Search

Exploring non-genetic evolution of cell states during cancer treatments has become attainable by recent advances in lineage-tracing methods coupling cell states to future fates. However, transcriptional changes that drive pre-treatment cells into resistant fates may be subtle, necessitating high resolution analysis. We developed ReSisTrace that uses shared transcriptomic features of synchronised sister cells to predict the states that prime treatment resistance, and allows identification of asymmetric features that drive phenotypic heterogeneity. Applying ReSisTrace in ovarian cancer cells revealed that BRCAness transcriptional signatures are associated with pre-existing vulnerability not only to olaparib and carboplatin treatments, but also to natural killer cell-mediated cytotoxicity. This novel connection between DNA repair defect and susceptibility to natural killer cells was further validated both functionally and in a clinical cohort. The high-resolution analysis by ReSisTrace enables resolving pre-existing transcriptional features of treatment vulnerability, facilitating molecular patient stratification for personalised therapies.

Motivation A major challenge in analyzing cancer patient transcriptomes is that the tumors are inherently heterogeneous and evolving. We analyzed 214 bulk RNA samples of a longitudinal, prospective ovarian cancer cohort and found that the sample composition changes systematically due to chemotherapy and between the anatomical sites, preventing direct comparison of treatment-naive and treated samples. Results To overcome this, we developed PRISM, a latent statistical framework to simultaneously extract the sample composition and cell-type-specific whole-transcriptome profiles adapted to each individual sample. Our results indicate that the PRISM-derived composition-free transcriptomic profiles and signatures derived from them predict the patient response better than the composite raw bulk data. We validated our findings in independent ovarian cancer and melanoma cohorts, and verified that PRISM accurately estimates the composition and cell-type-specific expression through whole-genome sequencing and RNA in situ hybridization experiments. Availabilityand implementation https://bitbucket.org/anthakki/prism. Supplementary information Supplementary data are available at Bioinformatics online.

High-grade serous ovarian cancer (HGSC) is the most prevalent and lethal ovarian cancer type. While HGSC usually responds well to primary treatment, most cases eventually relapse. Functional precision medicine - tailoring individualized treatments based on functional in vitro assays on patient-derived cells - has been recently employed in cancer clinical trials. Cancer organoids - three-dimensional, self-organizing, self-renewing cell cultures that recapitulate original tissue structure and function - have been applied as cellular models in these trials. However, in case of HGSC, organoid derivation has proven time consuming and inefficient, hindering their application in functional precision medicine due to a short time window, in which therapy for each patient needs to be selected. To address this problem, we aimed to establish whether drug vulnerabilities at HGSC relapse could be predicted using organoids derived from the primary disease cells. We derived sequential organoid models from material sampled during primary treatment and at relapse. Then, for organoid pairs (primary-relapse), we performed large-scale drug response profiling of a library of 370 compounds (approved drugs or drugs in clinical development), in 384-well microplate format, alone or in combination with a standard HGSC chemotherapeutic agent carboplatin. First, we found that HGSC organoid responses to standard chemotherapeutics retrospectively correlated to observed clinical treatment outcomes. But further, for each patient we identified compounds with pronounced cytotoxicity both in the primary and in the relapsed model, amounting to 66% of all hits (7% were primary-specific and 27% relapse-specific). We then focused on identifying patient-specific hits rather than compounds displaying general toxicity in all patient models. Based on a potential clinical applicability, for three patients we selected compounds for validation in organoid outgrowth assay, with prolonged (>1 month) drug-free period post-treatment. In two patients, AZD4573, a selective CDK9 inhibitor in clinical development for hematological malignancies, at nanomolar concentrations caused eradication of organoids when combined with carboplatin. Organoids from the third patient were vulnerable to nitazoxanide, an approved anti-helminthic agent and an inhibitor of mitochondrial oxidative phosphorylation. Importantly, the selected final hits were identified solely based on screening in organoid models from primary disease. In summary, we here demonstrate that HGSC organoids derived from primary disease material predict a majority of patient-specific drug vulnerabilities of organoids derived from the relapsed HGSC lesions. This indicates that patient stratification in functional precision medicine for treatment of HGSC relapse could be prospectively performed at the primary disease stage. Citation Format: Wojciech Senkowski, Laura Gall-Mas, Matias M. Falco, Yilin Li, Kari Lavikka, Mette C. Kriegbaum, Jaana Oikkonen, Daria Bulanova, Elin J. Pietras, Karolin Voßgröne, Yan-Jun Chen, Erdogan P. Erkan, Mia K. Høg, Ida M. Larsen, Tarja Lamminen, Katja Kaipio, Jutta Huvila, Anni Virtanen, Lars H. Engelholm, Pernille Christiansen, Eric Santoni Rugiu, Kaisa Huhtinen, Olli Carpén, Johanna Hynninen, Sampsa Hautaniemi, Anna Vähärautio, Krister Wennerberg. A platform utilizing high-grade serous ovarian cancer organoids for prospective patient stratification in functional precision medicine. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5779.

The transcription factor Myc drives cell growth across animal phyla and is activated in most forms of human cancer. However, it is unclear which Myc target genes need to be regulated to induce growth and whether multiple targets act additively or if induction of each target is individually necessary. Here, we identified Myc target genes whose regulation is conserved between humans and flies and deleted Myc-binding sites (E-boxes) in the promoters of fourteen of these genes in Drosophila. E-box mutants of essential genes were homozygous viable, indicating that the E-boxes are not required for basal expression. Eight E-box mutations led to Myc-like phenotypes; the strongest mutant, ppan(Ebox-/-), also made the flies resistant to Myc-induced cell growth without affecting Myc-induced apoptosis. The ppan(Ebox-/-) flies are healthy and display only a minor developmental delay, suggesting that it may be possible to treat or prevent tumorigenesis by targeting individual downstream targets of Myc.

