Abstract Understanding the spatial and temporal dynamics of tumour evolution is crucial for determining the drivers of cancer progression. Linking genotype to phenotype across a tumour remains challenging, however. Here, we use single-cell and spatial multi-omics to comprehensively profile a primary malignant peripheral nerve sheath tumour (MPNST) and its multifocal recurrence. Combining the native barcoding system from extensive heterogeneity in copy number alterations with mutation data, we resolve the evolutionary tree of this tumour, revealing a branching structure suggestive of ongoing chromosomal instability. We show that gene dosage effects contribute to phenotypic diversity and scale predominantly linearly with copy number. Using spatial genomics assisted by laser capture microdissection and spatial transcriptomics, we performed in situ lineage tracing in this human tumour, elucidating the relationship between local expansions and interactions between tumour cells and the microenvironment. These findings demonstrate the potential of combined bulk, single-cell and spatial techniques to dissect cancer evolution in time and space and link genotype to phenotype in detail.

BACKGROUND:The number and type of genetic alterations required to initiate breast and ovarian cancer remain unclear. While germline BRCA1/2 carriers show markedly elevated cancer risk, it is uncertain whether point mutations or copy number alterations constitute the rate-limiting events of tumourigenesis. METHODS:We developed a statistical framework extending prior incidence-mutation models to estimate the minimal number and type of driver events required for cancer initiation. Somatic mutation and copy-number data from >3000 breast and ovarian cancers in TCGA and METABRIC were compared between germline BRCA1/2 carriers and non-carriers matched on subtypes. Results were validated through analyses of evolutionary timing data, as well as single-cell whole genome sequencing (scWGS) data of genetically-engineered and patient-derived cancer/pre-cancerous cells. FINDINGS:Deletions, rather than single-nucleotide variants (SNVs), emerged as the likely rate-limiting events. Modeling indicated that 1-3 deletions are sufficient to initiate tumourigenesis, whereas SNVs alone could not explain observed incidence ratios. BRCA1/2-driven and sporadic tumours converged on similar deletion profiles, including early recurrent deletions of chromosomes 13q and 17, though carriers accumulated them more rapidly. INTERPRETATION:Deletion-associated chromosomal instability likely represents the central trigger for breast and ovarian cancer initiation. These results explain why certain somatic driver mutations detected in normal tissues may not predict malignant progression, and that early detection strategies should instead prioritize testing deletions as potential biomarkers. FUNDING:NIH/NCI (P30CA016042; 1U01CA214194-01), NIH NIGMS (R35GM138113, 2R35GM138113), ACS (RSG-22-115-01-DMC), CIHR Vanier Fellowship, and the Francis Crick Institute with core funding from Cancer Research UK, UK Medical Research Council, and Wellcome Trust.

Tumor subclonal architecture shapes cancer evolution, yet subclonal reconstruction from bulk sequencing remains difficult to scale due to computational cost and model complexity. We present CliPP, a penalized-likelihood framework that jointly estimates cellular prevalence with pairwise fusion penalties, automatically identifying subclones without requiring extensive priors. Across simulations and 2,778 whole-genome tumors with external consensus reconstructions, CliPP achieves consistently good performances when compared to state-of-the-art approaches while providing substantial runtime reductions. Applied to 7,000+ tumors across >30 cancer types, CliPP quantifies pervasive subclonality and delineates cohort-level subclone landscapes. CliPP enables fast, reproducible large-scale subclonal analysis and is freely available to the community through GitHub and a shiny app.

