Comprehensive genomic characterization of squamous cell lung cancers

Summary

Lung squamous cell carcinoma (lung SqCC) is a common type of lung cancer, causing approximately 400,000 deaths per year worldwide. Genomic alterations in lung SqCC have not been comprehensively characterized and no molecularly targeted agents have been developed specifically for its treatment. As part of The Cancer Genome Atlas (TCGA), we profiled 178 lung SqCCs to provide a comprehensive landscape of genomic and epigenomic alterations. Lung SqCC is characterized by complex genomic alterations, with a mean of 360 exonic mutations, 165 genomic rearrangements, and 323 segments of copy number alteration per tumor. We found statistically recurrent mutations in 18 genes in including mutation of TP53 in nearly all specimens. Previously unreported loss-of-function mutations were seen in the HLA-A class I major histocompatibility gene. Significantly altered pathways included NFE2L2/KEAP1 in 34%, squamous differentiation genes in 44%, PI3K/AKT in 47%, and CDKN2A/RB1 in 72% of tumors. We identified a potential therapeutic target in the majority of tumors, offering new avenues of investigation for lung SqCC treatment.

Introduction

Lung cancer is the leading cause of cancer-related mortality worldwide, leading to an estimated 1.4 million deaths in 2010¹. The discovery of recurrent mutations in the Epidermal Growth Factor Receptor (EGFR) kinase as well as fusions involving the Anaplastic Lymphoma Kinase (ALK), has led to a dramatic change in the treatment of patients with lung adenocarcinoma, the most common type of lung cancer^2–5. More recent data have suggested that targeting mutations in BRAF, AKT1, ERBB2 and PIK3CA and fusions that involve ROS1 and RET may also be successful^6,7. Unfortunately, activating mutations in EGFR and ALK fusions are typically not present in the second most common type of lung cancer, lung squamous cell carcinoma (lung SqCC),⁸ and targeted agents developed for lung adenocarcinoma are largely ineffective against lung SqCC.

Although no comprehensive genomic analysis of lung SqCCs has been reported, single-platform studies have identified regions of somatic copy number alteration in lung SqCCs, including amplification of SOX2, PDGFRA and FGFR1/WHSC1L1 and deletion of CDKN2A^9,10. DNA sequencing studies of lung SqCCs have reported recurrent mutations in a number of genes, including TP53, NFE2L2, KEAP1, BAI3, FBXW7, GRM8, MUC16, RUNX1T1, STK11 and ERBB4^11,12. DDR2 mutations and FGFR1 amplification have been nominated as therapeutic targets^13–15.

We have conducted a comprehensive study of lung SqCCs from a large cohort of patients as part of The Cancer Genome Atlas (TCGA) project. The twin aims are to characterize the genomic/epigenomic landscape of lung SqCC and to identify potential opportunities for therapy. We report an integrated analysis based on DNA copy number, somatic exonic mutations, mRNA sequencing (RNA-seq), mRNA expression, and promoter methylation for 178 histopathologically reviewed lung SqCCs in addition to whole-genome sequencing (WGS) of 19 samples and miRNA sequencing of 159 samples (Supplementary Table S1.1). Demographic and clinical data and results of the genomic analyses can be downloaded from the TCGA data portal (https://tcga-data.nci.nih.gov/tcga/tcgaDownload.jsp).

Samples and clinical data

Tumor samples were obtained from 178 patients with previously untreated stage I–IV lung SqCC. Germline DNA was obtained from adjacent, histologically normal tissues resected at the time of surgery (n=137) or peripheral blood (n=41). All patients provided written informed consent to conduct genomic studies in accordance with local Institutional Review Boards. The demographic characteristics are described in Supplementary Table S1.2. The median follow-up for the cohort was 15.8 months, and 60% of patients were alive at the time of last follow-up (data updated in November 2011). 96% of the patients had a history of tobacco use, similar to previous reports for North American lung SqCC patients¹⁶. DNA and RNA were extracted from patient specimens and were measured by several genomic assays, which included standard quality control assessments (Supplementary Methods, sections 2–8). A committee of experts in lung cancer pathology performed an additional review of all samples to confirm the histological subtype (Supplementary Figure 1.1 and Supplementary Methods, section 1).

Somatic DNA Alterations

The lung SqCCs analyzed in this study display a large number and variety of DNA alterations, with a mean 360 exonic mutations, 323 altered copy number segments and 165 genomic rearrangements per tumor.

Copy number alterations were analyzed using multiple platforms. Analysis of SNP 6.0 array data across the set of 178 lung SqCCs identified a high rate of copy number alteration (mean of 323 segments) when compared to other TCGA projects (as of February 1, 2012) including ovarian cancer (477 segments),¹⁷ glioblastoma multiforme (282 segments),¹⁸ colorectal carcinoma (213 segments), breast carcinoma (282 segments) and renal cell carcinoma (156 segments). These segments gave rise to regions of both focal and broad somatic copy number alterations (SCNAs) with a mean of 47 focal and 23 broad events per tumor (broad events defined as ≥50% of the length of the chromosome arm). There was strong concordance between the three independent copy number assays for all regions of SCNA (Supplementary Figures S2.1–2.4).