Chemotherapy resistance is a critical contributor to cancer mortality and thus an urgent unmet challenge in oncology. To characterize chemotherapy resistance processes in high-grade serous ovarian cancer, we prospectively collected tissue samples before and after chemotherapy and analyzed their transcriptomic profiles at a single-cell resolution. After removing patient-specific signals by a novel analysis approach, PRIMUS, we found a consistent increase in stress-associated cell state during chemotherapy, which was validated by RNA in situ hybridization and bulk RNA sequencing. The stress-associated state exists before chemotherapy, is subclonally enriched during the treatment, and associates with poor progression-free survival. Co-occurrence with an inflammatory cancer–associated fibroblast subtype in tumors implies that chemotherapy is associated with stress response in both cancer cells and stroma, driving a paracrine feed-forward loop. In summary, we have found a resistant state that integrates stromal signaling and subclonal evolution and offers targets to overcome chemotherapy resistance.

Each patient’s cancer consists of multiple cell subpopulations that are inherently heterogeneous and may develop differing phenotypes such as drug sensitivity or resistance. A personalized treatment regimen should therefore target multiple oncoproteins in the cancer cell populations that are driving the treatment resistance or disease progression in a given patient to provide maximal therapeutic effect, while avoiding severe co-inhibition of non-malignant cells that would lead to toxic side effects. To address the intra- and inter-tumoral heterogeneity when designing combinatorial treatment regimens for cancer patients, we have implemented a machine learning-based platform to guide identification of safe and effective combinatorial treatments that selectively inhibit cancer-related dysfunctions or resistance mechanisms in individual patients. In this case study, we show how the platform enables prediction of cancer-selective drug combinations for patients with high-grade serous ovarian cancer using single-cell imaging cytometry drug response assay, combined with genome-wide transcriptomic and genetic profiles. The platform makes use of drug-target interaction networks to prioritize those combinations that warrant further preclinical testing in scarce patient-derived primary cells. During the case study in ovarian cancer patients, we investigated (i) the relative performance of various ensemble learning algorithms for drug response prediction, (ii) the use of matched single-cell RNA-sequencing data to deconvolute cell population-specific transcriptome profiles from bulk RNA-seq data, (iii) and whether multi-patient or patient-specific predictive models lead to better predictive accuracy. The general platform and the comparison results are expected to become useful for future studies that use similar predictive approaches also in other cancer types.

Single-cell RNA-sequencing (scRNA-seq) has made it possible to profile gene expression in tissues at high resolution. An important preprocessing step prior to performing downstream analyses is to identify and remove cells with poor or degraded sample quality using quality control (QC) metrics. Two widely used QC metrics to identify a ‘low-quality’ cell are (i) if the cell includes a high proportion of reads that map to mitochondrial DNA (mtDNA) encoded genes and (ii) if a small number of genes are detected. Current best practices use these QC metrics independently with either arbitrary, uniform thresholds (e.g. 5%) or biological context-dependent (e.g. species) thresholds, and fail to jointly model these metrics in a data-driven manner. Current practices are often overly stringent and especially untenable on certain types of tissues, such as archived tumor tissues, or tissues associated with mitochondrial function, such as kidney tissue [1]. We propose a data-driven QC metric (miQC) that jointly models both the proportion of reads mapping to mtDNA genes and the number of detected genes with mixture models in a probabilistic framework to predict the low-quality cells in a given dataset. We demonstrate how our QC metric easily adapts to different types of single-cell datasets to remove low-quality cells while preserving high-quality cells that can be used for downstream analyses. Our software package is available at https://bioconductor.org/packages/miQC., Author summary We developed the miQC package to predict the low-quality cells in a given scRNA-seq dataset by jointly modeling both the proportion of reads mapping to mitochondrial DNA (mtDNA) genes and the number of detected genes using mixture models in a probabilistic framework. We demonstrate how our QC metric easily adapts to different types of single-cell datasets to remove low-quality cells while preserving high-quality cells that can be used for downstream analyses.

Extensive utilization of organoids from high-grade serous ovarian carcinoma (HGSOC), the most common and lethal ovarian cancer, has been hampered by low success rates of long-term culture and scarcity of fresh tumor material. Here we present the development of a novel method for efficient generation, expansion and use of HGSOC organoids from cryopreserved tumor material. First, we assessed commonly used organoid media components and found that supplements such as FGF-2, R-Spondin1, Wnt or Noggin had negative impact on the HGSOC organoid derivation. But further, we found that supplementation with FGF-4, which has not been used in cancer organoid culture before, is beneficial for HGSOC organoid growth. Through extensive testing of various supplements and their combinations, we designed two novel HGSOC organoid media formulations - Medium 1 (M1) and Medium 2 (M2). Using M1 and M2 enabled generation and long-term expansion of living HGSOC organoid biobank with markedly improved success rate than in previous reports (55% vs. 23-38%). The organoids were established from cryopreserved tumor material, demonstrating the feasibility of using frozen tissue biobanks for HGSOC organoid derivation. Overall, we generated a collection of 18 expandable HGSOC organoid lines from 11 patients, encompassing samples from different tissue sites and disease progression stages. We validated the organoids using whole-genome sequencing, immunohistochemistry and single-cell RNA sequencing and demonstrated that they are genetically and phenotypically representative of original patient samples over long-term culture. Based on available patient consents, we deposited 3 organoid lines in a publicly accessible biobank. Finally, we investigated whether organoid drug responses correlate to those observed earlier in the clinic in the corresponding patients. Organoid-based drug-response profiling of clinically used HGSOC drug collection was performed in 384-well microplate format. To explore whether growth conditions impact correlation between organoid drug responses and clinical response, we compared the organoid drug responses in the nutrient-rich M1/M2 growth media to the ones observed in human plasma-like medium (HPLM), supplemented with relevant niche factors from M1/M2. Organoid drug responses correlated with clinical treatment outcomes, but only for organoids maintained in HPLM (Spearman r = 0.987, p=0.007 in HPLM vs 0.607, p=0.167 in growth medium, n=7), highlighting the importance of culture conditions in organoid-based functional assays. Taken together, we introduce a resource for efficient development and use of HGSOC organoids from cryopreserved material in ovarian cancer research. Citation Format: Wojciech Senkowski, Laura Gall Mas, Matias M. Falco, Yilin Li, Kari Lavikka, Mette C. Kriegbaum, Jaana Oikkonen, Daria Bulanova, Elin J. Pietras, Karolin Voßgröne, Erdogan P. Erkan, Terese K. Høj, Mia K. Høg, Tarja Lamminen, Katja Kaipio, Anni Virtanen, Lars H. Engelholm, Pernille Christiansen, Eric Santoni-Rugiu, Kaisa Huhtinen, Olli Carpén, Johanna Hynninen, Sampsa Hautaniemi, Anna Vähärautio, Krister Wennerberg. Efficient establishment and utilization of a high-grade serous ovarian cancer organoid biobank [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 3069.

