent signaling components including MAPK signaling as compared with other GSK3 inhibitors. GS87 is a novel GSK3 inhibitor with therapeutic potential as a differentiation agent for non-promyelocytic AML. Mol Cancer Ther; 15(7); 1485–94. ©2016 AACR.

Representative flow histograms showing CD11b staining of AML cells after GS87 treatment

Heatmap of Z-scores of genes that are significantly dysregulated

ATRA leads to marked induction of CD11b in Acute Promyelocytic Cells

GS87 does not lead to observable alterations in normal hematopoiesis

Heatmap of Z-scores of genes that are significantly dysregulated

Kinase profiling

Supplementary Figure Legends

GS87 does not significantly impact the viability of AML cells

Primer sequences

Microarray differential expression

The prognosis of most patients with AML is poor due to frequent disease relapse. The cause of relapse is thought to be from the persistence of leukemia initiating cells (LIC's) following treatment. Here we assessed RNA based changes in LICs from matched patient diagnosis and relapse samples using single-cell RNA sequencing. Previous studies on AML progression have focused on genetic changes at the DNA mutation level mostly in bulk AML cells and demonstrated the existence of DNA clonal evolution. Here we identified in LICs that the phenomenon of RNA clonal evolution occurs during AML progression. Despite the presence of vast transcriptional heterogeneity at the single cell level, pathway analysis identified common signaling networks involving metabolism, apoptosis and chemokine signaling that evolved during AML progression and become a signature of relapse samples. A subset of this gene signature was validated at the protein level in LICs by flow cytometry from an independent AML cohort and functional studies were performed to demonstrate co-targeting BCL2 and CXCR4 signaling may help overcome therapeutic challenges with AML heterogeneity. It is hoped this work will facilitate a greater understanding of AML relapse leading to improved prognostic biomarkers and therapeutic strategies to target LIC's.

Background and AimsInflammatory bowel disease is associated with an increased risk of skin cancer. The aims of this study were to determine whether genomic variants associated with IBD susceptibility are also associated with skin cancer susceptibility and if such risk is augmented by the use of immune-suppressive therapy.MethodsThe discovery cohort included participants in the UK Biobank (n=408,381). The validation cohort included participants in the Michigan Genomics Initiative (n=51,405). The primary outcome of interest was skin cancer, sub-grouped into non-melanoma (NMSC) and melanoma skin cancers (MSC). Multivariable logistic regression was performed to identify genomic predictors of skin malignancy. Validated SNPs were evaluated for effect modification by immune-suppressive medication.ResultsThe discovery cohort included 11,079 cases of NMSC and 2,054 cases of MSC. The validation cohort included 7,334 cases of NMSC and 3,304 cases of MSC. Thirty variants were associated with risk of NMSC in the discovery cohort, of which six replicated in the validation cohort [Increased risk: rs7773324-A (DUSP22; IRF4), rs2476601-G (PTPN22), rs1847472-C (BACH2), rs72810983-A (CPEB4); Decreased risk: rs6088765-G (PROCR; MMP24), rs11229555-G (ZFP91-CNTF; GLYAT)]. Twelve variants were associated with risk of MSC in the discovery cohort, of which three replicated in the validation cohort (Increased risk: rs61839660-T (IL2RA); Decreased risk: rs17391694-C (GIPC2; MGC27382), rs6088765-G (PROCR; MMP24)]. No effect modification was observed.ConclusionThe results of this study highlight shared genetic susceptibility across IBD and skin cancer, with increased risk of NMSC in those who carry risk variants in IRF4, PTPN22, CPEB4, and BACH2 and increased risk of MSC in those who carry a risk variant in IL2RA.

Background Gliomas are the most common type of primary brain tumor and one of many cancers where males are diagnosed with greater frequency than females. However, little is known about the sex-based molecular differences in glioblastomas (GBMs) or lower grade glioma (non-GBM) subtypes. DNA methylation is an epigenetic mechanism involved in regulating gene transcription. In glioma and other cancers, hypermethylation of specific gene promoters downregulates transcription and may have a profound effect on patient outcome. The purpose of this study was to determine if sex-based methylation differences exist in different glioma subtypes. Methods Molecular and clinical data from glioma patients were obtained from The Cancer Genome Atlas and grouped according to tumor grade and molecular subtype (IDH1/2 mutation and 1p/19q chromosomal deletion). Sex-specific differentially methylated probes (DMPs) were identified in each subtype and further analyzed to determine if they were part of differentially methylated regions (DMRs) or associated with differentially methylated DNA transcription regulatory binding motifs. Results Analysis of methylation data in 4 glioma subtypes revealed unique sets of both sex-specific DMPs and DMRs in each subtype. Motif analysis based on DMP position also identified distinct sex-based sets of DNA-binding motifs that varied according to glioma subtype. Downstream targets of 2 of the GBM-specific transcription binding sites, NFAT5 and KLF6, showed differential gene expression consistent with increased methylation mediating downregulation. Conclusion DNA methylation differences between males and females in 4 glioma molecular subtypes suggest an important, sex-specific role for DNA methylation in epigenetic regulation of gliomagenesis.