DNA ploidy is an important predictor of tumor behavior and prognosis, and its accurate estimation is essential for robust genomic analysis in translational cancer research and diagnostics. However, the most common in silico methods for ploidy estimation using next-generation-sequencing-based copy-number aberration (CNA)-calling algorithms are often inaccurate due to inherent ambiguity in fitting ploidy solutions. This study evaluates the accuracy of state-of-the-art CNA callers using whole-genome sequencing by comparing their ploidy estimates with gold-standard ploidy measurements derived by flow cytometry (FC). We demonstrate that CNA callers are up to 38% inaccurate in cancers with complex genomes, which impacts the accurate estimation of copy number of cancer genes and could have clinical implications and impacts on inferences of tumor evolution. Critically, flow-cytometry-based calibration of CNA callers yields highly accurate ploidy estimates (ρPearson = 0.92, p < 0.001), providing a robust solution to the substantial inaccuracies that compromise clinical decision-making and evolutionary inference in complex cancers.

Abstract Immune escape is a fundamental hallmark of cancer, facilitating disease progression and resistance to therapy. The temporal dynamics and genetic evolution underlying immune escape across diverse cancers remain incompletely characterized. Here we developed an integrative framework that systematically identifies immunomodulatory genes from functional genomic screens and reconstructs tumor evolutionary history using whole-genome sequencing data to infer mutation timing. Applying this approach to 2,658 tumors across 38 cancer types, we generated the first pan-cancer atlas of immune escape evolution, revealing distinct trajectories such as late amino acid metabolism mutations in pancreatic adenocarcinoma, early neuroactive ligand-receptor and IFNγ pathway mutations in esophageal adenocarcinoma, and late protein methylation mutations in breast adenocarcinoma. These findings provide a comprehensive map of genetic immune evasion evolution, offering insights that may inform early detection strategies, immunoprevention, and therapeutic interventions across cancers. To facilitate easy access to the results, we further built a user-friendly website to interactively interrogate the evolution of immune escape. Citation Format: Shengqing Gu, Wenjie Chen, Toby Baker, Zhihui Zhang, Huw A. Ogilvie, Peter Van Loo. Pan-cancer mapping of genetic immune escape evolutionary trajectories [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 688.

Abstract Background: The aromatase inhibitors (AI) letrozole and exemestane are often used in sequence in targeting ER+ breast cancers. However, resistance to AI poses a major barrier to sustained clinical benefit, while the biological mechanisms underlying the phenomenon remain largely unknown. In this study, we build on our clinical NeoLetExe trial, with the aim to investigate the molecular basis of resistance to AI, by analysing subclonal evolutionary dynamics during sequential treatment. Methods: We use whole-exome sequencing (WES) data from 11 ER+ breast cancer patients and 3 timepoints of the Neoletexe trial to reconstruct cancer cell fraction-based subclonal composition. Single-cell DNA sequencing from matched tumour samples is generated on the MissionBio platform to validate the identified clones and variants. Subclonal variants were annotated to genes by integrating evidence from public data and ExpectoSc. Pathway enrichment analysis using Human Base was conducted. Results: Higher cancer cell fraction clone trajectories were significantly associated with reduced treatment response (p = 0.023). Clones reconstructed by WES were validated at 81% using single-cell DNA sequencing. Clones resistant to both letrozole and exemestane demonstrated PIK3CA/AKT/mTOR signaling activation, KRAS pathway dysregulation, hedgehog signaling, and androgen receptor pathways, alongside extensive immune activation and metabolic reprogramming. Drug-specific resistance patterns showed exemestane-resistant clones enriched for epigenetic control and miRNA-mediated silencing, while letrozole-resistant clones displayed metabolic dysregulation but notably lacked immune pathway activation. In contrast, treatment-sensitive clones maintained coordinated cell cycle control, preserved DNA damage responses, and retained immune signaling capacity. Analysis of FDA-approved breast cancer targets identified actionable alterations in PIK3CA (4 patients) and AKT1 (1 patient) that persisted through AI treatment, with RNA expression analysis revealing 48 additional therapeutic targets spanning PI3K/AKT/mTOR, CDK4/6, DNA repair (BRCA1/2, ATM), and immune checkpoint pathways. Conclusion: WES-based cancer cell fraction analysis successfully captured subclonal evolutionary trajectories during AI treatment, revealing drug-specific mechanisms and identifying key molecular players in endocrine therapy resistance. This work establishes a framework for precision oncology approaches by providing actionable therapeutic targets and advancing our understanding of resistance mechanisms to improve clinical outcomes in sequential AI therapy. Citation Format: Vessela N. Kristensen, Denise G. O`Mahony, Tom Lesluyes, Ina S. Brorson, Patrik H. Vernhoff, Ksenia Sokolova, Miriam R. Aure, Grethe G. Alnæs, Rebecca M. Hoøen, Arvind Y. Sundaram, Chandra Theesfeld, Stephanie B. Geisler, Torill Sauer Sauer, Nazli Bahrami, Andliena Tahiri, Torben Lüders, Olga Troyanskaya, Charles Vaske, Peter Van Loo, Jürgen Geisler. Single cell DNA sequencing reveals clonal selection, hormonal adaptation and treatment resistance in neoadjuvant clinical trial of Aromatase inhibition [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(8_Suppl):Abstract nr LB190.