In this work, we map the transcriptional targets of 107 previously identified Drosophila genes whose loss caused the strongest cell-cycle phenotypes in a genome-wide RNA interference screen and mine the resulting data computationally. Besides confirming existing knowledge, the analysis revealed several regulatory systems, among which were two highly-specific and interconnected feedback circuits, one between the ribosome and the proteasome that controls overall protein homeostasis, and the other between the ribosome and Myc/Max that regulates the protein synthesis capacity of cells. We also identified a set of genes that alter the timing of mitosis without affecting gene expression, indicating that the cyclic transcriptional program that produces the components required for cell division can be partially uncoupled from the cell division process itself. These genes all have a function in a pathway that regulates the phosphorylation state of Cdk1. We provide evidence showing that this pathway is involved in regulation of cell size, indicating that a Cdk1-regulated cell size checkpoint exists in metazoans.

Chemotherapy resistance is the greatest contributor to cancer mortality and the most urgent unmet challenge in oncology. In order to reveal transcriptomics changes due to platinum-based chemotherapy we analyzed single-cell RNA-seq data from fresh tissue samples taken at the time of diagnosis and after neoadjuvant chemotherapy from 11 high-grade serous ovarian cancer (HGSOC) patients. With a novel clustering method accounting for patient-specific variability and technical confounders, we identified 12 clusters. Of these, a stress-related transcriptional profile was enriched during chemotherapy and associated significantly to poor progression-free survival (PFS) and disease-free interval (DFI) in deconvoluted bulk RNA-seq data analysis of treatment-naive samples in TCGA cohort. Pan-cancer cell line analysis suggests that patients with high stress-related transcriptional profile may benefit from MEK1/2 inhibitors instead of platinum. Further, high proportion of stromal components and high interaction score between tumor and stromal suggest the tumor cells with high-stress profile actively interact with and potentially recruit stromal cells to their microenvironment already prior to chemotherapy, potentially facilitating protection from chemotherapeutic treatments. In summary, we have identified a stress-related transcriptional profile, which is present at the time of diagnosis, enriched during platinum treatment, independent predictor for poor PFS and DFI, and, based on in vitro data, targetable with MEK1/2 inhibitors.Translational relevanceWe discovered a stress-related transcriptional profile that is significantly enriched in fresh tissue samples after chemotherapy and is significantly associated with poor progression-free survival in an independent patient cohort. The survival association is independent of age, tumor purity or BRCAness. Therefore, this chemotherapy resistance associated profile is intrinsic and could thus be targeted already in treatment-naive patients. The translation potential of the stress-related transcriptomics profile was further supported by pan-cancer cell line analysis that showed that cell lines with high stress-related transcriptional profile are not affected by platinum, corroborating our results, whereas they were more sensitive to MEK1/2 inhibitors. Taken together, the stress-related transcriptional profile, quantifiable with a set of 35 marker genes, provides a basis for improved prediction of platinum response as well as novel avenues to treat this patient group more effectively.

The transcription factor Myc is activated in most major forms of human cancer. Myc regulates a large set of target genes, and drives cell growth across animal phyla. However, it has not been clear which target genes are required for Myc-induced growth, and whether the targets are individually necessary or act in an additive fashion. Here, we have used comparative functional genomics to identify a core set of Myc target genes whose regulation is conserved between humans andDrosophila melanogaster. Most of these targets are essential genes involved in ribosome biogenesis and ribonucleotide metabolism. To identifyDrosophilagenes whose upregulation is necessary for Myc induced growth, we deleted the Myc binding sites (E-boxes) in the promoter regions of four genes using CRISPR/Cas9. All mutant flies were homozygous viable, indicating that E-box sequences are not required for basal expression of the Myc target genes. E-Box deletions in RpS20, RpS24 and Nop56 did not cause strong growth phenotypes. However, deletion of the E-box in the rRNA processing factorPeter Pan(ppan) made the flies resistant to Myc-induced cell growth, without affecting Myc-induced apoptosis. Despite their failure to respond to Myc, theppanEbox−/−flies are healthy and display only a minor developmental delay, suggesting that it may be possible to treat or prevent tumorigenesis by targeting individual downstream targets of Myc.

A major challenge in analyzing cancer patient transcriptomes is that the tumors are inherently heterogeneous and evolving. We analyzed 214 bulk RNA samples of a longitudinal, prospective ovarian cancer cohort and found that the sample composition changes systematically due to chemotherapy and between the anatomical sites, preventing direct comparison of treatment-naive and treated samples. To overcome this, we developed PRISM, a latent statistical framework to simultaneously extract the sample composition and cell type specific whole-transcriptome profiles adapted to each individual sample. Our results indicate that the PRISM-derived composition-free transcriptomic profiles and signatures derived from them predict the patient response better than the composite raw bulk data. We validated our findings in independent ovarian cancer and melanoma cohorts, and verified that PRISM accurately estimates the composition and cell type specific expression through whole-genome sequencing and RNA in situ hybridization experiments. PRISM is freely available with full source code and documentation.