Background The median survival for patients with glioblastoma (GBM), the most common primary malignant brain tumor in adults, has remained approximately 1 year for more than 2 decades. Recent advances in the field have identified GBM as a sexually dimorphic disease. It is less prevalent in females and they have better survival compared to males. The molecular mechanism of this difference has not yet been established. Iron is essential for many biological processes supporting tumor growth and its regulation is impacted by sex. Therefore, we interrogated the expression of a key component of cellular iron regulation, the HFE (homeostatic iron regulatory) gene, on sexually dimorphic survival in GBM. Methods We analyzed TCGA microarray gene expression and clinical data of all primary GBM patients (IDH-wild type) to compare tumor mRNA expression of HFE with overall survival, stratified by sex. Results In low HFE expressing tumors (below median expression, n = 220), survival is modulated by both sex and MGMT status, with the combination of female sex and MGMT methylation resulting in over a 10-month survival advantage (P < .0001) over the other groups. Alternatively, expression of HFE above the median (high HFE, n = 240) is associated with significantly worse overall survival in GBM, regardless of MGMT methylation status or patient sex. Gene expression analysis uncovered a correlation between high HFE expression and expression of genes associated with immune function. Conclusions The level of HFE expression in GBM has a sexually dimorphic impact on survival. Whereas HFE expression below the median imparts a survival benefit to females, high HFE expression is associated with significantly worse overall survival regardless of established prognostic factors such as sex or MGMT methylation.

Background Models of epigenetic aging (epigenetic clocks) have been implicated as potentially useful markers for cancer risk and prognosis. Using 2 previously published methods for modeling epigenetic age, Horvath's clock and epiTOC, we investigated epigenetic aging patterns related to World Health Organization grade and molecular subtype as well as associations of epigenetic aging with glioma survival and recurrence. Methods Epigenetic ages were calculated using Horvath's clock and epiTOC on 516 lower-grade glioma and 141 glioblastoma cases along with 136 nontumor (normal) brain samples. Associations of tumor epigenetic age with patient chronological age at diagnosis were assessed with correlation and linear regression, and associations were validated in an independent cohort of 203 gliomas. Contribution of epigenetic age to survival prediction was assessed using Cox proportional hazards modeling. Sixty-three samples from 18 patients with primary-recurrent glioma pairs were also analyzed and epigenetic age difference and rate of epigenetic aging of primary-recurrent tumors were correlated to time to recurrence. Results Epigenetic ages of gliomas were near-universally accelerated using both Horvath's clock and epiTOC compared with normal tissue. The 2 independent models of epigenetic aging were highly associated with each other and exhibited distinct aging patterns reflective of molecular subtype. EpiTOC was found to be a significant independent predictor of survival. Epigenetic aging of gliomas between primary and recurrent tumors was found to be highly variable and not significantly associated with time to recurrence. Conclusions We demonstrate that epigenetic aging reflects coherent modifications of the epigenome and can potentially provide additional prognostic power for gliomas.

Background Glioblastoma (GBM) is the most common and most malignant glioma. Nonglioblastoma (non-GBM) gliomas (WHO Grades II and III) are invasive and also often fatal. The goal of this study is to determine whether sex differences exist in glioma survival. Methods Data were obtained from the National Cancer Database (NCDB) for years 2010 to 2014. GBM (WHO Grade IV; N = 2073) and non-GBM (WHO Grades II and III; N = 2963) were defined using the histology grouping of the Central Brain Tumor Registry of the United States. Non-GBM was divided into oligodendrogliomas/mixed gliomas and astrocytomas. Sex differences in survival were analyzed using Kaplan–Meier and multivariable Cox proportional hazards models adjusted for known prognostic variables. Results There was a female survival advantage in patients with GBM both in the unadjusted (P = .048) and adjusted (P = .003) models. Unadjusted, median survival was 20.1 months (95% CI: 18.7-21.3 months) for women and 17.8 months (95% CI: 16.9-18.7 months) for men. Adjusted, median survival was 20.4 months (95% CI: 18.9-21.6 months) for women and 17.5 months (95% CI: 16.7-18.3 months) for men. When stratifying by age group (18-55 vs 56+ years at diagnosis), this female survival advantage appeared only in the older group, adjusting for covariates (P = .017). Women (44.1%) had a higher proportion of methylated MGMT (O6-methylguanine-DNA methyltransferase) than men (38.4%). No sex differences were found for non-GBM. Conclusions Using the NCDB data, there was a statistically significant female survival advantage in GBM, but not in non-GBM.

