Characterizing genetic intra-tumor heterogeneity across 2,658 human cancer genomes.

Intra-tumor heterogeneity (ITH) is a mechanism of therapeutic resistance and therefore an important clinical challenge. However, the extent, origin, and drivers of ITH across cancer types are poorly understood. To address this, we extensively characterize ITH across whole-genome sequences of 2,658 cancer samples spanning 38 cancer types. Nearly all informative samples (95.1%) contain evidence of distinct subclonal expansions with frequent branching relationships between subclones. We observe positive selection of subclonal driver mutations across most cancer types and identify cancer type-specific subclonal patterns of driver gene mutations, fusions, structural variants, and copy number alterations as well as dynamic changes in mutational processes between subclonal expansions. Our results underline the importance of ITH and its drivers in tumor evolution and provide a pan-cancer resource of comprehensively annotated subclonal events from whole-genome sequencing data.

A pan-cancer compendium of chromosomal instability.

Chromosomal instability (CIN) results in the accumulation of large-scale losses, gains and rearrangements of DNA1. The broad genomic complexity caused by CIN is a hallmark of cancer2; however, there is no systematic framework to measure different types of CIN and their effect on clinical phenotypes pan-cancer. Here we evaluate the extent, diversity and origin of CIN across 7,880 tumours representing 33 cancer types. We present a compendium of 17 copy number signatures that characterize specific types of CIN, with putative aetiologies supported by multiple independent data sources. The signatures predict drug response and identify new drug targets. Our framework refines the understanding of impaired homologous recombination, which is one of the most therapeutically targetable types of CIN. Our results illuminate a fundamental structure underlying genomic complexity in human cancers and provide a resource to guide future CIN research.

Multi-omic machine learning predictor of breast cancer therapy response.

Breast cancers are complex ecosystems of malignant cells and the tumour microenvironment1. The composition of these tumour ecosystems and interactions within them contribute to responses to cytotoxic therapy2. Efforts to build response predictors have not incorporated this knowledge. We collected clinical, digital pathology, genomic and transcriptomic profiles of pre-treatment biopsies of breast tumours from 168 patients treated with chemotherapy with or without HER2 (encoded by ERBB2)-targeted therapy before surgery. Pathology end points (complete response or residual disease) at surgery3 were then correlated with multi-omic features in these diagnostic biopsies. Here we show that response to treatment is modulated by the pre-treated tumour ecosystem, and its multi-omics landscape can be integrated in predictive models using machine learning. The degree of residual disease following therapy is monotonically associated with pre-therapy features, including tumour mutational and copy number landscapes, tumour proliferation, immune infiltration and T cell dysfunction and exclusion. Combining these features into a multi-omic machine learning model predicted a pathological complete response in an external validation cohort (75 patients) with an area under the curve of 0.87. In conclusion, response to therapy is determined by the baseline characteristics of the totality of the tumour ecosystem captured through data integration and machine learning. This approach could be used to develop predictors for other cancers.

The evolutionary history of 2,658 cancers.

Cancer develops through a process of somatic evolution1,2. Sequencing data from a single biopsy represent a snapshot of this process that can reveal the timing of specific genomic aberrations and the changing influence of mutational processes3. Here, by whole-genome sequencing analysis of 2,658 cancers 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)4, we reconstruct the life history and evolution of mutational processes and driver mutation sequences of 38 types of cancer. Early oncogenesis is characterized by mutations in a constrained set of driver genes, and specific copy number gains, such as trisomy 7 in glioblastoma and isochromosome 17q in medulloblastoma. The mutational spectrum changes significantly throughout tumour evolution in 40% of samples. A nearly fourfold diversification of driver genes and increased genomic instability are features of later stages. Copy number alterations often occur in mitotic crises, and lead to simultaneous gains of chromosomal segments. Timing analyses suggest that driver mutations often precede diagnosis by many years, if not decades. Together, these results determine the evolutionary trajectories of cancer, and highlight opportunities for early cancer detection.

Allele-specific multi-sample copy number segmentation in ASCAT.

