Large-scale Pan-cancer Cell Line Screening Identifies Actionable and Effective Drug Combinations.

Oncology drug combinations can improve therapeutic responses and increase treatment options for patients. The number of possible combinations is vast and responses can be context-specific. Systematic screens can identify clinically relevant, actionable combinations in defined patient subtypes. We present data for 109 anticancer drug combinations from AstraZeneca's oncology small molecule portfolio screened in 755 pan-cancer cell lines. Combinations were screened in a 7 × 7 concentration matrix, with more than 4 million measurements of sensitivity, producing an exceptionally data-rich resource. We implement a new approach using combination Emax (viability effect) and highest single agent (HSA) to assess combination benefit. We designed a clinical translatability workflow to identify combinations with clearly defined patient populations, rationale for tolerability based on tumor type and combination-specific "emergent" biomarkers, and exposures relevant to clinical doses. We describe three actionable combinations in defined cancer types, confirmed in vitro and in vivo, with a focus on hematologic cancers and apoptotic targets. SIGNIFICANCE: We present the largest cancer drug combination screen published to date with 7 × 7 concentration response matrices for 109 combinations in more than 750 cell lines, complemented by multi-omics predictors of response and identification of "emergent" combination biomarkers. We prioritize hits to optimize clinical translatability, and experimentally validate novel combination hypotheses. This article is featured in Selected Articles from This Issue, p. 695.

Effective drug combinations in breast, colon and pancreatic cancer cells.

Combinations of anti-cancer drugs can overcome resistance and provide new treatments1,2. The number of possible drug combinations vastly exceeds what could be tested clinically. Efforts to systematically identify active combinations and the tissues and molecular contexts in which they are most effective could accelerate the development of combination treatments. Here we evaluate the potency and efficacy of 2,025 clinically relevant two-drug combinations, generating a dataset encompassing 125 molecularly characterized breast, colorectal and pancreatic cancer cell lines. We show that synergy between drugs is rare and highly context-dependent, and that combinations of targeted agents are most likely to be synergistic. We incorporate multi-omic molecular features to identify combination biomarkers and specify synergistic drug combinations and their active contexts, including in basal-like breast cancer, and microsatellite-stable or KRAS-mutant colon cancer. Our results show that irinotecan and CHEK1 inhibition have synergistic effects in microsatellite-stable or KRAS-TP53 double-mutant colon cancer cells, leading to apoptosis and suppression of tumour xenograft growth. This study identifies clinically relevant effective drug combinations in distinct molecular subpopulations and is a resource to guide rational efforts to develop combinatorial drug treatments.

Integrated cross-study datasets of genetic dependencies in cancer.

CRISPR-Cas9 viability screens are increasingly performed at a genome-wide scale across large panels of cell lines to identify new therapeutic targets for precision cancer therapy. Integrating the datasets resulting from these studies is necessary to adequately represent the heterogeneity of human cancers and to assemble a comprehensive map of cancer genetic vulnerabilities. Here, we integrated the two largest public independent CRISPR-Cas9 screens performed to date (at the Broad and Sanger institutes) by assessing, comparing, and selecting methods for correcting biases due to heterogeneous single-guide RNA efficiency, gene-independent responses to CRISPR-Cas9 targeting originated from copy number alterations, and experimental batch effects. Our integrated datasets recapitulate findings from the individual datasets, provide greater statistical power to cancer- and subtype-specific analyses, unveil additional biomarkers of gene dependency, and improve the detection of common essential genes. We provide the largest integrated resources of CRISPR-Cas9 screens to date and the basis for harmonizing existing and future functional genetics datasets.

Project Score database: a resource for investigating cancer cell dependencies and prioritizing therapeutic targets.