Table S1 shows the number of possible routes for different copy numbers states.

Leiomyosarcoma is a smooth muscle-derived malignancy marked by significant clinical heterogeneity. The extent and nature of cellular heterogeneity and molecular underpinnings remain poorly understood. To address this at transcriptomic and epigenomic levels, we performed single-nucleus multiome sequencing on untreated primary leiomyosarcoma tissues. Malignant cells segregated almost exclusively into two previously unrecognized and epigenetically distinct states: a dedifferentiated, mesenchymal-like subtype (MES) and a differentiated smooth muscle-enriched subtype (SMC). Chromatin accessibility profiling revealed strong enrichment of nuclear factor I (NFI) transcription factor motifs in MES cells, whereas AP-1 family motifs-most prominently FOSL2-were selectively accessible in SMC cells. Established leiomyosarcoma cell lines faithfully recapitulated these subtypes, and targeted depletion of NFI or AP-1 factors suppressed proliferation, invasion, and in vivo tumor growth, demonstrating functional dependency on these transcriptional programs. Spatial transcriptomics across 328 tissue cores from 128 leiomyosarcomas showed that immunosuppressive macrophages preferentially cluster around MES regions, revealing a subtype-specific tumor-immune niche. Clinically, MES-dominant tumors were associated with significantly worse patient outcomes. Through an epigenetic inhibitor screen, we identify and validate SMARCA4/2 inhibition as a promising therapeutic vulnerability for MES leiomyosarcomas. Together, this work defines two epigenetically driven, transcription factor-regulated, and clinically relevant states of leiomyosarcoma, revealing mechanistic underpinnings of tumor heterogeneity and uncovering actionable therapeutic strategies.

Abstract Epigenetic remodeling is a hallmark of tumor evolution, yet the timing and clonality of promoter methylation events remain poorly defined. We investigated the spatial architecture of promoter DNA methylation in multiregion lung adenocarcinoma (LUAD) to identify clonal epigenetic alterations and evaluate their prognostic and translational relevance across cancers. Reduced representation bisulfite sequencing (RRBS) was performed on 151 tumor regions from 32 TRACERx LUAD patients with matched normal tissue. Differentially methylated regions (DMRs) were quantified for intratumor (ITH) and intertumor (ITeH) heterogeneity. Clonality was evaluated through three complementary metrics of methylation ubiquity, followed by univariate and LASSO Cox modeling. The resulting 30DMR panel, termed PROMISE, was derived through clustering concordance with TCGA LUAD and independently validated in CPTAC 3. PROMISE was subsequently tested for pancancer specificity across 18 tumor types from several publicly available sources.We identified 21,358 promoter DMRs showing a continuum of ITH and ITeH patterns. Promoters with low ITH but high ITeH, ubiquitously hypermethylated within tumors yet variable across patients, represented early clonal events significantly associated with poor survival. LASSO Cox selection yielded a 30-DMR signature encompassing genes involved in immune regulation, epithelial polarity, and TGFβ signaling. PROMISE robustly stratified patients into high and low risk groups in both TCGA LUAD and CPTAC 3 cohorts (logrank P < 0.01) and remained independent of clinicopathologic variables. Cross-cancer analyses revealed strong prognostic associations in kidney, thyroid, liver, and colorectal cancers, but minimal signal in squamous, stromal rich, or hematologic malignancies.PROMISE captures a panepithelial program of early, clonal promoter hypermethylation recurrent across multiple carcinomas and predictive of outcome in selected epithelial tumors. These findings highlight clonal methylation remodeling as an early determinant of tumor evolution and nominate PROMISE as a clinically actionable biomarker for cancer risk stratification. Citation Format: Francisco Gimeno-Valiente, Constantino De La Vega, Yun-Hsin Liu, Carla Castignani, Ieva Usaite, Martín Arana Jorge, Elrick Hillary, Stephan Beck, Miljana Tanic, Jonas Demeulemeester, Peter Van Loo, Charles Swanton, Mariam Jamal-Hanjani, Nnennaya Kanu. PROMISE, a clonal promoter methylation signature capturing early epigenetic evolution and prognostic programs across epithelial cancers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 3197.

