Search

SF1:Super-enhancer profiling identifies three subtypes of LUSC. SF2: Brn2 is expressed in a subset of human primary LUSC tumors. SF3: Brn2 and Sox2 co-localize and collaborate in the 'neural' LUSC cells. SF4: DNp63 overexpression in the 'neural' LUSC cells suppresses Brn2 expression and induces phenotypic changes. SF5: Brn2 overexpression and p63 ablation in the 'classical' LUSC cells and Brn2 ablation in the 'neural' LUSC cells. SF6: DNp63 induces a classical squamous-cell transcriptional program and G1 phase arrest in the 'neural' LK2 cells. SF7: DNp63 suppresses the 'neural' lineage program and induces the classical squamous-cell transcriptional program. SF8: DNp63 alters ErbB family signaling in the 'neural' LK2 cells.

Molecular characterization of lung squamous cell carcinoma (LUSC), one of the major subtypes of lung cancer, has not sufficiently improved its nonstratified treatment strategies over decades. Accumulating evidence suggests that lineage-specific transcriptional regulators control differentiation states during cancer evolution and underlie their distinct biological behaviors. In this study, by investigating the super-enhancer landscape of LUSC, we identified a previously undescribed “neural” subtype defined by Sox2 and a neural lineage factor Brn2, as well as the classical LUSC subtype defined by Sox2 and its classical squamous partner p63. Robust protein–protein interaction and genomic cooccupancy of Sox2 and Brn2, in place for p63 in the classical LUSC, indicated their transcriptional cooperation imparting this unique lineage state in the “neural” LUSC. Forced expression of p63 downregulated Brn2 in the “neural” LUSC cells and invoked the classical LUSC lineage with more squamous/epithelial features, which were accompanied by increased activities of ErbB/Akt and MAPK–ERK pathways, suggesting differential dependency. Collectively, our data demonstrate heterogeneous cell lineage states of LUSC featured by Sox2 cooperation with Brn2 or p63, for which distinct therapeutic approaches may be warranted.Significance:Epigenomic profiling reveals a novel subtype of lung squamous cell carcinoma with neural differentiation.

ST1: Original sources for the cell lines used in this study. ST2: sgRNA sequences for CRISPR-Cas9 system and primers for quantitative reverse transcription PCR. ST3: Characteristics of human primary LUSC tumor specimens for immunohistochemistry. ST4: Gene loci for super-enhancers near transcriptional regulators shared by all the LUSC cell lines in 'classical' subgroup. ST5: Gene loci for super-enhancers with highest signals near transcriptional regulators in LK2 and NCI-H520 cells. ST6: Associations between gene signatures from DNp63-overexpressed LK2 model and those from human TCGA-LUSC dataset based on POU3F2/ TP63 expression.

There is increased incidence of prostate cancer (PC) among World Trade Center (WTC)-exposed responders and community members, with preliminary evidence suggestive of more aggressive disease. While previous research is supportive of differences in DNA methylation and gene expression as a consequence of WTC exposure, as measured in blood of healthy individuals, the epigenetics of WTC PC tissues has yet to be explored. Patients were recruited from the World Trade Center Health Program. Non-WTC PC samples were frequency matched on age, race/ethnicity and Gleason score. Bisulfite-treated DNA was extracted from tumor tissue blocks and used to assess global DNA methylation with the MethylationEPIC BeadChip. Differential and pathway enrichment analyses were conducted. RNA from the same tumor blocks was used for gene expression analysis to further support DNA methylation findings. Methylation data were generated for 28 samples (13 WTC and 15 non-WTC). Statistically significant differences in methylation were observed for 3,586 genes; on average WTC samples were statistically significantly more hypermethylated (P = 0.04131). Pathway enrichment analysis revealed hypermethylation in epithelial mesenchymal transition (EMT), hypoxia, mitotic spindle, TNFA signaling via NFKB, WNT signaling, and TGF beta signaling pathways in WTC compared to non-WTC samples. The androgen response, G2M and MYC target pathways were hypomethylated. These results correlated well with RNA gene expression. In conclusion, long-term epigenic changes associated with WTC dust exposure were observed in PC tissues. These occurred in genes of critical pathways, likely increasing prostate tumorigenesis potential. This warrants analysis of larger WTC groups and other cancer types.

Lung cancer is the most common cause of cancer-related deaths in both men and women, accounting for one-quarter of total cancer-related mortality globally. Lung adenocarcinoma is the major subtype of non-small cell lung cancer (NSCLC) and accounts for around 40% of lung cancer cases. Lung adenocarcinoma is a highly heterogeneous disease and patients often display variable histopathological morphology, genetic alterations, and genomic aberrations. Recent advances in transcriptomic and genetic profiling of lung adenocarcinoma by investigators, including our group, has provided better stratification of this heterogeneous disease, which can facilitate devising better treatment strategies suitable for targeted patient cohorts. In a recent study we have shown gene expression profiling identified novel clustering of early stage LUAD patients and correlated with tumor invasiveness and patient survival. In this study, we focused on copy number alterations in LUAD patients. SNP array data identified amplification at chromosome 12q15 on MDM2 locus and protein overexpression in a subclass of LUAD patients with an invasive subtype of the disease. High copy number amplification and protein expression in this subclass correlated with poor overall survival. We hypothesized that MDM2 copy number and overexpression predict response to MDM2-targeted therapy. In vitro functional data on a panel of LUAD cells showed that MDM2-targeted therapy effectively suppresses cell proliferation, migration, and invasion in cells with MDM2 amplification/overexpression but not in cells without MDM2 amplification, independent of p53 status. To determine the key signaling mechanisms, we used RNA sequencing (RNA seq) to examine the response to therapy in MDM2-amplified/overexpressing p53 mutant and wild-type LUAD cells. RNA seq data shows that in MDM2-amplified/overexpression with p53 wild-type condition, the E2F → PEG10 → MMPs pathway is operative, while in p53 mutant genetic background, MDM2-targeted therapy abrogates tumor progression in LUAD cells by suppressing epithelial to mesenchymal transition (EMT) signaling. Our study provides a potentially clinically relevant strategy of selecting LUAD patients for MDM2-targeted therapy that may provide for increased response rates and, thus, better survival.