Context Glycogen storage diseases are rare. Increased glycogen in the liver results in increased attenuation. Objective Investigate the association and function of a noncoding region associated with liver attenuation but not histologic nonalcoholic fatty liver disease. Design Genetics of Obesity-associated Liver Disease Consortium. Setting Population-based. Main Outcome Computed tomography measured liver attenuation. Results Carriers of rs4841132-A (frequency 2%-19%) do not show increased hepatic steatosis; they have increased liver attenuation indicative of increased glycogen deposition. rs4841132 falls in a noncoding RNA LOC157273 ~190 kb upstream of PPP1R3B. We demonstrate that rs4841132-A increases PPP1R3B through a cis genetic effect. Using CRISPR/Cas9 we engineered a 105-bp deletion including rs4841132-A in human hepatocarcinoma cells that increases PPP1R3B, decreases LOC157273, and increases glycogen perfectly mirroring the human disease. Overexpression of PPP1R3B or knockdown of LOC157273 increased glycogen but did not result in decreased LOC157273 or increased PPP1R3B, respectively, suggesting that the effects may not all occur via affecting RNA levels. Based on electronic health record (EHR) data, rs4841132-A associates with all components of the metabolic syndrome (MetS). However, rs4841132-A associated with decreased low-density lipoprotein (LDL) cholesterol and risk for myocardial infarction (MI). A metabolic signature for rs4841132-A includes increased glycine, lactate, triglycerides, and decreased acetoacetate and beta-hydroxybutyrate. Conclusions These results show that rs4841132-A promotes a hepatic glycogen storage disease by increasing PPP1R3B and decreasing LOC157273. rs4841132-A promotes glycogen accumulation and development of MetS but lowers LDL cholesterol and risk for MI. These results suggest that elevated hepatic glycogen is one cause of MetS that does not invariably promote MI.

Despite the fact that AML is the most common acute leukemia in adults, patient outcomes are poor necessitating the development of novel therapies. We identified that inhibition of Thioredoxin Reductase (TrxR) is a promising strategy for AML and report a highly potent and specific inhibitor of TrxR, S-250. Both pharmacologic and genetic inhibition of TrxR impairs the growth of human AML in mouse models. We found that TrxR inhibition leads to a rapid and marked impairment of metabolism in leukemic cells subsequently leading to cell death. TrxR was found to be a major and direct regulator of metabolism in AML cells through impacts on both glycolysis and the TCA cycle. Studies revealed that TrxR directly regulates GAPDH leading to a disruption of glycolysis and an increase in flux through the pentose phosphate pathway (PPP). The combined inhibition of TrxR and the PPP led to enhanced leukemia growth inhibition. Overall, TrxR abrogation, particularly with S-250, was identified as a promising strategy to disrupt AML metabolism.

The growth of precision medicine has made access to biobanks with high-quality, well-annotated neuro-oncology biospecimens critical. Developing and maintaining neuro-oncology biobanks is best accomplished through multidisciplinary collaboration between clinicians and researchers. Balancing the needs and leveraging the skills of all stakeholders in this multidisciplinary effort is of utmost importance. Collaboration with a multidisciplinary team of clinicians, health care team members, and institutions, as well as patients and their families, is essential for access to participants in order to obtain informed consent, collect samples under strict standard operating procedures, and accurate and relevant clinical annotation. Once a neuro-oncology biobank is established, development and implementation of policies related to governance and distribution of biospecimens (both within and outside the institution) is of critical importance for sustainability. Proper implementation of a governance process helps to ensure that the biospecimens and data can be utilized in research with the largest potential benefit. New NIH and peer-reviewed journal policies related to public sharing of ‘omic’ data generated from stored biospecimens create new ethical challenges that must be addressed in developing informed consents, protocols, and standard operating procedures. In addition, diversification of sources of funding for the biobanks is needed for long-term sustainability.

Gliomas are the most common primary intracranial neoplasms, which cause significant mortality and morbidity that is disproportionate to their relatively rare incidence. Many potential risk factors for glioma have been studied to date, but only few provide explanation for the number of brain tumor cases identified. The most significant findings include increased risk due to exposure to ionizing radiation and decreased risk with the history of allergy or atopic diseases. The potential effect of the cellular phone usage has been evaluated extensively, but the results remain inconclusive. A very small proportion of gliomas can be attributed to inherited genetic disorders. Additionally, recent analyses using the genome-wide association study design have identified several inherited genomic risk variants.

Gliomas are the most common primary intracranial neoplasms, which cause significant mortality and morbidity that is disproportionate to their relatively rare incidence. Many potential risk factors for glioma have been studied to date, but only few provide explanation for the number of brain tumor cases identified. The most significant findings include increased risk due to exposure to ionizing radiation and decreased risk with the history of allergy or atopic diseases. The potential effect of the cellular phone usage has been evaluated extensively, but the results remain inconclusive. A very small proportion of gliomas can be attributed to inherited genetic disorders. Additionally, recent analyses using the genome-wide association study design have identified several inherited genomic risk variants.

SUMMARYHistomorphology and current grading schemes are unable to predict glioma relapse and malignant tumor progression. We reported that the IDH-mutant associated Glioma-CpG Island Methylator Phenotype (G-CIMP) can be further divided into two clinically distinct subtypes independent of histopathological grading (G-CIMP-high and -low) with evidence of correlation with tumor progression. Here we performed a comprehensive epigenomic analysis of 74 longitudinally collected glioma samples (grade II-IV) to understand malignant recurrence from G-CIMP-high to G-CIMP-low. G-CIMP-low recurrence appeared in 12% of all gliomas and resemble IDH-wildtype primary glioblastoma. G-CIMP-low recurrence can be characterized by distinct epigenetic changes at candidate functional tissue enhancers with AP-1/SOX binding elements, stem cell-like epigenomic phenotype, and genomic instability. Finally, we defined a set of candidate biomarker signatures that predict recurrence of G-CIMP-low with clinically relevance on patient outcomes. Our study provides opportunity for refined clinical trial designs and therapeutic targets that limit progression to more aggressive G-CIMP-low phenotype.HIGHLIGHTSIndolent G-CIMP-high progresses to aggressive G-CIMP-low phenotypeIncidence of G-CIMP-low recurrent tumors are 3 times greater than G-CIMP-low primaryG-CIMP-low recurrent tumors share epigenomic features with IDH-wildtype primary GBMPredictive biomarkers of G-CIMP-low progression at primary diagnosis

