Chromatin is folded into successive layers to organize linear DNA. Genes within the same topologically associating domains (TADs) demonstrate similar expression and histone-modification profiles, and boundaries separating different domains have important roles in reinforcing the stability of these features. Indeed, domain disruptions in human cancers can lead to misregulation of gene expression. However, the frequency of domain disruptions in human cancers remains unclear. Here, as part of the Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium of the International Cancer Genome Consortium (ICGC) and The Cancer Genome Atlas (TCGA), which aggregated whole-genome sequencing data from 2,658 cancers across 38 tumor types, we analyzed 288,457 somatic structural variations (SVs) to understand the distributions and effects of SVs across TADs. Notably, SVs can lead to the fusion of discrete TADs, and complex rearrangements markedly change chromatin folding maps in the cancer genomes. Notably, only 14% of the boundary deletions resulted in a change in expression in nearby genes of more than twofold.

A key mutational process in cancer is structural variation, in which rearrangements delete, amplify or reorder genomic segments that range in size from kilobases to whole chromosomes^1-7. Here we develop methods to group, classify and describe somatic structural variants, using data from the Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium of the International Cancer Genome Consortium (ICGC) and The Cancer Genome Atlas (TCGA), which aggregated whole-genome sequencing data from 2,658 cancers across 38 tumour types⁸. Sixteen signatures of structural variation emerged. Deletions have a multimodal size distribution, assort unevenly across tumour types and patients, are enriched in late-replicating regions and correlate with inversions. Tandem duplications also have a multimodal size distribution, but are enriched in early-replicating regions-as are unbalanced translocations. Replication-based mechanisms of rearrangement generate varied chromosomal structures with low-level copy-number gains and frequent inverted rearrangements. One prominent structure consists of 2-7 templates copied from distinct regions of the genome strung together within one locus. Such cycles of templated insertions correlate with tandem duplications, and-in liver cancer-frequently activate the telomerase gene TERT. A wide variety of rearrangement processes are active in cancer, which generate complex configurations of the genome upon which selection can act.

Mitochondrial genomes are separated from the nuclear genome for most of the cell cycle by the nuclear double membrane, intervening cytoplasm, and the mitochondrial double membrane. Despite these physical barriers, we show that somatically acquired mitochondrial-nuclear genome fusion sequences are present in cancer cells. Most occur in conjunction with intranuclear genomic rearrangements, and the features of the fusion fragments indicate that nonhomologous end joining and/or replication-dependent DNA double-strand break repair are the dominant mechanisms involved. Remarkably, mitochondrial-nuclear genome fusions occur at a similar rate per base pair of DNA as interchromosomal nuclear rearrangements, indicating the presence of a high frequency of contact between mitochondrial and nuclear DNA in some somatic cells. Transmission of mitochondrial DNA to the nuclear genome occurs in neoplastically transformed cells, but we do not exclude the possibility that some mitochondrial-nuclear DNA fusions observed in cancer occurred years earlier in normal somatic cells.

Pancreatic cancer, a highly aggressive tumour type with uniformly poor prognosis, exemplifies the classically held view of stepwise cancer development. The current model of tumorigenesis, based on analyses of precursor lesions, termed pancreatic intraepithelial neoplasm (PanINs) lesions, makes two predictions: first, that pancreatic cancer develops through a particular sequence of genetic alterations (KRAS, followed by CDKN2A, then TP53 and SMAD4); and second, that the evolutionary trajectory of pancreatic cancer progression is gradual because each alteration is acquired independently. A shortcoming of this model is that clonally expanded precursor lesions do not always belong to the tumour lineage, indicating that the evolutionary trajectory of the tumour lineage and precursor lesions can be divergent. This prevailing model of tumorigenesis has contributed to the clinical notion that pancreatic cancer evolves slowly and presents at a late stage. However, the propensity for this disease to rapidly metastasize and the inability to improve patient outcomes, despite efforts aimed at early detection, suggest that pancreatic cancer progression is not gradual. Here, using newly developed informatics tools, we tracked changes in DNA copy number and their associated rearrangements in tumour-enriched genomes and found that pancreatic cancer tumorigenesis is neither gradual nor follows the accepted mutation order. Two-thirds of tumours harbour complex rearrangement patterns associated with mitotic errors, consistent with punctuated equilibrium as the principal evolutionary trajectory. In a subset of cases, the consequence of such errors is the simultaneous, rather than sequential, knockout of canonical preneoplastic genetic drivers that are likely to set-off invasive cancer growth. These findings challenge the current progression model of pancreatic cancer and provide insights into the mutational processes that give rise to these aggressive tumours.

