Cell-of-Origin Patterns Dominate the Molecular Classification of 10,000 Tumors from 33 Types of Cancer.

We conducted comprehensive integrative molecular analyses of the complete set of tumors in The Cancer Genome Atlas (TCGA), consisting of approximately 10,000 specimens and representing 33 types of cancer. We performed molecular clustering using data on chromosome-arm-level aneuploidy, DNA hypermethylation, mRNA, and miRNA expression levels and reverse-phase protein arrays, of which all, except for aneuploidy, revealed clustering primarily organized by histology, tissue type, or anatomic origin. The influence of cell type was evident in DNA-methylation-based clustering, even after excluding sites with known preexisting tissue-type-specific methylation. Integrative clustering further emphasized the dominant role of cell-of-origin patterns. Molecular similarities among histologically or anatomically related cancer types provide a basis for focused pan-cancer analyses, such as pan-gastrointestinal, pan-gynecological, pan-kidney, and pan-squamous cancers, and those related by stemness features, which in turn may inform strategies for future therapeutic development.

Perspective on Oncogenic Processes at the End of the Beginning of Cancer Genomics.

The Cancer Genome Atlas (TCGA) has catalyzed systematic characterization of diverse genomic alterations underlying human cancers. At this historic junction marking the completion of genomic characterization of over 11,000 tumors from 33 cancer types, we present our current understanding of the molecular processes governing oncogenesis. We illustrate our insights into cancer through synthesis of the findings of the TCGA PanCancer Atlas project on three facets of oncogenesis: (1) somatic driver mutations, germline pathogenic variants, and their interactions in the tumor; (2) the influence of the tumor genome and epigenome on transcriptome and proteome; and (3) the relationship between tumor and the microenvironment, including implications for drugs targeting driver events and immunotherapies. These results will anchor future characterization of rare and common tumor types, primary and relapsed tumors, and cancers across ancestry groups and will guide the deployment of clinical genomic sequencing.

Comprehensive Molecular Characterization of Muscle-Invasive Bladder Cancer.

We report a comprehensive analysis of 412 muscle-invasive bladder cancers characterized by multiple TCGA analytical platforms. Fifty-eight genes were significantly mutated, and the overall mutational load was associated with APOBEC-signature mutagenesis. Clustering by mutation signature identified a high-mutation subset with 75% 5-year survival. mRNA expression clustering refined prior clustering analyses and identified a poor-survival "neuronal" subtype in which the majority of tumors lacked small cell or neuroendocrine histology. Clustering by mRNA, long non-coding RNA (lncRNA), and miRNA expression converged to identify subsets with differential epithelial-mesenchymal transition status, carcinoma in situ scores, histologic features, and survival. Our analyses identified 5 expression subtypes that may stratify response to different treatments.

Molecular Profiling Reveals Biologically Discrete Subsets and Pathways of Progression in Diffuse Glioma.

Therapy development for adult diffuse glioma is hindered by incomplete knowledge of somatic glioma driving alterations and suboptimal disease classification. We defined the complete set of genes associated with 1,122 diffuse grade II-III-IV gliomas from The Cancer Genome Atlas and used molecular profiles to improve disease classification, identify molecular correlations, and provide insights into the progression from low- to high-grade disease. Whole-genome sequencing data analysis determined that ATRX but not TERT promoter mutations are associated with increased telomere length. Recent advances in glioma classification based on IDH mutation and 1p/19q co-deletion status were recapitulated through analysis of DNA methylation profiles, which identified clinically relevant molecular subsets. A subtype of IDH mutant glioma was associated with DNA demethylation and poor outcome; a group of IDH-wild-type diffuse glioma showed molecular similarity to pilocytic astrocytoma and relatively favorable survival. Understanding of cohesive disease groups may aid improved clinical outcomes.

Multiplatform analysis of 12 cancer types reveals molecular classification within and across tissues of origin.

Recent genomic analyses of pathologically defined tumor types identify "within-a-tissue" disease subtypes. However, the extent to which genomic signatures are shared across tissues is still unclear. We performed an integrative analysis using five genome-wide platforms and one proteomic platform on 3,527 specimens from 12 cancer types, revealing a unified classification into 11 major subtypes. Five subtypes were nearly identical to their tissue-of-origin counterparts, but several distinct cancer types were found to converge into common subtypes. Lung squamous, head and neck, and a subset of bladder cancers coalesced into one subtype typified by TP53 alterations, TP63 amplifications, and high expression of immune and proliferation pathway genes. Of note, bladder cancers split into three pan-cancer subtypes. The multiplatform classification, while correlated with tissue-of-origin, provides independent information for predicting clinical outcomes. All data sets are available for data-mining from a unified resource to support further biological discoveries and insights into novel therapeutic strategies.