Uterine leiomyomas are common benign smooth muscle tumors that impose a major burden on women's health. Recent sequencing studies have revealed recurrent and mutually exclusive mutations in leiomyomas, suggesting the involvement of molecularly distinct pathways. In this study, we explored transcriptional differences among leiomyomas harboring different genetic drivers, including high mobility group AT-hook 2 (HMGA2) rearrangements, mediator complex subunit 12 (MED12) mutations, biallelic inactivation of fumarate hydratase (FH), and collagen, type IV, alpha 5 and collagen, type IV, alpha 6 (COL4A5-COL4A6) deletions. We also explored the transcriptional consequences of 7q22, 22q, and 1p deletions, aiming to identify possible target genes. We investigated 94 leiomyomas and 60 corresponding myometrial tissues using exon arrays, whole genome sequencing, and SNP arrays. This integrative approach revealed subtype-specific expression changes in key driver pathways, including Wnt/β-catenin, Prolactin, and insulin-like growth factor (IGF)1 signaling. Leiomyomas with HMGA2 aberrations displayed highly significant up-regulation of the proto-oncogene pleomorphic adenoma gene 1 (PLAG1), suggesting that HMGA2 promotes tumorigenesis through PLAG1 activation. This was supported by the identification of genetic PLAG1 alterations resulting in expression signatures as seen in leiomyomas with HMGA2 aberrations. RAD51 paralog B (RAD51B), the preferential translocation partner of HMGA2, was up-regulated in MED12 mutant lesions, suggesting a role for this gene in the genesis of leiomyomas. FH-deficient leiomyomas were uniquely characterized by activation of nuclear factor erythroid 2-related factor 2 (NRF2) target genes, supporting the hypothesis that accumulation of fumarate leads to activation of the oncogenic transcription factor NRF2. This study emphasizes the need for molecular stratification in leiomyoma research and possibly in clinical practice as well. Further research is needed to determine whether the candidate biomarkers presented herein can provide guidance for managing the millions of patients affected by these lesions.

A key event in cancer progression is the establishment of an immunosuppressive environment that prevents the immune system from effectively attacking the cancer cells. In this study we aim to characterize the cellular determinants and the spatial features of immune suppressive mechanisms in high-grade serous ovarian cancer (HGSOC) via integrating tissue Cyclic Immunofluorescence (tCycIF), a novel highly multiplexed single-cell spatial proteomics platform with multi-omics data, including state-of-the-art single-cell RNA sequencing. We have conducted a comprehensive analysis of a unique dataset consisting of 23 HGSOC samples, including 7 pairs of treatment-naïve and post-chemotherapy samples, using tCycIF, single-cell, and bulk RNA sequencing coupled with novel deconvolution algorithms and whole-genome sequencing. All samples were prospectively collected from patients treated at Turku University Hospital, Finland, as a part of the HERCULES consortium. Our preliminary findings suggest that tCycIF can reliably capture tumor microenvironment composition in HGSOC. Overall, there was a good correlation with tumor purity between all the methods used. In tCycIF, scRNAseq and bulk deconvolution macrophages were the most common immune cell type in the HGSOC microenvironment. However, specific immune subpopulations, such as CD8+T-cells, had a larger proportion in tCycIF compared to single-cell RNA sequencing and bulk-deconvoluted RNAseq data. Further, the immune profiling based on RNAseq deconvolution algorithm showed highly variable results based on the algorithm used. We are in the process of using our complete dataset to compare state-of-the-art methods and improve algorithms for HGSOC immunophenotyping. tCycIF’s spatial proteomic data in single-cell resolution allows us to gain new insights into the tumor microenvironment and discover the previously latent cell-to-cell interactions that make up the immunosuppressive milieus in HGSOC tumors. The improvement of novel reliable single-cell immunophenotyping methods is critical to develop more effective immunotherapies to improve the outcomes and survival of patients with HGSOC. Citation Format: Miikka K. Kilkkila, Julia Casado, Connor Jacobson, Antti Hakkinen, Erdogan Erkai, Jun Dai, Kaiyang Zhan, Zoltan Maliga, Olli Carpen, Rainer Lehtonen, Sampsa Hautaniemi, Peter Sorger, Anna Vaharautio, Anniina Farkkila. Integrating highly-multiplexed imaging with multi-omics data to uncover immunologic vulnerabilities in high-grade serous ovarian cancer [abstract]. In: Proceedings of the AACR Special Conference on Advances in Ovarian Cancer Research; 2019 Sep 13-16, 2019; Atlanta, GA. Philadelphia (PA): AACR; Clin Cancer Res 2020;26(13_Suppl):Abstract nr B62.