At the level of whole chromosome arm SCNAs, lung SqCCs exhibit many similarities to 205 cases of lung adenocarcinoma analyzed by TCGA (Supplementary Figure S2.1a). The most striking difference between these cancers is selective amplification of chromosome 3q in lung SqCC, as has been reported^9,19 (p<1×10⁻¹⁵ by Fisher’s exact test). Using the SNP 6.0 array platform and GISTIC 2.0^20,21, we identified regions of significant copy number alteration (Supplementary Methods, section 2). There were 50 peaks of significant amplification or deletion (q<0.05), several of which included SCNAs previously seen in lung SqCCs including SOX2, PDGFRA/KIT, EGFR, FGFR1/WHSC1L1, CCND1 and CDKN2A^9,10,19 (Supplementary Figure S2.1b and Supplementary Data S2.1 and S2.2). Other peaks defined regions of SCNA reported for the first time including amplifications of chromosomal segments containing NFE2L2, MYC, CDK6, MDM2, BCL2L1 and EYS and deletions of FOXP1, PTEN and NF1 (Supplementary Figure S2.1b).

Whole exome sequencing of 178 lung SqCCs and matched germline DNA, targeted 193,094 exons from 18,863 genes. Mean sequencing coverage across targeted bases in was 121X with 83% of target bases above 30X coverage. We identified a total of 48,690 non-silent mutations with a mean of 228 non-silent and 360 total exonic mutations per tumor, corresponding to a mean somatic mutation rate of 8.1 mutations per megabase (Mb) and median of 8.4/Mb. That rate is higher than rates observed in other TCGA projects including acute myelogenous leukemia (0.56/Mb), breast carcinoma (1.0/Mb), ovarian cancer¹⁷ (2.1/Mb), glioblastoma multiforme¹⁸ (2.3/Mb) and colorectal carcinoma (3.2/Mb) (data as of February 1, 2012, p<2.2 x10⁻¹⁶ by t-test or Wilcoxon rank sum test for lung SqCC versus all others). In lung SqCC, CpG transitions and transversions were the most commonly observed mutation types, with mean rates of 9.9 and 10.7 per sequenced megabase of CpG context, respectively, for a total mutation rate of 20.6/Mb. At non-CpG sites, transversions at C:G sites were more common than transitions (7.3 vs. 2.9 per Mb; total = 10.2/Mb) and more common than transversions or transitions at A:T sites (1.5 vs. 1.3 per Mb; total = 2.8/Mb).

Significantly mutated genes were identified using a modified version of the MutSig algorithm (Supplementary Methods, section 3)^22,23. We identified 10 genes with a false discovery rate (FDR) q value <0.1 (Supplementary Table S3.1): TP53, CDKN2A, PTEN, PIK3CA, KEAP1, MLL2, HLA-A, NFE2L2, NOTCH1 and RB1, all of which demonstrated robust evidence of gene expression as defined by Reads Per Kilobase of exon model per Million mapped reads (RPKM) >1 (Figure 1). TP53 mutation was observed in 81% of samples by automated analysis; visual review of sequencing reads identified an additional 9% of samples with potential mutations in regions of sub-optimal coverage or in samples with low purity. The majority of observed mutations in NOTCH1 (8 of 17) were truncating alterations, suggesting loss-of-function, as has recently been reported for head and neck SCCs^22,24. Mutations in HLA-A were also almost exclusively nonsense or splice site events (7 of 8).

Figure 1. Significantly mutated genes in lung SqCC.

Significantly mutated genes (q-value <0.1) identified by exome sequencing are listed vertically by q-value. The percentage of lung SqCC samples with a mutation detected by automated calling is noted at the left. Samples displayed as columns, with the overall number of mutations plotted at the top and samples arranged to emphasize mutual exclusivity among mutations.

To increase our statistical power to detect mutated genes in the setting of the observed high background mutation rate, we performed a secondary MutSig analysis only considering genes previously observed to be mutated in cancer according to the COSMIC database. This yielded 12 additional genes with FDR <0.1: FAM123B (WTX), HRAS, FBXW7, SMARCA4, NF1, SMAD4, EGFR, APC, TSC1, BRAF, TNFAIP3 and CREBBP (Supplementary Table S3.1). Both the spectrum and the frequency of EGFR mutations differed from those seen in lung adenocarcinomas. The two most common alterations in lung adenocarcinoma, L858R and in-frame deletions in exon 19, were absent, whereas two L861Q mutations were detected in EGFR.

As described in Supplemental Figure S3.1 we verified somatic mutations by performing an independent hybrid-recapture of 76 genes in all samples. 1289 mutations were assayed and we achieved satisfactory coverage to have power to verify at 1283 positions. 1235 mutations were validated (96.2%) (Supplementary Figure S3.1 and Supplementary Methods, section 3). We also verified mutation calls using WGS and RNA-sequencing data with similar results (Supplementary Figures S3.1 and S4.3 and Supplementary Methods, sections 3 and 4).

Whole genome sequencing was performed for 19 tumor/normal pairs with a mean computed coverage of 54X. A mean of 165 somatic rearrangements was found per lung SqCC tumor pair (Supplementary Figure S3.2), a value in excess of that reported for WGS studies of other tumor types including colorectal carcinoma (75)²⁵, prostate carcinoma (108)²⁶, multiple myeloma (21)²³, and breast cancer (90)²⁷. Although the majority of in-frame coding fusions detected in WGS were validated by RNA-seq, no recurrent rearrangements predicted to generate fusion proteins were identified (Supplementary Data S3.1 and S4.1).

