Asaf Rotem

Dissecting the multicellular ecosystem of metastatic melanoma by single-cell RNA-seq

Single-cell expression profiles of melanoma Tumors harbor multiple cell types that are thought to play a role in the development of resistance to drug treatments. Tirosh et al. used single-cell sequencing to investigate the distribution of these differing genetic profiles within melanomas. Many cells harbored heterogeneous genetic programs that reflected two different states of genetic expression, one of which was linked to resistance development. Following drug treatment, the resistance-linked expression state was found at a much higher level. Furthermore, the environment of the melanoma cells affected their gene expression programs. Science, this issue p. 189 Melanoma cells show transcriptional heterogeneity. To explore the distinct genotypic and phenotypic states of melanoma tumors, we applied single-cell RNA sequencing (RNA-seq) to 4645 single cells isolated from 19 patients, profiling malignant, immune, stromal, and endothelial cells. Malignant cells within the same tumor displayed transcriptional heterogeneity associated with the cell cycle, spatial context, and a drug-resistance program. In particular, all tumors harbored malignant cells from two distinct transcriptional cell states, such that tumors characterized by high levels of the MITF transcription factor also contained cells with low MITF and elevated levels of the AXL kinase. Single-cell analyses suggested distinct tumor microenvironmental patterns, including cell-to-cell interactions. Analysis of tumor-infiltrating T cells revealed exhaustion programs, their connection to T cell activation and clonal expansion, and their variability across patients. Overall, we begin to unravel the cellular ecosystem of tumors and how single-cell genomics offers insights with implications for both targeted and immune therapies.

Adaptive resistance of melanoma cells to RAF inhibition via reversible induction of a slowly dividing de‐differentiated state

Treatment of BRAF‐mutant melanomas with MAP kinase pathway inhibitors is paradigmatic of the promise of precision cancer therapy but also highlights problems with drug resistance that limit patient benefit. We use live‐cell imaging, single‐cell analysis, and molecular profiling to show that exposure of tumor cells to RAF/MEK inhibitors elicits a heterogeneous response in which some cells die, some arrest, and the remainder adapt to drug. Drug‐adapted cells up‐regulate markers of the neural crest (e.g., NGFR), a melanocyte precursor, and grow slowly. This phenotype is transiently stable, reverting to the drug‐naïve state within 9 days of drug withdrawal. Transcriptional profiling of cell lines and human tumors implicates a c‐Jun/ECM/FAK/Src cascade in de‐differentiation in about one‐third of cell lines studied; drug‐induced changes in c‐Jun and NGFR levels are also observed in xenograft and human tumors. Drugs targeting the c‐Jun/ECM/FAK/Src cascade as well as BET bromodomain inhibitors increase the maximum effect (Emax) of RAF/MEK kinase inhibitors by promoting cell killing. Thus, analysis of reversible drug resistance at a single‐cell level identifies signaling pathways and inhibitory drugs missed by assays that focus on cell populations.

Alternative to the soft-agar assay that permits high-throughput drug and genetic screens for cellular transformation

Significance The paper describes an assay for cellular transformation that involves growth in low attachment (GILA). This assay is comparable to the gold-standard soft-agar assay, but it is much easier to perform and is suitable for high-throughput drug and genetic screens. We describe such screens for drugs and genes that selectively inhibit or increase transformation, but not proliferation. Such molecules are unlikely to be found through conventional drug screening. Lastly, we demonstrate the ability of Food and Drug Administration-approved noncancer drugs to selectively kill ovarian cancer cells derived from patients with chemotherapy-resistant disease, suggesting this approach may provide useful information for personalized cancer treatment. Colony formation in soft agar is the gold-standard assay for cellular transformation in vitro, but it is unsuited for high-throughput screening. Here, we describe an assay for cellular transformation that involves growth in low attachment (GILA) conditions and is strongly correlated with the soft-agar assay. Using GILA, we describe high-throughput screens for drugs and genes that selectively inhibit or increase transformation, but not proliferation. Such molecules are unlikely to be found through conventional drug screening, and they include kinase inhibitors and drugs for noncancer diseases. In addition to known oncogenes, the genetic screen identifies genes that contribute to cellular transformation. Lastly, we demonstrate the ability of Food and Drug Administration-approved noncancer drugs to selectively kill ovarian cancer cells derived from patients with chemotherapy-resistant disease, suggesting this approach may provide useful information for personalized cancer treatment.