MOTIVATION: Allele-specific copy number alterations are commonly used to trace the evolution of tumours. A key step of the analysis is to segment genomic data into regions of constant copy number. For precise phylogenetic inference, breakpoints shared between samples need to be aligned to each other. RESULTS: Here, we present asmultipcf, an algorithm for allele-specific segmentation of multiple samples that infers private and shared segment boundaries of phylogenetically related samples. The output of this algorithm can directly be used for allele-specific copy number calling using ASCAT. AVAILABILITY AND IMPLEMENTATION: asmultipcf is available as part of the ASCAT R package (version ≥2.5) from github.com/Crick-CancerGenomics/ascat/.

Ultrasound-guided targeted biopsies of CT-based radiomic tumour habitats: technical development and initial experience in metastatic ovarian cancer.

PURPOSE: To develop a precision tissue sampling technique that uses computed tomography (CT)-based radiomic tumour habitats for ultrasound (US)-guided targeted biopsies that can be integrated in the clinical workflow of patients with high-grade serous ovarian cancer (HGSOC). METHODS: Six patients with suspected HGSOC scheduled for US-guided biopsy before starting neoadjuvant chemotherapy were included in this prospective study from September 2019 to February 2020. The tumour segmentation was performed manually on the pre-biopsy contrast-enhanced CT scan. Spatial radiomic maps were used to identify tumour areas with similar or distinct radiomic patterns, and tumour habitats were identified using the Gaussian mixture modelling. CT images with superimposed habitat maps were co-registered with US images by means of a landmark-based rigid registration method for US-guided targeted biopsies. The dice similarity coefficient (DSC) was used to assess the tumour-specific CT/US fusion accuracy. RESULTS: We successfully co-registered CT-based radiomic tumour habitats with US images in all patients. The median time between CT scan and biopsy was 21 days (range 7-30 days). The median DSC for tumour-specific CT/US fusion accuracy was 0.53 (range 0.79 to 0.37). The CT/US fusion accuracy was high for the larger pelvic tumours (DSC: 0.76-0.79) while it was lower for the smaller omental metastases (DSC: 0.37-0.53). CONCLUSION: We developed a precision tissue sampling technique that uses radiomic habitats to guide in vivo biopsies using CT/US fusion and that can be seamlessly integrated in the clinical routine for patients with HGSOC. KEY POINTS: • We developed a prevision tissue sampling technique that co-registers CT-based radiomics-based tumour habitats with US images. • The CT/US fusion accuracy was high for the larger pelvic tumours (DSC: 0.76-0.79) while it was lower for the smaller omental metastases (DSC: 0.37-0.53).

Estimating the predictability of cancer evolution.

MOTIVATION: How predictable is the evolution of cancer? This fundamental question is of immense relevance for the diagnosis, prognosis and treatment of cancer. Evolutionary biologists have approached the question of predictability based on the underlying fitness landscape. However, empirical fitness landscapes of tumor cells are impossible to determine in vivo. Thus, in order to quantify the predictability of cancer evolution, alternative approaches are required that circumvent the need for fitness landscapes. RESULTS: We developed a computational method based on conjunctive Bayesian networks (CBNs) to quantify the predictability of cancer evolution directly from mutational data, without the need for measuring or estimating fitness. Using simulated data derived from >200 different fitness landscapes, we show that our CBN-based notion of evolutionary predictability strongly correlates with the classical notion of predictability based on fitness landscapes under the strong selection weak mutation assumption. The statistical framework enables robust and scalable quantification of evolutionary predictability. We applied our approach to driver mutation data from the TCGA and the MSK-IMPACT clinical cohorts to systematically compare the predictability of 15 different cancer types. We found that cancer evolution is remarkably predictable as only a small fraction of evolutionary trajectories are feasible during cancer progression. AVAILABILITY AND IMPLEMENTATION: https://github.com/cbg-ethz/predictability\_of\_cancer\_evolution. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Intestinal region-specific Wnt signalling profiles reveal interrelation between cell identity and oncogenic pathway activity in cancer development.