Somatically altered pathways

Many of the somatic alterations we have identified in lung SqCCs appear to be drivers of pathways important to the initiation or progression of the cancer. Specifically, genes involved in oxidative stress response and squamous differentiation were frequently altered by mutation or SCNA. We observed mutations and copy number alterations of NFE2L2 and KEAP1 and/or deletion or mutation of CUL3 in 34% of cases (Figure 2). NFE2L2 and KEAP1 code for proteins that bind each other, have been shown to regulate the cell’s response to oxidative damage, chemo- and radiotherapy and are somatically altered in a variety of cancer types^28,29. We found mutations in NFE2L2 almost exclusively in one of two KEAP1 interaction motifs, DLG or ETGE. Mutations in KEAP1 and CUL3 showed a pattern consistent with loss-of-function and were mutually exclusive with mutations in NFE2L2 (Figures 1c and 2). PARADIGM³⁰ analysis predicts that mutations in NFE2L2 and KEAP1 exert a significant functional impact (Supplementary Figure S7.C.1 and S.7.C.2 and Supplementary Methods, section 7).

Figure 2. Somatically altered pathways in squamous cell lung cancer.

Left, Alterations in oxidative response pathway genes by somatic mutation as defined by somatic mutation, copy number alteration or up- or down-regulation. Frequencies of alteration are expressed as a percentage of all cases, with background in red for activated genes and blue for inactivated genes. Right, Alterations in genes that regulate squamous differentiation, as defined in the left panel.

We also found alterations in genes with known roles in squamous cell differentiation in 44% of samples, including overexpression and amplification of SOX2 and TP63, loss-of-function mutations in NOTCH1, NOTCH2, and ASCL4 and focal deletions in FOXP1 (Figure 2). Although NOTCH1 has been well characterized as an oncogene in hematologic cancers³¹, NOTCH1 and NOTCH2 truncating mutations have been reported in cutaneous SCCs and lung SqCCs³². Truncating mutations in ASCL4 are the first to be reported in human cancer and may have a lineage role given the requirement for ASCL1 for survival of small cell lung cancer cells³³. Alterations in NOTCH1, NOTCH2 and ASCL4 were mutually exclusive and exhibited minimal overlap with amplification of TP63 and/or SOX2 (Figure 2), suggesting that aberrations in those modulators of squamous cell differentiation have overlapping functional consequences.

mRNA expression profiling and subtype classification

Whole-transcriptome expression profiles were generated by RNA-sequencing for the entire cohort and by microarrays for a 121-sample subset. Of 20,502 genes analyzed, the mean RNA coverage indices were 19X and 6420 RPKM (Supplementary Figure S4.1 and Supplementary Methods, section 4). Previously reported lung SqCC gene expression-subtype signatures³⁴ were applied to both of the expression platforms, yielding four subtypes designated as classical (36%), basal (25%), secretory (24%) and primitive (15%). The concordance of subtypes between the two platforms was high (94% agreement) (Supplementary Figure S4.2). Significant correlations were found between the expression subtypes and genomic alterations in copy number, mutation and methylation (Figure 3). The classical subtype was characterized by alterations in KEAP1, NFE2L2, and PTEN, as well as pronounced hypermethylation and chromosomal instability. The 3q26 amplicon was present in all of the subtypes, but it was most characteristic of the classical subtype, which also showed the greatest overexpression of three known oncogenes on 3q: SOX2, TP63, and PIK3CA. RNA sequencing data suggested that high expression levels of TP63, in samples with and without amplification of TP63, was associated with dominant expression of the deltaN isoform (also called p40), which lacks the N-terminal transactivation domain, compared to the longer isoform, called tap63 (89% of tumors overexpressed deltaN compared to tap63; p<2.2e-16). That short deltaN isoform is thought to function as an oncogene^35,36, and its expression was most enriched in the classical subtype. In contrast, the primitive expression-subtype more commonly exhibited RB1 and PTEN alterations, and the basal expression-subtype showed NF1 alterations (Figure 3). Amplification of FGFR1 and WHSC1L1 was anti-correlated with the classical subtype and specifically with NFE2L2 or KEAP1 mutated samples. While CDKN2A alterations are common in lung SqCCs, they are not associated with any particular expression subtype (Figure 3).

Figure 3. Gene expression subtypes integrated with genomic alterations.

Tumors are displayed as columns, grouped by gene expression subtype. Subtypes were compared by Kruskal-Wallis tests for continuous features and by Fisher’s exact tests for categorical features. Displayed features displayed showed significant association with gene expression subtype (P<0.05), except for CDKN2A alterations. deltaN expression percentage represents transcript isoform usage between the TP63 isoforms, deltaN and tap63, as determined by RNA-sequencing. Chromosomal instability (CIN) is defined by the mean of the absolute values of chromosome arm copy numbers from the GISTIC^23,24 output. Absolute values are used so that amplification and deletion alterations are counted equally. Hypermethylation scores and iCluster assignments are described in Supplementary Figure S6.1 and S7.A1, respectively. CIN, methylation, gene expression, and deltaN values were standardized for display using z-score transformation.

Independent clustering of miRNA and methylation data indicated association with expression subtypes. The highest overall methylation was seen in the classical subtype (Figure 3, Supplementary Figures S5.1 and S6.1, Supplementary Methods, sections 5 and 6, Supplementary Data S6.1 and S6.2 and Supplementary Table 5.1). Integrative clustering (iCluster)³⁷ of mRNA, miRNA, methylation, SCNA, and mutation data demonstrated concordance with the mRNA expression-subtypes and associated alterations. (Figure 3, Supplementary Figure S7. A.1 and Supplementary Methods, section 7). Independent correlation of somatic mutations, copy number alterations and gene expression signatures revealed significant subtype associations with alterations in the TP53, PI3K, RB1 and NFE2L2/KEAP1 pathways (Supplementary Figure S7. B.1 and Supplementary Methods, section 7).