Ex Vivo Profiling of PD-1 Blockade Using Organotypic Tumor Spheroids

Ex vivo systems that incorporate features of the tumor microenvironment and model the dynamic response to immune checkpoint blockade (ICB) may facilitate efforts in precision immuno-oncology and the development of effective combination therapies. Here, we demonstrate the ability to interrogate ex vivo response to ICB using murine- and patient-derived organotypic tumor spheroids (MDOTS/PDOTS). MDOTS/PDOTS isolated from mouse and human tumors retain autologous lymphoid and myeloid cell populations and respond to ICB in short-term three-dimensional microfluidic culture. Response and resistance to ICB was recapitulated using MDOTS derived from established immunocompetent mouse tumor models. MDOTS profiling demonstrated that TBK1/IKKε inhibition enhanced response to PD-1 blockade, which effectively predicted tumor response in vivo Systematic profiling of secreted cytokines in PDOTS captured key features associated with response and resistance to PD-1 blockade. Thus, MDOTS/PDOTS profiling represents a novel platform to evaluate ICB using established murine models as well as clinically relevant patient specimens.Significance: Resistance to PD-1 blockade remains a challenge for many patients, and biomarkers to guide treatment are lacking. Here, we demonstrate feasibility of ex vivo profiling of PD-1 blockade to interrogate the tumor immune microenvironment, develop therapeutic combinations, and facilitate precision immuno-oncology efforts. Cancer Discov; 8(2); 196-215. ©2017 AACR.See related commentary by Balko and Sosman, p. 143See related article by Deng et al., p. 216This article is highlighted in the In This Issue feature, p. 127.

Dicer loss and recovery induce an oncogenic switch driven by transcriptional activation of the oncofetal Imp1–3 family

MicroRNAs (miRNAs) are post-transcriptional regulators of gene expression critical for organismal viability. Changes in miRNA activity are common in cancer, but how these changes relate to subsequent alterations in transcription and the process of tumorigenesis is not well understood. Here, we report a deep transcriptional, oncogenic network regulated by miRNAs. We present analysis of the gene expression and phenotypic changes associated with global miRNA restoration in miRNA-deficient fibroblasts. This analysis uncovers a miRNA-repressed network containing oncofetal genes Imp1, Imp2, and Imp3 (Imp1-3) that is up-regulated primarily transcriptionally >100-fold upon Dicer loss and is resistant to resilencing by complete restoration of miRNA activity. This Dicer-resistant epigenetic switch confers tumorigenicity to these cells. Let-7 targets Imp1-3 are required for this tumorigenicity and feed back to reinforce and sustain expression of the oncogenic network. Together, these Dicer-resistant genes constitute an mRNA expression signature that is present in numerous human cancers and is associated with poor survival.

Abstract S1-01: Whole exome and transcriptome sequencing of resistant ER+ metastatic breast cancer

