Cancer progression is an evolutionary process driven by the selection of cells adapted to gain growth advantage. We present the first formal study on the adaptation of gene expression in subclonal evolution. We model evolutionary changes in gene expression as stochastic Ornstein–Uhlenbeck processes, jointly leveraging the evolutionary history of subclones and single-cell expression data. Applying our model to sublines derived from single cells of a mouse melanoma revealed that sublines with distinct phenotypes are underlined by different patterns of gene expression adaptation, indicating non-genetic mechanisms of cancer evolution. Interestingly, sublines previously observed to be resistant to anti-CTLA-4 treatment showed adaptive expression of genes related to invasion and non-canonical Wnt signaling, whereas sublines that responded to treatment showed adaptive expression of genes related to proliferation and canonical Wnt signaling. Our results suggest that clonal phenotypes emerge as the result of specific adaptivity patterns of gene expression.

Most current studies rely on short-read sequencing to detect somatic structural variation (SV) in cancer genomes. Long-read sequencing offers the advantage of better mappability and long-range phasing, which results in substantial improvements in germline SV detection. However, current long-read SV detection methods do not generalize well to the analysis of somatic SVs in tumor genomes with complex rearrangements, heterogeneity, and aneuploidy. Here, we present Severus: a method for the accurate detection of different types of somatic SVs using a phased breakpoint graph approach. To benchmark various short- and long-read SV detection methods, we sequenced five tumor/normal cell line pairs with Illumina, Nanopore, and PacBio sequencing platforms; on this benchmark Severus showed the highest F1 scores (harmonic mean of the precision and recall) as compared to long-read and short-read methods. We then applied Severus to three clinical cases of pediatric cancer, demonstrating concordance with known genetic findings as well as revealing clinically relevant cryptic rearrangements missed by standard genomic panels.

Melanoma is characterized by significant intratumoral heterogeneity and complex evolutionary dynamics. This diversity in genomic alterations leads to the emergence of various subclonal populations within a single tumor. In our research, we established a new model system composed of 24 single-cell-derived clonal sublines (C1-C24), originating from the M4 melanoma model. This model was developed using a genetically engineered hepatocyte growth factor (HGF)-transgenic mouse. We employed Trisicell (Triple-toolkit for single-cell intratumor heterogeneity inference), a cutting-edge computational tool for scalable analysis of intratumor heterogeneity and evaluation based on single-cell RNA mutations. This enabled us to construct a phylogeny tree, revealing melanoma’s intricate branching evolutionary patterns. These patterns show ancestral clones evolving into genetically distinct subclones, which demonstrate varied phenotypic traits such as drug sensitivity or resistance, cellular plasticity, and immunogenicity. In our study, we conducted long-read sequencing on these clonal sublines in the phylogeny and identified structural variants (SVs) using Severus, a tool optimized for phasing in long-read sequencing. The types of SVs we discovered include deletions, insertions, amplifications, translocations, and inversions. We explored their roles in subclonal evolution, particularly focusing on how they disrupt genes and accumulate during melanoma progression. Our initial data from eleven sublines indicated a higher prevalence of ancestral SVs, shared by all sublines, compared to subline-specific SVs, representing later events. Notably, the individual sublines showed a higher rate of gene disruption by SVs, hinting at potential functional selection. Our analysis further revealed that SVs common to all sublines are linked with genes in key cell growth pathways, such as Rap1, Hippo, and calcium signaling pathways. In contrast, subline-specific SVs primarily affect genes involved in neurophysiological pathways, such as glutamatergic synapse and morphine addiction pathways. These findings suggest that different genes, associated with various pathways, are disrupted at distinct stages of melanoma progression, providing insights into the genetic factors that may predispose individual melanocytes to melanomagenesis. This methodology presents a comprehensive tool for characterizing tumor genomes and understanding their relationship with disease progression and therapy resistance. We are continuing our analysis to map SVs across the entire mutation-based phylogeny of all sublines, which will further our understanding of melanoma’s genomic landscape. Citation Format: Xiwen Cui, Ayse G. Keskus, Farid R. Mehrabadi, Salem Malikic, Mikhail Kolmogorov, Chi-Ping Day, Glenn Merlino, S. Cenk Sahinalp. Structural variant dynamics in melanoma: Unraveling tumor heterogeneity and evolution [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6924.

