Clonal Architecture Research Articles

Revisiting ramet connection under homogeneous water availability environments

There is little consistency on effects of ramet physical connection in homogeneous environments. To revisit this issue, we chose Cynodon dactylon (Linn.) Pers. (Bermuda grass) as a model plant to establish three experiments. In the water availability experiment (I), stolon severing had no significant effects on plant biomass; however, it significantly affected plant morphology by increasing main stolon length of C. dactylon grass under the homogeneous high water (i.e., less stress) condition, possibly due to an escape strategy from disturbance (stolon severing). Apical dominance and root competition are common factors shaping these plant performance measures. We followed these results with two additional experiments to test if such results could be explained by the partial release of apical dominance or root competition. In the apical dominance release experiment (II), biomass and number of branches of distal sections were significantly increased. However, at the whole clonal fragment level, main stolon and inter-node length were both significantly decreased after apical ends were disconnected. In the root competition experiment (III), no significant effect was found indicating weak root competition in homogeneous environments. We concluded that stolon connection had no significant effects on the biomass of C. dactylon while it affected plant morphology in homogeneous environments. Secondly, clonal architecture is partially determined by the presence of ramet connection and apical dominance. Removal of the apical end may have enhanced development of the distal sections, but the loss of stolon connection with distal sections inhibited the performance of apical sections leading to negative effects to the whole clonal fragment. Apical dominance seems more important than root competition for plant performance of C. dactylon under homogeneous environments, but this is only one clonal species in one environmental condition.

Discovering significant evolutionary trajectories in cancer phylogenies.

Tumors are the result of a somatic evolutionary process leading to substantial intra-tumor heterogeneity. Single-cell and multi-region sequencing enable the detailed characterization of the clonal architecture of tumors and have highlighted its extensive diversity across tumors. While several computational methods have been developed to characterize the clonal composition and the evolutionary history of tumors, the identification of significantly conserved evolutionary trajectories across tumors is still a major challenge. We present a new algorithm, MAximal tumor treeS TRajectOries (MASTRO), to discover significantly conserved evolutionary trajectories in cancer. MASTRO discovers all conserved trajectories in a collection of phylogenetic trees describing the evolution of a cohort of tumors, allowing the discovery of conserved complex relations between alterations. MASTRO assesses the significance of the trajectories using a conditional statistical test that captures the coherence in the order in which alterations are observed in different tumors. We apply MASTRO to data from nonsmall-cell lung cancer bulk sequencing and to acute myeloid leukemia data from single-cell panel sequencing, and find significant evolutionary trajectories recapitulating and extending the results reported in the original studies. MASTRO is available at https://github.com/VandinLab/MASTRO. Supplementary data are available at Bioinformatics online.

Mutational signature and clonal relatedness of recurrent urothelial carcinomas with aristolochic acid.

Urothelial carcinomas (UCs) are malignant tumors that arise from the lower and upper urinary tract and are characterized by multiple recurrences. Aristolochic acid (AA) is a potent nephrotoxin and human carcinogen associated with UC. East Asian populations with a high UC prevalence have an unusual genome-wide AA-induced mutational pattern. To address the genomic differences and clonal relatedness between primary and recurrent tumors in the UCs with AA pattern, we investigated the genomic differences and tumor microenvironment (TME) of AA and non-AA UCs. 17 UC patients were recruited, with nine documented AA exposure. Eleven of them showed recurrence. After-surgery tissues of primary and paired recurrent tumors were collected. Capture-based targeted deep sequencing was performed using a commercial panel consisting of 520 cancer-related genes. Tumor-infiltrating lymphocytes (TILs) were identified with an immunofluorescence-based microenvironment analysis panel (MAP). Hierarchical clustering based on the COSMIC signatures confirmed two significant subtypes: AA Sig and non-AA Sig. AA Sig was associated with AA-containing herbal drug intake, recurrence, and higher tumor mutation burden (TMB). The clonal architecture of UCs revealed three types of clonal evolution patterns. Non-AA Sig cohort showed shared clonal origin of primary and recurrent tumors. AA Sig showed heterogeneity and had multiple independent origins. Recurrent tumors as second primary tumors in AA Sig showed immunoreactive TME, indicating a better response with immune checkpoint inhibitor therapy. The AA mutational signature and unique immune profiles are helpful molecular markers to distinguish AA exposure from other carcinogens. These results also provide new insights into the origin of recurrent UCs that could affect treatment strategies.

Comprehensive characterization of clonality of driver genes revealing their clinical relevance in colorectal cancer

BackgroundGenomic studies of colorectal cancer have revealed the complex genomic heterogeneity of the tumor. The acquisition and selection of genomic alterations may be critical to understanding the initiation and progression of this disease.MethodsIn this study, we have systematically characterized the clonal architecture of 97 driver genes in 536 colorectal cancer patients from TCGA.ResultsA high proportion of clonal mutations in 93 driver genes were observed. 40 genes showed significant associations between their clonality and multiple clinicopathologic factors. Kaplan–Meier analysis suggested that the mutation clonality of ANK1, CASP8, SMAD2, and ARID1A had a significant impact on the CRC patients' outcomes. Multivariable analysis revealed that subclonal ANK1 mutations, clonal CASP8 mutations, and clonal SMAD2 mutations independently predicted for shorter overall survival after adjusting for clinicopathological factors. The poor outcome of the subclonal ANK1 mutation may be caused by upregulation of IL4I1, IDO1, IFNG and MAPK12 which showed potential roles in tumor immune evasion through accumulation of immunosuppressive cells such as regulatory T cells and myeloid derived suppressor cells.ConclusionThese results suggested that the clonality of driver genes could act as prognostic markers and potential therapeutic targets in human colorectal cancer.