Summary: Glioma diagnosis is based on histomorphology and grading; however, such classification does not have predictive clinical outcome after glioblastomas have developed. To date, no bona fide biomarkers that significantly translate into a survival benefit to glioblastoma patients have been identified. We previously reported that the IDH mutant G-CIMP-high subtype would be a predecessor to the G-CIMP-low subtype. Here, we performed a comprehensive DNA methylation longitudinal analysis of diffuse gliomas from 77 patients (200 tumors) to enlighten the epigenome-based malignant transformation of initially lower-grade gliomas. Intra-subtype heterogeneity among G-CIMP-high primary tumors allowed us to identify predictive biomarkers for assessing the risk of malignant recurrence at early stages of disease. G-CIMP-low recurrence appeared in 9.5% of all gliomas, and these resembled IDH-wild-type primary glioblastoma. G-CIMP-low recurrence can be characterized by distinct epigenetic changes at candidate functional tissue enhancers with AP-1/SOX binding elements, mesenchymal stem cell-like epigenomic phenotype, and genomic instability. Molecular abnormalities of longitudinal G-CIMP offer possibilities to defy glioblastoma progression. : IDH-mutant lower-grade glioma glioblastoma often progresses to a more aggressive phenotype upon recurrence. de Souza et al. examines the intra-subtype heterogeneity of initial G-CIMP-high and use this information to identify predictive biomarkers for assessing the risk of recurrence and malignant transformation. Keywords: longitudinal gliomas, DNA methylation, IDH mutation, G-CIMP-high, intra-subtype heterogeneity, malignant transformation and recurrence, G-CIMP-low, stem cell-like glioblastoma, predictive biomarkers

Gliomas are diffusely invasive brain tumors with fatal outcomes and few effective treatments. Precision medicine focuses on targeting the genetics of individual tumors, but not host genetics, despite studies that have linked germline polymorphisms with glioma risk. Accordingly, glioma survival studies in mice utilize genetically variable tumors on identical host genetic backgrounds, which fails to differentiate between cancer cell-autonomous (CCA) and tumor microenvironment (TME) effects on glioma progression and host survival. The Collaborative Cross (CC) is a panel of genetically diverse mouse strains derived from both wild- and traditional inbred laboratory strains that facilitates high-resolution genetic mapping in models of complex disease. Here, we implement a novel platform to discover genetic modifiers of both CCA and TME phenotypes using genetically defined orthotopic murine allograft gliomas and CC hosts. We stereotactically injected Nf1;Trp53-/-oligodendrocyte progenitor-derived mouse tumor cells into syngeneic C57BL/6 control mice and 14 different CC strains. Seven strains survived significantly longer than controls (P Citation Format: Kasey Skinner, Martin Ferris, Ryan Bash, Abigail Shelton, Erin Smithberger, Steve Angus, Brian Golitz, Noah Sciaky, Jeremy Simon, Jason Stein, Glenn Matsushima, Quinn Ostrom, Lindsay Stetson, Jill Barnholtz-Sloan, Harshil Dhruv, Michael Berens, Fernando Pardo Manuel de Villena, C. Ryan Miller. Tumor microenvironment and host genetics impact glioma progression in a Collaborative Cross-based orthotopic allograft model [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 2745.

Glioblastoma multiforme (GBM) is a genomically complex and aggressive primary adult brain tumor, with a median survival time of 12-14 months. The heterogeneous nature of this disease has made the identification and validation of prognostic biomarkers difficult. Using reverse phase protein array data from 203 primary untreated GBM patients, we have identified a set of 13 proteins with prognostic significance. Our protein signature predictive of glioblastoma (PROTGLIO) patient survival model was constructed and validated on independent data sets and was shown to significantly predict survival in GBM patients (log-rank test: p = 0.0009). Using a multivariate Cox proportional hazards, we have shown that our PROTGLIO model is distinct from other known GBM prognostic factors (age at diagnosis, extent of surgical resection, postoperative Karnofsky performance score (KPS), treatment with temozolomide (TMZ) chemoradiation, and methylation of the MGMT gene). Tenfold cross-validation repetition of our model generation procedure confirmed validation of PROTGLIO. The model was further validated on an independent set of isocitrate dehydrogenase wild-type (IDHwt) lower grade gliomas (LGG)-a portion of these tumors progress rapidly to GBM. The PROTGLIO model contains proteins, such as Cox-2 and Annexin 1, involved in inflammatory response, pointing to potential therapeutic interventions. The PROTGLIO model is a simple and effective predictor of overall survival in glioblastoma patients, making it potentially useful in clinical practice of glioblastoma multiforme.