Small cell lung cancer (SCLC), which comprises about 15% of lung cancer cases, is the most aggressive and deadliest type of lung cancer with extremely poor clinical outcomes (about 6% of 5-year overall survival). For about the last three decades, combinatory chemotherapy of etoposide and platinum (EP) treatment has been used as the standard first-line treatment for SCCL. While tumors are generally responsive to the EP treatment, in most cases, they rapidly relapse and acquire resistance after the treatment. However, detailed mechanisms underlying the acquired chemoresistance are not well understood. In this study, we constructed a SCLC comprehensive regulatory network using 135 SCLC tumors, projected chemo-resistant signature genes derived from patient-derived xenografts and genetically engineered mouse models on the network, and identified Granulin (GRN) as a key regulator of the chemo-resistant genes. In multiple independent SCLC datasets, expression levels of GRN and its associated genes increase with the EP treatment and show anti-correlation with neuroendocrine (NE) features of SCLC. Yap activation in a SCLC mouse model increases Grn expression suggesting Yap1 as a potential upstream regulator of Grn. But, on the other hand, the expression levels of GRN and its associated genes are up-regulated in EP treated patient-derived CDX models compared to treatment naïve ones, in which YAP1 expression is depleted in both groups, suggesting YAP1 independent GRN functions associated with chemoresistance. Our observations were validated using 4 SCLC cell lines having different GRN expressions (GRNhigh: SHP77 and H841 and GRNlow: H524 and H2081). The GRNlow showed better responses to the EP treatment compared to the GRNhigh cells (IC50: GRNhigh > 1𝜇M; GRNlow ≈ 1nM). Furthermore, the GRNlow cells with GRN over-expression acquired resistance to the treatment suggesting that GRN expression in SCLC is sufficient for chemo-resistance regardless of YAP1 activation. When we stratified SCLC patients using GRN and its associated genes, the patients in the GRN low group received clear benefits of the chemotherapy with better survival than ones without the treatment (LRT p=0.004) while there were no survival differences among patients regardless of the treatment in the GRN high group. Interestingly, immune checkpoint blockade marker genes were significantly up-regulated in patients from the GRN high group (p= 1.8×10-5, 4.8×10-5, 3.5×10-5, and 0.0002 for PDCD1, CD274, PDCD1LG2, and CTLA4, respectively). Combining this with an observation of that GRN and its associated genes were associated with high PDL1 expression in non-NE SCLC mouse models, immunotherapies might be a potentially effective treatment option for the GRN high group. Our study suggests GRN as a novel key regulator modulating chemo-resistance as well as a potential biomarker for immunotherapy response in SCLC and, hence, provide valuable information in the clinical decisions for better diagnosis, prognosis, and treatment for the purpose of precision medicine. Citation Format: Seungyeul Yoo, Ayushi Patel, Yi Zhong, Feng Jiang, Wenhui Wang, Hideo Watanabe, Jun Zhu. Molecular network analysis identifies GRN as a key regulator of chemotherapy resistance in small cell lung cancer [abstract]. In: Proceedings of the AACR Special Conference: Tumor Immunology and Immunotherapy; 2022 Oct 21-24; Boston, MA. Philadelphia (PA): AACR; Cancer Immunol Res 2022;10(12 Suppl):Abstract nr B53.

Summary Pancreatic ductal adenocarcinoma (PDAC) is a highly aggressive cancer with poor patient survival. Toward understanding the underlying molecular alterations that drive PDAC oncogenesis, we conducted comprehensive proteogenomic analysis of 140 pancreatic cancers, 67 normal adjacent tissues, and 9 normal pancreatic ductal tissues. Proteomic, phosphoproteomic, and glycoproteomic analyses were used to characterize proteins and their modifications. In addition, whole-genome sequencing, whole-exome sequencing, methylation, RNA sequencing (RNA-seq), and microRNA sequencing (miRNA-seq) were performed on the same tissues to facilitate an integrated proteogenomic analysis and determine the impact of genomic alterations on protein expression, signaling pathways, and post-translational modifications. To ensure robust downstream analyses, tumor neoplastic cellularity was assessed via multiple orthogonal strategies using molecular features and verified via pathological estimation of tumor cellularity based on histological review. This integrated proteogenomic characterization of PDAC will serve as a valuable resource for the community, paving the way for early detection and identification of novel therapeutic targets.

Introduction: The current molecular classification of small cell lung cancer (SCLC) based on expression of four lineage transcription factors, SCLC-A (ASCL1), SCLC-N (NEUROD1), SCLC-P (POU2F3), and SCLC-Y (YAP1) still leaves its major subtype SCLC-A as a large heterogeneous group, necessitating more precise characterization of lineage subclasses. Experimental procedure: To refine the current SCLC classification and to identify specific lineage features of the SCLC subtypes, we performed unsupervised hierarchical clustering of H3K27ac profiles on transcriptional regulators from 25 SCLC cell lines and determined the epigenomic features for each cluster. Functional significance of the transcriptional lineage regulators for the identified cluster was evaluated by cell growth, apoptosis and xenograft using CRISPR-Cas9-mediated deletion. The specific cistromic profiles by ChIP-seq and its functional transcriptional partners using co-immunoprecipitation followed by mass spectrometry were determined to reveal their functional output in the identified subtype. Rb1fl/flTrp53fl/fl and Rb1fl/flTrp53fl/flNkx2-1fl/fl genetic engineered mouse model were generated to explore the function of Nkx2-1 in tumor initiation and differentiation. H3K27ac profiles were further analyzed to reveal 6 human SCLC specimen and 20 mice tumors epigenomic landscapes. Summary: We identified previously uncharacterized epigenomic sub-clusters of the major SCLC-A subtype, named SCLC-A1 and SCLC-A2. SCLC-A1 was characterized by the presence of a super-enhancer at the NKX2-1 locus, which was observed in human SCLC specimens and a murine SCLC model. We found NKX2-1, a dual lung and neural lineage factor, is uniquely relevant in SCLC-A1, where it maintains neural lineage rather than pulmonary epithelial identity. We further found maintenance of this neural identity in SCLC-A1 is mediated by collaborative transcriptional activity with another neuronal transcriptional factor SOX1. ? Conclusions: We comprehensively describe an additional epigenomic heterogeneity of the major SCLC-A subtype, and define SCLC-A1 subtype by the core regulatory circuitry representing NKX2-1 and SOX1 super-enhancers and their functional collaborations to maintain a neuronal linage state. Citation Format: ranran kong, Ayushi S. Patel, Takashi Sato, Seungyeul Yoo, Abhilasha Sinha, Yang Tian, Feng Jiang, Charles A. Powell, Eric Snyder, Jiantao Jiang, Shaomin Li, Hideo Watanabe. Transcriptional circuitry of NKX2-1 and SOX1 defines a previously unrecognized lineage subtype of small cell lung cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 5709.

BackgroundHepatitis B virus (HBV) related hepatocellular carcinoma (HCC) is heterogeneous and frequently contains multifocal tumors, but how the multifocal tumors relate to each other in terms of HBV integration and other genomic patterns is not clear.MethodsTo interrogate heterogeneity of HBV-HCC, we developed a HBV genome enriched single cell sequencing (HGE-scSeq) procedure and a computational method to identify HBV integration sites and infer DNA copy number variations (CNVs).ResultsWe performed HGE-scSeq on 269 cells from four tumor sites and two tumor thrombi of a HBV-HCC patient. HBV integrations were identified in 142 out of 269 (53%) cells sequenced, and were enriched in two HBV integration hotspots chr1:34,397,059 (CSMD2) and chr8:118,557,327 (MED30/EXT1). There were also 162 rare integration sites. HBV integration sites were enriched in DNA fragile sites and sequences around HBV integration sites were enriched for microhomologous sequences between human and HBV genomes. CNVs were inferred for each individual cell and cells were grouped into four clonal groups based on their CNVs. Cells in different clonal groups had different degrees of HBV integration heterogeneity. All of 269 cells carried chromosome 1q amplification, a recurrent feature of HCC tumors, suggesting that 1q amplification occurred before HBV integration events in this case study. Further, we performed simulation studies to demonstrate that the sequential events (HBV infecting transformed cells) could result in the observed phenotype with biologically reasonable parameters.ConclusionOur HGE-scSeq data reveals high heterogeneity of HCC tumor cells in terms of both HBV integrations and CNVs. There were two HBV integration hotspots across cells, and cells from multiple tumor sites shared some HBV integration and CNV patterns.