BACKGROUND: Cancer results from the accumulation of mutations leading to the acquisition of cancer promoting characteristics such as increased proliferation and resistance to cell death. In colorectal cancer, an early mutation leading to such features usually occurs in the APC or CTNNB1 genes, thereby activating Wnt signalling. However, substantial phenotypic differences between cancers originating within the same organ, such as molecular subtypes, are not fully reflected by differences in mutations. Indeed, the phenotype seems to result from a complex interplay between the cell-intrinsic features and the acquired mutations, which is difficult to disentangle when established tumours are studied. METHODS: We use a 3D in vitro organoid model to study the early phase of colorectal cancer development. From three different murine intestinal locations we grow organoids. These are transformed to resemble adenomas after Wnt activation through lentiviral transduction with a stable form of ß-Catenin. The gene expression before and after Wnt activation is compared within each intestinal origin and across the three locations using RNA sequencing. To validate and generalize our findings, we use gene expression data from patients. RESULTS: In reaction to Wnt activation we observe downregulation of location specific genes and differentiation markers. A similar effect is seen in patient data, where genes with significant differential expression between the normal left and right colon are downregulated in the cancer samples. Furthermore, the signature of Wnt target genes differs between the three intestinal locations in the organoids. The location specific Wnt signatures are dominated by genes which have been lowly expressed in the tissue of origin, and are the targets of transcription factors that are activated following enhanced Wnt signalling. CONCLUSION: We observed that the region-specific cell identity has a substantial effect on the reaction to Wnt activation in a simple intestinal adenoma model. These findings provide a way forward in resolving the distinct biology between left- and right-sided human colon cancers with potential clinical relevance.

All biology is computational biology.

Here, I argue that computational thinking and techniques are so central to the quest of understanding life that today all biology is computational biology. Computational biology brings order into our understanding of life, it makes biological concepts rigorous and testable, and it provides a reference map that holds together individual insights. The next modern synthesis in biology will be driven by mathematical, statistical, and computational methods being absorbed into mainstream biological training, turning biology into a quantitative science.

Parallel factor ChIP provides essential internal control for quantitative differential ChIP-seq.

A key challenge in quantitative ChIP combined with high-throughput sequencing (ChIP-seq) is the normalization of data in the presence of genome-wide changes in occupancy. Analysis-based normalization methods were developed for transcriptomic data and these are dependent on the underlying assumption that total transcription does not change between conditions. For genome-wide changes in transcription factor (TF) binding, these assumptions do not hold true. The challenges in normalization are confounded by experimental variability during sample preparation, processing and recovery. We present a novel normalization strategy utilizing an internal standard of unchanged peaks for reference. Our method can be readily applied to monitor genome-wide changes by ChIP-seq that are otherwise lost or misrepresented through analytical normalization. We compare our approach to normalization by total read depth and two alternative methods that utilize external experimental controls to study TF binding. We successfully resolve the key challenges in quantitative ChIP-seq analysis and demonstrate its application by monitoring the loss of Estrogen Receptor-alpha (ER) binding upon fulvestrant treatment, ER binding in response to estrodiol, ER mediated change in H4K12 acetylation and profiling ER binding in patient-derived xenographs. This is supported by an adaptable pipeline to normalize and quantify differential TF binding genome-wide and generate metrics for differential binding at individual sites.

Immunophenotypes of pancreatic ductal adenocarcinoma: Meta-analysis of transcriptional subtypes.

Pancreatic ductal adenocarcinoma (PDAC) is the most common malignancy of the pancreas and has one of the highest mortality rates of any cancer type with a 5-year survival rate of <5%. Recent studies of PDAC have provided several transcriptomic classifications based on separate analyses of individual patient cohorts. There is a need to provide a unified transcriptomic PDAC classification driven by therapeutically relevant biologic rationale to inform future treatment strategies. Here, we used an integrative meta-analysis of 353 patients from four different studies to derive a PDAC classification based on immunologic parameters. This consensus clustering approach indicated transcriptomic signatures based on immune infiltrate classified as adaptive, innate and immune-exclusion subtypes. This reveals the existence of microenvironmental interpatient heterogeneity within PDAC and could serve to drive novel therapeutic strategies in PDAC including immune modulation approaches to treating this disease.