Tumor heterogeneity in high-grade serous ovarian cancer (HGSOC) stems from highly variable copy number changes that lack shared actionable drug targets and contribute to inevitable recurrence and death. To unveil the vast heterogeneity in HGSOC, and understand the molecular changes that occur during chemotherapy, we analyzed single-cell transcriptomes of 49 dissociated HGSOC tumor or ascites specimens, along with six normal postmenopausal fimbria/tuba specimens as controls. This set contains 13 pairs of samples from patients who underwent neoadjuvant chemotherapy, taken before treatment (primary), and after three cycles of standard carboplatin-paclitaxel chemotherapy (interval). Clustering single-cell transcriptomes based on acknowledged cell type markers revealed epithelial cancer cells, as well as diverse stromal and immune cell types. The cell type proportions displayed high variability across HGSOC tumors, intriguingly with lowest tumor purity in patients with short platinum-free interval. To identify the impact of chemotherapy on transcriptional programs, we extracted shared transcriptomic differences in the cancer cells of seven primary-interval pairs. Regardless of the clinical response, three cycles of chemotherapy induced an oxidative stress response program that inhibits apoptosis, while inducing genes involved in MHC II antigen presentation to T helper cells. Next, to study how cell states change during chemotherapy, we ordered the cells in pseudotime based on their transcriptomic similarity, revealing a mixture of primary and interval cells in all but two primary-interval pairs. Furthermore, based on inferred copy number analysis, we estimate that genomic differences contribute variably, between 20%, to the overall transcriptomic differences between the primary and interval pairs. These results suggest that pre- and post-chemotherapy states are not completely separate and that genomic evolution has a highly variable effect on the state differences, providing support for an intrinsic chemoresistance model with genomic and nongenomic components. To further the understanding of intrinsic resistance, we next focused on patients’ clinical response mechanisms in HGSOC cancer cells. Genes encoding well-known serum markers of HGSOC, WFDC2 and MUC16, were expressed at higher levels in primary specimens of patients with incomplete response, whereas patients with complete response displayed higher expression of genes related to chromatin repression/heterochromatin. Finally, we focused on tumor resectability to understand the transcriptomic patterns underlying HGSOC metastasis. Already prior to chemotherapy, patients with residual disease showed induced FOXO3-driven regulon linked to quiescence, stress resistance, and stemness, whereas patients with no residual disease express increased level of E2F1/DP1-driven regulon linked to BRCAness, high proliferation, and high metabolic rate. Citation Format: Erdogan Pekcan Erkan, Jun Dai, Kaiyang Zhang, Katja Kaipio, Tarja Lamminen, Naziha Mansuri, Lasse Suominen, Amjad Alkodsi, Kaisa Huhtinen, Sakari Hietanen, Johanna Hynninen, Seija Grénman, Olli Carpén, Sampsa Hautaniemi, Anna Vähärautio. Singling out tumor heterogeneity and chemoresistance in high-grade serous ovarian cancer [abstract]. In: Proceedings of the AACR Special Conference on Advances in Ovarian Cancer Research; 2019 Sep 13-16, 2019; Atlanta, GA. Philadelphia (PA): AACR; Clin Cancer Res 2020;26(13_Suppl):Abstract nr A40.

Introduction: To identify genes and pathways that drive platinum-taxane resistance in high-grade serous ovarian cancer (HGSOC), we analyzed 108 samples subjected to RNA-seq before and after chemotherapy. Experimental Procedure: We collected prospectively 108 tissue samples from various anatomic locations (omentum, ovary, mesothelium, peritoneum) for 53 HGSOC patients with high enough tumor purity (>25%) and follow-up information to calculate platinum free interval (PFI). These samples were subjected to RNA-seq to obtain gene expression data for known protein coding genes. Additionally, data from 16,826 samples (cells) with single-cell RNA-seq were used in the decomposition analysis to overcome cell heterogeneity in the samples. For validation we used 306 The Cancer Genome Atlas (TCGA) HGSOC samples that were analyzed with the decompose algorithm. Differential expressions for selected resistance associated genes were validated with immunohistochemistry (IHC). Results: We examined firstly whether gene expressions values are systematically influenced by anatomic location of the samples and chemotherapy. Our results show that bias due to anatomic locations can be controlled by relatively simple computational adjusting. However, chemotherapy-treated samples have significantly lower tumor purity than treatment-naive samples and samples have varying cell compositions. We developed a computational decomposition method to correct these biases. Secondly, we compared gene expression values of cancer cells between 1) well and poor responding patients, and 2) before and after chemotherapy followed by pathway analysis. We identified several genes that were significantly upregulated in the poor-responder group, such as FOXK1 and FBXO32, which also had significant survival association in the TCGA cohort. Our results highlight genes whose expression values are significantly enriched after chemotherapy, such as BTG2, CITED2, CTGF, FOS, DUSP1, and EGR1, which have been shown to promote resistance to platinum or paclitaxel in HGSOC or other cancers. In addition to these known chemoresistant-associated genes, we identified several hitherto unknown genes and validated them with IHC. Pathway analysis revealed actionable targets in the B-cell receptor and MAPK signaling pathways, which were the most significant enriched signaling pathways after chemotherapy. Conclusions: We have shown that successful comparison of gene expression values before and after chemotherapy requires heavy computational correction of the strong and systematic bias caused by chemotherapy and tumor heterogeneity. To overcome these biases, we have developed a novel decomposition algorithm. This method allows to obtain reliable data for the subsequent analyses. The gene expression analyses herein identified several genes and pathways that play central roles in chemoresistance and are rational targets for overcoming platinum-taxane resistance in HGSOC. Citation Format: Sanaz Jamalzadeh, Kaiyang Zhang, Antti Häkkinen, Kaisa Huhtinen, Johanna Hynninen, Noora Andersson, Naziha Mansuri, Olli Carpén, Erkan Erdogan, Jun Dai, Anna Vähärautio, Sakari Hietanen, Jaana Oikkonen, Rainer Lehtonen, Sampsa Hautaniemi. RNA-seq analysis of high-grade serous ovarian cancer patients before and after chemotherapy reveals chemoresistance-associated genes and pathways [abstract]. In: Proceedings of the AACR Special Conference on Advances in Ovarian Cancer Research; 2019 Sep 13-16, 2019; Atlanta, GA. Philadelphia (PA): AACR; Clin Cancer Res 2020;26(13_Suppl):Abstract nr A59.