CRISPR genetic screens in cancer cell models are a powerful tool to elucidate oncogenic mechanisms and to identify promising therapeutic targets. The Project Score database (https://score.depmap.sanger.ac.uk/) uses genome-wide CRISPR-Cas9 dropout screening data in hundreds of highly annotated cancer cell models to identify genes required for cell fitness and prioritize novel oncology targets. The Project Score database currently allows users to investigate the fitness effect of 18 009 genes tested across 323 cancer cell models. Through interactive interfaces, users can investigate data by selecting a specific gene, cancer cell model or tissue type, as well as browsing all gene fitness scores. Additionally, users can identify and rank candidate drug targets based on an established oncology target prioritization pipeline, incorporating genetic biomarkers and clinical datasets for each target, and including suitability for drug development based on pharmaceutical tractability. Data are freely available and downloadable. To enhance analyses, links to other key resources including Open Targets, COSMIC, the Cell Model Passports, UniProt and the Genomics of Drug Sensitivity in Cancer are provided. The Project Score database is a valuable new tool for investigating genetic dependencies in cancer cells and the identification of candidate oncology targets.

Prioritization of cancer therapeutic targets using CRISPR-Cas9 screens.

Functional genomics approaches can overcome limitations-such as the lack of identification of robust targets and poor clinical efficacy-that hamper cancer drug development. Here we performed genome-scale CRISPR-Cas9 screens in 324 human cancer cell lines from 30 cancer types and developed a data-driven framework to prioritize candidates for cancer therapeutics. We integrated cell fitness effects with genomic biomarkers and target tractability for drug development to systematically prioritize new targets in defined tissues and genotypes. We verified one of our most promising dependencies, the Werner syndrome ATP-dependent helicase, as a synthetic lethal target in tumours from multiple cancer types with microsatellite instability. Our analysis provides a resource of cancer dependencies, generates a framework to prioritize cancer drug targets and suggests specific new targets. The principles described in this study can inform the initial stages of drug development by contributing to a new, diverse and more effective portfolio of cancer drug targets.

Cell Model Passports-a hub for clinical, genetic and functional datasets of preclinical cancer models.

In vitro cancer cell cultures are facile experimental models used widely for research and drug development. Many cancer cell lines are available and efforts are ongoing to derive new models representing the histopathological and molecular diversity of tumours. Cell models have been generated by multiple laboratories over decades and consequently their annotation is incomplete and inconsistent. Furthermore, the relationships between many patient-matched and derivative cell lines have been lost, and accessing information and datasets is time-consuming and difficult. Here, we describe the Cell Model Passports database; cellmodelpassports.sanger.ac.uk, which provides details of cell model relationships, patient and clinical information, as well as access to associated genetic and functional datasets. The Passports database currently contains curated details and standardized annotation for >1200 cell models, including cancer organoid cultures. The Passports will be updated with newly derived cell models and datasets as they are generated. Users can navigate the database via tissue, cancer-type, genetic feature and data availability to select a model most suitable for specific applications. A flexible REST-API provides programmatic data access and exploration. The Cell Model Passports are a valuable tool enabling access to high-dimensional genomic and phenotypic cancer cell model datasets empowering diverse research applications.

Successive bacterial colonisation of pork and its implications for forensic investigations.

AIMS: Bacteria are considered one of the major driving forces of the mammalian decomposition process and have only recently been recognised as forensic tools. At this point, little is known about their potential use as 'post-mortem clocks'. This study aimed to establish the proof of concept for using bacterial identification as post-mortem interval (PMI) indicators, using a multi-omics approach. METHODS AND RESULTS: Pieces of pork were placed in the University's outdoor facility and surface swabs were taken at regular intervals up to 60 days. Terminal restriction fragment length polymorphism (T-RFLP) of the 16S rDNA was used to identify bacterial taxa. It succeeded in detecting two out of three key contributors involved in decomposition and represents the first study to reveal Vibrionaceae as abundant on decomposing pork. However, a high fraction of present bacterial taxa could not be identified by T-RFLP. Proteomic analyses were also performed at selected time points, and they partially succeeded in the identification of precise strains, subspecies and species of bacteria that colonized the body after different PMIs. CONCLUSION: T-RFLP is incapable of reliably and fully identifying bacterial taxa, whereas proteomics could help in the identification of specific strains of bacteria. Nevertheless, microbial identification by next generation sequencing might be used as PMI clock in future investigations and in conjunction with information provided by forensic entomologists. SIGNIFICANCE AND IMPACT OF THE STUDY: To the best of our knowledge, this work represents the first attempt to find a cheaper and easily accessible, culture-independent alternative to high-throughput techniques to establish a 'microbial clock', in combination with proteomic strategies to address this issue.