Multi-parametric and multi-regional histogram analysis of MRI: modality integration reveals imaging phenotypes of glioblastoma.

OBJECTIVES: Integrating multiple imaging modalities is crucial for MRI data interpretation. The purpose of this study is to determine whether a previously proposed multi-view approach can effectively integrate the histogram features from multi-parametric MRI and whether the selected features can offer incremental prognostic values over clinical variables. METHODS: Eighty newly-diagnosed glioblastoma patients underwent surgery and chemoradiotherapy. Histogram features of diffusion and perfusion imaging were extracted from contrast-enhancing (CE) and non-enhancing (NE) regions independently. An unsupervised patient clustering was performed by the multi-view approach. Kaplan-Meier and Cox proportional hazards regression analyses were performed to evaluate the relevance of patient clustering to survival. The metabolic signatures of patient clusters were compared using multi-voxel spectroscopy analysis. The prognostic values of histogram features were evaluated by survival and ROC curve analyses. RESULTS: Two patient clusters were generated, consisting of 53 and 27 patients respectively. Cluster 2 demonstrated better overall survival (OS) (p = 0.007) and progression-free survival (PFS) (p < 0.001) than Cluster 1. Cluster 2 displayed lower N-acetylaspartate/creatine ratio in NE region (p = 0.040). A higher mean value of anisotropic diffusion in NE region was associated with worse OS (hazard ratio [HR] = 1.40, p = 0.020) and PFS (HR = 1.36, p = 0.031). The seven features selected by this approach showed significantly incremental value in predicting 12-month OS (p = 0.020) and PFS (p = 0.022). CONCLUSIONS: The multi-view clustering method can provide an effective integration of multi-parametric MRI. The histogram features selected may be used as potential prognostic markers. KEY POINTS: • Multi-parametric magnetic resonance imaging captures multi-faceted tumor physiology. • Contrast-enhancing and non-enhancing tumor regions represent different tumor components with distinct clinical relevance. • Multi-view data analysis offers a method which can effectively select and integrate multi-parametric and multi-regional imaging features.

Inferring modulators of genetic interactions with epistatic nested effects models.

Maps of genetic interactions can dissect functional redundancies in cellular networks. Gene expression profiles as high-dimensional molecular readouts of combinatorial perturbations provide a detailed view of genetic interactions, but can be hard to interpret if different gene sets respond in different ways (called mixed epistasis). Here we test the hypothesis that mixed epistasis between a gene pair can be explained by the action of a third gene that modulates the interaction. We have extended the framework of Nested Effects Models (NEMs), a type of graphical model specifically tailored to analyze high-dimensional gene perturbation data, to incorporate logical functions that describe interactions between regulators on downstream genes and proteins. We benchmark our approach in the controlled setting of a simulation study and show high accuracy in inferring the correct model. In an application to data from deletion mutants of kinases and phosphatases in S. cerevisiae we show that epistatic NEMs can point to modulators of genetic interactions. Our approach is implemented in the R-package 'epiNEM' available from https://github.com/cbg-ethz/epiNEM and https://bioconductor.org/packages/epiNEM/.

The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups.