Fusion Gene Detection and Quantification by Asymmetric Capture Sequencing (aCAP-Seq)

Several fusion genes such as BCR::ABL1, FIP1L1::PDGFRA, and PML::RARA are now efficiently targeted by specific therapies in patients with leukemia. Although these therapies have significantly improved patient outcomes, leukemia relapse and progression remain clinical concerns. Most myeloid next-generation sequencing (NGS) panels do not detect or quantify these fusions. It therefore remains difficult to decipher the clonal architecture and dynamics of myeloid malignancy patients, although these factors can affect clinical decisions and provide pathophysiologic insights. An asymmetric capture sequencing strategy (aCAP-Seq) and a bioinformatics algorithm (HmnFusion) were developed to detect and quantify MBCR::ABL1, μBCR::ABL1, PML::RARA, and FIP1L1::PDGFRA fusion genes in an NGS panel targeting 41 genes. One-hundred nineteen DNA samples derived from 106 patients were analyzed by conventional methods at diagnosis or on follow-up and were sequenced with this NGS myeloid panel. The specificity and sensitivity of fusion detection by aCAP-Seq were 100% and 98.1%, respectively, with a limit of detection estimated at 0.1%. Fusion quantifications were linear from 0.1% to 50%. Breakpoint locations and sequences identified by NGS were concordant with results obtained by Sanger sequencing. Finally, this new sensitive and cost-efficient NGS method allowed integrated analysis of resistant chronic myeloid leukemia patients and thus will be of interest to elucidate the mutational landscape and clonal architecture of myeloid malignancies driven by these fusion genes at diagnosis, relapse, or progression.

Autopolyploid establishment depends on life-history strategy and the mating outcomes of clonal architecture.

Polyploidy is a significant component in the evolution of many taxa, particularly plant groups. However, new polyploids face substantial fitness disadvantages due to a lack of same-cytotype mates, and the factors promoting or preventing polyploid establishment in natural populations are often unclear. We develop spatially explicit agent-based simulation models to test the hypothesis that a perennial life history and clonal propagation facilitate the early stages of polyploid establishment and persistence. Our models show that polyploids are more likely to establish when they have longer life spans than diploids, especially when self-fertilization rates are high. Polyploids that combine sexual and clonal reproduction can establish across a wide range of life histories, but their success is moderated by clonal strategy. By tracking individuals and mating events, we reveal that clonal architecture has a substantial impact on the spatial structure of the mixed diploid-polyploid population during polyploid establishment: altering patterns of mating within or between cytotypes via geitonogamous self-fertilization, the mechanisms through which polyploid establishment proceeds, and the final composition of the polyploid population. Overall, our findings provide novel insight into the role of clonal structure in modulating the complex relationship between polyploidy, perenniality, and clonality and offer testable predictions for future empirical work.

In This Issue

Current murine models of myeloproliferative neoplasms (MPN) cannot address how MPN can invade hematopoietic organs from a defined single anatomical origin as observed in patients. Murakami et al. developed an immunocompetent wild-type mouse model of Polycythemia Vera (PV) spreading from a single bone marrow site. PV development was abrogated in mice with splenectomy or lacking carbonic anhydrase 1 (CA1) in mutant hematopoietic progenitors. Increased expression of CA1 was found in CD34+ cells from PV patients, and its pharmacologic inhibition decreased variant allele frequency in colony-forming assays, suggesting CA1 as a potential therapeutic target for PV.See article, p. 285.Approximately every fifth patient with myeloproliferative neoplasms (MPN) harbors a CALR mutation of either type I or type II, which differ in clinical features and outcomes. Ibarra et al. show that type I, but not type II, mutant CALR proteins lose Ca2+ binding, causing Ca2+ depletion from the endoplasmic reticulum and activation of the IRE1α/XBP1 pathway. IRE1α inhibition prevented type I-mutant CALR cell survival in culture and thrombocythemia development in mice. These results identify IRE1α/XBP1 as a potentially targetable dependency specific to type I CALR-mutated MPNs.See article, p. 298.By single-cell DNA, RNA and proteogenomic sequencing of matched patient samples, Guess et al. uncovered three clonal evolution patterns associated with clinical progression from myelodysplastic syndrome (MDS) to secondary acute myeloid leukemia (sAML). The Static pattern, characterized by minimal clonal change, had founding clones with mutations in DNA methylation genes and transitioned to sAML in the absence of further genomic aberrations, suggesting nongenetic evolution. Clonal architecture changed dramatically in Dynamic groups, often involving acquisition of mutations in signaling pathways (Dynamic-S) or progressive structural aberrations along with TP53 mutations (Dynamic-C). Collectively, these findings elucidate the interplay of genetic and nongenetic factors in leukemic transformation.See article, p. 316.To investigate patterns of subclone evolution during progression from myelodysplastic syndrome (MDS) to secondary acute myeloid leukemia (sAML), Menssen et al. sequenced paired MDS and sAML samples from a cohort of patients. Signaling gene mutations were found to occur universally in subclones and undergo complex patterns of clonal evolution, including acquisition, expansion, persistence, and contraction of mutations during progression, with multiple patterns often coexisting in the same patient. The presence of at least one signaling gene mutation at MDS was associated with future progression to sAML in patients with lower risk MDS.See article, p. 330.Nakauchi et al. report the genome-wide landscape of DNA hydroxymethylation (5hmC) in human hematopoietic lineages. The authors map TET2-dependent 5hmC, epigenetic and transcriptional programs matched to competitive advantage, myeloid skewing, and reduced erythroid output in TET2-deficient hematopoietic stem and progenitor cells (HSPC). Vitamin C and azacitidine restore the 5hmC landscape and phenotypes in TET2-mutant HSPCs. These findings offer a comprehensive resource for TET-dependent transcriptional regulation of human hematopoiesis and shed light on the potential mechanisms by which TET deficiency contributes to clonal hematopoiesis and malignancies.See article, p. 346.