Integrated analysis of the tumor suppressor locus for CDKN2A

Integrated multi-platform analyses showed that CDKN2A, a known tumor suppressor gene in lung SqCC³⁸ that encodes the INK4A/p16 and ARF/p14 proteins, is inactivated in 72% of cases of lung SqCC (Figure 4a and Supplementary Data S7.1)—by epigenetic silencing by methylation (21%), inactivating mutation (18%), exon 1β skipping (4%), and homozygous deletion (29%).

Figure 4. Multi-faceted characterization of mechanisms of CDKN2A loss.

a, Schematic view of the exon structure of CDKN2A demonstrating the types of alterations identified in the study. The locations of point mutation are denoted by black and green circles. b, CDKN2A expression (y-axis) versus CDKN2A copy number (x-axis). Samples are represented by circles and colored-coded by specific type of CDKN2A alteration. c, Diagram of the KIAA1797-CDKN2A fusion identified by whole genome sequencing. d, CDKN2A alterations and expression levels (binary) in each sample.

Analysis of mRNA expression across the CDKN2A locus revealed four distinct patterns of p16INK4 and ARF expression: complete absence of both p16INK4 and ARF (33%); expression of high levels of both p16INK4 and ARF (31%); high expression of ARF and absent p16INK4 (31%); or expression of a transcript that represents a splicing of exon E1b from ARF with the shared E3 of ARF and p16INK4, generating a premature stop codon (4%) (Supplementary Figure S4.4). Almost all of the cases completely lacking p16INK4 and ARF expression showed homozygous deletion (Figure 4b and Supplemental Data S7.1). In one case, p16INK4 expression was detected but analysis of WGS data demonstrated an intergenic fusion event that resulted in detectable transcription between exon 1a of p16INK4 and exon 18 of KIAA1797 (Figures 4b, 4c). Interestingly, combined analysis of WGS and RNA-sequencing data identified tumor suppressor gene inactivation by intra- or interchromosomal rearrangement in PTEN, NOTCH1, ARID1A, CTNNA2, VHL and NF1, in 8 additional cases (Supplementary Data S3.1 and S4.1).

In addition to homozygous deletion, there are frequent mutational events in CDKN2A (Figure 4b and Supplementary Data 7.1). These account for 45% of the 56 cases with high p16INK4 and ARF expression. Furthermore, methylation of the exon 1a promoter accounts for many other cases of CDKN2A inactivation (70% of lung SqCCs with ARF expression in the absence of detectable p16INK4). Seven additional tumors in the high-ARF/low-INK4A group had documented mutations of INK4A, primarily nonsense, suggesting nonsense-mediated decay as a mechanism. Of the 28% of tumors without CDKN2A alterations, RB1 mutations were identified in eight cases and CDK6 amplification in one case (Figure 4d).

Therapeutic targets

Molecularly targeted agents are now commonly used in patients with adenocarcinoma of the lung while no effective targeted agents have been developed specifically for lung SqCCs¹³. We analyzed our genomic data for evidence of the two common genomic alterations in adenocarcinomas of the lung: EGFR and KRAS mutation. Only one sample had a KRAS codon 61 mutation, and there were no exon 19 deletions or L858R mutations in EGFR. However, amplifications of EGFR were found in 7% of cases as were two instances of the L861Q EGFR mutation, which confers sensitivity to erlotinib and gefitinib³⁹.

The presence of new potential therapeutic targets in lung SqCC was suggested by the observation that 96% of tumors (171 of 178) contain one or more mutations in tyrosine kinases, serine/threonine kinases, PI3K catalytic and regulatory subunits, nuclear hormone receptors, G protein-coupled receptors, proteases and tyrosine phosphatases (Supplementary Figure S7.D.1a and Supplementary Data S7.2 and S7.3). 50–77% of the mutations were predicted to have a medium or high functional impact as determined by Mutation Assessor score⁴⁰ (Supplementary Figure S7.D.1a) and 39% of tyrosine and 42% of serine/threonine kinase mutations were located in the kinase domain. Many of the alterations were in known oncogenes and tumor suppressors, as defined in the COSMIC database (Supplementary Data S7.3).

We selected potential therapeutic targets based on several features, including (i) availability of an FDA-approved targeted therapeutic agent or one under study in current clinical trials (Supplementary Data S7.2) (ii) confirmation of the altered allele in RNA-sequencing; and (iii) Mutation Assessor score⁴⁰. Using those criteria, we identified 114 cases with somatic alteration of a potentially targetable gene (64%) (Supplementary Figure S7.D.1b and Supplementary Data S7.4). Among these we identified three families of tyrosine kinases, the ERBBs, FGFRs and JAKs, all of which were found to be mutated and/or amplified⁴¹. As discussed for EGFR, the mutational spectra in these potential therapeutic targets differed from those in lung adenocarcinoma (Supplementary Figure S7.D.2).⁴²

To complement a gene-centered search for potential therapeutic targets, we analyzed core cellular pathways known to represent potential therapeutic vulnerabilities: PI3K/AKT, RTK and RAS. Analysis of the 178 lung SqCCs revealed alteration in at least one of those pathways in 69% of samples after restriction of the analysis to mutations confirmed by RNA-sequencing and to amplifications associated with overexpression of the target gene (Figure 5). Mutational events that have been curated in COSMIC are additionally shown in Supplementary Figure S7D.2 as is the distribution of mutations, amplifications and overexpression of the genes depicted in Figure 5C. Specifically, one of three components of the PI3K/Akt pathway (PIK3CA, PTEN or AKT3) was altered in 47% of tumors and receptor-tyrosine kinase (RTK) signaling likely altered by EGFR amplification, BRAF mutation or FGFR amplification or mutation in 26% of tumors. (Figures 5 and Supplementary Figure S7.D.3). Alterations in the PI3K/Akt pathway genes were mutually exclusive with EGFR alterations as identified by MEMo⁴³ (Supplementary Figure S7.D.4.). While the dependence of lung SqCC on many of these individual alterations remains to be defined functionally, this analysis suggests new areas for potential therapeutic development in this cancer.