Here we focus on the molecular characterization of clinically significant histological subtypes of early-stage lung adenocarcinoma (esLUAD), which is the most common histological subtype of lung cancer. Within lung adenocarcinoma, histology is heterogeneous and associated with tumor invasion and diverse clinical outcomes. We present a gene signature distinguishing invasive and non-invasive tumors among esLUAD. Using the gene signatures, we estimate an Invasiveness Score that is strongly associated with survival of esLUAD patients in multiple independent cohorts and with the invasiveness phenotype in lung cancer cell lines. Regulatory network analysis identifies aurora kinase as one of master regulators of the gene signature and the perturbation of aurora kinases in vitro and in a murine model of invasive lung adenocarcinoma reduces tumor invasion. Our study reveals aurora kinases as a therapeutic target for treatment of early-stage invasive lung adenocarcinoma.

Historically well-known tumor-promoting gene family plays a key role in cell identity regulation in small cell lung cancer., Comprehensive genomic analyses of small cell lung cancer (SCLC) have revealed frequent mutually exclusive genomic amplification of MYC family members. Hence, it has been long suggested that they are functionally equivalent; however, more recently, their expression has been associated with specific neuroendocrine markers and distinct histopathology. Here, we explored a previously undescribed role of L-Myc and c-Myc as lineage-determining factors contributing to SCLC molecular subtypes and histology. Integrated transcriptomic and epigenomic analyses showed that L-Myc and c-Myc impart neuronal and non-neuroendocrine–associated transcriptional programs, respectively, both associated with distinct SCLC lineage. Genetic replacement of c-Myc with L-Myc in c-Myc–SCLC induced a neuronal state but was insufficient to induce ASCL1-SCLC. In contrast, c-Myc induced transition from ASCL1-SCLC to NEUROD1–SCLC characterized by distinct large-cell neuroendocrine carcinoma–like histopathology. Collectively, we characterize a role of historically defined general oncogenes, c-Myc and L-Myc, for regulating lineage plasticity across molecular and histological subtypes.

Glioblastoma (GBM) is the most aggressive nervous system cancer. Understanding its molecular pathogenesis is crucial to improving diagnosis and treatment. Integrated analysis of genomic, proteomic, post-translational modification and metabolomic data on 99 treatment-naive GBMs provides insights to GBM biology. We identify key phosphorylation events (e.g., phosphorylated PTPN11 and PLCG1) as potential switches mediating oncogenic pathway activation, as well as potential targets for EGFR-, TP53-, and RB1-altered tumors. Immune subtypes with distinct immune cell types are discovered using bulk omics methodologies, validated by snRNA-seq, and correlated with specific expression and histone acetylation patterns. Histone H2B acetylation in classical-like and immune-low GBM is driven largely by BRDs, CREBBP, and EP300. Integrated metabolomic and proteomic data identify specific lipid distributions across subtypes and distinct global metabolic changes in IDH-mutated tumors. This work highlights biological relationships that could contribute to stratification of GBM patients for more effective treatment. Wang et al. perform integrated proteogenomic analysis of adult glioblastoma (GBM), including metabolomics, lipidomics, and single nuclei RNA-Seq, revealing insights into the immune landscape of GBM, cell-specific nature of EMT signatures, histone acetylation in classical GBM, and the existence of signaling hubs which could provide therapeutic vulnerabilities.

Deep proteomics profiling using labeled LC-MS/MS experiments has been proven to be powerful to study complex diseases. However, due to the dynamic nature of the discovery mass spectrometry, the generated data contain a substantial fraction of missing values. This poses great challenges for data analyses, as many tools, especially those for high dimensional data, cannot deal with missing values directly. To address this problem, the NCI-CPTAC Proteogenomics DREAM Challenge was carried out to develop effective imputation algorithms for labeled LC-MS/MS proteomics data through crowd learning. The final resulting algorithm, DreamAI, is based on an ensemble of six different imputation methods. The imputation accuracy of DreamAI, as measured by Pearson correlation, is about 15%-50% greater than existing tools among less abundant proteins, which are more vulnerable to be missed in proteomics data sets. This new tool notably enhances data analysis capabilities in proteomics research.

Background Patient stratification based on molecular subtypes is an important strategy for cancer precision medicine. Deriving clinically informative cancer molecular subtypes from transcriptomic data generated on whole tumor tissue samples is a non-trivial task, especially given the various non-cancer cellular elements intertwined with cancer cells in the tumor microenvironment. Methods We developed a computational deconvolution method, DeClust, that stratifies patients into subtypes based on cancer cell-intrinsic signals identified by distinguishing cancer-type-specific signals from non-cancer signals in bulk tumor transcriptomic data. DeClust differs from most existing methods by directly incorporating molecular subtyping of solid tumors into the deconvolution process and outputting molecular subtype-specific tumor reference profiles for the cohort rather than individual tumor profiles. In addition, DeClust does not require reference expression profiles or signature matrices as inputs and estimates cancer-type-specific microenvironment signals from bulk tumor transcriptomic data. Results DeClust was evaluated on both simulated data and 13 solid tumor datasets from The Cancer Genome Atlas (TCGA). DeClust performed among the best, relative to existing methods, for estimation of cellular composition. Compared to molecular subtypes reported by TCGA or other similar approaches, the subtypes generated by DeClust had higher correlations with cancer-intrinsic genomic alterations (e.g., somatic mutations and copy number variations) and lower correlations with tumor purity. While DeClust-identified subtypes were not more significantly associated with survival in general, DeClust identified a poor prognosis subtype of clear cell renal cancer, papillary renal cancer, and lung adenocarcinoma, all of which were characterized by CDKN2A deletions. As a reference profile-free deconvolution method, the tumor-type-specific stromal profiles and cancer cell-intrinsic subtypes generated by DeClust were supported by single-cell RNA sequencing data. Conclusions DeClust is a useful tool for cancer cell-intrinsic molecular subtyping of solid tumors. DeClust subtypes, together with the tumor-type-specific stromal profiles generated by this pan-cancer study, may lead to mechanistic and clinical insights across multiple tumor types.