Comprehensive analysis of intratumoural heterogeneity of somatic copy number alterations in diffuse glioma reveals clonality-dependent prognostic patterns.

Intratumoural heterogeneity (ITH) has been implicated in tumour growth and progression as well as therapy resistance. However, the extent of ITH of somatic copy number alterations (ITH-SCNAs) as a result of tumour evolution and its influence on clinical outcomes in diffuse glioma (DG) remain poorly understood. We used an integrated computational method to infer clonal and subclonal SCNAs in 760 untreated primary DGs from The Cancer Genome Atlas. ITH-SCNAs at the genome-wide, peak (region with recurrent SCNAs) and gene level were calculated. We used the Kaplan-Meier estimators and Cox proportional hazards models to examine the associations of ITH-SCNA with patient outcomes. An independent cohort of 243 patients with paired initial and recurrent tumours from the Glioma Longitudinal Analysis Consortium was used for validation. DGs showed widespread ITH-SCNA, with a median of 25.5% of SCNAs identified as subclonal. We found that clonal SCNA burden had stronger prognostic power than total SCNAs in IDH-mutant astrocytoma. Coamplifications of receptor tyrosine kinases (RTKs) tended to be subclonal, and subclonal RTK amplification was significantly associated with high tumour proliferative potential and unfavourable clinical outcomes in IDH-wild-type glioblastoma. In addition, we found that the prognostic values of the peak-level SCNAs were related to their mutated clonal architecture, from which three clonality-dependent prognostic patterns of SCNAs were proposed, including clonal-dominant, subclonal-dominant and clonality-independent schemas. The systematic analysis of ITH-SCNAs in large cohorts of DGs highlighted the importance of considering the clonality of SCNA in discovery of tumour prognostic markers.

P349: CLONAL ARCHITECTURE DISSECTING AT SINGLE-CELL RESOLUTION REVEALS MECHANISMS OF CHEMOTHERAPY RESISTANCE AND RELAPSE IN T CELL ACUTE LYMPHOBLASTIC LEUKEMIA

Background: T-lineage acute lymphoblastic leukemia (T-ALL) comprises approximately 10-15% of pediatric ALL cases with distinct feature in biology and largely inferior outcome compared to B-ALL. Growing evidence has reflected pivotal roles of clonal evolution in T-ALL recurrence, but bulk sequencing may not serve as the perfect model to reliably infer clonal heterogeneities and their immunomodulatory milieu during leukemia development. In this study, single-cell sequencing was applied to uncover leukemic clonal relationships with relapse throughout chemotherapy in T-ALL at a more accurate resolution. Aims: We performed bulk and single cell multi omics analysis for diagnostic and relapse T-ALL, with the aim to dissect mechanisms of chemotherapy resistance and relapse in T cell acute lymphoblastic leukemia. Methods: We performed bulk whole-exome sequencing for sorted CD7+ BMMCs from 5 pairs of diagnosis-relapse (Dx_Rel) samples, revealing a series of well-reported hotspot mutations in T-ALL. To dissect clonal diversities within and across the 5 Dx_Rel T-ALL pairs, we carried out high-throughput droplet-based 5’-single-cell RNA-seq (scRNA-seq) and paired T cell receptor sequencing (scTCR-seq). Results: By performing unsupervised clustering of scRNA- seq profiles encompassing 10 samples, we identified 23 distinct T-lineage clusters (Cluster 0-22) based on the two-dimensional UMAP visualization. In 2 out of 5 patients, diffusion map of T-lineage sub-clusters between diagnostic and relapsed samples appeared to be almost identical, while distinct shifts from diagnosis to relapse in the compositions have been observed in the other 3 out of 5 patients. We sought to further deconvolute the clonal architecture trajectory for T-ALL Dx_Rel pairs. We observed that dominant diagnostic clones of 4 patients diminished or vanished at relapse, sparing newly emerged subclones predominantly substituted at relapse. We clearly depicted two distinct patterns of evolutionary trajectories in these 4 Dx_Rel pairs by comprehensively mapping hierarchical TCR clonotypes onto leukemic clonotypes at single cell levels. Specifically, in T956 and T723, we observed significant outgrowth of incidental diagnostic sub-clones at relapse, whereby surrogate TCR repertoires correspondingly enumerated, suggestive of dynamic shifts in dominant clone over continuous chemo-exposure. Whereas in T593 and T856, expanding clones at relapse were showed up with completely different gene signatures from the diagnostic ones, but dominant clones at diagnosis and relapse were surprisingly presented with identical TCR repertoires. This was undoubtedly informative of leukemic “clonal drift” within which hypothetical intrinsic transformation happened to the same subclones over persistent chemotherapy. Image:Summary/Conclusion: Collectively, our presented study accurately distinguished leukemic cells from normal T cells in T-ALL at a single-cell resolution. By tracking transcriptomic profiles within and across Dx_Rel T-ALL pairs, we further identified distinct clonal evolutionary patterns, which may determine diversified fates of leukemic clones in response to therapeutic pressures, extending significant implications for future precise chemotherapies.