Cancer is an evolutionary process characterized by profound intratumor heterogeneity (ITH), which can be quantified using in silico estimates of cancer cell fractions (CCFs) of tumor-specific somatic mutations. We demonstrate a data-driven approach based on CCF distributions to identify 4 robust pan-cancer evolutionary signatures from 4,146 tumors across 17 cancer types in The Cancer Genome Atlas (TCGA). These signatures define a continuum of cancer cell fractions reflecting neutral evolution, clonal expansion, and clonal fixation. Correlating evolutionary signatures with mutational and biological programs reveals that tumors enriched for clonal expansion and fixation are associated with immune evasion and distinct changes in the tumor immune microenvironment. Our analysis reveals a dynamic shift from adaptive to innate immune programs as tumors progress toward clonal fixation and escape immune surveillance, accompanied by the clonal expansion of driver genes modulating tumor-stroma interactions. These evolutionary dynamic subtypes are further associated with clinical outcomes and immunotherapy responses.

LINE-1 (L1) retrotransposition generates somatic genomic variation in human cancer, but short-read sequencing has limited our understanding of its structural consequences and dynamics. Using long-read sequencing, we analyzed 10 tumors with exceptionally high retrotransposition activity, comprising more than 6000 somatic events. We reveal that L1-mediated reciprocal translocations occur frequently, typically driven by two concurrent L1 retrotransposition events on nonhomologous chromosomes. Using an independent tumor cohort spanning low to high L1 activity, we estimate that retrotransposon-mediated rearrangements arise at a frequency of one event per 60 somatic retrotranspositions. Molecular timing analyses indicate that these events arise early in tumorigenesis, establishing L1 activity as an early driver of chromosomal instability. Our findings demonstrate that L1 contributes substantially to cancer genome evolution in certain tumors.

Cancer arises from an evolutionary process that can be reconstructed from DNA sequencing and modeled by tumor phylogenies. High coverage bulk DNA sequencing (bulk DNA-seq) is widely available, but tumor phylogeny inference requires deconvolution, often resulting in non-uniqueness in the solution space. Single-cell DNA sequencing (scDNA-seq) holds potential to yield higher resolution tumor phylogenies, but the sparsity of emerging low-pass sequencing technologies poses challenges for the study of single-nucleotide variants. Increasing availability of data sequenced with both modalities provides an opportunity to capitalize on the advantages of these technologies. While inference methods exist for bulk DNA-seq and for low-pass scDNA-seq, no joint inference methods currently exist. As a first step, we propose a method named ARBORIST that prioritizes tumor phylogenies inferred via bulk DNA-seq using low-pass scDNA-seq data. ARBORIST takes as input a candidate set of trees with corresponding SNV clustering, along with variant and total read count data from scDNA-seq and uses variational inference to approximate a lower bound on the marginal likelihood of each tree in the candidate set. On simulated data, matching characteristics of current scDNA-seq data, ARBORIST outperforms both bulk and low-pass single-cell reconstruction methods. On a biological dataset, ARBORIST conclusively resolves the evolutionary relationship between different SNV clusters on a malignant peripheral nerve sheath tumor, which is supported by orthogonal validation via a proxy for copy number. ARBORIST provides a principled framework for integrating bulk DNA-seq and low-pass scDNA-seq data, improving confidence in tumor phylogeny reconstruction. Availability: https://github.com/VanLoo-lab/Arborist