Next-generation characterization of the Cancer Cell Line Encyclopedia.

Large panels of comprehensively characterized human cancer models, including the Cancer Cell Line Encyclopedia (CCLE), have provided a rigorous framework with which to study genetic variants, candidate targets, and small-molecule and biological therapeutics and to identify new marker-driven cancer dependencies. To improve our understanding of the molecular features that contribute to cancer phenotypes, including drug responses, here we have expanded the characterizations of cancer cell lines to include genetic, RNA splicing, DNA methylation, histone H3 modification, microRNA expression and reverse-phase protein array data for 1,072 cell lines from individuals of various lineages and ethnicities. Integration of these data with functional characterizations such as drug-sensitivity, short hairpin RNA knockdown and CRISPR-Cas9 knockout data reveals potential targets for cancer drugs and associated biomarkers. Together, this dataset and an accompanying public data portal provide a resource for the acceleration of cancer research using model cancer cell lines.

RPPA SPACE: an R package for normalization and quantitation of Reverse-Phase Protein Array data.

SUMMARY: Reverse-Phase Protein Array (RPPA) is a robust high-throughput, cost-effective platform for quantitatively measuring proteins in biological specimens. However, converting raw RPPA data into normalized, analysis-ready data remains a challenging task. Here, we present the RPPA SPACE (RPPA Superposition Analysis and Concentration Evaluation) R package, a substantially improved successor to SuperCurve, to meet that challenge. SuperCurve has been used to normalize over 170 000 samples to date. RPPA SPACE allows exclusion of poor-quality samples from the normalization process to improve the quality of the remaining samples. It also features a novel quality-control metric, 'noise', that estimates the level of random errors present in each RPPA slide. The noise metric can help to determine the quality and reliability of the data. In addition, RPPA SPACE has simpler input requirements and is more flexible than SuperCurve, it is much faster with greatly improved error reporting. AVAILABILITY AND IMPLEMENTATION: The standalone RPPA SPACE R package, tutorials and sample data are available via https://rppa.space/, CRAN (https://cran.r-project.org/web/packages/RPPASPACE/index.html) and GitHub (https://github.com/MD-Anderson-Bioinformatics/RPPASPACE). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Consensus subtypes of hepatocellular carcinoma associated with clinical outcomes and genomic phenotypes.

BACKGROUND AND AIMS: Although many studies revealed transcriptomic subtypes of HCC, concordance of the subtypes are not fully examined. We aim to examine a consensus of transcriptomic subtypes and correlate them with clinical outcomes. APPROACH AND RESULTS: By integrating 16 previously established genomic signatures for HCC subtypes, we identified five clinically and molecularly distinct consensus subtypes. STM (STeM) is characterized by high stem cell features, vascular invasion, and poor prognosis. CIN (Chromosomal INstability) has moderate stem cell features, but high genomic instability and low immune activity. IMH (IMmune High) is characterized by high immune activity. BCM (Beta-Catenin with high Male predominance) is characterized by prominent ß-catenin activation, low miRNA expression, hypomethylation, and high sensitivity to sorafenib. DLP (Differentiated and Low Proliferation) is differentiated with high hepatocyte nuclear factor 4A activity. We also developed and validated a robust predictor of consensus subtype with 100 genes and demonstrated that five subtypes were well conserved in patient-derived xenograft models and cell lines. By analyzing serum proteomic data from the same patients, we further identified potential serum biomarkers that can stratify patients into subtypes. CONCLUSIONS: Five HCC subtypes are correlated with genomic phenotypes and clinical outcomes and highly conserved in preclinical models, providing a framework for selecting the most appropriate models for preclinical studies.

SAMMI: a semi-automated tool for the visualization of metabolic networks.

SUMMARY: Here we present a browser-based Semi-Automated Metabolic Map Illustrator (SAMMI) for the visualization of metabolic networks. While automated features allow for easy network partitioning, navigation, and node positioning, SAMMI also offers a wide array of manual map editing features. This combination allows for fast, context-specific visualization of metabolic networks as well as the development of standardized, large-scale, visually appealing maps. The implementation of SAMMI with popular constraint-based modeling toolboxes also allows for effortless visualization of simulation results of genome-scale metabolic models. AVAILABILITY AND IMPLEMENTATION: SAMMI has been implemented as a standalone web-based tool and as plug-ins for the COBRA and COBRApy toolboxes. SAMMI and its COBRA plugins are available under the GPL 3.0 license and are available along with documentation, tutorials, and source code at www.SammiTool.com. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

NExUS: Bayesian simultaneous network estimation across unequal sample sizes.