Rise of the Clones: Myelodysplastic Syndrome to Secondary Acute Myeloid Leukemia.

Myelodysplastic syndrome (MDS) describes a family of blood disorders driven by the clonal expansion of mutated blood cells that can evolve into secondary acute myeloid leukemia (sAML). Two new studies use single-cell and deep sequencing to elucidate the progression of MDS to AML, revealing discrete clonal architectures and the driving role of signaling mutations.

Abstract 3625: Clonal architecture and evolution of treatment-resistant prostate cancer via deep whole-genome ctDNA sequencing

Abstract Background: Circulating tumor DNA (ctDNA) in blood plasma is an emerging tool for clinical cancer genotyping and longitudinal disease monitoring. However, integration of ctDNA tests into clinical management is critically impeded by the poor understanding of the distinct somatic populations comprising bulk ctDNA—including their relationship to synchronous metastatic tissue, and their temporal dynamics during standard-of-care treatment. Prior approaches relying on targeted and/or low-resolution techniques (e.g. targeted exon sequencing, low-pass (shallow) whole-genome sequencing; WGS) do not permit comprehensive dissection of clonal architecture and unbiased analysis of putative resistance mechanisms. Methods: We performed deep WGS on serial plasma ctDNA (median depth: 185×) and synchronous metastatic tissue biopsies with high tumor purity from 35 patients with metastatic castration-resistant prostate cancer. We developed a subclonal reconstruction algorithm optimized for our data enabling resolution of per-patient evolutionary histories and ctDNA clonal composition. ctDNA nucleosome footprinting was used to infer mRNA abundance in synchronously biopsied metastases and androgen receptor (AR) transcription factor activity at 3224 AR binding sites (ARBS). Results: We comprehensively assess all classes of genomic alterations and demonstrate that ctDNA harbors greater populational heterogeneity than metastatic tissue (p<0.001). The evolutionary histories of ctDNA populations indicate frequent whole-genome doubling and attenuation of C>T aging-associated mutation signature during subclonal differentiation. Although driver alterations were largely concordant between tissue and ctDNA, each individual metastasis contributed only a minor share of total ctDNA (average ctDNA contribution: 17%). By comparing serial ctDNA before and after clinical progression on potent AR pathway inhibitors, we reveal population restructuring converging solely on AR copy augmentation as the dominant genomic driver of acquired treatment-resistance. Nucleosome depletion at transcription start-sites is highly correlated with same-patient metastatic tissue mRNA abundance, indicating that ctDNA fragmentomics can recapitulate transcriptomic patterns in metastatic lesions. Most ctDNA samples exhibited strong ARBS nucleosome depletion which correlated with AR gene copy number (R=0.36, p=0.003). Finally, serial ctDNA nucleosome profiling at ARBS revealed adaptive transcriptomic resistance to AR pathway inhibitors, including lineage switch to a neuroendocrine-like (AR-low) state. Conclusions: We show that the populations comprising ctDNA are typically complex and more heterogeneous than those found in bulk WGS of a synchronous metastasis. Our work advocates for liquid biopsy as a comprehensive multi-omic discovery tool for cancers with high ctDNA fractions. Citation Format: Cameron Herberts, Matti Annala, Joonatan Sipola, Sarah W. Ng, Xinyi E. Chen, Anssi Nurminen, Olga Korhonen, Aslı D. Munzur, Kevin Beja, Elena Schönlau, Cecily Q. Bernales, Elie Ritch, Jack V. Bacon, Nathan A. Lack, Matti Nykter, Rahul Aggarwal, Eric J. Small, Martin E. Gleave, David A. Quigley, Felix Y. Feng, Kim N. Chi, Alexander W. Wyatt. Clonal architecture and evolution of treatment-resistant prostate cancer via deep whole-genome ctDNA sequencing [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 3625.