Figure 5. Alterations in targetable oncogenic pathways in lung SqCCs.

Pathway diagram showing the percentage of samples with alterations in the PI3K/RTK/RAS pathways. Alterations are defined by somatic mutations, homozygous deletions, high-level, focal amplifications, and, in some cases, by significant up- or down-regulation of gene expression (AKT3, FGFR1, PTEN).

Discussion

Lung SqCCs are characterized by a high overall mutation rate of 8.1 mutations/Mb and marked genomic complexity. Similar to high-grade serous ovarian carcinoma¹⁷, almost all lung SqCCs display somatic mutation of TP53. There were also frequent alterations in the CDKN2A/RB1, NFE2L2/KEAP1/CUL3, PI3K/AKT and SOX2/TP63/NOTCH1 pathways, providing evidence of common dysfunction in cell cycle control, response to oxidative stress, apoptotic signaling and/or squamous cell differentiation. Pathway alterations clustered according to expression-subtype in many cases, suggesting that those subtypes have a biological basis.

A role for somatic mutation in the cancer hallmark of “avoiding immune destruction⁴⁴ is suggested by the presence of inactivating mutations in the HLA-A gene. Somatic loss-of-function alterations of HLA-A have not been reported previously in genomic studies of lung cancer. Given the recently reported efficacy of anti-Programmed Death 1 (PD1)⁴⁵ and anti-Cytotoxic T-Lymphocyte Antigen 4 (CTLA4) antibodies in NSCLC⁴⁶, these HLA-A mutations suggest a possible role for genotypic selection of patients for immunotherapies.

Targeted kinase inhibitors have been successfully used for treatment of lung adenocarcinoma but minimally so in lung SqCC. The observations reported here suggest that a detailed understanding of the possible targets in lung SqCCs may identify targeted therapeutic approaches. While EGFR and KRAS mutations, the two most common oncogenic aberrations in lung adenocarcinoma, are extremely rare in lung SqCC, alterations in the FGFR kinase family are common in lung SqCC. Lung SqCCs also share many alterations in common with head and neck squamous cell carcinomas without evidence of human papilloma virus (HPV) infection, including mutation in PIK3CA, PTEN, TP53, CDKN2A, NOTCH1 and HRAS^22,24, suggesting that the biology of these two diseases may be similar.

The current study has identified a potentially targetable gene or pathway alteration in the majority of the lung SqCC samples studied. The data presented here can help organize efforts to analyze lung SqCC clinical tumor specimens for a panel of specific, actionable mutations to select patients for appropriately targeted clinical trials. These data could thereby help to facilitate effective personalized therapy for this deadly disease.

Methods Summary

All specimens were obtained from patients with appropriate consent from the relevant institutional review board. DNA and RNA were collected from samples using the Allprep kit (Qiagen). We used commercial technology for capture and sequencing of exomes from tumor DNA and normal DNA and whole-genome shotgun sequencing. Significantly mutated genes were identified by comparing them with expectation models based on the exact measured rates of specific sequence lesions. GISTIC^23,24 analysis of the circular-binary-segmented Affymetrix SNP 6.0 copy number data was used to identify recurrent amplification and deletion peaks. Consensus clustering approaches were used to analyze mRNA, miRNA and methylation subtypes using previous approaches^{20,21,34,38,41,44}.

TCGA Network Authors and Affiliations

Genome Sequencing Centers

Broad Institute – Peter S Hammerman(¹,⁵), Michael S. Lawrence(¹), Douglas Voet(¹), Rui Jing(¹), Kristian Cibulskis(¹), Andrey Sivachenko(¹), Petar Stojanov(¹), Aaron McKenna(¹), Eric S Lander(¹,²,³), Stacey Gabriel(⁴), Gad Getz(¹,⁴), Carrie Sougnez(⁴), Marcin Imielinski (¹), Elena Helman (¹), Bryan Hernandez (¹), Nam H. Pho (¹), Matthew Meyerson(¹,⁵,⁶)