MOTIVATION: Network-based analyses of high-throughput genomics data provide a holistic, systems-level understanding of various biological mechanisms for a common population. However, when estimating multiple networks across heterogeneous sub-populations, varying sample sizes pose a challenge in the estimation and inference, as network differences may be driven by differences in power. We are particularly interested in addressing this challenge in the context of proteomic networks for related cancers, as the number of subjects available for rare cancer (sub-)types is often limited. RESULTS: We develop NExUS (Network Estimation across Unequal Sample sizes), a Bayesian method that enables joint learning of multiple networks while avoiding artefactual relationship between sample size and network sparsity. We demonstrate through simulations that NExUS outperforms existing network estimation methods in this context, and apply it to learn network similarity and shared pathway activity for groups of cancers with related origins represented in The Cancer Genome Atlas (TCGA) proteomic data. AVAILABILITY AND IMPLEMENTATION: The NExUS source code is freely available for download at https://github.com/priyamdas2/NExUS. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

TCPA v3.0: An Integrative Platform to Explore the Pan-Cancer Analysis of Functional Proteomic Data.

Reverse-phase protein arrays represent a powerful functional proteomics approach to characterizing cell signaling pathways and understanding their effects on cancer development. Using this platform, we have characterized ∼8,000 patient samples of 32 cancer types through The Cancer Genome Atlas and built a widely used, open-access bioinformatic resource, The Cancer Proteome Atlas (TCPA). To maximize the utility of TCPA, we have developed a new module called "TCGA Pan-Cancer Analysis," which provides comprehensive protein-centric analyses that integrate protein expression data and other TCGA data across cancer types. We further demonstrate the value of this module by examining the correlations of RPPA proteins with significantly mutated genes, assessing the predictive power of somatic copy-number alterations, DNA methylation, and mRNA on protein expression, inferring the regulatory effects of miRNAs on protein expression, constructing a co-expression network of proteins and pathways, and identifying clinically relevant protein markers. This upgraded TCPA (v3.0) will provide the cancer research community with a more powerful tool for studying functional proteomics and making translational impacts.

Credentialing Individual Samples for Proteogenomic Analysis.

An integrated analysis of DNA, RNA and protein, so called proteogenomic studies, has the potential to greatly increase our understanding of both normal physiology and disease development. However, such studies are challenged by a lack of a systematic approach to credential individual samples resulting in the introduction of noise into the system that limits the ability to identify important biological signals. Indeed, a recent proteogenomic CPTAC study identified 26% of samples as unsatisfactory, resulting in a marked increase in cost and loss of information content. Based on a large-scale analysis of RNA-seq and proteomic data generated by reverse phase protein arrays (RPPA) and by mass spectrometry, we propose a protein-mRNA correlation-based (PMC) score as a robust metric to credential single samples for integrated proteogenomic studies. Samples with high PMC scores have significantly higher protein-mRNA correlation, total protein content and tumor purity. Our results highlight the importance of credentialing individual samples prior to proteogenomic analysis.

Comprehensive Molecular Characterization of Papillary Renal-Cell Carcinoma.

BACKGROUND: Papillary renal-cell carcinoma, which accounts for 15 to 20% of renal-cell carcinomas, is a heterogeneous disease that consists of various types of renal cancer, including tumors with indolent, multifocal presentation and solitary tumors with an aggressive, highly lethal phenotype. Little is known about the genetic basis of sporadic papillary renal-cell carcinoma, and no effective forms of therapy for advanced disease exist. METHODS: We performed comprehensive molecular characterization of 161 primary papillary renal-cell carcinomas, using whole-exome sequencing, copy-number analysis, messenger RNA and microRNA sequencing, DNA-methylation analysis, and proteomic analysis. RESULTS: Type 1 and type 2 papillary renal-cell carcinomas were shown to be different types of renal cancer characterized by specific genetic alterations, with type 2 further classified into three individual subgroups on the basis of molecular differences associated with patient survival. Type 1 tumors were associated with MET alterations, whereas type 2 tumors were characterized by CDKN2A silencing, SETD2 mutations, TFE3 fusions, and increased expression of the NRF2-antioxidant response element (ARE) pathway. A CpG island methylator phenotype (CIMP) was observed in a distinct subgroup of type 2 papillary renal-cell carcinomas that was characterized by poor survival and mutation of the gene encoding fumarate hydratase (FH). CONCLUSIONS: Type 1 and type 2 papillary renal-cell carcinomas were shown to be clinically and biologically distinct. Alterations in the MET pathway were associated with type 1, and activation of the NRF2-ARE pathway was associated with type 2; CDKN2A loss and CIMP in type 2 conveyed a poor prognosis. Furthermore, type 2 papillary renal-cell carcinoma consisted of at least three subtypes based on molecular and phenotypic features. (Funded by the National Institutes of Health.).