Additional file 1 : Figure S1. A schematic illustration of DeClust algorithm. Figure S2 Flowchart of the study. Figure S3. Simulation results when gene expression was simulated under negative binomial distribution. Details can be found in legend of Fig. 1. Figure S4. Simulation results when gene expression was simulated under negative binomial distribution. Details can be found in legend of Fig. 1. Figure S5. Comparing cell fraction estimations by different gene expression deconvolution methods with the ones based on MethyCIBERSORT (treated as “ground truth”) for immune compartment (AB) and stromal compartment (CD) using Spearman’s correlation coefficients (AC) or Median Absolute Deviation (BD). The p-values above are the difference between DeClust and other methods according the two-sided paired t-test. Figure S6. Kaplan-Meier curves of patients with high/low stromal compartment fractions as defined by DeClust (left) and ESTIMATE (right) in TCGA dataset (top) and non-TCGA dataset (bottom) for KIRC (A) and BLCA (B). Figure S7. Pathway analysis of stromal proles across 13 TCGA datasets using Canonical pathways (left) or hallmark_cancer pathways (right) from MsigDB. The color indicates the significance of the up/down-regulation of that pathway in that stromal prole as compared to other stromal proles (−log10 (p-value of Wilcoxon rank-sum test), see Method). Red color represents up-regulation and blue color represents down-regulation. Figure S8. Number of genes with subtype-specific methylation before (A) and after (B) methylation correlated with immune or stromal cell frequencies were removed. Figure S9. Overlap between CRIS subtypes and subtypes defined by different methods. Figure S10. Plots of the correlation between tumor purity (Consensus Purity Estimation*) and subtypes defined by different methods, grouped by methods (A) or by cancer types (B). Figure S11. Kaplan-Meier curves of DeClust subtypes (left) and TCGA subtypes (right) for TCGA dataset (top) and non-TCGA dataset (bottom) for KIRP (A), KIRC (B) and LUAD (C). Figure S12. (A) Spearman’s correlation coefficient between immune and stromal proles estimated by DeClust (y-axis) and reference proles used by EPIC (x-axis). (B) Exemplary scatter plot of reference expression prole versus immune and stromal prole estimated by Declust using TCGA BLCA dataset.(C) Correlation between different reference proles(x-axis) and bulk tissue expression proles (gray), immune proles (red) and stromal proles (blue) estimated by DeClust using TCGA BLCA dataset. (D) Cell type-specific markers identified in EPIC reference proles and their expression in immune and stromal proles estimated by DeClust. Figure S13. t-SNE plot of scRNAseq data for pRCC samples. Cell-type specific markers used to annotate the cell clusters were shown below. Figure S14. Correlation between mean expression prole of each epithelial cell cluster and subtype-specific cancer proles estimated by DeClust or by TCGA for ccRCC scRNAseq data (A) and pRCC scRNAseq data(B). Error bars indicated 95% confidence intervals. Figure S15. Comparison of inferred cancer proles and scRNAseq data. Figure S16. Violin plot of estimated immune (left) and stromal (fraction) fraction in each subtype of BLCA. Figure S17. Proportions of different immune subtypes within each subtypes defined by DeClust. Figure S18. Comparison of immune subtypes and subtypes defined by DeClust in association with overall survival. Plots are grouped by methods (A) and by cancer types(B). Figure S19. (A) The association of overall survival and immune subtypes within each cancer cell-intrinsic subtype defined by Declust.(B) Kaplan-Meier curves of immune subtypes within Atypical_1 subtype of HNSC. Figure S20. (A) The association of overall survival and immune/stromal cell fraction within each subtype defined by Declust. The color in the heatmap represents the –log10(p-value) of the association (log rank test). Red means the higher cell fraction corresponds to worse survival, and blue means the opposite. Only cancer subtypes and cell fractions with at least one significant association (p

A large amount of panomic data has been generated in populations for understanding causal relationships in complex biological systems. Both genetic and temporal models can be used to establish causal relationships among molecular, cellular, or phenotypical traits, but with limitations. To fully utilize high-dimension temporal and genetic data, we develop a multivariate polynomial temporal genetic association (MPTGA) approach for detecting temporal genetic loci (teQTLs) of quantitative traits monitored over time in a population and a temporal genetic causality test (TGCT) for inferring causal relationships between traits linked to the locus. We apply MPTGA and TGCT to simulated data sets and a yeast F2 population in response to rapamycin, and demonstrate increased power to detect teQTLs. We identify a teQTL hotspot locus interacting with rapamycin treatment, infer putative causal regulators of the teQTL hotspot, and experimentally validate RRD1 as the causal regulator for this teQTL hotspot., Temporal omics data have the potential to dissect complex biological networks. Here the authors develop methods for detecting temporal genetic loci (teQTLs) of quantitative traits monitored over time and inferring causal relationships between traits linked to the locus.

Comprehensive genomic analyses of small cell lung cancer (SCLC), the most aggressive form of lung cancer, have revealed near universal loss of tumor suppressors (RB1 and TP53) and frequent genomic amplification of all three MYC family members. The amplification of each Myc family member is mutually exclusive; hence it had been long suggested that they are functionally equivalent. However, their expression has more recently been associated with specific neuroendocrine markers and distinct histopathology. In this study, we explored a novel role of c-Myc and L-Myc as lineage determining factors contributing to SCLC molecular subtypes and histology. Integrated analyses of a gene regulatory network generated from mRNA expression of primary SCLC tumor and chromatin state profiling of SCLC cell lines showed that Myc family members impart distinct transcriptional programs associated with lineage state; wherein the L-Myc signature was enriched for neuronal pathways while the c-Myc signature was enriched for Notch signaling and epithelial-to-mesenchymal transition. We investigated the functional redundancy and distinction of c-Myc and L-Myc, and noted the insufficiency of L-Myc to induce lineage switch in contrast to the potential of c-Myc to induce trans-differentiation. c-Myc rewires the Myc-accessible landscape and activates neuron al repressor, Rest to mediate transition from ASCL1-SCLC to NeuroD1-SCLC characterized by distinct LCNEC-like histopathology. Collectively, our findings reveal a previously undescribed role of historically defined general oncogenes, c-Myc and L-Myc, for regulating lineage plasticity across molecular subtypes as well as histological subclasses.

Multiple myeloma (MM) is the second most prevalent hematological cancer. MM is a complex and heterogeneous disease, and thus, it is essential to leverage omics data from large MM cohorts to understand the molecular mechanisms underlying MM tumorigenesis, progression, and drug responses, which may aid in the development of better treatments. In this study, we analyzed gene expression, copy number variation, and clinical data from the Multiple Myeloma Research Consortium (MMRC) dataset and constructed a multiple myeloma molecular causal network (M3CN). The M3CN was used to unify eight prognostic gene signatures in the literature that shared very few genes between them, resulting in a prognostic subnetwork of the M3CN, consisting of 178 genes that were enriched for genes involved in cell cycle (fold enrichment = 8.4, p value = 6.1 &times, 10&minus, 26). The M3CN was further used to characterize immunomodulators and proteasome inhibitors for MM, demonstrating the pleiotropic effects of these drugs, with drug-response signature genes enriched across multiple M3CN subnetworks. Network analyses indicated potential links between these drug-response subnetworks and the prognostic subnetwork. To elucidate the structure of these important MM subnetworks, we identified putative key regulators predicted to modulate the state of these subnetworks. Finally, to assess the predictive power of our network-based models, we stratified MM patients in an independent cohort, the MMRF-CoMMpass study, based on the prognostic subnetwork, and compared the performance of this subnetwork against other signatures in the literature. We show that the M3CN-derived prognostic subnetwork achieved the best separation between different risk groups in terms of log-rank test p-values and hazard ratios. In summary, this work demonstrates the power of a probabilistic causal network approach to understanding molecular mechanisms underlying the different MM signatures.