Genome Characterization Centers

BC Cancer Agency – Andy Chu(⁷), Hye-Jung E. Chun(⁷), Andrew J. Mungall(⁷), Erin Pleasance(⁷), A. Gordon Robertson(⁷), Payal Sipahimalani(⁷), Dominik Stoll(⁷), Miruna Balasundaram(⁷), Inanc Birol(⁷), Yaron S.N. Butterfield(⁷), Eric Chuah(⁷), Robin J.N. Coope(⁷), Richard Corbett(⁷), Noreen Dhalla(⁷), Ranabir Guin(⁷), An He(⁷), Carrie Hirst(⁷), Martin Hirst(⁷), Robert A. Holt(⁷), Darlene Lee(⁷), Haiyan I. Li(⁷), Michael Mayo(⁷), Richard A. Moore(⁷), Karen Mungall(⁷), Ka Ming Nip(⁷), Adam Olshen(⁸), Jacqueline E. Schein(⁷), Jared R. Slobodan(⁷), Angela Tam(⁷), Nina Thiessen(⁷), Richard Varhol(⁷), Thomas Zeng(⁷), Yongjun Zhao(⁷), Steven J.M. Jones(⁷), Marco A. Marra(⁷) Broad Institute – Gordon Saksena(¹), Andrew D. Cherniack(¹), Stephen E. Schumacher(¹,⁵), Barbara Tabak(¹,⁵), Scott L. Carter(¹), Nam H. Pho(¹), Huy Nguyen(¹), Robert C. Onofrio(⁴), Andrew Crenshaw(¹), Kristin Ardlie(⁴), Rameen Beroukhim(¹,⁵), Wendy Winckler(¹,⁴), Peter S Hammerman (¹,⁵) Gad Getz(¹,⁴), Matthew Meyerson(¹,⁵,⁶), Brigham and Women’s Hospital and Harvard Medical School – Alexei Protopopov (⁹,⁹¹), Jianhua Zhang (⁹,⁹¹), Angela Hadjipanayis (¹⁰,¹¹), Semin Lee (¹²), Ruibin Xi (¹²), Lixing Yang(¹²), Xiaojia Ren(⁹,¹⁰,¹¹), Hailei Zhang (¹,⁹), Sachet Shukla (¹,⁹), Peng-Chieh Chen (¹⁰,¹¹), Psalm Haseley (¹¹,¹²), Eunjung Lee (¹¹,¹²), Lynda Chin (¹,⁵,⁹,¹⁴,⁹¹), Peter J. Park (¹¹,¹²,¹³), Raju Kucherlapati (¹⁰,¹¹), Memorial Sloan-Kettering Cancer Center (TCGA Pilot Phase only) Nicholas D. Socci (²⁷), Yupu Liang (²⁷), Nikolaus Schultz (²⁷), Laetitia Borsu (²⁷), Alex E. Lash (²⁷), Agnes Viale (²⁷), Chris Sander (²⁷), Marc Ladanyi (²⁸, ³⁰), University of North Carolina, Chapel Hill: J. Todd Auman(¹⁵,¹⁶), Katherine A. Hoadley(¹⁷,¹⁸,¹⁹), Matthew D. Wilkerson(¹⁹), Yan Shi(¹⁹), Christina Liquori (¹⁹), Shaowu Meng(¹⁹), Ling Li(¹⁹), Yidi J. Turman(¹⁹), Michael D. Topal(¹⁸,¹⁹), Donghui Tan(²⁰), Scot Waring(¹⁹), Elizabeth Buda(¹⁹), Jesse Walsh(¹⁹), Corbin D. Jones(²¹), Piotr A. Mieczkowski(¹⁷), Darshan Singh(²⁴), Junyuan Wu(¹⁹), Anisha Gulabani(¹⁹), Peter Dolina(¹⁹), Tom Bodenheimer(¹⁹), Alan P. Hoyle(¹⁹), Janae V. Simons(¹⁹), Matthew G. Soloway(¹⁹), Lisle E. Mose(¹⁸), Stuart R. Jefferys(¹⁸), Saianand Balu(¹⁹), Brian D. O’Connor(¹⁹), Jan F. Prins(²²), Jinze Liu (²³), Derek Y. Chiang(¹⁷,¹⁹), D. Neil Hayes(¹⁹,²⁴), Charles M. Perou(¹⁷,¹⁸,¹⁹), University of Southern California / Johns Hopkins – Leslie Cope (²⁶), Ludmila Danilova (²⁶), Daniel J. Weisenberger(²⁵), Dennis T. Maglinte(²⁵), Fei Pan(²⁵), David J. Van Den Berg(²⁵), Timothy Triche Jr(²⁵), Stephen B. Baylin(²⁶), & Peter W. Laird(²⁵)