Abstract 5232: Next-generation sequencing revealed clonal architecture and faciliated diagnosis of patients with multiple lung tumors: An analysis of 12 cases

Abstract Diagnosing multiple lung tumors as multiple primary lung cancer (MPLC) or intrapulmonary metastasis (IPM) has important therapeutic implications. The clinical applicability of next-generation sequencing (NGS) has been explored, although there is currently no consensus regarding how to integrate its findings for cancer staging. We performed a retrospective analysis of the mutational landscape of 13 patients with multiple tumors and underwent surgery at Renmin Hospital of Wuhan University from Sep. 2018 to Dec. 2020.The cohort consisted mostly of women (11/13, 84.6%) and non-smokers (12/13, 92.3%). Median age was 58 years (range 26-71). No patient manifested relapse after a median follow-up duration of 11.2 months (range 8.2-36.0). NGS was performed with a large panel (≥ 350 genes) in most cases (11/13, 84.6%). Most nodules were invasive adenocarcinomas (25/41, 61.0%), followed by adenocarcinomas in situ (13/41, 31.7%), and minimally invasive adenocarcinomas (3/41, 7.3%). Four patients were clinically diagnosed with IPM, of whom 1 had only one nodule sequenced. Of the 12 patients with genomic profiles from multiple nodules, 10 (83.3%) had consistent NGS inference and clinical diagnosis of MPLC or IPM. Of the remaining two, P1 was clinically diagnosed with IPM but NGS did not detect any genomic aberrations from one nodule, which may have missed evidence that could have led to inference of clonal relationship. P8 was diagnosed with MPLC but had two nodules that harbored EGFR L858R mutation. NGS revealed clonal relationships that enabled high-confidence differential diagnosis in the concordant cases. For instance, the 3 nodules in P10 each harbored an oncogenic driver: EGFR exon 19 deletion, EGFR L858R, and ALK fusion, which were suggestive of tumor clones of separate origins. Similar mutual exclusion was also observed in P6 (EGFR exon 19 deletion vs L858R), P2 (EGFR exon 19 deletion vs MAP2K1), and P7 (2 EGFR exon 18 mutations G719A and E709 in one nodule vs EGFR E746_A750del in another). On the other hand, P9 illustrates a case of complex clonal architecture, in which 8 nodules were derived from 5 origins. Further analysis found that alterations in KRAS were associated with smaller nodule size, whereas those in TP53, CDKN2A, NKX2-1, or GNAS were associated with greater size.In summary, NGS achieved highly concordant MPLC/IPM differentiation with clinical diagnosis based on histopathologic and radiographic evidence. However, caution is warranted when NGS detects no aberration from any of the multiple sequenced nodules. Examination of the mutational landscape found that altered proto-oncogene KRAS was associated with smaller tumors whereas altered tumor suppressors TP53 and CDKN2A associated with smaller ones, suggesting different roles in these molecules play in tumor growth. Citation Format: Wenyang Jiang, Wei Wang, Luting Qiu, Xiao Zou, Zhiqiu Chen, Qing Geng. Next-generation sequencing revealed clonal architecture and faciliated diagnosis of patients with multiple lung tumors: An analysis of 12 cases [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 5232.

Abstract 6171: A single-cell solution for solid tumors to detect mutations and quantify copy number variations

Abstract Cancer is a highly heterogeneous disease characterized by the presence of chromosomal structural copy number variation (CNV) and driver gene mutation events, which appear somatically at the early stages of oncogenesis and can drive tumor progression. Previously, we have developed a high throughput single cell DNA analysis platform that leverages droplet microfluidics and a multiplex-PCR based targeted DNA sequencing approach. It demonstrates high sensitivity detection of SNVs and indels in the same cells and generation of high resolution maps of clonal architecture based on mutational profiling. Here we expand upon the same approach and create an algorithm to accurately quantify CNVs along with genetic mutations from the same single cells. To estimate the CNV the amplicon read counts were corrected for artefacts related to cell to cell variability due sequencing efficiencies along with amplicon to amplicon variability that occur due to amplification efficiency differences. We used a spiked in diploid cell line as a reference to normalize all the cells. The spiked cell line was identified using its mutational profile. We validated the system using cell lines with known CNVs and found that we could at least accurately estimate copy numbers ranging from 0 through 5. Some of the regions were also validated orthogonally using droplet digital PCR. Another set of validation was performed using well characterized cell lines from NIST. We enable multiple visualizations of the copy number estimates in karyotype plots and line plots projected on SNV clones. With improved biochemistry, panel design and novel data analysis algorithms we develop a complete solution to detect amplification or loss of function in oncogenes and tumor suppressors reliably. Integration of CNVs and SNVs facilitates more accurate reconstruction of tumor evolution to better understand cancer progression mechanisms as well for quality control of gene edited cells, to further advance cancer research and therapy. Citation Format: Saurabh Parikh, Khushali Patel, Alex Li, Anup Parikh. A single-cell solution for solid tumors to detect mutations and quantify copy number variations [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 6171.