The elucidation of breast cancer subgroups and their molecular drivers requires integrated views of the genome and transcriptome from representative numbers of patients. We present an integrated analysis of copy number and gene expression in a discovery and validation set of 997 and 995 primary breast tumours, respectively, with long-term clinical follow-up. Inherited variants (copy number variants and single nucleotide polymorphisms) and acquired somatic copy number aberrations (CNAs) were associated with expression in ~40% of genes, with the landscape dominated by cis- and trans-acting CNAs. By delineating expression outlier genes driven in cis by CNAs, we identified putative cancer genes, including deletions in PPP2R2A, MTAP and MAP2K4. Unsupervised analysis of paired DNA­RNA profiles revealed novel subgroups with distinct clinical outcomes, which reproduced in the validation cohort. These include a high-risk, oestrogen-receptor-positive 11q13/14 cis-acting subgroup and a favourable prognosis subgroup devoid of CNAs. Trans-acting aberration hotspots were found to modulate subgroup-specific gene networks, including a TCR deletion-mediated adaptive immune response in the 'CNA-devoid' subgroup and a basal-specific chromosome 5 deletion-associated mitotic network. Our results provide a novel molecular stratification of the breast cancer population, derived from the impact of somatic CNAs on the transcriptome.

Master Regulators of Oncogenic KRAS Response in Pancreatic Cancer: An Integrative Network Biology Analysis.

BACKGROUND: KRAS is the most frequently mutated gene in pancreatic ductal adenocarcinoma (PDAC), but the mechanisms underlying the transcriptional response to oncogenic KRAS are still not fully understood. We aimed to uncover transcription factors that regulate the transcriptional response of oncogenic KRAS in pancreatic cancer and to understand their clinical relevance. METHODS AND FINDINGS: We applied a well-established network biology approach (master regulator analysis) to combine a transcriptional signature for oncogenic KRAS derived from a murine isogenic cell line with a coexpression network derived by integrating 560 human pancreatic cancer cases across seven studies. The datasets included the ICGC cohort (n = 242), the TCGA cohort (n = 178), and five smaller studies (n = 17, 25, 26, 36, and 36). 55 transcription factors were coexpressed with a significant number of genes in the transcriptional signature (gene set enrichment analysis [GSEA] p < 0.01). Community detection in the coexpression network identified 27 of the 55 transcription factors contributing to three major biological processes: Notch pathway, down-regulated Hedgehog/Wnt pathway, and cell cycle. The activities of these processes define three distinct subtypes of PDAC, which demonstrate differences in survival and mutational load as well as stromal and immune cell composition. The Hedgehog subgroup showed worst survival (hazard ratio 1.73, 95% CI 1.1 to 2.72, coxPH test p = 0.018) and the Notch subgroup the best (hazard ratio 0.62, 95% CI 0.42 to 0.93, coxPH test p = 0.019). The cell cycle subtype showed highest mutational burden (ANOVA p < 0.01) and the smallest amount of stromal admixture (ANOVA p < 2.2e-16). This study is limited by the information provided in published datasets, not all of which provide mutational profiles, survival data, or the specifics of treatment history. CONCLUSIONS: Our results characterize the regulatory mechanisms underlying the transcriptional response to oncogenic KRAS and provide a framework to develop strategies for specific subtypes of this disease using current therapeutics and by identifying targets for new groups.

Patterns of Immune Infiltration in Breast Cancer and Their Clinical Implications: A Gene-Expression-Based Retrospective Study.

BACKGROUND: Immune infiltration of breast tumours is associated with clinical outcome. However, past work has not accounted for the diversity of functionally distinct cell types that make up the immune response. The aim of this study was to determine whether differences in the cellular composition of the immune infiltrate in breast tumours influence survival and treatment response, and whether these effects differ by molecular subtype. METHODS AND FINDINGS: We applied an established computational approach (CIBERSORT) to bulk gene expression profiles of almost 11,000 tumours to infer the proportions of 22 subsets of immune cells. We investigated associations between each cell type and survival and response to chemotherapy, modelling cellular proportions as quartiles. We found that tumours with little or no immune infiltration were associated with different survival patterns according to oestrogen receptor (ER) status. In ER-negative disease, tumours lacking immune infiltration were associated with the poorest prognosis, whereas in ER-positive disease, they were associated with intermediate prognosis. Of the cell subsets investigated, T regulatory cells and M0 and M2 macrophages emerged as the most strongly associated with poor outcome, regardless of ER status. Among ER-negative tumours, CD8+ T cells (hazard ratio [HR] = 0.89, 95% CI 0.80-0.98; p = 0.02) and activated memory T cells (HR 0.88, 95% CI 0.80-0.97; p = 0.01) were associated with favourable outcome. T follicular helper cells (odds ratio [OR] = 1.34, 95% CI 1.14-1.57; p < 0.001) and memory B cells (OR = 1.18, 95% CI 1.0-1.39; p = 0.04) were associated with pathological complete response to neoadjuvant chemotherapy in ER-negative disease, suggesting a role for humoral immunity in mediating response to cytotoxic therapy. Unsupervised clustering analysis using immune cell proportions revealed eight subgroups of tumours, largely defined by the balance between M0, M1, and M2 macrophages, with distinct survival patterns by ER status and associations with patient age at diagnosis. The main limitations of this study are the use of diverse platforms for measuring gene expression, including some not previously used with CIBERSORT, and the combined analysis of different forms of follow-up across studies. CONCLUSIONS: Large differences in the cellular composition of the immune infiltrate in breast tumours appear to exist, and these differences are likely to be important determinants of both prognosis and response to treatment. In particular, macrophages emerge as a possible target for novel therapies. Detailed analysis of the cellular immune response in tumours has the potential to enhance clinical prediction and to identify candidates for immunotherapy.