Analysis of Genomes and Transcriptomes of Hepatocellular Carcinomas Identifies Mutations and Gene Expression Changes in the Transforming Growth Factor-ß Pathway.

BACKGROUND & AIMS: Development of hepatocellular carcinoma (HCC) is associated with alterations in the transforming growth factor-beta (TGF-ß) signaling pathway, which regulates liver inflammation and can have tumor suppressor or promoter activities. Little is known about the roles of specific members of this pathway at specific of HCC development. We took an integrated approach to identify and validate the effects of changes in this pathway in HCC and identify therapeutic targets. METHODS: We performed transcriptome analyses for a total of 488 HCCs that include data from The Cancer Genome Atlas. We also screened 301 HCCs reported in the Catalogue of Somatic Mutations in Cancer and 202 from Cancer Genome Atlas for mutations in genome sequences. We expressed mutant forms of spectrin beta, non-erythrocytic 1 (SPTBN1) in HepG2, SNU398, and SNU475 cells and measured phosphorylation, nuclear translocation, and transcriptional activity of SMAD family member 3 (SMAD3). RESULTS: We found somatic mutations in at least 1 gene whose product is a member of TGF-ß signaling pathway in 38% of HCC samples. SPTBN1 was mutated in the largest proportion of samples (12 of 202, 6%). Unsupervised clustering of transcriptome data identified a group of HCCs with activation of the TGF-ß signaling pathway (increased transcription of genes in the pathway) and a group of HCCs with inactivation of TGF-ß signaling (reduced expression of genes in this pathway). Patients with tumors with inactivation of TGF-ß signaling had shorter survival times than patients with tumors with activation of TGF-ß signaling (P = .0129). Patterns of TGF-ß signaling correlated with activation of the DNA damage response and sirtuin signaling pathways. HepG2, SNU398, and SNU475 cells that expressed the D1089Y mutant or with knockdown of SPTBN1 had increased sensitivity to DNA crosslinking agents and reduced survival compared with cells that expressed normal SPTBN1 (controls). CONCLUSIONS: In genome and transcriptome analyses of HCC samples, we found mutations in genes in the TGF-ß signaling pathway in almost 40% of samples. These correlated with changes in expression of genes in the pathways; up-regulation of genes in this pathway would contribute to inflammation and fibrosis, whereas down-regulation would indicate loss of TGF-ß tumor suppressor activity. Our findings indicate that therapeutic agents for HCCs can be effective, based on genetic features of the TGF-ß pathway; agents that block TGF-ß should be used only in patients with specific types of HCCs.

Identification and validation of a prognostic proteomic signature for cervical cancer.

OBJECTIVE: To date, The Cancer Genome Atlas (TCGA) has provided the most extensive molecular characterization of invasive cervical cancer (ICC). Analysis of reverse phase protein array (RPPA) data from TCGA samples showed that cervical cancers could be stratified into 3 clusters exhibiting significant differences in survival outcome: hormone, EMT, and PI3K/AKT. The goals of the current study were to: 1) validate the TCGA RPPA results in an independent cohort of ICC patients and 2) to develop and validate an algorithm encompassing a small antibody set for clinical utility. METHODS: Subjects consisted of 2 ICC patient cohorts with accompanying RPPA and clinical-pathologic data: 155 samples from TCGA (TCGA-155) and 61 additional, unique samples (MCW-61). Using data from 173 common RPPA antibodies, we replicated Silhouette clustering analysis in both ICC cohorts. Further, an index score for each patient was calculated from the survival-associated antibodies (SAAs) identified using Random survival forests (RSF) and the Cox proportional hazard regression model. Kaplan-Meier survival analysis and the log-rank test were performed to assess and compare cluster or risk group survival outcome. RESULTS: In addition to validating the prognostic ability of the proteomic clusters reported by TCGA, we developed an algorithm based on 22 unique antibodies (SAAs) that stratified women with ICC into low-, medium-, or high-risk survival groups. CONCLUSIONS: We provide a signature of 22 antibodies which accurately predicted survival outcome in 2 separate groups of ICC patients. Future studies examining these candidate biomarkers in additional ICC cohorts is warranted to fully determine their clinical potential.