Genome Data Analysis Centers

Broad Institute – Gad Getz(¹), Michael Noble(¹), Doug Voet(¹), Gordon Saksena(¹), Nils Gehlenborg(¹,¹²), Daniel DiCara(¹), Jinhua Zhang(⁹, ⁹¹), Hailei Zhang(¹), Chang-Jiun Wu(⁵,⁹¹), Spring Yingchun Liu(¹), Michael S. Lawrence(¹), Lihua Zou(¹), Andrey Sivachenko(¹), Pei Lin(¹), Petar Stojanov(¹), Rui Jing(¹), Juok Cho(¹), Marc-Danie Nazaire(¹), Jim Robinson(¹), Helga Thorvaldsdottir(¹), Jill Mesirov(¹), Peter J. Park(¹¹,¹²,¹³), Lynda Chin(¹,⁵,⁹,¹⁴,⁹¹) Memorial Sloan-Kettering Cancer Center – Nikolaus Schultz(²⁷), Rileen Sinha(²⁷), Giovanni Ciriello(²⁷), Ethan Cerami(²⁷), Benjamin Gross(²⁷), Anders Jacobsen(²⁷), Jianjiong Gao(²⁷), B. Arman Aksoy(²⁷), Nils Weinhold(²⁷), Ricardo Ramirez(²⁷), Barry S. Taylor(²⁷), Yevgeniy Antipin(²⁷), Boris Reva(²⁷), Ronglai Shen(²⁹), Qianxing Mo(²⁹), Venkatraman Seshan(²⁹), Paul K Paik (³¹), Marc Ladanyi(²⁸, ³⁰), Chris Sander(²⁷) The University of Texas MD Anderson Cancer Center – Rehan Akbani(³²), Nianxiang Zhang(³²), Bradley M. Broom(³²), Tod Casasent(³²), Anna Unruh(³²), Chris Wakefield(³²), R. Craig Cason (³³), Keith A. Baggerly(³²), John N. Weinstein(³²,³⁴), University of California, Santa Cruz / Buck Institute – David Haussler(³⁵,³⁶), Christopher C. Benz(³⁷), Joshua M. Stuart(³⁵), Jingchun Zhu(³⁵), Christopher Szeto(³⁵), Gary K. Scott(³⁷), & Christina Yau(³⁷). Sam Ng(³⁵), Ted Goldstein(³⁵), Peter Waltman(³⁵), Artem Sokolov(³⁵), Kyle Ellrott(³⁵), Eric A. Collisson(³⁸), Daniel Zerbino(³⁵), Christopher Wilks(³⁵), Singer Ma(³⁵), Brian Craft(³⁵), University of North Carolina, Chapel Hill: Matthew D. Wilkerson(¹⁹), J. Todd Auman(¹⁵,¹⁶), Katherine A. Hoadley(¹⁷,¹⁸,¹⁹), Ying Du(¹⁹), Christopher Cabanski (¹⁹), Vonn Walter (¹⁹), Darshan Singh (¹⁹) Junyuan Wu(¹⁹), Anisha Gulabani(¹⁹), Tom Bodenheimer(¹⁹), Alan P. Hoyle(¹⁹), Janae V. Simons(¹⁹), Matthew G. Soloway(¹⁹), Lisle E. Mose(¹⁸), Stuart R. Jefferys(¹⁸), Saianand Balu(¹⁹), J.S. Marron (⁹⁴), Yufeng Liu (⁹⁵), Kai Wang (²³), Jinze Liu (²³), Jan F. Prins(¹⁹), D. Neil Hayes(¹⁹,²⁴), Charles M. Perou(¹⁷,¹⁸,¹⁹), Baylor College of Medicine - Chad J Creighton(⁴⁰), Yiqun Zhang(⁴⁰)

Pathology Committee

William D. Travis (⁴¹), Natasha Rekhtman(⁴¹), Joanne Yi(⁴²), Marie C. Aubry(⁴²), Richard Cheney(⁴³), Sanja Dacic(⁴⁴), Douglas Flieder(⁴⁵), William Funkhouser(⁴⁶), Peter Illei(⁴⁷), Jerome Myers (⁴⁸), Ming Sound Tsao(⁴⁹)

Biospecimen Core Resources

International Genomics Consortium – Robert Penny(⁵⁰), David Mallery (⁵⁰), Troy Shelton(⁵⁰), Martha Hatfield(⁵⁰), Scott Morris(⁵⁰), Peggy Yena(⁵⁰), Candace Shelton(⁵⁰), Mark Sherman(⁵⁰),& Joseph Paulauskis(⁵⁰)

Disease working group

Matthew Meyerson(¹,⁵,⁶), Stephen B. Baylin(²⁶), Ramaswamy Govindan(⁵¹), Rehan Akbani(³²), Ijeoma Azodo(⁶⁰), David Beer(⁵²), Ron Bose(⁵¹), Lauren A. Byers (⁵⁴) David Carbone(⁵³), Li-Wei Chang (⁵¹), Derek Chiang(¹⁷,¹⁹), Andy Chu(⁷), Elizabeth Chun(⁷), Eric Collisson(³⁸), Leslie Cope(²⁶), Chad J Creighton(⁴⁰), Ludmila Danilova(²⁶), Li Ding(⁵¹), Gad Getz(¹,⁴), Peter S Hammerman(¹,⁵), D Neil Hayes(¹⁹,²³), Bryan Hernandez(¹), James Herman(²⁶), John Heymach(⁵⁴), Cristiane Ida(⁴²), Marcin Imielinski(¹,⁶), Bruce Johnson(⁵), Igor Jurisica(⁵⁵), Jacob Kaufman(⁵³), Farhad Kosari(⁶⁰), Raju Kucherlapati(¹⁰,¹¹), David Kwiatkowski(⁵), Marc Ladanyi(²⁸, ³⁰), Michael Lawrence(¹), Christopher A Maher(⁵¹), Andy Mungall(⁷), Sam Ng(³⁵), William Pao(⁵³), Martin Peifer(⁵⁶, ⁹²), Robert Penny(⁵⁰), Gordon Robertson(⁷), Valerie Rusch(⁵⁷) Chris Sander(²⁷), Nikolaus Schultz(²⁷), Ronglai Shen(²⁹), Jill Siegfried(⁵⁸), Rileen Sinha(²⁷), Andrey Sivachenko(¹), Carrie Sougnez(⁴), Dominik Stoll(⁷), Joshua Stuart(³⁵), Roman K Thomas(⁵⁶, ⁹², ⁹³), Sandra Tomaszek(⁶⁰), Ming-Sound Tsao(⁴⁹), William D Travis(⁴¹), Charles Vaske(³⁵), John N Weinstein(³²,³⁴), Daniel Weisenberger(²⁵), David Wheeler(⁵⁹), Dennis A Wigle(⁶⁰), Matthew D Wilkerson(¹⁹), Christopher Wilks(²⁵), Ping Yang(⁶⁰), Jianjua (John) Zhang(⁹)