Glioblastoma displays strong sexual dimorphism with male having an increased prevalence and poorer prognosis. While sex based methylation differences in the MGMT promoter have been described in GBM patients, we expanded on these findings with a full genome interrogation of sex-based DNA methylation differences in GBM patients (n = 56 females, n = 77 males). Using Illumina 450k DNA methylation data from The Cancer Genome Atlas (TCGA), we found 359 probes and 12 regions significantly differentially methylated between males and females (sex chromosomes were excluded from the analysis). Males had three times the number of probes significantly differentially hypermethylated than females. In males hypermethylated probes occurred in known enhancer regions at a rate of 4:1 as compared to females. Areas of hypermethylation in females predominately (65%) occurred in CpG islands. An analysis of DNA motif binding sites showed multiple zinc finger transcription factor binding sites located in genomic regions hypermethylated males. Binding sites of KLF6, an important tumor suppressor known to increase p21 expression through a p53 independent pathway, were located in areas significantly hypermethylated in males. We established significant correlation between expression of p21 and methylation of KLF6 binding sites. These findings provide a potential biological basis for our previous observation of increased p21 activity and cell cycle arrest in response to DNA damage in female, but not male GBM astrocytes. Additionally, we found that NFAT5 transcription factor binding sites were significantly hypermethylated in females. These results point to possible protective biology in females, where repression of NFAT5 results in reduced integrin-induced cell dispersion and increased p21 expression results in decreased proliferation. Understanding the molecular basis for sex-based differences will improve our understanding of tumor biology and pave the way for novel diagnostics and a personalized approach for the development of more effective therapies.

Gliomas have fatal outcomes and few effective treatments. Precision medicine promises to target somatic tumor mutations, but the impact of host genetics on glioma biology remains unclear. Germline polymorphisms have been associated with glioma risk, but such studies cannot disambiguate loci affecting cancer cell-autonomous versus tumor microenvironment (TME) phenotypes, since germline variation is shared by both. Here, we implement a novel platform to discover genetic modifiers of both cancer cell autonomous and TME phenotypes using genetically defined non-germline genetically-engineered mouse (nGEM) glioma models and genetically diverse hosts from the Collaborative Cross (CC). We stereotactically injected Nf1;Trp53(-/-) oligodendrocyte progenitor-derived tumor cells into syngeneic C57/BL6 control mice and 14 different CC strains. Survival analysis showed 7 strains with significantly prolonged survival relative to controls (P

Iron homeostasis is essential for cellular energy production particularly in rapidly dividing cells. The HFE protein is critical for regulating cellular iron uptake. To explore the impact of HFE expression in glioblastoma (GBM) and other brain tumors, we utilized TCGA GBM and GBMLGG databases from GlioVis. Within the GBMLGG database which includes lower grade tumors, high expression of HFE was associated with significantly poorer survival. In GBM patients with high HFE expression survival was also significantly worse compared to low HFE expression. There was a strong positive correlation between high HFE expression and tumor grade, with the highest expression in GBM. HFE expression was also correlated to GBM subtype. The lowest expression was in the proneural subtype and the highest in the mesenchymal subtype consistent again with high expression associated with worse prognosis. Normally MGMT methylation is associated with better outcome. However, when combined with high HFE expression there is a significant reduction in survival time (12.9 vs. 20.8 months). To identify the possible molecular basis for these survival differences, we interrogated the reverse phase protein array data (RPPA) for TCGA datasets and discovered a correlation between high HFE expression and expression of a number of proteins. Annexin-1 has the highest correlation of expression with HFE. This protein plays a critical role in innate immune response through the generation of proinflammatory mediators and modulates migration and the phagocytotic ability of neutrophils and macrophages. We have previously shown that mutations in the HFE protein alter macrophage phenotype. Mutations in the HFE gene, the most common autosomal recessive polymorphism found in Caucasians, reportedly impacts the progression of a number of diseases including GBMs. This study identifies HFE as a negative modulator of outcome in brain tumors.

Background: Glioblastoma (GBM) is the most common and lethal primary malignant brain tumor in adults. The heterogeneity of the disease leads to significant variability in response to standard therapy (surgery plus concurrent radiation and temozolomide). An accurate and reliable predictor of patient prognosis represents an unmet need to improve the care of GBM patients. While protein markers are an effective readout of cellular function, proteomics has rarely been utilized in GBM prognostic marker discovery. Experimental Procedures: GBM patients were prospectively recruited (Ohio Brain Tumor Study) and proteomics discovery using liquid chromatography mass spectrometry analysis (LC MS/MS) was performed in a discovery set of 27 patients including 13 short-term survivors (< 9 months, STS) and 14 long-term survivors (>= 18 months, LTS). Statistically significant proteins were evaluated in two independent datasets, including in18 samples micro-dissected from multiple tumor areas of 6 GBM patients. Results: Proteomics discovery identified 11,941 peptides in 2,495 unique proteins, with 172 proteins exhibiting significant dysregulation when comparing STS and LTS. Proteins involved in glycolysis/TCA cycle were up-regulated in STS compared to LTS by examination of individual targets as well as upon application of a novel protein and peptide pathway enrichment analysis. Validation of these dysregulated proteins and other protein markers from the literature in our first validation set (18 samples micro-dissected from n=6 patients) demonstrated that they were very unevenly distributed throughout individual patients' tumors. Several proteins overcame the heterogeneous nature of the tumors and were both prognostic markers differentially expressed between LTS and STS, as well as potential drug targets owing to their even distribution throughout the tumor. Protein abundance of significant marker proteins were verified in a second independent validation set using Western immunoblots. Conclusion: The current study verified the importance of metabolism in GBM pathology and demonstrated the heterogeneous nature of GBMs at the protein level. Citation Format: Lindsay C. Stetson, Quinn T. Ostrom, Peter Liao, Andrew E. Sloan, Mark R. Chance, Jill S. Barnholtz-Sloan. Heterogeneous distribution of prognostic protein markers in glioblastoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 4220.