Melanoma is the most invasive skin cancer caused by the malignant melanocytes. The use of immune checkpoint blockade (ICB) improved the survival rate in advanced melanoma. Yet, the response rate to ICB varies across patients due to the highly heterogeneous nature of melanoma. Recent studies reported genomic and epigenetic factors contributing to the therapeutic response. Identifying these factors involved in clonal evolution in melanoma is a key to better understanding the tumor progression and divergence in the therapy response. To study melanoma heterogeneity, we generated twenty-four clonal sublines, each derived from a single cell isolated from a parental cell line derived from the M4 mouse melanoma model. DNA prepared from each subline and a spleen from a healthy mouse were subjected to long-read sequencing. Importantly, long-read sequencing allows direct detection of methylation states, small variants, and structural variants in the same run. The results thus provide excellent means to study genetic and epigenetic factors in clonal evolution. Each of these sublines was also implanted in distinct mice for survival and tumor growth analysis and further functional evaluation. We developed Severus, a structural variation (SV) caller for long reads, which works with single (germline), paired (tumor-normal), and multiple samples (e.g., multi-site or time series). Severus takes advantage of improved phasing in long-read sequencing, attributes somatic variants to germline haplotype, and builds a haplotype-specific breakpoint graph that is used to cluster multi-break rearrangements and represent the derived chromosomal structure. We benchmarked Severus using multiplatform validated COLO829 truthset and CHM1/CHM13 haploid genomes, and Severus outperformed all other methods in terms of recall and precision. The structural variation (SV) analysis using Severus with all the sublines showed enrichment of possibly clonal SVs in chr4, chr11, and mainly in chr13. Further investigation revealed SVs in chr13 were not distributed across the chromosome but clustered in certain regions. Analysis of somatic SNVs in chr13 corroborates the localized hypermutation profile, which indicates a possible kataegis. Interestingly, a loss of the chr13 copy affected by kataegis in one of the sublines coincides with a better survival rate, as compared to other sublines, in an allograft melanoma mouse model, which further supports the association between the kataegis and tumor progression. We also identified other clonal and subclonal events, including the homozygous deletion of Cdkn2a in most of the sublines, associated with their aggressiveness in vivo. Our analysis allowed us to identify clonal and subclonal genetic and epigenetic factors involved in tumor progression and evolution in a melanoma model, which can potentially translate to human disease. Citation Format: Ayse Keskus, Anton Goretsky, Yuelin Liu, Xiwen Cui, Tanveer Ahmad, Eva Perez Guijarro, Asher Bryant, Erin Malloy, Salem Malikic, Glenn Merlino, Chi-Ping Day, Cenk Sahinalp, Mikhail Kolmogorov. Melanoma clonal subline analysis reveals genetic factors driving intra-tumor heterogeneity [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7407.

Intratumoral heterogeneity arises as a result of genetically distinct subclones emerging during tumor progression. These subclones are characterized by various types of somatic genomic aberrations, with single nucleotide variants (SNVs) and copy number aberrations (CNAs) being the most prominent. While single-cell sequencing provides powerful data for studying tumor progression, most existing and newly generated sequencing datasets are obtained through conventional bulk sequencing. Most of the available methods for studying tumor progression from multi-sample bulk sequencing data are either based on the use of SNVs from genomic loci not impacted by CNAs or designed to handle a small number of SNVs via enumerating their possible copy number trees. In this paper, we introduce DETOPT, a combinatorial optimization method for accurate tumor progression tree inference that places SNVs impacted by CNAs on trees of tumor progression with minimal distortion on their variant allele frequencies observed across available samples of a tumor. We show that on simulated data DETOPT provides more accurate tree placement of SNVs impacted by CNAs than the available alternatives. When applied to a set of multi-sample bulk exome-sequenced tumor metastases from a treatment-refractory, triple-positive metastatic breast cancer, DETOPT reports biologically plausible trees of tumor progression, identifying the tree placement of copy number state gains and losses impacting SNVs, including those in clinically significant genes.

Intratumoral heterogeneity (ITH) can promote cancer progression and treatment failure, but the complexity of the regulatory programs and contextual factors involved complicates its study. To understand the specific contribution of ITH to immune checkpoint blockade (ICB) response, we generated single cell-derived clonal sublines from an ICB-sensitive and genetically and phenotypically heterogeneous mouse melanoma model, M4. Genomic and single cell transcriptomic analyses uncovered the diversity of the sublines and evidenced their plasticity. Moreover, a wide range of tumor growth kinetics were observed in vivo, in part associated with mutational profiles and dependent on T cell-response. Further inquiry into melanoma differentiation states and tumor microenvironment (TME) subtypes of untreated tumors from the clonal sublines demonstrated correlations between highly inflamed and differentiated phenotypes with the response to anti-CTLA-4 treatment. Our results demonstrate that M4 sublines generate intratumoral heterogeneity at both levels of intrinsic differentiation status and extrinsic TME profiles, thereby impacting tumor evolution during therapeutic treatment. These clonal sublines proved to be a valuable resource to study the complex determinants of response to ICB, and specifically the role of melanoma plasticity in immune evasion mechanisms.

Recent studies exploring the impact of methylation in tumor evolution suggest that while the methylation status of many of the CpG sites are preserved across distinct lineages, others are altered as the cancer progresses. Since changes in methylation status of a CpG site may be retained in mitosis, they could be used to infer the progression history of a tumor via single-cell lineage tree reconstruction. In this work, we introduce the first principled distance-based computational method, Sgootr, for inferring a tumor’s single-cell methylation lineage tree and jointly identifying lineage-informative CpG sites which harbor changes in methylation status that are retained along the lineage. We apply Sgootr on the single-cell bisulfite-treated whole genome sequencing data of multiregionally-sampled tumor cells from 9 metastatic colorectal cancer patients made available by Bian et al., as well as multiregionally-sampled single-cell reduced-representation bisulfite sequencing data from a glioblastoma patient made available by Chaligne et al.. We demonstrate that the tumor lineages constructed reveal a simple model underlying colorectal tumor progression and metastatic seeding. A comparison of Sgootr against alternative approaches shows that Sgootr can construct lineage trees with fewer migration events and more in concordance with the sequential-progression model of tumor evolution, in time a fraction of that used in prior studies. Interestingly, lineage-informative CpG sites identified by Sgootr are in inter-CpG island (CGI) regions, as opposed to CGI’s, which have been the main regions of interest in genomic methylation-related analyses. Sgootr is implemented as a Snakemake workflow, available at https://github.com/algo-cancer/Sgootr.