Background Data errors, including sample swapping and mis-labeling, are inevitable in the process of large-scale omics data generation. Data errors need to be identified and corrected before integrative data analyses where different types of data are merged on the basis of the annotated labels. Data with labeling errors dampen true biological signals. More importantly, data analysis with sample errors could lead to wrong scientific conclusions. We developed a robust probabilistic multi-omics data matching procedure, proMODMatcher, to curate data and identify and correct data annotation and errors in large databases. Results Application to simulated datasets suggests that proMODMatcher achieved robust statistical power even when the number of cis-associations was small and/or the number of samples was large. Application of our proMODMatcher to multi-omics datasets in The Cancer Genome Atlas and International Cancer Genome Consortium identified sample errors in multiple cancer datasets. Our procedure was not only able to identify sample-labeling errors but also to unambiguously identify the source of the errors. Our results demonstrate that these errors should be identified and corrected before integrative analysis. Conclusions Our results indicate that sample-labeling errors were common in large multi-omics datasets. These errors should be corrected before integrative analysis.

Molecular characterization of lung squamous cell carcinoma (LUSC), one of the major subtypes of lung cancer, has not sufficiently improved its nonstratified treatment strategies over decades. Accumulating evidence suggests that lineage-specific transcriptional regulators control differentiation states during cancer evolution and underlie their distinct biological behaviors. In this study, by investigating the super-enhancer landscape of LUSC, we identified a previously undescribed “neural” subtype defined by Sox2 and a neural lineage factor Brn2, as well as the classical LUSC subtype defined by Sox2 and its classical squamous partner p63. Robust protein–protein interaction and genomic cooccupancy of Sox2 and Brn2, in place for p63 in the classical LUSC, indicated their transcriptional cooperation imparting this unique lineage state in the “neural” LUSC. Forced expression of p63 downregulated Brn2 in the “neural” LUSC cells and invoked the classical LUSC lineage with more squamous/epithelial features, which were accompanied by increased activities of ErbB/Akt and MAPK–ERK pathways, suggesting differential dependency. Collectively, our data demonstrate heterogeneous cell lineage states of LUSC featured by Sox2 cooperation with Brn2 or p63, for which distinct therapeutic approaches may be warranted. Significance: Epigenomic profiling reveals a novel subtype of lung squamous cell carcinoma with neural differentiation.

Summary Sample mislabeling or misannotation has been a long-standing problem in scientific research, particularly prevalent in large-scale, multi-omic studies due to the complexity of multi-omic workflows. There exists an urgent need for implementing quality controls to automatically screen for and correct sample mislabels or misannotations in multi-omic studies. Here, we describe a crowdsourced precisionFDA NCI-CPTAC Multi-omics Enabled Sample Mislabeling Correction Challenge, which provides a framework for systematic benchmarking and evaluation of mislabel identification and correction methods for integrative proteogenomic studies. The challenge received a large number of submissions from domestic and international data scientists, with highly variable performance observed across the submitted methods. Post-challenge collaboration between the top-performing teams and the challenge organizers has created an open-source software, COSMO, with demonstrated high accuracy and robustness in mislabeling identification and correction in simulated and real multi-omic datasets., Highlights • A community effort to combat sample mislabeling in multi-omic studies • Computational solutions received show a wide range of accuracy • The final collaborative product, COSMO, achieves high performance • Applying COSMO to published datasets demonstrates biological impact of the tool, The bigger picture Sample mislabeling is a well-recognized problem in scientific research, particularly prevalent in large-scale, multi-omic studies, due to the complexity of multi-omic workflows. Here, we describe a crowdsourced precisionFDA NCI-CPTAC Multi-omics Enabled Sample Mislabeling Correction Challenge, which provides a framework for systematic benchmarking and evaluation of mislabel identification and correction methods for integrative proteogenomic studies. Individual solutions submitted by the challenge participants, even those from the same team, show a wide range of accuracy, underscoring the importance of the benchmarking effort. Post-challenge collaboration between the top-performing teams and the challenge organizers has created an open-source software, COSMO, with demonstrated high accuracy and robustness in mislabeling identification and correction in simulated and real multi-omic datasets., A community effort to combat sample mislabeling in multi-omic studies leads to an open-source software, COSMO, with demonstrated high accuracy and robustness in mislabeling identification and correction in simulated and real multi-omic datasets.

We present a proteogenomic study of 108 human papilloma virus (HPV)-negative head and neck squamous cell carcinomas (HNSCCs). Proteomic analysis systematically catalogs HNSCC-associated proteins and phosphosites, prioritizes copy number drivers, and highlights an oncogenic role for RNA processing genes. Proteomic investigation of mutual exclusivity between FAT1 truncating mutations and 11q13.3 amplifications reveals dysregulated actin dynamics as a common functional consequence. Phosphoproteomics characterizes two modes of EGFR activation, suggesting a new strategy to stratify HNSCCs based on EGFR ligand abundance for effective treatment with inhibitory EGFR monoclonal antibodies. Widespread deletion of immune modulatory genes accounts for low immune infiltration in immune-cold tumors, whereas concordant upregulation of multiple immune checkpoint proteins may underlie resistance to anti-programmed cell death protein 1 monotherapy in immune-hot tumors. Multi-omic analysis identifies three molecular subtypes with high potential for treatment with CDK inhibitors, anti-EGFR antibody therapy, and immunotherapy, respectively. Altogether, proteogenomics provides a systematic framework to inform HNSCC biology and treatment.

We report a comprehensive proteogenomics analysis, including whole-genome sequencing, RNA sequencing, and proteomics and phosphoproteomics profiling, of 218 tumors across 7 histological types of childhood brain cancer: low-grade glioma (n = 93), ependymoma (32), high-grade glioma (25), medulloblastoma (22), ganglioglioma (18), craniopharyngioma (16), and atypical teratoid rhabdoid tumor (12). Proteomics data identify common biological themes that span histological boundaries, suggesting that treatments used for one histological type may be applied effectively to other tumors sharing similar proteomics features. Immune landscape characterization reveals diverse tumor microenvironments across and within diagnoses. Proteomics data further reveal functional effects of somatic mutations and copy number variations (CNVs) not evident in transcriptomics data. Kinase-substrate association and co-expression network analysis identify important biological mechanisms of tumorigenesis. This is the first large-scale proteogenomics analysis across traditional histological boundaries to uncover foundational pediatric brain tumor biology and inform rational treatment selection.