Cancer evolution: mathematical models and computational inference.

Cancer is a somatic evolutionary process characterized by the accumulation of mutations, which contribute to tumor growth, clinical progression, immune escape, and drug resistance development. Evolutionary theory can be used to analyze the dynamics of tumor cell populations and to make inference about the evolutionary history of a tumor from molecular data. We review recent approaches to modeling the evolution of cancer, including population dynamics models of tumor initiation and progression, phylogenetic methods to model the evolutionary relationship between tumor subclones, and probabilistic graphical models to describe dependencies among mutations. Evolutionary modeling helps to understand how tumors arise and will also play an increasingly important prognostic role in predicting disease progression and the outcome of medical interventions, such as targeted therapy.

Spatial and temporal heterogeneity in high-grade serous ovarian cancer: a phylogenetic analysis.

BACKGROUND: The major clinical challenge in the treatment of high-grade serous ovarian cancer (HGSOC) is the development of progressive resistance to platinum-based chemotherapy. The objective of this study was to determine whether intra-tumour genetic heterogeneity resulting from clonal evolution and the emergence of subclonal tumour populations in HGSOC was associated with the development of resistant disease. METHODS AND FINDINGS: Evolutionary inference and phylogenetic quantification of heterogeneity was performed using the MEDICC algorithm on high-resolution whole genome copy number profiles and selected genome-wide sequencing of 135 spatially and temporally separated samples from 14 patients with HGSOC who received platinum-based chemotherapy. Samples were obtained from the clinical CTCR-OV03/04 studies, and patients were enrolled between 20 July 2007 and 22 October 2009. Median follow-up of the cohort was 31 mo (interquartile range 22-46 mo), censored after 26 October 2013. Outcome measures were overall survival (OS) and progression-free survival (PFS). There were marked differences in the degree of clonal expansion (CE) between patients (median 0.74, interquartile range 0.66-1.15), and dichotimization by median CE showed worse survival in CE-high cases (PFS 12.7 versus 10.1 mo, p = 0.009; OS 42.6 versus 23.5 mo, p = 0.003). Bootstrap analysis with resampling showed that the 95% confidence intervals for the hazard ratios for PFS and OS in the CE-high group were greater than 1.0. These data support a relationship between heterogeneity and survival but do not precisely determine its effect size. Relapsed tissue was available for two patients in the CE-high group, and phylogenetic analysis showed that the prevalent clonal population at clinical recurrence arose from early divergence events. A subclonal population marked by a NF1 deletion showed a progressive increase in tumour allele fraction during chemotherapy. CONCLUSIONS: This study demonstrates that quantitative measures of intra-tumour heterogeneity may have predictive value for survival after chemotherapy treatment in HGSOC. Subclonal tumour populations are present in pre-treatment biopsies in HGSOC and can undergo expansion during chemotherapy, causing clinical relapse.