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.

Computational Approaches for the Investigation of Intra-tumor Heterogeneity and Clonal Evolution from Bulk Sequencing Data in Precision Oncology Applications.

While the clonal model of cancer evolution was first proposed over 40 years ago, only recently next-generation sequencing has allowed a more precise and quantitative assessment of tumor clonal and subclonal landscape. Consequently, a plethora of computational approaches and tools have been developed to analyze this data with the goal of inferring the clonal landscape of a tumor and characterize its temporal or spatial evolution. This chapter introduces intra-tumor heterogeneity (ITH) in the context of precision oncology applications and provides an overview of the basic concepts, algorithms, and tools for the dissection, analysis, and visualization of ITH from bulk DNA sequencing.

Computational Analysis of DNA and RNA Sequencing Data Obtained from Liquid Biopsies.

Next-generation sequencing of DNA and RNA obtained from liquid biopsies of cancer patients may reveal important insights into disease progression and metastasis formation, and it holds the promise to enable new methods for noninvasive screening and clinical decision support. However, implementing liquid biopsy sequencing protocols is challenged by capturing circulating tumor cells or cell-free tumor DNA from blood samples, by amplifying genomic DNA and RNA in a reliable and unbiased manner, and by extracting biologically meaningful signals from the noisy sequencing data. In this chapter, we discuss computational methods for the analysis of DNA and RNA sequencing data obtained from liquid biopsies, addressing these challenges.

BAMSE: Bayesian model selection for tumor phylogeny inference among multiple samples.

BACKGROUND: Intra-tumor heterogeneity is known to contribute to cancer complexity and drug resistance. Understanding the number of distinct subclones and the evolutionary relationships between them is scientifically and clinically very important and still a challenging problem. RESULTS: In this paper, we present BAMSE (BAyesian Model Selection for tumor Evolution), a new probabilistic method for inferring subclonal history and lineage tree reconstruction of heterogeneous tumor samples. BAMSE uses somatic mutation read counts as input and can leverage multiple tumor samples accurately and efficiently. In the first step, possible clusterings of mutations into subclones are scored and a user defined number are selected for further analysis. In the next step, for each of these candidates, a list of trees describing the evolutionary relationships between the subclones is generated. These trees are sorted by their posterior probability. The posterior probability is calculated using a Bayesian model that integrates prior belief about the number of subclones, the composition of the tumor and the process of subclonal evolution. BAMSE also takes the sequencing error into account. We benchmarked BAMSE against state of the art software using simulated datasets. CONCLUSIONS: In this work we developed a flexible and fast software to reconstruct the history of a tumor's subclonal evolution using somatic mutation read counts across multiple samples. BAMSE software is implemented in Python and is available open source under GNU GLPv3 at https://github.com/HoseinT/BAMSE .

Single-cell sequencing maps gene expression to mutational phylogenies in PDGF- and EGF-driven gliomas.

Glioblastoma multiforme (GBM) is the most common and aggressive type of primary brain tumor. Epidermal growth factor (EGF) and platelet-derived growth factor (PDGF) receptors are frequently amplified and/or possess gain-of-function mutations in GBM However, clinical trials of tyrosine-kinase inhibitors have shown disappointing efficacy, in part due to intra-tumor heterogeneity. To assess the effect of clonal heterogeneity on gene expression, we derived an approach to map single-cell expression profiles to sequentially acquired mutations identified from exome sequencing. Using 288 single cells, we constructed high-resolution phylogenies of EGF-driven and PDGF-driven GBMs, modeling transcriptional kinetics during tumor evolution. Descending the phylogenetic tree of a PDGF-driven tumor corresponded to a progressive induction of an oligodendrocyte progenitor-like cell type, expressing pro-angiogenic factors. In contrast, phylogenetic analysis of an EGFR-amplified tumor showed an up-regulation of pro-invasive genes. An in-frame deletion in a specific dimerization domain of PDGF receptor correlates with an up-regulation of growth pathways in a proneural GBM and enhances proliferation when ectopically expressed in glioma cell lines. In-frame deletions in this domain are frequent in public GBM data.

ConDoR: tumor phylogeny inference with a copy-number constrained mutation loss model.