Supplementary Figures S1 to S52 S1 WGD frequencies across cancer types and stage. S2 Effect of WGD constraint on timing accuracy. S3 Measuring timing accuracy on simulated data. S4-7 Measuring timing accuracy on simulated data by copy number state. S8-11 Measuring inferred route probabilities on simulated data. S12 Timing of gains in multi-region tumors. S13-14 Difference in timing between different gain routes. S15 Non-parsimony in copy number gain evolution. S16 Non-parsimony by copy number state. S17-18 Calibrating a penalty on non-parsimony. S19 Clear-cell sample gain timing. S20 Gain route agreement within chromosomes. S21 Probability of pre-WGD gains in different chromosomes and copy number states. S22 Distribution of gain timing by major copy number. S23 Single-cell copy number profiles of an undifferentiated sarcoma. S24 Distribution of gain rates relative to WGD by cancer type. S25 Distribution of gain rates relative to WGD compared to simulations. S26 Example sample gain timing posterior. S27 Combined distribution over gain timing by WGD status. S28 The timing of gains relative to WGD. S29 The timing of gains relative to WGD by cancer type. S30 Proportion of copy number events post-WGD. S31 The relationship between genome gained post-WGD and WGD timing by cancer type. S32 The relationship between genome gained pre-WGD and WGD timing by cancer type. S33 The relationship between fraction of genome lost pre and post-WGD and WGD timing by cancer type. S34 Punctuated gains in WGD tumors. S35 Association between chromothripsis and punctuated gains. S36 Genomic features of punctuated gains. S37 Frequency of arm gains pre and post-WGD and in non-WGD tumors. S38 Frequency of arm gains pre and post-WGD and in non-WGD tumors by cancer type. S39 Frequency of arm losses pre and post-WGD and in non-WGD tumors by cancer type. S40 Effect of oncogene and tumor suppressor gene density on arm gain rates. S41 Effect of oncogene and tumor suppressor gene density on arm loss rates. S42-46 Pan-genome frequencies of pre and post-WGD gains by cancer type. S47-49 Pan-genome frequencies of pre and post-WGD losses by cancer type. S50 The effect of NRPCC and mutation count on gain timing inference. S51 WGD status calling in GRITIC. S52 The effect of the non-parsimony penalty on event timing.

Somatic mutations accumulate independently in the two parental genome copies of our cells throughout life and shape cancer evolution. Although local mutation rates are influenced by allele-specific features such as DNA sequence, epigenetic marks, and chromatin structure, whether these translate into genome-wide differences in mutation accrual between the two parental copies is unknown. Cancer genomics analyses, including copy-number gain timing and molecular archaeology, assume that mutations accrue symmetrically on the two homologous parental genomes, yet this assumption has never been tested. Here we present PhaSoMix, a framework exploiting the genetic differentiation between parental haplotypes in admixed cancer patients to assign somatic mutations to their parent of origin without parent or parent-surrogate sequencing. Applying it with explicit modeling and propagation of phasing and ancestry-inference uncertainty across 21 tumor whole genomes from the Pan-Cancer Analysis of Whole Genomes cohort, we find mutation burdens highly symmetric between maternal and paternal genomes, across cancer types, genomic annotations, clonal timing categories, and mutational processes including clock-like CpG sites, bounding any asymmetry to within 4-5%. Simulations show that violations would substantially bias gain-timing estimates in late evolutionary windows. This provides the first quantification of parental mutation-burden symmetry in vivo, validating a key assumption of cancer evolutionary analyses.