Integrated genomic characterization of endometrial carcinoma.

We performed an integrated genomic, transcriptomic and proteomic characterization of 373 endometrial carcinomas using array- and sequencing-based technologies. Uterine serous tumours and ∼25% of high-grade endometrioid tumours had extensive copy number alterations, few DNA methylation changes, low oestrogen receptor/progesterone receptor levels, and frequent TP53 mutations. Most endometrioid tumours had few copy number alterations or TP53 mutations, but frequent mutations in PTEN, CTNNB1, PIK3CA, ARID1A and KRAS and novel mutations in the SWI/SNF chromatin remodelling complex gene ARID5B. A subset of endometrioid tumours that we identified had a markedly increased transversion mutation frequency and newly identified hotspot mutations in POLE. Our results classified endometrial cancers into four categories: POLE ultramutated, microsatellite instability hypermutated, copy-number low, and copy-number high. Uterine serous carcinomas share genomic features with ovarian serous and basal-like breast carcinomas. We demonstrated that the genomic features of endometrial carcinomas permit a reclassification that may affect post-surgical adjuvant treatment for women with aggressive tumours.

Generation of Raw RPPA Data and Their Conversion to Analysis-Ready Data.

Reverse phase protein array (RPPA) provides investigators with a powerful high-throughput, quantitative, cost-effective technology for functional proteomics studies. It is an antibody-based technique with procedures similar to that of Western blots. RPPA has a wide variety of applications that range from pharmacodynamics and drug sensitivity assessment to biomarker discovery, subtype classification, and prediction of patient prognosis and response to targeted therapy. In this paper, we describe the technology, its limitations, and some solutions to overcome them. We discuss the steps necessary to obtain raw RPPA data and convert them into robust, high-quality, analysis-ready data. We then illustrate the utility of the platform by highlighting some biomarkers and drug responses of cancer cell lines that confirm previous findings, as a means to validate the platform and the methods presented here.

Analytical Platforms 3: Processing Samples via the RPPA Pipeline to Generate Large-Scale Data for Clinical Studies.

Reverse phase protein array (RPPA) is a functional proteomics technology amenable to moderately high throughputs of samples and antibodies. The University of Texas MD Anderson Cancer Center RPPA Core Facility has implemented various processes and techniques to maximize RPPA throughput; key among them are maximizing array configuration and relying on database management and automation. One major tool used by the RPPA Core is a semi-automated RPPA process management system referred to as the RPPA Pipeline. The RPPA Pipeline, developed with the aid of MD Avnderson's Department of Bioinformatics and Computational Biology and InSilico Solutions, has streamlined sample and antibody tracking as well as advanced quality control measures of various RPPA processes. This chapter covers RPPA Core processes associated with the RPPA Pipeline workflow from sample receipt to sample printing to slide staining and RPPA report generation that enables the RPPA Core to process at least 13,000 samples per year with approximately 450 individual RPPA-quality antibodies. Additionally, this chapter will cover results of large-scale clinical sample processing, including The Cancer Genome Atlas Project and The Cancer Proteome Atlas.

Spectral Clustering via sparse graph structure learning with application to Proteomic Signaling Networks in Cancer.

Clustering methods for multivariate data exploiting the underlying geometry of the graphical structure between variables are presented. As opposed to standard approaches for graph clustering that assume known graph structures, the edge structure of the unknown graph is first estimated using sparse regression based approaches for sparse graph structure learning. Subsequently, graph clustering on the lower dimensional projections of the graph is performed based on Laplacian embeddings using a penalized k-means approach, motivated by Dirichlet process mixture models in Bayesian nonparametrics. In contrast to standard algorithmic approaches for known graphs, the proposed method allows estimation and inference for both graph structure learning and clustering. More importantly, the arguments for Laplacian embeddings as suitable projections for graph clustering are formalized by providing theoretical support for the consistency of the eigenspace of the estimated graph Laplacians. Fast computational algorithms are proposed to scale the method to large number of nodes. Extensive simulations are presented to compare the clustering performance with standard methods. The methods are applied to a novel pan-cancer proteomic data set, and protein networks and clusters are evaluated across multiple different cancer types.