Background: This study aims to determine the impact of intracellular hepatitis B virus (HBV) DNA, covalently closed circular DNA (cccDNA) and viral replicative activity in both tumour and non-neoplastic liver on prognosis and to determine the relationship of viral replicative activity and Ishak fibrosis in predicting outcome following resection. Methods: A total of 99 prospectively enrolled patients treated with primary liver resection for HBV-HCC are included. Intracellular HBV DNA and cccDNA were quantitated by real-time PCR. The RNA-sequencing (RNA-seq) was performed in a subset of 21 patients who had either minimal liver fibrosis (Ishak stages 0–2) or end-stage fibrosis (Ishak stage 6). Results: Tumour tissue contained a lower cccDNA copy number compared with paired non-neoplastic liver, and larger tumours (>3 cm) had less cccDNA compared with small tumours (⩽3 cm). High viral replicative activity in non-neoplastic liver was associated with higher HCC recurrence rate independent of Ishak fibrosis stage. Genes correlated with viral replicative activity in non-neoplastic liver (620 genes) were distinct from those associated with end-stage fibrosis (1226 genes). Genes associated with viral replicative activity were preferentially distributed in regions on chr3, chr16 and chr19. Conclusions: Viral replicative activity in non-neoplastic liver is associated with HCC recurrence through mechanisms that are distinct from and independent of Ishak fibrosis stage.

Patients with head and neck squamous cell carcinomas (HNSCCs) are treated with surgery, radiation, chemotherapy, and limited targeted therapies. Compared to human papillomavirus (HPV)-positive HNSCCs, HPV-negative cases have worse treatment response and prognosis and represent an unmet clinical need. We performed comprehensive proteogenomic characterization of tumor specimens, matched normal adjacent tissues (NATs), and blood samples from 109 HPV-negative HNSCC patients. This cohort is dominated by tumors from oral cavity (45, 41%) and larynx (49, 45%). Somatic mutation and somatic copy number analyses validated previously reported genomic aberrations in HPV-negative HNSCC. Proteomics analysis linked p53 loss of heterozygosity to increased expression of EPCAM, a stemness marker. Additionally, FAT1 truncation mutations were associated with increased expression of proteins involved in keratinization, a key feature of SCC differentiation. Deletions of 3p and 9p led to the loss of genes encoding p16, chemokine receptors, and interferon/JAK/STAT signaling pathway proteins, whereas amplifications of 3q and 11q led to overexpression of proteins involved in cell proliferation and anti-apoptosis pathways. Comparative analysis of tumor and NAT proteomes and phosphoproteomes identified putative diagnostic biomarkers and druggable targets, and proteogenomic integration further identified putative neoantigens. Tumor site-specific characterization associated epigenetic silencing of neurofilaments with laryngeal but not oral cavity SCC. Protein targets of FDA approved or investigational drugs for HNSCC treatment showed high inter-tumor heterogeneity in their protein abundances. DNA copy number and RNA expression level were good surrogates of protein abundance for some targets, such as EGFR and PD-L1, but they failed to reflect protein levels or kinase activities for other targets, such as MMP9 and MTOR. Thus, there is a critical need for protein biomarker-driven treatment stratification. Deconvolution of bulk tumor gene expression data revealed an immune-hot subgroup and an immune-cold subgroup. Immune-hot tumors broadly overexpressed multiple immune checkpoints including PD-L1, IDO1, and CTLA4, underscoring the necessity of combination immune checkpoint inhibition to improve treatment efficacy. Immune-cold tumors were characterized by smoking, chromosomal instability, and activation of the CDK4/6-pRb axis, suggesting they could be targeted by CDK4/6 inhibitors. We also noted that EGFR-amplified tumors frequently harbor copy number aberrations of downstream signaling components of the EGFR pathway. This may explain the low response rate of EGFR-amplified tumors to EGFR inhibitors, and targeting multiple pathway components, including EGFR, PIK3CA and STAT3, may be required for these tumors. In summary, our integrative proteogenomic characterization revealed multiple novel insights into the pathogenesis and treatment of HPV-negative HNSCCs. Citation Format: Chen Huang, Lijun Chen, Yize Li, Sara Savage, Michael Schnaubelt, Felipe V. Leprevost, Marcin Cieslik, Yongchao Dou, Bo Wen, Jonathan T. Lei, Kai Li, Eric Jaehnig, Zhiao Shi, Meenakshi Anurag, Jianbo Pan, Yingwei Hu, Rodrigo V. Eguez, David J. Clark, Matthew Wyczalkowski, Saravana M. Dhanasekaran, Chandan Kumar, Antonio Colaprico, Karsten Krug, Michael Gillette, D. R. Mani, Seungyeul Yoo, Jiayi Ji, Xiaoyu Song, Weiping Ma, Xi Steven Chen, Alex Pico, Nathan J. Edwards, Scott D. Jewell, Mathangi Thiagarajan, Emily S. Boja, Henry Rodriguez, Andrew Sikora, Pei Wang, Matthew Ellis, Gilbert S. Omenn, Li Ding, Alexey I. Nesvizhskii, Adel K. EI-Naggar, Daniel W. Chan, Hui Zhang, Bing Zhang, Clinical Proteomic Tumor Analysis Consortium. Proteogenomics characterization of HPV-negative head and neck squamous cell carcinomas [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 5118.

Small cell lung cancer (SCLC) is the most aggressive subtype of lung cancer with a dismal prognosis. The standard-of-care remains to be uniform treatment with chemotherapy and radiotherapy, while emerging evidence suggests its molecular heterogeneity that has been previously under-appreciated. In primary SCLC, the gene loci for Myc family members are amplified mutually exclusively, their expression is correlated with unique neuroendocrine markers and distinct histopathology of xenografts from SCLC cell lines and murine SCLC. In this study, we use integrative genomic and epigenomic analyses to explore a novel role for c-Myc and L-Myc as lineage determining factors to bridge the gap between SCLC molecular subtypes and histological classification. First, we built a novel network using the Bayesian estimation from combined mRNA expression profile datasets for a total of 135 primary SCLC tumors. This revealed distinct transcriptional networks for c-Myc and L-Myc; wherein L-Myc was enriched for neuronal pathways and c-Myc for Notch signaling and epithelial-to-mesenchymal transition. The assay for transposase accessible-chromatin profiling of 3 representative cell lines for each c-Myc and L-Myc, revealed enrichment of biological processes involved in neuronal development for L-Myc expressing cell lines and active Notch signaling in c-Myc expressing cell lines. Together, these analyses implied that c-Myc and L-Myc govern distinct transcriptional programs to impart respective transcriptional networks associated with features unique to SCLC molecular subtypes. Next, we genetically engineered c-Myc amplified SCLC to exchange c-Myc with L-Myc and found L-Myc regulates neuronal associated pathways but was insufficient to induce lineage switch, however; c-Myc was required for the maintenance of NeuroD1-driven lineage state. In contrast, exogenous expression of c-Myc in classical-ASCL1-positive SCLC revealed incompatibility of c-Myc expression in this subtype, and led to suppression of neuronal associated pathways, trans-differentiation to NeuroD1-SCLC accompanied by variant histopathological features. Pharmacological inhibition of neuroendocrine-low associated Notch signaling and its target RE-1 silencing transcription factor (REST), revealed c-Myc-induced suppression of ASCL1 is not mediated by Notch signaling but rather by direct activation of REST expression. Collectively, our findings reveal a previously undescribed role for historically defined general oncogenes, c-Myc and L-Myc, in regulating lineage plasticity across SCLC molecular subtypes as well as histological classes. Citation Format: Ayushi S. Patel, Seungyeul Yoo, Ranran Kong, Takashi Sato, Maya Fridrikh, German Nudelman, Charles A. Powell, Jun Zhu, Hideo Watanabe. Myc family members differentially regulate lineage plasticity in small cell lung cancer [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 1295.