Supplementary methods, including full description of the GRITIC method

Abstract Somatic copy-number alterations (SCNAs) are pervasive across cancers, arising through diverse mutational mechanisms and subsequently shaped by selection according to fitness advantages [1–4]. These alterations can drive tumourigenesis via gene-dosage changes in oncogenes and tumour suppressors [5, 6]. However, observed SCNA profiles do not uniquely define the genomic events that led to them, which is a major obstacle to understanding the selective forces shaping cancer genomes. Here we present SPICE, Selection Patterns In somatic Copy-number Events, an event-level framework that infers discrete copy-number events from allele-specific profiles and models focal selection from first principles, incorporating uniform breakpoint formation and locus-specific selective pressures. Applied to 5,966 samples from the TCGA dataset, SPICE quantifies the full spectrum of SCNA events, recapitulates known mutational processes, and tracks systematic shifts in event types before and after whole-genome duplication (WGD). Next, SPICE employs a generative selection model to identify the location of oncogenes and tumour-suppressors. Unlike previous approaches that detected peaks in aggregated copy-number signals, our model operates directly on inferred events and treats genome-wide breakpoint formation as a neutral reference against which locus-specific selection is detected. This analysis reveals 460 loci under selection which are highly enriched for known oncogenes and tumour suppressors, recapitulating most previously reported sites and uncovering many novel regions, and simultaneously showing that most internal copy-number events are not subject to focal selection. These results establish a unified framework that deconvolves copy-number profiles into their underlying evolutionary events and greatly expand the catalogue of loci implicated in cancer development. Graphical abstract

Polyploid giant cancer cells (PGCCs), characterised by multinucleation and atypical nuclear morphology, are a common feature of undifferentiated pleomorphic sarcomas. While PGCCs may be a critical substrate for cancer evolution, their formation pathways and genomic consequences remain underexplored. In this study, we characterise PGCCs in ten pleomorphic sarcomas and use topographic single-cell DNA sequencing (scDNA-seq) to investigate their genomic landscape. We selected PGCCs based on their nuclear morphology, including mononucleated or multinucleated bizarre, misshapen nuclei, and analysed them at single-cell resolution. Histopathological analysis showed that PGCCs were often randomly distributed throughout the tumour and did not appear in clusters, suggesting that they arise de novo rather than through clonal expansion. scDNA-seq revealed that PGCCs originate from the dominant tumour population and exhibit extensive copy number heterogeneity, either due to subsequent or ongoing chromosomal instability. Both clonal and subclonal chromothripsis-like events were identified in PGCCs, indicating that chromothripsis is a key driver of heterogeneity in these cells and is linked to multinucleation rather than mononuclear PGCC formation. FACS-based ploidy analysis of one undifferentiated pleomorphic sarcoma (UPS) revealed a twice whole-genome-duplicated population (6.2n) distinct from the bulk tumour (3.3n). This population contained all clonal, but none of the subclonal chromothripsis-like events observed in PGCCs. Our findings highlight PGCCs as a highly heterogeneous and evolutionarily dynamic component of UPSs. The recurrent chromothripsis-like events observed in PGCCs suggest ongoing genomic reshaping that may drive tumour progression and the poor clinical outcomes observed for these tumours.