Point mutations in cancer have been extensively studied but chromosomal gains and losses have been more challenging to interpret due to their unspecific nature. Here we examine high-resolution allelic imbalance (AI) landscape in 1699 colorectal cancers, 256 of which have been whole-genome sequenced (WGSed). The imbalances pinpoint 38 genes as plausible AI targets based on previous knowledge. Unbiased CRISPR-Cas9 knockout and activation screens identified in total 79 genes within AI peaks regulating cell growth. Genetic and functional data implicate loss of TP53 as a sufficient driver of AI. The WGS highlights an influence of copy number aberrations on the rate of detected somatic point mutations. Importantly, the data reveal several associations between AI target genes, suggesting a role for a network of lineage-determining transcription factors in colorectal tumorigenesis. Overall, the results unravel the contribution of AI in colorectal cancer and provide a plausible explanation why so few genes are commonly affected by point mutations in cancers., In this study the authors examine the allelic imbalance (AI) landscape of colorectal cancer, reporting loss of TP53 as a driver of AI. They use CRISPR-Cas9 screens to identify 79 genes (within AI regions) regulating cell growth and identify a network of transcription factors that may contribute to colorectal tumorigenesis.

Purpose: Homologous recombination deficiency (HRD) correlates with platinum sensitivity in patients with ovarian cancer, which clinically is the most useful predictor of sensitivity to PARPi. To date, there are no reliable diagnostic tools to anticipate response to platinum-based chemotherapy, thus we aimed to develop an ex vivo functional HRD detection test that could predict both platinum-sensitivity and patient eligibility to targeted drug treatments. Experimental Design: We obtained a functional HR score by quantifying homologous recombination (HR) repair after ionizing radiation-induced DNA damage in primary ovarian cancer samples (n = 32). Samples clustered in 3 categories: HR-deficient, HR-low, and HR-proficient. We analyzed the HR score association with platinum sensitivity and treatment response, platinum-free interval (PFI) and overall survival (OS), and compared it with other clinical parameters. In parallel, we performed DNA-sequencing of HR genes to assess if functional HRD can be predicted by currently offered genetic screening. Results: Low HR scores predicted primary platinum sensitivity with high statistical significance (P = 0.0103), associated with longer PFI (HR-deficient vs. HR-proficient: 531 vs. 53 days), and significantly correlated with improved OS (HR score Conclusions: We developed an ex vivo assay that detects tumor functional HRD and an HR score able to predict platinum sensitivity, which holds the clinically relevant potential to become the routine companion diagnostic in the management of patients with ovarian cancer. Clin Cancer Res; 24(18); 4482–93. ©2018 AACR.

PURPOSE: Homologous recombination deficiency (HRD) correlates with platinum sensitivity in ovarian cancer patients, which clinically is the most useful predictor of sensitivity to PARPi. For this reason, there is clearly a critical need to identify homologous recombination deficiency (HRD) as early as possible. Currently, HRD is mostly diagnosed by genetic testing, which however fails to identify a large proportion of HR-deficient tumors and predict a patient's response to chemotherapy, thus we aimed to develop an ex vivo functional HRD detection test that could predict both platinum-sensitivity and patient eligibility to targeted drug treatments. METHODS: We obtained a functional HR score by quantifying homologous recombination (HR) repair after ionizing radiation-induced DNA damage in primary ovarian cancer samples (n=32). Samples clustered in 3 categories: HR-deficient, HR-low and HR-proficient. We analysed the HR score association with platinum sensitivity and treatment response, platinum-free interval (PFI) and overall survival (OS), and compared it with other clinical parameters. In parallel, we performed whole-genome sequencing and targeted DNA-sequencing of HR genes to assess if functional HRD can be predicted by currently offered genetic screening. RESULTS: Low HR scores predicted primary platinum sensitivity with high statistical significance (p=0.0103), associated with longer PFI (HR-deficient vs HR-proficient: 531 vs 53 days), and significantly correlated with improved OS (HR score CONCLUSIONS: We developed an ex vivo assay that detects tumor functional HRD and an HR score able to predict platinum sensitivity, which holds the clinically relevant potential to become the routine companion diagnostic in the management of ovarian cancer patients. Citation Format: Manuela Tumiati, Sakari Hietanen, Johanna Hynninen, Elina Pietilä, Anniina Färkkilä, Amjad Alkodsi, Yilin Li, Rainer Lehtonen, Erdogan Pekcan Erkan, Minna M. Tuominen, Kaisa Lehti, Sampsa K. Hautaniemi, Anna Vähärautio, Seija Grénman, Olli Carpén, and Liisa Kauppi. A FUNCTIONAL HOMOLOGOUS RECOMBINATION ASSAY PREDICTS PRIMARY CHEMOTHERAPY RESPONSE AND LONG-TERM SURVIVAL IN OVARIAN CANCER PATIENTS [abstract]. In: Proceedings of the 12th Biennial Ovarian Cancer Research Symposium; Sep 13-15, 2018; Seattle, WA. Philadelphia (PA): AACR; Clin Cancer Res 2019;25(22 Suppl):Abstract nr GMM-057.