Background: While great strides have been made in the treatment of estrogen receptor-positive (ER+) metastatic breast cancer (MBC), therapeutic resistance invariably occurs. A better understanding of the underlying resistance mechanisms is critical to enable durable control of this disease. Methods: We performed whole exome sequencing (WES) and transcriptome sequencing (RNA-seq) on metastatic tumor biopsies from 88 patients with ER+ MBC who had developed resistance to one or more ER-directed therapies. For 27 of these patients, we sequenced the treatment-naive primary tumors for comparison to the resistant specimens. Tumors were analyzed for point mutations, insertions/deletions, copy number alterations, translocations, and gene expression. Detailed clinicopathologic data was collected for each patient and linked to the genomic information. Results: WES of all metastatic samples demonstrated several recurrently altered genes whose incidence differed significantly from primary, treatment-naive ER+ breast cancers sequenced in the TCGA study (TCGA). These include ESR1 mutations (n=17, 19.3%; 32.86 fold enrichment, q.value Comparing to matched primary samples from the same patient, many alterations were found to be acquired in several cases, including for ESR1, ERBB2, PIK3CA, PTEN, RB1, AKT1, and others. Initial analysis of RNA-seq data from metastatic samples (n=59) allowed classification of individual resistance mechanisms into broader resistance modes based on the observed transcriptional state. Conclusions: We present a genomic landscape of resistant ER+ MBC using WES and RNA-seq. Multiple genes were recurrently altered in these tumors at significantly higher rates than in ER+ primary breast cancer. When compared with matched primary tumors from the same patient, alterations in these and other genes were often found to be acquired after treatment, suggesting a role in resistance to ER-directed therapies and/or metastasis. Potential resistance mechanisms appear to fall into several categories; integrating RNA-seq data may enhance the ability to identify these categories even when genomic alterations are not identified. Multiple clinically relevant genomic and molecular alterations are identified in metastatic biopsies– with implications for choice of next therapy, clinical trial eligibility, and novel drug targets. Citation Format: Cohen O, Kim D, Oh C, Waks A, Oliver N, Helvie K, Marini L, Rotem A, Lloyd M, Stover D, Adalsteinsson V, Freeman S, Ha G, Cibulskis C, Anderka K, Tamayo P, Johannessen C, Krop I, Garraway L, Winer E, Lin N, Wagle N. Whole exome and transcriptome sequencing of resistant ER+ metastatic breast cancer [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr S1-01.

Genome-scale identification of transcription factors that mediate an inflammatory network during breast cellular transformation

Transient activation of Src oncoprotein in non-transformed, breast epithelial cells can initiate an epigenetic switch to the stably transformed state via a positive feedback loop that involves the inflammatory transcription factors STAT3 and NF-κB. Here, we develop an experimental and computational pipeline that includes 1) a Bayesian network model (AccessTF) that accurately predicts protein-bound DNA sequence motifs based on chromatin accessibility, and 2) a scoring system (TFScore) that rank-orders transcription factors as candidates for being important for a biological process. Genetic experiments validate TFScore and suggest that more than 40 transcription factors contribute to the oncogenic state in this model. Interestingly, individual depletion of several of these factors results in similar transcriptional profiles, indicating that a complex and interconnected transcriptional network promotes a stable oncogenic state. The combined experimental and computational pipeline represents a general approach to comprehensively identify transcriptional regulators important for a biological process.Systematic analysis of the control of dynamic cellular processes remains a challenge. Here the authors introduce a pipeline enabling them to identify TFs involved in Src-induced cellular transformation, and find that a large number of TFs with diverse DNA binding specificities orchestrate the process.

Miniaturizing 3D assay for high-throughput drug and genetic screens for small patient-derived tumor samples (Conference Presentation)

Three-dimensional growth conditions reflect the natural environment of cancer cells and are crucial to be performed at drug screens. We developed a 3D assay for cellular transformation that involves growth in low attachment (GILA) conditions and is strongly correlated with the 50-year old benchmark assay-soft agar. Using GILA, we performed high-throughput screens for drugs and genes that selectively inhibit or increase transformation, but not proliferation. This phenotypic approach is complementary to our genetic approach that utilizes single-cell RNA-sequencing of a patient sample to identify putative oncogenes that confer sensitivity to drugs designed to specifically inhibit the identified oncoprotein. Currently, we are dealing with a big challenge in our field- the limited number of cells that might be extracted from a biopsy. Small patient-derived samples are hard to test in the traditional multiwell plate and it will be helpful to minimize the culture area and the experimental system. We managed to design a suitable microfluidic device for limited number of cells and perform the assay using image analysis. We aim to test drugs on tumor cells, outside of the patient body- and recommend on the ideal treatment that is tailored to the individual. This device will help to minimize biopsy-sampling volumes and minimize interventions in the patient’s tumor.