Abstract 2202: Contiguous airway sampling for basal cell isolation

Abstract Introduction: Airway basal cells isolated from histologically normal epithelium in current and former smokers demonstrate a significantly higher mutation burden than those in never-smokers. However, a proportion of basal cells with lower (“near normal”) mutation burden are observed, which show preferential expansion in response to smoking cessation(1). We believe this expansion represents a “cancer protective” process, reducing the risk of future malignancy. However, mechanisms determining the dynamic balance of high versus low mutant basal cells throughout the bronchial tree are unclear. We have developed protocols that enable systematic scanning of airway epithelium in resected lung specimens, through which we aim to elucidate the mutation burden and spatial clonal architecture across contiguous regions of normal epithelium through the bronchial tree. Study design: 5-10 current or former smokers undergoing curative lobectomy for peripheral lung cancer with histologically normal airways will be consented to the study. To date, two participants have donated their samples. Fresh lobectomies were collected at the point of resection and kept on wet ice until pathological examination. All samples undergo examination within 1 hour of resection. The lobe is orientated, and the tumour and segmental bronchi identified. A sterile metal cannula is inserted through the airway lumen and sterile pathology scissors are used to carefully isolate the airways from the surrounding parenchyma. Upon macro-dissection, airway branches are immediately placed into transport media containing antibiotics. Samples are gently washed using transport media to remove residual blood, then sectioned into 1 cm3 segments, which are orientated, embedded and frozen in room-temperature Optimal Cutting Temperature compound. An 8 μM section is cut and stained with haematoxylin and eosin from each block immediately after a “full-face” tissue confirmation is observed.These will be used as guiding slides for laser capture microdissection of the regions of epithelium prior to downstream sequencing pathway. Conclusion: We have developed a pathological approach in which we can define somatic mutations in normal epithelium along the bronchial tree. We will use these data to determine basal cell clonal relationships and interpret these in a spatial context, creating a more detailed map of the smoking-induced changes that occur in the airway epithelium and how these may contribute to early carcinogenesis.

Abstract 1692: Single-cell DNA sequencing from frozen endometrial tumors to address clonal evolution of somatic mutations

Abstract Background: Single-cell DNA sequencing with droplet microfluidics has enabled the characterization of genetic heterogeneity in hematologic malignancies and fresh solid tumors through the direct DNA sequencing of thousands of single cells from individual samples without the need of whole-genome amplification. The application of this technology in banked solid tumor samples has been rather limited to date, however. Here, we describe and validate a method to characterize clonal evolution of somatic mutations from frozen tumor specimens, utilizing endometrial carcinomas (ECs) as an example. Methods: Tissue cores were obtained from frozen ECs and the extracellular matrix was enzymatically digested. A subset of extracted nuclei were used for bulk DNA extraction and targeted sequencing. The remaining nuclei were encapsulated with cell lysis buffer, and the DNA was primed and barcoded with hydrogel beads containing 317 cancer gene amplicons using a Tapestri instrument (Mission Bio). Following droplet PCR, Illumina libraries were generated and sequenced on a NovaSeq6000. Sequencing reads were trimmed and mapped to the reference genome. Barcode sequences were error-corrected and cells were detected from barcodes based on number of reads assigned to each barcode and amplicon read completeness. Genotyping was performed using GATK and the allele drop-out (ADO) rate was calculated. Non-neoplastic cells were defined by the absence of somatic mutations. Allele-specific copy number (ASCN) profiles for each tumor cell were estimated by normalizing mean reads per cell per amplicon to reads of non-neoplastic cells. Results: We performed single-cell genotyping and ASCN profiling on two ECs with matched bulk sequencing data. For EC1, 5,004 tumor and 132 non-neoplastic cells were sequenced, for a 97% tumor purity. For EC2, 2,981 tumor and 165 non-neoplastic cells were obtained, for a 95% tumor purity. The average number of reads per amplicon per cell for EC1 and EC2 was 116 and 146 with an ADO rate of 6% and 8%, respectively. The data obtained from single-cell analysis were concordant with the matched bulk sequencing data in terms of clonal and subclonal mutations, and their respective cancer cell fraction. Single-cell sequencing further revealed the presence of loss of heterozygosity and emergence of minor subclones as late events in tumor evolution, which could not be inferred by bulk sequencing. In EC2, branch mutations displayed a spectrum consistent with microsatellite instability (MSI), reflecting its acquired MSI phenotype. Conclusions: Here we describe a robust method to perform single-cell targeted DNA sequencing with droplet microfluidics of individual nuclei isolated from frozen tissue samples, providing the opportunity to determine the repertoire of somatic mutations at single cell resolution and to infer cancer clonal architecture and dynamics utilizing banked solid cancer tissue samples. Citation Format: Arnaud Da Cruz Paula, Yingjie Zhu, Shirin Issa Bhaloo, Fresia Pareja, Timothy Hoang, Pier Selenica, Etta Hanlon, Thais Basili, Higinio Dopeso, Nadeem Abu-Rustum, Jorge Sergio Reis-Filho, Britta Weigelt. Single-cell DNA sequencing from frozen endometrial tumors to address clonal evolution of somatic mutations [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1692.

Abstract 2731: Longitudinally modeling anti-neoantigen immunity under paradigms of immunodominance

Abstract Neoantigen clonal architecture is known to influence the efficacy of immunotherapy. The immunodominance of clonal neoantigens over a canvas of heterogeneous subclonal neoantigens is particularly challenging to recapitulate in experimental models. We developed a computational model of neoantigen evolution that longitudinally tracks the “presentome” - that is, a cell’s aggregate neoantigen repertoire and HLA status - of simulated non-small cell lung cancer (NSCLC) prior to and during treatment with checkpoint blockade. Using neoantigen immunogenicity parameters from published reports, we first recapitulated the presentome of early-stage NSCLC. Upon simulating checkpoint blockade treatment of advanced NSCLC, we found salient correlates and consequences of response, including a profound clonal collapse among completely responding tumors. Our work could be adapted to longitudinally model the neoantigen architecture of other cancers and modalities of immunotherapy, ultimately informing cognizant strategies for future therapeutic development. Citation Format: Tim Qi, Benjamin Vincent, Yanguang Cao. Longitudinally modeling anti-neoantigen immunity under paradigms of immunodominance [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2731.