High-grade serous ovarian cancer (HGSOC) is the most common cause of gynecologic-cancer-associated death. Although HGSOC is initially extremely sensitive to platinum-based chemotherapy, relapse and progression to chemotherapy resistance are frequently seen, resulting in the low 5-year survival rate of this disease. Therefore, we aim to establish a feasible methodology that combines genome editing and single-cell RNA-seq to high-throughput screen new potential targets that could be utilized alone or combined with platinum for HGSOC treatment. Firstly, we constructed several inducible Cas9-expressed HGSOC cell lines, including both platinum-sensitive and -resistant cells. Secondly, we transduced these cell lines with a pooled sgRNA lentivirus consisting of 105 non-target scrambled sgRNAs and sgRNAs against 188 genes (10 sgRNAs per gene) that are currently targets of FDA/EMA approved drugs or drugs undergoing clinical trials in ovarian cancer treatment. Our analysis of multiple HGSOC dissociated tumors by single-cell RNA-seq has revealed intertumoral heterogeneity in expression of these drug targets in vivo. The majority of these 188 genes, approximately two thirds, are expressed in low level in most epithelial ovarian cancer (EOC) cells. Only 4 genes (AKT1, IL6, C1QA, INHBA) show relatively high expression, and only in a small subset of EOC cells. Also, the remaining drug target genes that are expressed show intertumoral heterogeneity, e.g., high expression of PARP-1 in tumors that are potentially deficient in homologous recombination repair. To elucidate the functionality of these drug targets, we will next analyze the transcriptomic effects of specific gene CRISPR/Cas9 knockouts in these HGSOC cell lines by droplet-based single-cell RNA-seq. From the data, we will compare the differences between platinum-sensitive and -resistant cell lines after the gene knockouts to identify the correlations between specific genetic perturbations and platinum resistance. As a future perspective, we will continue optimizing our platform for screening other potential targets. Overall, this study provides a comprehensive and unbiased approach for potential HGSOC target screening. Note: This abstract was not presented at the conference. Citation Format: Jun Dai, Erdogan Pekcan Erkan, Kaiyang Zhang, Katja Kaipio, Tarja Lamminen, Kaisa Huhtinen, Johanna Hynninen, Seija Grenman, Olli Mikael Carpen, Sampsa Hautaniemi, Anna Vähärautio. High-throughput screening of new potential targets for high-grade serous ovarian cancer treatment. [abstract]. In: Proceedings of the AACR Conference: Addressing Critical Questions in Ovarian Cancer Research and Treatment; Oct 1-4, 2017; Pittsburgh, PA. Philadelphia (PA): AACR; Clin Cancer Res 2018;24(15_Suppl):Abstract nr B40.

In certain human cancers, the expression of critical oncogenes is driven from large regulatory elements, called super-enhancers, which recruit much of the cell’s transcriptional apparatus and are defined by extensive acetylation of histone H3 lysine 27 (H3K27ac). In a subset of T-cell acute lymphoblastic leukemia (T-ALL) cases, we found that heterozygous somatic mutations are acquired that introduce binding motifs for the MYB transcription factor in a precise noncoding site, which creates a super-enhancer upstream of the TAL1 oncogene. MYB binds to this new site and recruits it’s H3K27 acetylase binding partner CBP, as well as core components of a major leukemogenic transcriptional complex that contains RUNX1, GATA-3, and TAL1 itself. Additionally, most endogenous super-enhancers found in T-ALL cells are occupied by MYB and CBP, suggesting a general role for MYB in super-enhancer initiation. Thus, this study identifies a genetic mechanism responsible for the generation of oncogenic super-enhancers in malignant cells.

Resistance to chemotherapy poses a major clinical problem in the treatment of ovarian cancer. The efficacy of platinum- based chemotherapy relies on defective DNA repair mechanisms, and increased expression of homologous recombination (HR)/DNA repair genes in primary tumor samples is associated with improved chemoresponse. However, the effects of neo-adjuvant chemotherapy (NACT) on the expression DNA repair genes in cancer cells are poorly understood. In addition, high-grade serous ovarian cancers (HGSOCs) display significant inter- and intratumoral heterogeneity, and the identification of mechanisms of drug resistance has been hampered by the lack of data at the single cell resolution. We collected 34 primary and 20 interval tumor samples from 15 patients treated for HGSOC at the Turku University Hospital, Finland. Tumor tissue total RNA was sequenced with Illumina HiSeq. For this study, we selected 363 genes selected due to their validated role in DNA repair pathways and compared their expression profiles between primary and interval samples. The treatment naïve primary tumor samples highly expressed genes related to HR, such as BRCA2, RAD54, PARPBP, FANCD2, and POLQ. Whereas after NACT the interval samples highly expressed genes regulating cell cycle, inflammatory response, and the drug efflux pump ABCB1. Moreover, higher expression of BRCA2 or POLQ in the interval samples correlated with poor primary therapy outcome (p To reveal the transcriptomics landscapes of treatment naïve and NACT treated HGSOCs at the single cell level, we performed single cell mRNA sequencing of primary and interval samples using the Fluidigm C1 Platform. Unsupervised clustering revealed eight distinct cell subpopulations, of which two clusters were annotated HGSOC cells based on enriched expression of cancer specific genes. In global gene expression analysis of 98 HGSOC cells, the most differentially expressed genes in primary HGSOC cells were heat shock proteins reflecting active protein synthesis. By contrast, the cancer cells from interval samples highly expressed genes regulating the cell cycle and immune response. Of the DNA repair pathway genes, the interval samples expressed lower levels of HES1, a transcription factor that is known to promote cancer progression and stemness. The herein elucidated DNA repair pathway transcriptomics landscape at single-cell and whole tumor level provides novel targets for overcoming chemoresistance in HGSOC. Citation Format: Anniina Farkkila, Kaiyang Zhang, Katja Kaipio, Tarja Lamminen, Rainer Lehtonen, Johanna Hynninen, Seija Grénman, Olli Carpén, Sampsa Hautaniemi, Anna Vähärautio. Single-cell transcriptomics reveals differential DNA repair signatures after chemotherapy in high-grade serous ovarian cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 1428. doi:10.1158/1538-7445.AM2017-1428

Counting individual RNA or DNA molecules is difficult because they are hard to copy quantitatively for detection. To overcome this limitation, we applied unique molecular identifiers (UMIs), which make each molecule in a population distinct, to genome-scale human karyotyping and mRNA sequencing in Drosophila melanogaster. Use of this method can improve accuracy of almost any next-generation sequencing method, including chromatin immunoprecipitation-sequencing, genome assembly, diagnostics and manufacturing-process control and monitoring.