A tumor contains a diverse collection of somatic mutations that reflect its past evolutionary history and that range in scale from single nucleotide variants (SNVs) to large-scale copy-number aberrations (CNAs). However, no current single-cell DNA sequencing (scDNA-seq) technology produces accurate measurements of both SNVs and CNAs, complicating the inference of tumor phylogenies. We introduce a new evolutionary model, the constrained k-Dollo model, that uses SNVs as phylogenetic markers but constrains losses of SNVs according to clusters of cells. We derive an algorithm, ConDoR, that infers phylogenies from targeted scDNA-seq data using this model. We demonstrate the advantages of ConDoR on simulated and real scDNA-seq data.

Multiregional Sequencing of IDH-WT Glioblastoma Reveals High Genetic Heterogeneity and a Dynamic Evolutionary History.

Glioblastoma is one of the most common and lethal primary neoplasms of the brain. Patient survival has not improved significantly over the past three decades and the patient median survival is just over one year. Tumor heterogeneity is thought to be a major determinant of therapeutic failure and a major reason for poor overall survival. This work aims to comprehensively define intra- and inter-tumor heterogeneity by mapping the genomic and mutational landscape of multiple areas of three primary IDH wild-type (IDH-WT) glioblastomas. Using whole exome sequencing, we explored how copy number variation, chromosomal and single loci amplifications/deletions, and mutational burden are spatially distributed across nine different tumor regions. The results show that all tumors exhibit a different signature despite the same diagnosis. Above all, a high inter-tumor heterogeneity emerges. The evolutionary dynamics of all identified mutations within each region underline the questionable value of a single biopsy and thus the therapeutic approach for the patient. Multiregional collection and subsequent sequencing are essential to try to address the clinical challenge of precision medicine. Especially in glioblastoma, this approach could provide powerful support to pathologists and oncologists in evaluating the diagnosis and defining the best treatment option.

Response and recurrence correlates in individuals treated with neoadjuvant anti-PD-1 therapy for resectable oral cavity squamous cell carcinoma.

Neoadjuvant PD-1 blockade may be efficacious in some individuals with high-risk, resectable oral cavity head and neck cancer. To explore correlates of response patterns to neoadjuvant nivolumab treatment and post-surgical recurrences, we analyzed longitudinal tumor and blood samples in a cohort of 12 individuals displaying 33% responsiveness. Pretreatment tumor-based detection of FLT4 mutations and PTEN signature enrichment favors response, and high tumor mutational burden improves recurrence-free survival. In contrast, preexisting and/or acquired mutations (in CDKN2A, YAP1, or JAK2) correlate with innate resistance and/or tumor recurrence. Immunologically, tumor response after therapy entails T cell receptor repertoire diversification in peripheral blood and intratumoral expansion of preexisting T cell clones. A high ratio of regulatory T to T helper 17 cells in pretreatment blood predicts low T cell receptor repertoire diversity in pretreatment blood, a low cytolytic T cell signature in pretreatment tumors, and innate resistance. Our study provides a molecular framework to advance neoadjuvant anti-PD-1 therapy for individuals with resectable head and neck cancer.

Parameter, noise, and tree topology effects in tumor phylogeny inference.

BACKGROUND: Accurate inference of the evolutionary history of a tumor has important implications for understanding and potentially treating the disease. While a number of methods have been proposed to reconstruct the evolutionary history of a tumor from DNA sequencing data, it is not clear how aspects of the sequencing data and tumor itself affect these reconstructions. METHODS: We investigate when and how well these histories can be reconstructed from multi-sample bulk sequencing data when considering only single nucleotide variants (SNVs). Specifically, we examine the space of all possible tumor phylogenies under the infinite sites assumption (ISA) using several approaches for enumerating phylogenies consistent with the sequencing data. RESULTS: On noisy simulated data, we find that the ISA is often violated and that low coverage and high noise make it more difficult to identify phylogenies. Additionally, we find that evolutionary trees with branching topologies are easier to reconstruct accurately. We also apply our reconstruction methods to both chronic lymphocytic leukemia and clear cell renal cell carcinoma datasets and confirm that ISA violations are common in practice, especially in lower-coverage sequencing data. Nonetheless, we show that an ISA-based approach can be relaxed to produce high-quality phylogenies. CONCLUSIONS: Consideration of practical aspects of sequencing data such as coverage or the model of tumor evolution (branching, linear, etc.) is essential to effectively using the output of tumor phylogeny inference methods. Additionally, these factors should be considered in the development of new inference methods.