Abstract 1700: Serial monitoring of single-cell circulating tumor cell genomics in metastatic lobular breast cancer to identify precision and immuno-oncology biomarkers with therapeutic implications

Abstract Clinical decisions for precision and immuno-oncology therapies are based on predictive biomarkers commonly obtained from a single metastatic biopsy, or from archived primary tumor material. Circulating genomic biomarkers present a minimally invasive way to monitor the intra-patient tumor heterogeneity and its fluctuations in order to provide a real-time evaluation of the changing clonal architecture with potential therapeutic implications. Single-cell DNA next generation sequencing (scNGS) of circulating tumor cells (CTC) is a particularly well-suited method of unraveling and monitoring that heterogeneity to complement biomarker information obtained from tissue and cell-free circulating tumor DNA (ctDNA). In this proof-of-concept study we analyzed 123 CTC, 15 white blood cells (WBC), and ctDNA from 15 CTC-positive lobular breast cancer patients, five of whom had CTC available at both metastatic baseline and after progression on a variety of therapies chosen at their physician’s discretion. CTC were enriched with the CellSearch® system and isolated as single cells with the DEPArray™ system. Whole genome amplified CTC DNA underwent scNGS with the Oncomine Comprehensive Assay covering ~500 genes and 1.1Mb of genomic space to detect mutations, copy number alterations, tumor mutation burden (TMB) and microsatellite instability (MSI). 99.1% of cells were informative, with a mean sequencing depth of 664x. Using our previously developed, CTC-based precision medicine reporting platform, MI-CTCSeq, multiple CTC in seven of 15 patients (47%) had mutations that were actionable by FDA-approved targeted therapies including in the oncogenes PIK3CA (alpelisib) and FGFR2 (erdafitinib). 13 patients (87%) displayed intra-patient, inter-CTC genomic heterogeneity of putative driver mutations. Two of five (40%) patients with CTC at both baseline and progression displayed fluctuations in their CTC subclonal makeup between timepoints. One of the two harbored a baseline ESR1 (estrogen receptor α) p.D538G activating mutation that largely disappeared at progression and was replaced by a CTC subclone with a different ESR1 activating mutation, p.Y537S. Intriguingly, this patient’s CTC also harbored an FGFR2 p.K659M mutation in an actionable “hotspot” at progression, which was absent at baseline, suggesting potential utility of serial monitoring by CTC scNGS. TMB scores and MSI status in CTC were highly concordant with those measured in clinical tissue biopsies. Taken together, these data suggest the non-invasive interrogation of the CTC genomic landscape and its serial monitoring to inform precision and immuno-oncology treatments in real time. Citation Format: Andi K. Cani, Emily M. Dolce, Elizabeth P. Darga, Kevin Hu, Chia-Jen Liu, James M. Rae, Daffyd G. Thomas, Scott A. Tomlins, Arul M. Chinnaiyan, Aaron M. Udager, Costanza Paoletti, Erin F. Cobain, Daniel F. Hayes. Serial monitoring of single-cell circulating tumor cell genomics in metastatic lobular breast cancer to identify precision and immuno-oncology biomarkers with therapeutic implications [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1700.

Abstract 476: Predicting response to treatment in early breast cancer using dynamic integrative multi-omic profiling

Abstract Aim: We recently published the first machine learning framework that integrates multi-omic data derived from the pre-therapy breast tumor ecosystem to accurately predict response to neoadjuvant systemic therapy (Sammut et al, Nature 2021). We aimed to extend this framework to incorporate serially acquired multi-omic data to further improve response predictions and model tumor biology as it is perturbed by treatment. Methods: Breast tumor core biopsies were acquired at diagnosis from 168 women that went on to receive pre-operative chemotherapy (or chemotherapy plus anti-HER2 targeted therapy). Serial tumor core biopsies were obtained midway (n=78) and on completion (n=69) of neoadjuvant systemic therapy. Response was assessed at surgery using the Residual Cancer Burden score. Core biopsies were molecularly profiled by shallow whole genome, exome and RNA sequencing and their histological architecture characterized using digital pathology. Baseline clinical, molecular, and digital pathology imaging features associated with response were identified and their dynamics modelled throughout therapy. Results: At baseline, a total of 34 features derived from multi-omic data were associated with response to neoadjuvant therapy. These included: tumor mutation and neoantigen burden, subclonal diversity, HRD and chromosomal instability. A suppressed immune response, typified by the presence of T-cell dysfunction and immune exclusion, was associated with extensive chemoresistance. The changes in abundance of these features across the serially sampled on-therapy tumors were then mapped to response outcomes. An early increase in adaptive and innate immune infiltration and activation was associated with a linear decrease in expressed neoantigens, tumor proliferation and loss of subclonal mutational and copy number diversity, indicating early response to therapy. Conversely, a stable tumor and microenvironment transcriptional landscape throughout treatment, corresponding to a minimal change in tumor clonal architecture and microenvironment composition, was associated with a poor response to therapy. Notably, tumors with LOH HLA failed to engage a productive immune response during treatment and this was associated with resistance. A dynamic framework that models the change of these on-therapy features and extends the functionality of the published static machine learning model is being developed. Conclusion: Response to neoadjuvant therapy is determined by the baseline characteristics of the tumor ecosystem. During therapy, both the tumor and its microenvironment follow distinct evolutionary trajectories that can be mapped to outcome. The change in enrichment of features derived from the tumor and its microenvironment can be integrated within machine learning frameworks that leverage dynamic data from the entire therapy time course for more accurate response prediction. Citation Format: Stephen-John Sammut, Mireia Crispin-Ortuzar, Suet-Feung Chin, Elena Provenzano, Wei Cope, Ali Dariush, Sarah-Jane Dawson, Paul D. Pharoah, Florian Markowetz, Oscar M. Rueda, Helena M. Earl, Carlos Caldas. Predicting response to treatment in early breast cancer using dynamic integrative multi-omic profiling [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 476.