Malignant peripheral nerve sheath tumors (MPNSTs) are a rare subtype of sarcomas arising from peripheral nerves, occurring either sporadically or in association with neurofibromatosis type 1 (NF1). Although comprising only 5-10% of soft tissue sarcomas, MPNSTs are highly aggressive and have limited treatment options, with 5-year survival rates of 34%-60%. Intra-Tumor Heterogeneity (ITH) is a key driver of tumor evolution and treatment resistance. However, prior genomic analyses of MPNSTs are often limited to a single sample per patient and have not explored ITH in detail. To fill this gap, we developed a multi-omics and single-cell integration pipeline to analyze bulk Whole-Genome Sequencing (WGS), single-nuclei DNA (snDNA) and RNA (snRNA) sequencing, and multi-regional shallow-coverage WGS derived from multiple Laser Capture Microdissection (LCM) spots within tissue sections. We applied this integrative multi-omics approach to analyze 63 samples collected from 17 patients. Somatic mutations were called in WGS using a two-out-of-three consensus approach (MuSE2, Strelka2, Mutect2), and Copy Number Aberrations (CNA) were analyzed using the Battenberg algorithm. Phylogenetic trees were reconstructed using Dirichlet Process-based methods and PyClone. We identified low-to-moderate mutational burden (median = 1.23 mutations/Mb, range: 0.52–2.34 mutations/Mb) but significant chromosomal aberrations and varied levels of genome instability. Whole genome doubling was highly prevalent (47/63 samples; 14/17 patients) and many samples showed widespread loss of heterozygosity (median = 26%, range: 4%-93%). Among the 17 patients, 5 exhibited a linear evolution pattern, while the majority demonstrated branching evolution. Cross-referencing CNA profiles across bulk WGS, snDNA (ASCAT.sc), LCM, and snRNA (inferCNV) data validated WGS-derived CNAs and demonstrated subclonal CNA co-occurrence within individual cells or subclones. Genotyping WGS-derived mutation clusters in snDNA revealed that co-occurrence of mutation pairs within single cells was exclusively observed when mutations belonged to the same cluster or lineage, further validating our bulk WGS-derived phylogenetic trees. Our findings reveal that MPNSTs are primarily driven by chromosomal instability, with ITH further elucidated through cross-referenced and refined phylogenetic tree reconstruction at single-cell resolution. Future work would focus on 3D phylogenetic tree reconstruction using spatial-genomics information derived from LCM and integrating single-cell transcriptomics. Additionally, incorporating spatial transcriptomics could provide deeper insights into the tumor microenvironment and its role in driving ITH. These efforts will further advance the understanding of MPNST biology and evolution, with potential implications for targeted therapeutic strategies. Yidan Pan, Yixiao Cheng, Chunxu Gao, Haixi Yan, Zhihui Zhang, Leah Weber, Tom Lesluyes, Cristina Cotobal Martin, Isidro Cortes-Ciriano, Nischalan Pillay, David T. Miller, Adrienne M. Flanagan, Maxime Tarabichi, Peter Van Loo. Exploring intra-tumor heterogeneity in malignant peripheral nerve sheath tumors through single-cell multi-omics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3888.

Intrahepatic cholangiocarcinoma (ICC) is a highly lethal cancer of the bile ducts that exhibits a diversity of driver oncogenic mutations, several of which have been shown to be frequently subclonal. However, existing tumor evolution studies in ICC have been limited by either the number of sub-regions assessed or a paucity of driver mutation heterogeneity. Here, we conducted a multi-omic analysis of 90 tumor regions from 11 discrete tumors in 10 patients, identifying subclonal driver mutations in 5 of the patients, in the genes PBRM1, PIK3CA, TP53, ARID1A, BRAF, MLH1, and CDKN2A. This high-number multi-region approach also allowed us to unambiguously identify distinct subclonal evolutionary branches and map them to the spatial location of the samples. Importantly, each subclonal driver mutation was tightly associated with these defined branches, implicating them as a frequent mechanism of driving subclonal evolution. Additionally, subclonal branching showed a spatial correlation on surgical resection depicting the strength of multi-region sequencing in treatment. In total, we conclude that while subclonal branching can be driven by a number of factors, the acquisition of bona fide driver mutations strongly propels subclonal evolution and spatial patterns of tumor growth. Evit John, Pooja Shukla, Anish Jain, Alexandria Lau, Tom Lesluyes, Huw Ogilvie, Maxime Tarabichi, Jason Roszik, Yinyin Yuan, Peter Van Loo, Lawrence Kwong, Yun Shin Chun. Driver mutations propel subclonal evolution and spatial patterns of tumor growth in intrahepatic cholangiocarcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 7499.