CALDER: Inferring Phylogenetic Trees from Longitudinal Tumor Samples.

Longitudinal DNA sequencing of cancer patients yields insight into how tumors evolve over time or in response to treatment. However, sequencing data from bulk tumor samples often have considerable ambiguity in clonal composition, complicating the inference of ancestral relationships between clones. We introduce Cancer Analysis of Longitudinal Data through Evolutionary Reconstruction (CALDER), an algorithm to infer phylogenetic trees from longitudinal bulk DNA sequencing data. CALDER explicitly models a longitudinally observed phylogeny incorporating constraints that longitudinal sampling imposes on phylogeny reconstruction. We show on simulated bulk tumor data that longitudinal constraints substantially reduce ambiguity in phylogeny reconstruction and that CALDER outperforms existing methods that do not leverage this longitudinal information. On real data from two chronic lymphocytic leukemia patients, we find that CALDER reconstructs more plausible and parsimonious phylogenies than existing methods, with CALDER phylogenies containing fewer tumor clones per sample. CALDER's use of longitudinal information will be advantageous in further studies of tumor heterogeneity and evolution.

Phylogenetic Copy-Number Factorization of Multiple Tumor Samples.

Cancer is an evolutionary process driven by somatic mutations. This process can be represented as a phylogenetic tree. Constructing such a phylogenetic tree from genome sequencing data is a challenging task due to the many types of mutations in cancer and the fact that nearly all cancer sequencing is of a bulk tumor, measuring a superposition of somatic mutations present in different cells. We study the problem of reconstructing tumor phylogenies from copy-number aberrations (CNAs) measured in bulk-sequencing data. We introduce the Copy-Number Tree Mixture Deconvolution (CNTMD) problem, which aims to find the phylogenetic tree with the fewest number of CNAs that explain the copy-number data from multiple samples of a tumor. We design an algorithm for solving the CNTMD problem and apply the algorithm to both simulated and real data. On simulated data, we find that our algorithm outperforms existing approaches that either perform deconvolution/factorization of mixed tumor samples or build phylogenetic trees assuming homogeneous tumor samples. On real data, we analyze multiple samples from a prostate cancer patient, identifying clones within these samples and a phylogenetic tree that relates these clones and their differing proportions across samples. This phylogenetic tree provides a higher resolution view of copy-number evolution of this cancer than published analyses.

Analysis of gene copy number changes in tumor phylogenetics.

BACKGOUND: Evolution of cancer cells is characterized by large scale and rapid changes in the chromosomal  landscape. The fluorescence in situ hybridization (FISH) technique provides a way to measure the copy numbers of preselected genes in a group of cells and has been found to be a reliable source of data to model the evolution of tumor cells. Chowdhury et al. (Bioinformatics 29(13):189-98, 23; PLoS Comput Biol 10(7):1003740, 24) recently develop a computational model for tumor progression driven by gains and losses in cell count patterns obtained by FISH probes. Their model aims to find the rectilinear Steiner minimum tree (RSMT) (Chowdhury et al. in Bioinformatics 29(13):189-98, 23) and the duplication Steiner minimum tree (DSMT) (Chowdhury et al. in PLoS Comput Biol 10(7):1003740, 24) that describe the progression of FISH cell count patterns over its branches in a parsimonious manner. Both the RSMT and DSMT problems are NP-hard and heuristics are required to solve the problems efficiently. METHODS: In this paper we propose two approaches to solve the RSMT problem, one inspired by iterative methods to address the "small phylogeny" problem (Sankoff et al. in J Mol Evol 7(2):133-49, 27; Blanchette et al. in Genome Inform 8:25-34, 28), and the other based on maximum parsimony phylogeny inference. We further show how to extend these heuristics to obtain solutions to the DSMT problem, that models large scale duplication events. RESULTS: Experimental results from both simulated and real tumor data show that our methods outperform previous heuristics (Chowdhury et al. in Bioinformatics 29(13):189-98, 23; Chowdhury et al. in PLoS Comput Biol 10(7):1003740, 24) in obtaining solutions to both RSMT and DSMT problems. CONCLUSION: The methods introduced here are able to provide more parsimony phylogenies compared to earlier ones which are consider better choices.