Standard therapies used for the treatment of acute myeloid leukemia (AML) are cytotoxic agents that target rapidly proliferating cells. Unfortunately, this therapeutic approach has limited efficacy and significant toxicity and the majority of AML patients still die of their disease. In contrast to the poor prognosis of most AML patients, most individuals with a rare subtype of AML, acute promyelocytic leukemia, can be cured by differentiation therapy using regimens containing all-trans retinoic acid. GSK3 has been previously identified as a therapeutic target in AML where its inhibition can lead to the differentiation and growth arrest of leukemic cells. Unfortunately, existing GSK3 inhibitors lead to suboptimal differentiation activity making them less useful as clinical AML differentiation agents. Here, we describe the discovery of a novel GSK3 inhibitor, GS87. GS87 was discovered in efforts to optimize GSK3 inhibition for AML differentiation activity. Despite GS87's dramatic ability to induce AML differentiation, kinase profiling reveals its high specificity in targeting GSK3 as compared with other kinases. GS87 demonstrates high efficacy in a mouse AML model system and unlike current AML therapeutics, exhibits little effect on normal bone marrow cells. GS87 induces potent differentiation by more effectively activating GSK3-dependent signaling components including MAPK signaling as compared with other GSK3 inhibitors. GS87 is a novel GSK3 inhibitor with therapeutic potential as a differentiation agent for non-promyelocytic AML. Mol Cancer Ther; 15(7); 1485–94. ©2016 AACR.

A challenge in precision medicine is the transformation of genomic data into knowledge that can be used to stratify patients into treatment groups based on predicted clinical response. Although clinical trials remain the only way to truly measure drug toxicities and effectiveness, as a scientific community we lack the resources to clinically assess all drugs presently under development. Therefore, an effective preclinical model system that enables prediction of anticancer drug response could significantly speed the broader adoption of personalized medicine.Three large-scale pharmacogenomic studies have screened anticancer compounds in greater than 1000 distinct human cancer cell lines. We combined these datasets to generate and validate multi-omic predictors of drug response. We compared drug response signatures built using a penalized linear regression model and two non-linear machine learning techniques, random forest and support vector machine. The precision and robustness of each drug response signature was assessed using cross-validation across three independent datasets. Fifteen drugs were common among the datasets. We validated prediction signatures for eleven out of fifteen tested drugs (17-AAG, AZD0530, AZD6244, Erlotinib, Lapatinib, Nultin-3, Paclitaxel, PD0325901, PD0332991, PF02341066, and PLX4720).Multi-omic predictors of drug response can be generated and validated for many drugs. Specifically, the random forest algorithm generated more precise and robust prediction signatures when compared to support vector machines and the more commonly used elastic net regression. The resulting drug response signatures can be used to stratify patients into treatment groups based on their individual tumor biology, with two major benefits: speeding the process of bringing preclinical drugs to market, and the repurposing and repositioning of existing anticancer therapies.

Gliomas are the most common type of primary intracranial tumors. Some glioma subtypes cause significant mortality and morbidity that are disproportionate to their relatively rare incidence. A very small proportion of glioma cases can be attributed to inherited genetic disorders. Many potential risk factors for glioma have been studied to date, but few provide explanation for the number of brain tumors identified. The most significant of these factors includes increased risk due to exposure to ionizing radiation, and decreased risk with history of allergy or atopic disease. The potential effect of exposure to cellular phones has been studied extensively, but the results remain inconclusive. Recent genomic analyses, using the genome-wide association study (GWAS) design, have identified several inherited risk variants that are associated with increased glioma risk. The following chapter provides an overview of the current state of research in the epidemiology of intracranial glioma.

Gliomas are the most common type of primary intracranial tumors. Some glioma subtypes cause significant mortality and morbidity that are disproportionate to their relatively rare incidence. A very small proportion of glioma cases can be attributed to inherited genetic disorders. Many potential risk factors for glioma have been studied to date, but few provide explanation for the number of brain tumors identified. The most significant of these factors includes increased risk due to exposure to ionizing radiation, and decreased risk with history of allergy or atopic disease. The potential effect of exposure to cellular phones has been studied extensively, but the results remain inconclusive. Recent genomic analyses, using the genome-wide association study (GWAS) design, have identified several inherited risk variants that are associated with increased glioma risk. The following chapter provides an overview of the current state of research in the epidemiology of intracranial glioma.