Abstract 3037: Dissecting treatment resistance in patients with ovarian cancer and PDX-models using single-cell RNA-sequencing

Background: Ovarian cancer (OvCa) is frequently associated with malignant effusions, which are complex ecosystems with heterogeneous populations of malignant cells and non-malignant cells. Bulk RNA-seq or whole-exome sequencing (WES) only reflect average cellular behavior and thereby mask intrinsic cell diversity with potential relevance for treatment resistance. Approach: To overcome some of these barriers, we applied single-cell RNA-sequencing (scRNA-seq) to malignant and non-malignant cells isolated from patients with platinum treatment resistant disease. Furthermore, we used patient-derived xenograft (PDX) cohorts, in which we isolated cells for scRNA-seq from vehicle tumors (VEH), treated the other models with carboplatin, and harvested cells at the time of minimal residual disease (MRD) or disease progression (PROG). Results: To date, we have profiled ~12000 single cells from 12 patients with treatment naive (n=3) or platinum-resistant disease (n=9), including sequential sampling in 3 patients with resistant disease. We observed significant inter- and intra-individual transcriptional heterogeneity in malignant cells. A recurrent pattern across resistant patients was the differential expression of inflammatory pathways in a subset of cells. In a patient with three consecutive specimens, we observed increasing accumulation of cells expressing a cell state characterized by tumor necrosis factor alpha (TNF-a) signaling, Importantly, these cells were genetically identical to the entire population, supporting the hypothesis that non-encoded mechanisms conferred treatment resistance. In a BRCA-mutant patient, unbiased analysis identified a stemness program in a subpopulation of cells, which was genetically identical to other cells, indicating phenotypic conversion. To systemically interrogate mechanisms of resistance to platinum therapy, sequenced single cells isolated from PDX models at three time points (VEH, MRD and PROG). In a BRCA-WT PDX model, resistant cells isolated at MRD and PROG shared a transcriptional program that was dominated by expression of a STAT3 program. Ex vivo cultures from platinum-resistant patients were exquisitely sensitivity to JAK/STAT3-inhibitor. Live cell imaging revealed that STAT3-inhibition prevented spheroid formation, attachment and clearance through a mesothelial monolayer in vitro. Conclusion: Our results indicate that non-encoded mechanisms play an important role in the development of treatment resistance in ovarian cancer. Our initial studies indicate an important role of inflammatory pathways in treatment resistance, in particular STAT3 signaling, which can be overcome with specific inhibitors at nanomolar concentrations. These data suggests that single-cell profiling can be performed on clinical ovarian cancer specimens and may yield novel therapeutic avenues for patients with treatment-resistant ovarian cancer. Citation Format: Benjamin Izar, Itay Tirosh, Elizabeth Stover, Asaf Rotem, Parin Shah, Mike Cuoco, Chris Rodman, Joyce Liu, Ursula Matulonis, Orit Rozenblatt-Rosen, Levi Garraway, Aviv Regev. Dissecting treatment resistance in patients with ovarian cancer and PDX-models using single-cell RNA-sequencing [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 3037. doi:10.1158/1538-7445.AM2017-3037

GILA, a Replacement for the Soft‐Agar Assay that Permits High‐Throughput Drug and Genetic Screens for Cellular Transformation

For the last five decades, measuring the ability of cells to grow in soft agar has served as the gold standard assay for in vitro cellular transformation. Nevertheless, the soft agar colony formation assay is time consuming and ill‐suited for high‐throughput screens. This unit describes an equally qualitative and quantitative assay known as growth in low attachment or GILA. The GILA assay is suitable for high‐throughput pharmacological or genetic screens and allows the simultaneous examination of multiple cell lines and experimental perturbations. GILA conditions are specific and relevant to the transformed state because they depend on a property of cancer cells that is not shared by non‐transformed cells. The GILA assay enables ex vivo drug sensitivity testing of patient‐derived tumor cells to define precise treatments for individual patients.