To explore the biology of lung adenocarcinoma (LUAD) and identify new therapeutic opportunities, we performed comprehensive proteogenomic characterization of 110 tumors and 101 matched normal adjacent tissues (NATs) incorporating genomics, epigenomics, deep-scale proteomics, phosphoproteomics, and acetylproteomics. Multi-omics clustering revealed four subgroups defined by key driver mutations, country, and gender. Proteomic and phosphoproteomic data illuminated biology downstream of copy number aberrations, somatic mutations, and fusions and identified therapeutic vulnerabilities associated with driver events involving KRAS, EGFR, and ALK. Immune subtyping revealed a complex landscape, reinforced the association of STK11 with immune-cold behavior, and underscored a potential immunosuppressive role of neutrophil degranulation. Smoking-associated LUADs showed correlation with other environmental exposure signatures and a field effect in NATs. Matched NATs allowed identification of differentially expressed proteins with potential diagnostic and therapeutic utility. This proteogenomics dataset represents a unique public resource for researchers and clinicians seeking to better understand and treat lung adenocarcinomas.

Gastric cancer (GC) is a heterogeneous disease in which diverse genetic, genomic, and epigenetic alterations can accumulate in different molecular and histologic subtypes. Tumor microenvironment (TME) also contributes to the heterogeneity of GC. To investigate what molecular features of tumor cells drive GC heterogeneity, we developed an integrative causal model, called Integrative Sequential Causality Test (ISCT), to identify key regulators of GC by integrating DNA methylation, copy number variation, and transcriptomic data. Applying ISCT to three GC cohorts that contain methylation, CNV, and gene expression data, 11 common methylation-driven key regulators were identified: ADHFE1, CDO1, CRYAB, FSTL1, GTP, PKP3, PTPRCAP, RAB25, RHOH, SFN, and SORD. Based on these 11 genes, gastric tumors resolved into three groups that were associated with known molecular subtypes, Lauren classification, tumor stage, and patient survival, suggesting significance of the methylation-driven key regulators in molecular and histologic heterogeneity of GC. We also investigated the relationship between TME and the methylation-driven key regulators and showed that both immune/stromal proportions in TME and tumor cell genomics variations contributed to expression variations of the methylation-driven key regulators. Especially, FSTL1, significantly associated with patient survival and tumor progression as well as stromal proportion in TME, was expressed at high level in both stromal and cancer cells, indicating its potential role in mediating tumor-stroma interactions. In summary, this study suggests that genetic, genomic, and epigenetic alterations as well as their interactions with TME contribute to heterogeneity of GC. Citation Format: Seungyeul Yoo, Quan Chen, Li Wang, Wenhui Wang, Ankur Chakravarthy, Rita Busuttil, Alex Boussioutas, Tim R. Fenton, Jiangwen Zhang, Xiaodan Fan, Seut-Yi Leung, Jun Zhu. Molecular heterogeneity of gastric cancer explained by methylation-driven key regulators [abstract]. In: Proceedings of the AACR Special Conference on Tumor Immunology and Immunotherapy; 2019 Nov 17-20; Boston, MA. Philadelphia (PA): AACR; Cancer Immunol Res 2020;8(3 Suppl):Abstract nr B103.

SUMMARY To elucidate the deregulated functional modules that drive clear cell renal cell carcinoma (ccRCC), we performed comprehensive genomic, epigenomic, transcriptomic, proteomic, and phosphoproteomic characterization of treatment-naive ccRCC and paired normal adjacent tissue samples. Genomic analyses identified a distinct molecular subgroup associated with genomic instability. Integration of proteogenomic measurements uniquely identified protein dysregulation of cellular mechanisms impacted by genomic alterations, including oxidative phosphorylation-related metabolism, protein translation processes, and phospho-signaling modules. To assess the degree of immune infiltration in individual tumors, we identified microenvironment cell signatures that delineated four immune-based ccRCC subtypes characterized by distinct cellular pathways. This study reports a large-scale proteogenomic analysis of ccRCC to discern the functional impact of genomic alterations and provides evidence for rational treatment selection stemming from ccRCC pathobiology., Graphical Abstract, In Brief Comprehensive proteogenomic characterization in 103 treatment-naive clear cell renal cell carcinoma patient samples highlights tumor-specific alterations at the proteomic level that are unrevealed by transcriptomic profiling and proposes a revised subtyping scheme based on integrated omics analysis.

Osteosarcoma (OS), the most common primary bone tumor, is highly metastatic with high chemotherapeutic resistance and poor survival rates. Using induced pluripotent stem cells (iPSCs) generated from Li–Fraumeni syndrome (LFS) patients, we investigate an oncogenic role of secreted frizzled-related protein 2 (SFRP2) in p53 mutation-associated OS development. Interestingly, we find that high SFRP2 expression in OS patient samples correlates with poor survival. Systems-level analyses identified that expression of SFRP2 increases during LFS OS development and can induce angiogenesis. Ectopic SFRP2 overexpression in normal osteoblast precursors is sufficient to suppress normal osteoblast differentiation and to promote OS phenotypes through induction of oncogenic molecules such as FOXM1 and CYR61 in a β-catenin–independent manner. Conversely, inhibition of SFRP2, FOXM1, or CYR61 represses the tumorigenic potential. In summary, these findings demonstrate the oncogenic role of SFRP2 in the development of p53 mutation-associated OS and that inhibition of SFRP2 is a potential therapeutic strategy.

Osteosarcoma (OS), the most common primary bone tumor, is highly metastatic with high chemotherapeutic resistance and poor survival rates. Using induced pluripotent stem cells (iPSCs) generated from Li-Fraumeni syndrome (LFS) patients, we investigated an oncogenic role of secreted frizzled-related protein 2 (sFRP2) in P53 mutation-associated OS development. Interestingly, we found that high sFRP2 expression in OS patient samples correlates with poor survival. Systems-level analyses identified that expression of sFRP2 increases during LFS OS development and can induce angiogenesis. Ectopic sFRP2 overexpression in normal osteoblast precursors is sufficient to suppress normal osteoblast differentiation and to promote OS phenotypes through induction of oncogenic molecules such as FOXM1 and CYR61 in a β-catenin independent manner. Conversely, inhibition of sFRP2, FOXM1 or CYR61 represses the tumorigenic potential. In summary, these findings demonstrate the oncogenic role of sFRP2 in P53 mutation-associated OS development and that inhibition of sFRP2 is a potential therapeutic strategy.