HGG-42. Evolutionary selection of key oncogenic alterations in patient-derived models of paediatric diffuse high grade glioma (PDHGG) subtypesin vitro andin vivo

PDHGG are a diverse group of childhood brain tumours comprising multiple subgroups carrying distinct molecular drivers. Patient-derived models accurately recapitulating this underlying biology are critical for mechanistic/preclinical studies aimed at improving patient outcome, however their behaviour over time in the environments in which they are propagated, and how this relates to the human disease, is largely unknown. To explore this, we collected 94 models of PDHGG established as 2D/3D stem cell cultures in vitro, and generated patient-derived xenografts (PDX) in 33/62 specimens implanted orthotopically in vivo. We carried out exome/targeted sequencing, methylation profiling and RNAseq to profile cells through their first 25 passages in culture, and sequential implantation from p0-p2 in mice. In 15/83 cultures, we observed enrichment of gene expression signatures of non-malignant cells over the first 5 passages, with concurrent depletion of somatic mutations/CNAs, excluding them from further study. Validated models retained tumour-matched genotypes, CNAs and driver alterations including H3.3G34R, H3.3/H3.1K27M, BRAF and ACVR1 over time, however subclonal alterations underwent selection in culture which profoundly altered their response to targeted drug treatment. In 6/7 PDGFRA-mutant models, activating mutations were selected against between p5-20 in 2D and/or 3D, whilst MAPK pathway mutations in NF1/PIK3R1 similarly diverged over 15 passages under different growth conditions, resulting in isogenic models with differential signalling, in vivo tumorigenicity, and in vitro sensitivity to multiple MEK inhibitors. In PDXs, serial xenografting reduced the time to tumour formation by up to half, with a concomitant shift in clonal architecture. Multi-region sequencing of diffusely-infiltrating tumours showed selection for alterations such as PIK3CA/NF1 at distant sites, with evidence for convergent evolution of subclonal mutations, as in human tumours. Understanding the evolutionary dynamics of targetable/predictive alterations in PDHGG model systems is key to developing new and effective therapeutic interventions in this highly heterogenous disease.

Molecular parameters impacting the success rate of a lung cancer PDX model.

e20592 Background: Patient-derived xenograft (PDX) models can provide renewable cancer tissue resources, are being increasingly utilized for the molecular characterizations of cancers and preclinical studies on drug activities, and have the potential to advance biomarker identification. Previous research has demonstrated that PDX tumors preserve the histologic and morphologic characteristics and gene expression and mutation patterns of primary tumors, and correlate to patient survival. Tumor histological subtypes and differentiation grades are known to contribute to the success of PDX intake, but little is known about how molecular features associate with the success of PDX engraftment or whether PDX preserves subclonal architecture of primary tumors. Methods: To fill this void, we analyzed 252 primary tumor samples from the SPORE/ICON project at MD Anderson Cancer Center that were used for generating PDXs and had comprehensive genomic, transcriptomic, and immune profiling data available to evaluate the parameters impacting the intake success of lung cancer PDXs. In order to establish the impact of molecular features on PDX intake success we also assessed the fidelity of PDX in representing the molecular features of this cohort of primary tumors. Results: Only 36% of the primary tumors in this study successfully generated a PDX, with Squamous Cell Carcinoma PDX demonstrating an intake rate of 67% and Adenocarcinoma PDX demonstrating an intake rate of 21%. Genomic architecture derived from whole exome sequencing showed not only 43% (38-48%) of mutations shared, but also similar clonal architecture between primary tumors and PDX. Higher copy number aberration burden, high tumor purity and low immune infiltration in primary tumors were found to be associated with successful intake of PDX. Transcriptomics revealed 6,103 genes differentially expressed between primary tumors with successful PDX intake versus those which were unsuccessful. Pathway analysis of these genes indicated that inhibition of cell proliferation, signaling, and migration as well as DNA replication, recombination and repair were associated with PDX intake failure regardless of histology. Conclusions: These findings indicate that PDX intake success is regulated by molecular features and multiomic evaluation of lung cancer primary tumors can be used as a marker for determining which samples to use for PDX model generation.