Background A challenge in precision medicine is the transformation of genomic data into knowledge that can be used to stratify patients into treatment groups based on predicted clinical response. Although clinical trials remain the only way to truly measure drug toxicities and effectiveness, as a scientific community we lack the resources to clinically assess all drugs presently under development. Therefore, an effective preclinical model system that enables prediction of anticancer drug response could significantly speed the broader adoption of personalized medicine. Results Three large-scale pharmacogenomic studies have screened anticancer compounds in greater than 1000 distinct human cancer cell lines. We combined these datasets to generate and validate multi-omic predictors of drug response. We compared drug response signatures built using a penalized linear regression model and two non-linear machine learning techniques, random forest and support vector machine. The precision and robustness of each drug response signature was assessed using cross-validation across three independent datasets. Fifteen drugs were common among the datasets. We validated prediction signatures for eleven out of fifteen tested drugs (17-AAG, AZD0530, AZD6244, Erlotinib, Lapatinib, Nultin-3, Paclitaxel, PD0325901, PD0332991, PF02341066, and PLX4720). Conclusions Multi-omic predictors of drug response can be generated and validated for many drugs. Specifically, the random forest algorithm generated more precise and robust prediction signatures when compared to support vector machines and the more commonly used elastic net regression. The resulting drug response signatures can be used to stratify patients into treatment groups based on their individual tumor biology, with two major benefits: speeding the process of bringing preclinical drugs to market, and the repurposing and repositioning of existing anticancer therapies., National Science Foundation (U.S.). Graduate Research Fellowship Program (Grant DGEO 9517830), Case Comprehensive Cancer Center (Grant P30 CA043703)

Varying possibilities of tumor relapse for lower grade glioma (LGG, WHO grade II and III) and glioblastoma (GBM, WHO grade IV) have led to heterogenous clinical outcomes for patients. Our current study aims to establish a molecular profile of recurrence of matched primary and recurrent LGG (n = 28) and recurrent GBM (n = 24) tumor samples. The Cancer Genome Atlas (TCGA) has comprehensively profiled these matched primary/recurrent tumor sets; whole genomes, coding exomes, methylomes, and transcriptomes have been sequenced, and the samples have undergone targeted profiling of the proteome. While unsupervised analysis techniques have not led to a clear recurrence signature, supervised analysis methods have revealed interesting patterns. Protein profiling has shown that recurrent gliomas retain the overall molecular signature of their primary counterpart, but the DNA damage response, apoptosis and RTK pathways are downregulated in the recurrent gliomas, in contrast to RAS/MAPK, PI3K/AKT, and EMT pathways, which are upregulated. Whole genome sequencing and rearrangement analysis have revealed increased genomic complexity among most recurrent gliomas as well as new fusions of interest in recurrent LGG samples (PTPRZ1-MET and involving ATRX). Using genome-wide Illumina HumanMethylation 450K data we observed that 78.6% of LGGs showed depletion of DNA methylation at recurrence and 50% of GBM tumors showed an enrichment of DNA methylation at recurrence. Patient centric enrichment analysis allowed us to discover a candidate biological subgroup characterized by a subset of LGG recurrences (50%) exhibiting an aberrant CpG methylation loss at inintergenic opensea regions when compared with canonical CpG islands and shores (fold > 1.3 and confidence intervals of 99%). Importantly, inspection of CpG sites significantly hypomethylated at openseas showed that this pronounced epigenetic signature maps to candidate TSS distal and hypomethylated enhancers. The gene-targets of these hypomethylated CpG sites show a corresponding up-regulation of expression. Pathway analysis has demonstrated that these upregulated genes are involved in cellular growth and proliferation, cellular function and maintenance, and cell cycle regulation. Our results provide evidence that DNA methylation may represent a stable signature of glioma recurrence and that the crosstalk between DNA hypomethylation at openseas and chromosomal instability may be involved in glioma recurrence. We plan to further integrate our findings between data types and correlate with treatment and patient clinical outcome. Citation Format: Lindsay C. Stetson, Camila Ferreira de Souza, Tathiane Maistro Malta, Thais Sarraf Sabedot, Quinn Ostrom, Peter Liao, Daniela Pretti da Cunha Tirapelli, Luciano Neder, Carlos Gilberto Carlotti, Rehan Akbani, Sofie Salama, Laila Poisson, Daniel Brat, The Cancer Genome Atlas Network, Houtan Noushmehr, Jill Barnholtz-Sloan. Multi-omic profiling of gliomas reveals distinct DNA methylation changes at tumor recurrence. [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 780.

Clinical behavior of lower grade glioma (LGG) and glioblastoma (GBM) is characterized by recurrence and progression, yet with highly variable patterns and intervals. Our current study aims to establish a DNA methylation profile associated with the recurrence of LGG (n = 28) and GBM (n = 24) by comparing primary and recurrent matched samples. Illumina Methylation 450k data were used from The Cancer Genome Atlas (TCGA) project. Genome-wide DNA methylation remained almost unchanged at CpG islands and shores. Unsupervised DNA methylation consensus clustering of 2,000 CpG with highest standard deviation (sd > 0.25) demonstrated that recurrent LGG and GBM clustered next to their primary counterparts. Strikingly, 78.6% of recurrent LGG showed depletion of DNA methylation at recurrence and 50% of recurrent GBM showed enrichment of DNA methylation at recurrence. Enrichment analysis of CpG sites with most DNA methylation changes between primary and recurrent samples revealed that 14.3% of LGG and 33.3% of GBM samples showed enrichment of opensea located probes (fold > 1.3). A random permutation of 10,000 probes and confidence intervals (99%) demonstrated that these results do not occur by chance. More than 50% of CpG sites enriched at openseas map to candidate TSS distal and hypomethylated enhancers, and are related to adhesion, apoptotic, developmental, immune system, and metabolic processes. Compared to 10 paired non-glioma tumor (lung, colon, liver, sarcoma), the signature of glioma relapse appears distinctive. Our results provide evidence that DNA methylation may represent a stable signature of glioma recurrence and that the crosstalk between DNA hypomethylation at openseas and chromosomal instability is involved in glioma recurrence. We plan to integrate our findings with whole-genome & RNA sequencing and perform DNA motif analyses to identify networks disrupted and associated with such pronounced epigenetic signature.