Data Coordination Center

Mark A. Jensen(⁶¹), Robert Sfeir(⁶¹), Ari B. Kahn(⁶¹), Anna L. Chu(⁶¹), Prachi Kothiyal(⁶¹), Zhining Wang(⁶¹), Eric E. Snyder(⁶¹), Joan Pontius(⁶¹), Todd D. Pihl(⁶¹), Brenda Ayala(⁶¹), Mark Backus(⁶¹), Jessica Walton(⁶¹), Julien Baboud(⁶¹), Dominique L. Berton(⁶¹), Matthew C. Nicholls(⁶¹), Deepak Srinivasan(⁶¹), Rohini Raman(⁶¹), Stanley Girshik(⁶¹), Peter A. Kigonya(⁶¹), Shelley Alonso(⁶¹), Rashmi N. Sanbhadti(⁶¹), Sean P. Barletta(⁶¹), John M. Greene(⁶¹) & David A. Pot(⁶¹)

Tissue Source Sites

Ming Tsao(⁴⁹), Bizhan Bandarchi-Chamkhaleh(⁴⁹), Jeff Boyd(⁴⁵), JoEllen Weaver(⁴⁵), Dennis A. Wigle (⁶⁰), Ijeoma A. Azodo(⁶⁰), Sandra C. Tomaszek(⁶⁰), Marie Christine Aubry(⁶²), Christiane M. Ida(⁶²), Ping Yang(⁶³), Farhad Kosari(⁶⁴), Malcolm V. Brock(⁶⁵), Kristen Rogers(⁶⁵), Marian Rutledge(⁶⁶), Travis Brown(⁶⁵), Beverly Lee(⁶⁶), James Shin(⁶⁷), Dante Trusty(⁶⁷), Rajiv Dhir(⁶⁸), Jill M. Siegfried(⁶⁹), Olga Potapova(⁷⁰), Konstantin V. Fedosenko(⁷¹), Elena Nemirovich-Danchenko(⁷⁰), Valerie Rusch(⁵⁷), Maureen Zakowski(⁷²), Mary V. Iacocca (⁷³), Jennifer Brown(⁷³), Brenda Rabeno(⁷³), Christine Czerwinski(⁷³), Nicholas Petrelli(⁷³), Zhen Fan(⁷⁴), Nicole Todaro(⁷⁴), John Eckman(⁴⁸), Jerome Myers(⁴⁸), W. Kimryn Rathmell(¹⁹), Leigh B. Thorne(⁷⁵), Mei Huang(⁷⁵), Lori Boice(⁷⁵), Ashley Hill(¹⁹), Robert Penny(⁵⁰), David Mallery(⁵⁰), Erin Curley(⁵⁰), Candace Shelton(⁵⁰), Peggy Yena(⁵⁰), Carl Morrison(⁴³), Carmelo Gaudioso(⁴³), John M.S. Bartlett(⁷⁶), Sugy Kodeeswaran(⁷⁶), Brent Zanke(⁷⁶), Harman Sekhon(⁷⁷), Kerstin David(⁷⁸), Hartmut Juhl(⁷⁹), Xuan Van Le(⁸⁰), Bernard Kohl(⁸⁰), Richard Thorp(⁸⁰), Nguyen Viet Tien(⁸¹), Nguyen Van Bang(⁸²), Howard Sussman(⁸³), Bui Duc Phu(⁸²), Richard Hajek(⁸⁴), Nguyen Phi Hung(⁸⁵) & Khurram Z. Khan(⁸⁶), Thomas Muley (⁹⁶)

Project Team

National Cancer Institute – Kenna R. Mills Shaw(⁸⁷), Margi Sheth (⁸⁷), Liming Yang (⁸⁷), Ken Buetow(⁸⁸), Tanja Davidsen(⁸⁸), John A. Demchok(⁸⁷), Greg Eley(⁸⁸), Martin Ferguson(⁸⁹), Laura A. L. Dillon(⁸⁷) & Carl Schaefer(⁸⁸) National Human Genome Research Institute: Mark S. Guyer(⁹⁰), Bradley A. Ozenberger(⁹⁰), Jacqueline D. Palchik(⁹⁰), Jane Peterson(⁹⁰), Heidi J. Sofia(⁹⁰), & Elizabeth Thomson(⁹⁰)

Writing Committee

Peter S Hammerman(¹,⁵), D Neil Hayes(¹⁹,²³), Matthew D Wilkerson(¹⁹), Nikolaus Schultz(²⁷), Ron Bose(⁵¹), Andy Chu(⁷), Eric A Collisson(³⁸), Leslie Cope(²⁶), Chad J Creighton(⁴⁰), Gad Getz(¹,⁴), James Herman(²⁶), Bruce E Johnson(⁵), Raju Kucherlapati(¹⁰,¹¹), Marc Ladanyi(²⁸, ³⁰), Christopher A Maher(⁵¹), Gordon Robertson(⁷), Chris Sander(²⁷), Ronglai Shen(²⁷), Rileen Sinha(²⁷), Andrey Sivachenko(¹), Roman K Thomas(⁵⁶, ⁹², ⁹³), William D Travis(⁴¹), Ming-Sound Tsao(⁴⁹), John N Weinstein(³²,³⁴), Dennis A Wigle(⁶⁰), Stephen B Baylin(²⁶), Ramaswamy Govindan(⁵¹), and Matthew Meyerson(¹,⁵,⁶)