A collaborative exercise on DNA methylation based body fluid typing.

A collaborative exercise on DNA methylation based body fluid identification was conducted by seven laboratories. For this project, a multiplex methylation SNaPshot reaction composed of seven CpG markers was used for the identification of four body fluids, including blood, saliva, semen, and vaginal fluid. A total of 30 specimens were prepared and distributed to participating laboratories after thorough testing. The required experiments included four increasingly complex tasks: (1) CE of a purified single-base extension reaction product, (2) multiplex PCR and multiplex single-base extension reaction of bisulfite-modified DNA, (3) bisulfite conversion of genomic DNA, and (4) extraction of genomic DNA from body fluid samples. In tasks 2, 3 and 4, one or more mixtures were analyzed, and specimens containing both known and unknown body fluid sources were used. Six of the laboratories generated consistent body fluid typing results for specimens of bisulfite-converted DNA and genomic DNA. One laboratory failed to set up appropriate conditions for capillary analysis of reference single-base extension products. In general, variation in the values obtained for DNA methylation analysis between laboratories increased with the complexity of the required experiments. However, all laboratories concurred on the interpretation of the DNA methylation profiles produced. Although the establishment of interpretational guidelines on DNA methylation based body fluid identification has yet to be performed, this study supports the addition of DNA methylation profiling to forensic body fluid typing.

Differentiating between monozygotic twins through DNA methylation-specific high-resolution melt curve analysis.

Although short tandem repeat profiling is extremely powerful in identifying individuals from crime scene stains, it is unable to differentiate between monozygotic (MZ) twins. Efforts to address this include mutation analysis through whole genome sequencing and through DNA methylation studies. Methylation of DNA is affected by environmental factors; thus, as MZ twins age, their DNA methylation patterns change. This can be characterized by bisulfite treatment followed by pyrosequencing. However, this can be time-consuming and expensive; thus, it is unlikely to be widely used by investigators. If the sequences are different, then in theory the melting temperature should be different. Thus, the aim of this study was to assess whether high-resolution melt curve analysis can be used to differentiate between MZ twins. Five sets of MZ twins provided buccal swabs that underwent extraction, quantification, bisulfite treatment, polymerase chain reaction amplification and high-resolution melting curve analysis targeting two markers, Alu-E2F3 and Alu-SP. Significant differences were observed between all MZ twins targeting Alu-E2F3 and in four of five MZ twins targeting Alu-SP (P<0.05). Thus, it has been demonstrated that bisulfite treatment followed by high-resolution melting curve analysis could be used to differentiate between MZ twins.

Evaluation of stem-loop reverse transcription and poly-A tail extension in microRNA analysis of body fluids.

MicroRNA has been demonstrated to be a viable tool for body fluid identification purposes in forensic casework. Stem-loop reverse transcription (slRT) is regularly used for cDNA synthesis from mature miRNA, along with poly-A tail extension. Both have been used in a forensic context, but no direct comparison has been carried out. It has also not been shown whether poly-A tail extension can be used upon DNA extracts, as previously shown with slRT. Blood and saliva samples were collected and underwent DNA extraction with or without on-column DNA digestion. All samples were then aliquoted and underwent slRT and poly-A tail extension separately. qPCR was then conducted targeting microRNA markers hsa-miR-451 and hsa-miR-205. It was shown that the DNA digestion step did not affect the ability to differentiate between blood and saliva. It was also shown that this differentiation was possible using poly-A tail extension, and that poly-A tail extension exhibited more amplification than slRT. So whilst the choice of slRT and poly-A tail extension for the purpose of forensic body fluid identification is not critical, it may be best to use poly-A tail extension, particularly where there are low traces of sample.