Small cell lung cancer (SCLC) represents 15% of all lung carcinomas and is typically diagnosed when the disease has metastasized. Although most SCLC is initially sensitive to chemotherapy regimens, relapse is common and the survival benefit of second-line therapies, such as topotecan, is also limited. There are currently no molecular targeted approaches to treat SCLC similar to those used successfully against non-small-cell lung cancer (NSCLC). This failure is partly attributable to the empiric nature of these efforts. Thus, a more systematic approach to drug discovery for SCLC may lie in mining available databases on the drug sensitivities of SCLC cell lines. In this regard, two comprehensive studies recently published in Nature, the Cancer Cell Line Encyclopedia (CCLE) and the Cancer Genome Project (CGP), examined the drug sensitivities of cancer cell lines and attempted to link these to their genomic profiles. In the present study we have extracted data on SCLC cell lines from these two studies and mined it for new, promising therapeutic agents for SCLC. Our analyses demonstrate that the drug sensitivity of the SCLC cell lines reflects what is observed clinically for metastatic SCLC tumors. That is, most cells were extremely sensitive to topoisomerase (camptothecin) and microtubule (vinblastine and docetaxel) inhibitors, by contrast many tyrosine kinase inhibitors (erlotinib, sorafenib and imatinib) were ineffective. Importantly, we identified heat shock proteins (HSP), cyclin-dependent kinases (CDK) and polo-like kinases (PLK) as attractive molecular targets that have received little/no attention in clinical trials of SCLC. We further developed a gene signature for PLK inhibitor sensitivity and validated it in untested SCLC cell lines. This gene signature was incorporated into circos plots to yield a comprehensive view of how transcriptional, copy number and mutational elements affect PLK sensitivity in SCLC cell lines. Citation Format: Gary Wildey, Yanwen Chen, Ian Lent, Lindsay Stetson, John Pink, Jill Barnholtz-Sloan, Afshin Dowlati. Bioinformatic analysis identifies polo-like kinase as a therapeutic target in small-cell lung cancer. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 4182. doi:10.1158/1538-7445.AM2014-4182

Unrealized needs in the care of patients diagnosed with primary glioblastoma multiforme (GBM) are accurate and clinically relevant predictors of patient prognosis. Patients receiving standard therapy (surgery plus concurrent radiation and temozolomide) have variable clinical outcomes. Available prognostic indicators such as MGMT DNA methylation, IDH1 mutation, and the G-CIMP phenotype are only relevant for a small proportion of those diagnosed with primary GBM. Protein markers of GBM prognosis would not only provide a potential route for classifying and stratifying patients into treatment groups they could also provide an opportunity to identify potential novel drug targets. In this study, using reverse-phase protein array data from Cancer Genome Atlas (TCGA) GBMs, we have constructed and validated a PROTein signature predictive of GLIOblastoma survival (PROTGLIO). Using L1 penalized cox regression, we have identified six protein markers most associated with overall survival in a training set of 107 TCGA GBMs. Three proteins (annexin, cox-2, and FOX03) were associated with shorter overall survival, and three proteins (phosphorylated RPS6KA1, phosphorylated RB1, and TGM2) were associated with longer overall survival. PROTGLIO scores were defined as a linear combination of protein expression levels of the six proteins and the associated Cox regression coefficients, where a higher score indicates a worse prognosis. Based on PROTGLIO scores cases were classified into high and low risk groups. Kaplan-Meier survival analysis showed a significant difference (p = 0.01) in overall survival between the high and low risk groups in our training set. The performance of the PROTGLIO score was validated in a testing set, which consisted of 107 TCGA cases not included in the training set. In the validation set there was a significant difference in overall survival between the high and low risk groups (p = 0.02). We have verified that our signature is independent of known prognostic variables such as age, treatment pattern, and molecular subtype by applying multivariate analysis using a Cox proportional hazard model. The annexin family of proteins has been previously associated with glioma migration and growth and the cox-2 protein has been reported as a marker of poor prognosis among diffuse glioma patients. Our work points to these proteins, along with FOX03, as being negative prognosticators for GBMs and potential avenues for therapeutic intervention. Additionally, we report on the protective effects of phosphorylated RB1 expression, which is supported by previous studies recommending clinical stratification of patients based on RB1 alterations. Further work is needed to elucidate the protective mechanisms of RPS6KA1 and TGM2. Note: This abstract was not presented at the meeting. Citation Format: Lindsay C. Stetson, Jill S. Barnholtz-Sloan. Protein markers predict overall survival in glioblastoma multiforme. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 297. doi:10.1158/1538-7445.AM2014-297