e19526 Background: Multiple myeloma (MM) is the 2nd most prevalent blood cancer. For each molecular subtype of MM defined by molecular profiles, mechanisms underlying prognosis and drug response are largely unknown. Elucidating the mechanisms underlying differences in prognosis and drug response will enable a more personalized, precision medicine (PM) approach to treating patients. Methods: Big multiomics data are now widely available and contain the ingredients necessary to build causal models to uncover mechanisms of prognosis and drug response. However, data from resources like the MM Research Consortium (MMRC) dataset may contain errors (eg, sample mislabeling) that diminish modeling accuracy. We developed a probabilistic data matching method ( proMODMatcher) to correct such errors and applied it to the MMRC data. We identified labeling errors in 10 patients, including swaps in RNA and CNV profiles. Given the corrected MMRC data, we characterized recurrent genomic aberrations in MM. We found > 8000 genes significantly associated (FDR < 0.01) with CNVs spanning the gene locations. To distinguish expression correlations driven by causal biological relationships from those driven by coincident CNV influence, we constructed a causal MM network based on the MMRC dataset. Results: Overlaps among multiomic prognostic or drug response biomarkers are sparse. To identify common mechanisms of different biomarkers, we projected them onto our MM network to define the causal context in which they occur. Prognostic biomarkers were enriched in subnetworks associated with mitotic regulation and lipid metabolism, and predicted by our network to be regulated by ZWINT, BUB1B, DTL, TPX2 and NOP16. Proteasome inhibitor and immunomodulating drug responses were mediated by amino acid biosynthesis and cell surface protein complex subnetworks, respectively, with MTHFD2 and TMC8 inferred as the master regulators of these subnetworks, respectively. Regulators include genes (eg, BUB1B and MTHFD2) known to associate with tumor growth and drug response and novel therapeutic control points. Conclusions: Causal models can elucidate the molecular mechanisms underlying prognosis and drug response, enabling the design of more personalized MM treatments.

International audience; Oligodendrocytes derive from progenitors (OPCs) through the interplay of epigenomic and transcriptional events. By integrating high-resolution methylomics, RNA-sequencing, and multiple transgenic lines, this study defines the role of DNMT1 in developmental myelination. We detected hypermethylation of genes related to cell cycle and neurogenesis during differentiation of OPCs, yet genetic ablation of Dnmt1 resulted in inefficient OPC expansion and severe hypomyelination associated with ataxia and tremors in mice. This phenotype was not caused by lineage switch or massive apoptosis but was characterized by a profound defect of differentiation associated with changes in exon-skipping and intron-retention splicing events and by the activation of an endoplasmic reticulum stress response. Therefore, loss of Dnmt1 in OPCs is not sufficient to induce a lineage switch but acts as an important determinant of the coordination between RNA splicing and protein synthesis necessary for myelin formation.

Introduction: Lung cancer is the major cause of cancer-related deaths in United States and lung adenocarcinoma (LUAD) is its most common subtype exhibiting highly variable molecular signature. Identification of patients at high risk of developing invasive tumor is important for their better management. Experimental procedure: We obtained 53 early-stage LUAD patient's tumor samples, histopathologically classified into 32 indolent (adenocarcinoma in situ, minimally invasive and lepidic predominant adenocarcinoma) and 21 invasive (acinar predominant, papillary predominant, micropapillary predominant and solid predominant adenocarcinoma) subtypes and profiled their transcriptomes by RNA sequencing. We clustered these samples in an unsupervised manner followed by additional supervised clustering into 2 groups. Using our signature genes, we categorized 51 lung adenocarcinoma cell lines from Cancer Cell Line Encyclopedia database based on their relative invasiveness scores derived from elastic network trained in our dataset. We chose NCI-H1792, NCI-H1373, SK-LU-1 and NCI-H2009 representing high while HCC1833, Calu3 and NCI-H3255 representing low invasive signature for our in vitro functional assays. Summary: Our clustering significantly overlapped with the groups classified by histology and we defined 21 tumors as “invasive” and 32 as “non-invasive” early stage LUAD. We identified 1,322 differentially expressed genes between two groups with 526 upregulated and 796 downregulated genes in invasive tumors. Among the enriched signature genes, AURKA and AURKB were significantly upregulated in the “invasive” group. We used two potent aurora kinase inhibitors AMG 900 and PF-03814735 to determine the effect of aurora kinase inhibition on invasiveness of LUAD cells. AMG 900 and PF-03814735 significantly inhibited Aurora kinase (Aur-) A and B activity, but did not affect cell viability (IC50 ≥2μM) of all cell lines regardless of their invasiveness score. Of note, both drugs significantly suppressed wound healing, migration and invasion of all four “invasive” cell lines at much lower 100nM. Moreover, the cells with low invasive signature used in the study showed little migration and invasion thus confirming the validity of our elastic network analysis for their relative invasiveness. Conclusion: Here, we have identified a robust gene signature that distinguishes invasive to indolent non-invasive early-stage LUADs, which may lead to development of assays on clinical specimens such as biopsy to identify patients at high risk of developing invasive LUAD at an early stage. This gene signature also suggests Aur-A and B promote invasiveness in early stage LUAD. Our in vitro functional data suggest that targeting Aur-A and B could be a promising approach for management of early stage LUAD patients who are more likely to develop an invasive tumor. Citation Format: Abhilasha Sinha, Seungyeul Yoo, Hideo Watanabe, Jun Zhu, Charles A. Powell. Aurora kinase: A target modulating invasiveness of lung adenocarcinoma [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 814.

Introduction: Liver cancer is the second cause of cancer-related mortality worldwide. Currently, there are only two systemic agents able to increase survival in patients at advanced stages (i.e. sorafenib and regorafenib). Median survival of these patients is still poor, which highlights the need for new therapies. Our aim is to identify key regulatory gene networks with oncogenic properties amenable for therapeutic intervention through 1) integration of gene expression and DNA methylation data from human HCC samples followed by 2) functional validation in mice using shRNA screens. Methods: DNA methylation (Illumina HM450) and mRNA expression (Affymetrix human genome U219) data of 215 human HCC samples (Villanueva, Hepatology 2015) were analyzed to identify key gene regulatory networks. A causality test interrogated the impact of cis and trans regulation of promoter methylation on gene expression (Yoo, PLoS Genet 2015). A key regulator gene was defined when it regulated a substantial number of downstream genes (more than 2 standard deviations from the mean predicted trans-regulated downstream genes). Data analysis includes differential gene expression, topological overlap clustering (hierarchical, non-negative factorization [NMF]), and gene annotation. The tumorigenic potential of candidate tumor suppressors was experimentally validated through a positive selection shRNA (short-hairpin RNA) screen in mice (6 shRNAs/gene, 48 shRNAs/pool, 5 mice/library). Results: We identified 116 potential HCC key regulator genes, predicted to regulate expression of between 1,484 and 3,511 downstream genes for each one. Among the key regulators, 60 were classified as potential tumor maintenance genes and 56 as potential tumor suppressors genes based on differential expression with non-tumoral tissue (FDR Conclusions: We have identified 116 putative key regulators of hepatocarcinogenesis by integrative analysis of gene expression and DNA methylation. Functional validation indicates enrichment of bona fide tumor suppressors. Citation Format: Amanda J. Craig, Seungyeul Yoo, Verónica Miguela, Delia D'Avola, Pedro Molina-Sánchez, Josep Llovet, Amaia Lujambio, Jun Zhu, Augusto Villanueva. Integrative molecular analysis of gene expression and methylation reveals 116 putative key regulator genes of human hepatocarcinogenesis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 4393. doi:10.1158/1538-7445.AM2017-4393