SummarySomatic mutations in exon 2 of the RNA polymerase II transcriptional Mediator subunit MED12 occur at very high frequency (∼70%) in uterine leiomyomas. However, the influence of these mutations on Mediator function and the molecular basis for their tumorigenic potential remain unknown. To clarify the impact of these mutations, we used affinity-purification mass spectrometry to establish the global protein-protein interaction profiles for both wild-type and mutant MED12. We found that uterine leiomyoma-linked mutations in MED12 led to a highly specific decrease in its association with Cyclin C-CDK8/CDK19 and loss of Mediator-associated CDK activity. Mechanistically, this occurs through disruption of a MED12-Cyclin C binding interface that we also show is required for MED12-mediated stimulation of Cyclin C-dependent CDK8 kinase activity. These findings indicate that uterine leiomyoma-linked mutations in MED12 uncouple Cyclin C-CDK8/19 from core Mediator and further identify the MED12/Cyclin C interface as a prospective therapeutic target in CDK8-driven cancers.

Previous studies have reported that chromosome synteny in Lepidoptera has been well conserved, yet the number of haploid chromosomes varies widely from 5 to 223. Here we report the genome (393 Mb) of the Glanville fritillary butterfly (Melitaea cinxia; Nymphalidae), a widely recognized model species in metapopulation biology and eco-evolutionary research, which has the putative ancestral karyotype of n=31. Using a phylogenetic analyses of Nymphalidae and of other Lepidoptera, combined with orthologue-level comparisons of chromosomes, we conclude that the ancestral lepidopteran karyotype has been n=31 for at least 140 My. We show that fusion chromosomes have retained the ancestral chromosome segments and very few rearrangements have occurred across the fusion sites. The same, shortest ancestral chromosomes have independently participated in fusion events in species with smaller karyotypes. The short chromosomes have higher rearrangement rate than long ones. These characteristics highlight distinctive features of the evolutionary dynamics of butterflies and moths., Marie Curie International Fellowship (PIOF-GA-2011-303312)

Most recurring somatic mutations in cancer affect protein-coding regions, either through activating oncogenes or inactivating tumor suppressors ( 1 ). However, several classes of mutations have been identified that affect the much larger noncoding regions of the genome, leading to changes in gene expression. These include large-scale genomic rearrangements that bring a strongly active promoter next to an oncogene, or place oncogene promoters under the influence of strong transcriptional enhancers (see the figure, panel A). On page 1373 of this issue, Mansour et al. ( 2 ) identify a very small mutation in ∼5% of T cell acute lymphoblastic leukemias (T-ALL; see the figure, panel B) that generates a large, powerful enhancer capable of driving tumorigenesis.

From Man to Mouse Genome-wide association studies of humans have identified single-nucleotide polymorphisms (SNPs) that increase an individual's risk of developing common diseases like cancer. Most of these SNPs have only a modest effect on risk, and many map to noncoding regions of the genome. Sur et al. (p. 1360 , published online 1 November; see the Perspective by Lewis and Tomlinson ) used a mouse model to study the functional impact of a particular SNP that resides 300 kilobases upstream of the MYC oncogene on human chromosome 8q24 and has been linked to cancer risk in humans. When a sequence encompassing this SNP was deleted in mice that were predisposed to develop intestinal tumors, the mice displayed fewer tumors than control mice. This SNP may thus play a causal role in human cancer, presumably through altered regulation of MYC.

Advances in molecular biology have made it easy to identify different DNA or RNA species and to copy them. Identification of nucleic acid species can be accomplished by reading the DNA sequence; currently millions of molecules can be sequenced in a single day using massively parallel sequencing. Efficient copying of DNA-molecules of arbitrary sequence was made possible by molecular cloning, and the polymerase chain reaction. Differences in the relative abundance of a large number of different sequences between two or more samples can in turn be measured using microarray hybridization and/or tag sequencing. However, determining the relative abundance of two different species and/or the absolute number of molecules present in a single sample has proven much more challenging. This is because it is hard to detect individual molecules without copying them, and even harder to make defined number of copies of molecules. We show here that this limitation can be overcome by using unique molecular identifiers (umis), which make each molecule in the sample distinct.

The broad research use of organoids from high-grade serous ovarian carcinoma (HGSC) has been hampered by low culture success rates and limited availability of fresh tumor material. Here we describe a method for generation and long-term expansion of HGSC organoids with efficacy markedly improved over previous reports (55% vs. 23-38%). We established organoids from cryopreserved material, demonstrating the feasibility of using viably biobanked tissue for HGSC organoid derivation. Genomic, histologic and single-cell transcriptomic analyses revealed that organoids recapitulated genetic and phenotypic features of original tumors. Organoid drug responses correlated with clinical treatment outcomes, although in culture conditions-dependent manner and only in organoids maintained in human plasma-like medium (HPLM). Organoids from consenting patients are available to the research community through a public biobank and organoid genomic data explorable through an interactive online tool. Taken together, this resource facilitates the application of HGSC organoids in basic and translational ovarian cancer research.

To identify cell cycle regulators that enable cancer cells to replicate DNA and divide in an unrestricted manner, we performed a parallel genome-wide RNAi screen in normal and cancer cell lines. In addition to many shared regulators, we found that tumor and normal cells are differentially sensitive to loss of the histone genes transcriptional regulator CASP8AP2. In cancer cells, loss of CASP8AP2 leads to a failure to synthesize sufficient amount of histones in the S-phase of the cell cycle, resulting in slowing of individual replication forks. Despite this, DNA replication fails to arrest, and tumor cells progress in an elongated S-phase that lasts several days, finally resulting in death of most of the affected cells. In contrast, depletion of CASP8AP2 in normal cells triggers a response that arrests viable cells in S-phase. The arrest is dependent on p53, and preceded by accumulation of markers of DNA damage, indicating that nucleosome depletion is sensed in normal cells via a DNA-damage-like response that is defective in tumor cells.