About half of all cancers have somatic integrations of retrotransposons. Here, to characterize their role in oncogenesis, we analyzed the patterns and mechanisms of somatic retrotransposition in 2,954 cancer genomes from 38 histological cancer subtypes within the framework of the Pan-Cancer Analysis of Whole Genomes (PCAWG) project. We identified 19,166 somatically acquired retrotransposition events, which affected 35% of samples and spanned a range of event types. Long interspersed nuclear element (LINE-1; L1 hereafter) insertions emerged as the first most frequent type of somatic structural variation in esophageal adenocarcinoma, and the second most frequent in head-and-neck and colorectal cancers. Aberrant L1 integrations can delete megabase-scale regions of a chromosome, which sometimes leads to the removal of tumor-suppressor genes, and can induce complex translocations and large-scale duplications. Somatic retrotranspositions can also initiate breakage-fusion-bridge cycles, leading to high-level amplification of oncogenes. These observations illuminate a relevant role of L1 retrotransposition in remodeling the cancer genome, with potential implications for the development of human tumors.

The discovery of drivers of cancer has traditionally focused on protein-coding genes^1-4. Here we present analyses of driver point mutations and structural variants in non-coding regions across 2,658 genomes from the Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium⁵ of the International Cancer Genome Consortium (ICGC) and The Cancer Genome Atlas (TCGA). For point mutations, we developed a statistically rigorous strategy for combining significance levels from multiple methods of driver discovery that overcomes the limitations of individual methods. For structural variants, we present two methods of driver discovery, and identify regions that are significantly affected by recurrent breakpoints and recurrent somatic juxtapositions. Our analyses confirm previously reported drivers^6,7, raise doubts about others and identify novel candidates, including point mutations in the 5' region of TP53, in the 3' untranslated regions of NFKBIZ and TOB1, focal deletions in BRD4 and rearrangements in the loci of AKR1C genes. We show that although point mutations and structural variants that drive cancer are less frequent in non-coding genes and regulatory sequences than in protein-coding genes, additional examples of these drivers will be found as more cancer genomes become available.

Cancer is driven by genetic change, and the advent of massively parallel sequencing has enabled systematic documentation of this variation at the whole-genome scale^1-3. Here we report the integrative analysis of 2,658 whole-cancer genomes and their matching normal tissues across 38 tumour types from the Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium of the International Cancer Genome Consortium (ICGC) and The Cancer Genome Atlas (TCGA). We describe the generation of the PCAWG resource, facilitated by international data sharing using compute clouds. On average, cancer genomes contained 4-5 driver mutations when combining coding and non-coding genomic elements; however, in around 5% of cases no drivers were identified, suggesting that cancer driver discovery is not yet complete. Chromothripsis, in which many clustered structural variants arise in a single catastrophic event, is frequently an early event in tumour evolution; in acral melanoma, for example, these events precede most somatic point mutations and affect several cancer-associated genes simultaneously. Cancers with abnormal telomere maintenance often originate from tissues with low replicative activity and show several mechanisms of preventing telomere attrition to critical levels. Common and rare germline variants affect patterns of somatic mutation, including point mutations, structural variants and somatic retrotransposition. A collection of papers from the PCAWG Consortium describes non-coding mutations that drive cancer beyond those in the TERT promoter⁴; identifies new signatures of mutational processes that cause base substitutions, small insertions and deletions and structural variation^5,6; analyses timings and patterns of tumour evolution⁷; describes the diverse transcriptional consequences of somatic mutation on splicing, expression levels, fusion genes and promoter activity^8,9; and evaluates a range of more-specialized features of cancer genomes^8,10-18.