Genetic Alterations In Cancer Research Articles

Abstract 5903: MASTL small-molecule inhibition to target mitosis for cancer therapy

Abstract Cell cycle/tumor repressors can be pharmacologically activated to repress uncontrollable proliferation of cancer cells. However, targeting cell cycle regulators of G2/M and mitotic checkpoints to achieve maximum therapeutic benefit remains a challenge due to a limited therapeutic index. Here, we identified highly potent and selective oral bioavailable clinical candidate inhibitors to the first-in-class oncology target Microtubule Associated Serine/Threonine Kinase-Like (MASTL) aimed at expanding therapeutic control of the cell cycle for cancer therapy. Breath of efficacy screening in multiple tumor indications in vitro followed by computational analyses identified the regulatory subunit PPP2R2A as a potential patient selection biomarker that strongly correlates to MASTL inhibitor response irrespective of cancer genetic alterations. Functional genetics studies confirmed that sensitivity to MASTL inhibition is defined by PPP2R2A expression. To identify genetic mediators of MASTL inhibitor resistance, a whole genome CRISPR knockout screen with the MASTL inhibitor was performed and further established that the MASTL inhibitor response is dependent on PPP2R2A containing PP2A complexes. Mechanistic studies (proteomics, resistance, functional genetics, and time-lapse microscopy) demonstrated that MASTL inhibition disrupts mitosis by activating the tumor suppressor PP2A, leading to a switch from CDK1 activity to CDK1 inactivity, mitotic defects, and cancer cell death. MASTL inhibition treatment in tumor models in vivo and in patient-derived organoid models results in potent tumor growth inhibition. Our results demonstrate that the first identified highly selective MASTL clinical candidate inhibitors represent a potential new cancer therapeutic strategy to target mitosis and mitotic exit by activating the tumor repressor PP2A to counter CDK1 activity. All procedures performed on animals were in accordance with regulations and established guidelines and were reviewed and approved by an Institutional Animal Care and Use Committee or through an ethical review process. Citation Format: Chad M. Toledo, Koleen J. Eisele, Dong Hyun Kim, Xinmeng Mu, Sonja Brun, Carl Davis, Eric C. Greenwald, Fion Hui, Meilan He, Yanling Yang, Ana Flores-Bojorquez, Yuxin Li, Penney L. Khamphavong, Joseph H. Lee, Jian Li, Dana J. Ramms, Greg Jones, Marin Auth, Cory L. Painter, Andrew Nager, Jon Oyer, Sergei L. Timofeevski, William E. Pierceall, Anwar Murtaza, Jonathan R. Heyen, Rebecca Gallego, Casey Quinlan, Indrawan McAlpine, Todd L. VanArsdale. MASTL small-molecule inhibition to target mitosis for cancer therapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 5903.

Cancer-Alterome: a literature-mined resource for regulatory events caused by genetic alterations in cancer

It is vital to investigate the complex mechanisms underlying tumors to better understand cancer and develop effective treatments. Metabolic abnormalities and clinical phenotypes can serve as essential biomarkers for diagnosing this challenging disease. Additionally, genetic alterations provide profound insights into the fundamental aspects of cancer. This study introduces Cancer-Alterome, a literature-mined dataset that focuses on the regulatory events of an organism’s biological processes or clinical phenotypes caused by genetic alterations. By proposing and leveraging a text-mining pipeline, we identify 16,681 thousand of regulatory events records encompassing 21K genes, 157K genetic alterations and 154K downstream bio-concepts, extracted from 4,354K pan-cancer literature. The resulting dataset empowers a multifaceted investigation of cancer pathology, enabling the meticulous tracking of relevant literature support. Its potential applications extend to evidence-based medicine and precision medicine, yielding valuable insights for further advancements in cancer research.

Cell fate regulation governed by p53: Friends or reversible foes in cancer therapy.

Cancer is a leading cause of death worldwide. Targeted therapies aimed at key oncogenic driver mutations in combination with chemotherapy and radiotherapy as well as immunotherapy have benefited cancer patients considerably. Tumor protein p53 (TP53), a crucial tumor suppressor gene encoding p53, regulates numerous downstream genes and cellular phenotypes in response to various stressors. The affected genes are involved in diverse processes, including cell cycle arrest, DNA repair, cellular senescence, metabolic homeostasis, apoptosis, and autophagy. However, accumulating recent studies have continued to reveal novel and unexpected functions of p53 in governing the fate of tumors, for example, functions in ferroptosis, immunity, the tumor microenvironment and microbiome metabolism. Among the possibilities, the evolutionary plasticity of p53 is the most controversial, partially due to the dizzying array of biological functions that have been attributed to different regulatory mechanisms of p53 signaling. Nearly 40 years after its discovery, this key tumor suppressor remains somewhat enigmatic. The intricate and diverse functions of p53 in regulating cell fate during cancer treatment are only the tip of the iceberg with respect to its equally complicated structural biology, which has been painstakingly revealed. Additionally, TP53 mutation is one of the most significant genetic alterations in cancer, contributing to rapid cancer cell growth and tumor progression. Here, we summarized recent advances that implicate altered p53 in modulating the response to various cancer therapies, including chemotherapy, radiotherapy, and immunotherapy. Furthermore, we also discussed potential strategies for targeting p53 as a therapeutic option for cancer.

Semi-Supervised, Attention-Based Deep Learning for Predicting TMPRSS2:ERG Fusion Status in Prostate Cancer Using Whole Slide Images.

Our study illuminates the potential of deep learning in effectively inferring key prostate cancer genetic alterations from the tissue morphology depicted in routinely available histology slides, offering a cost-effective method that could revolutionize diagnostic strategies in oncology.

CMC: Cancer miRNA Census - a list of cancer-related miRNA genes.

A growing body of evidence indicates an important role of miRNAs in cancer; however, there is no definitive, convenient-to-use list of cancer-related miRNAs or miRNA genes that may serve as a reference for analyses of miRNAs in cancer. To this end, we created a list of 165 cancer-related miRNA genes called the Cancer miRNA Census (CMC). The list is based on a score, built on various types of functional and genetic evidence for the role of particular miRNAs in cancer, e.g. miRNA-cancer associations reported in databases, associations of miRNAs with cancer hallmarks, or signals of positive selection of genetic alterations in cancer. The presence of well-recognized cancer-related miRNA genes, such as MIR21, MIR155, MIR15A, MIR17or MIRLET7s, at the top of the CMC ranking directly confirms the accuracy and robustness of the list. Additionally, to verify and indicate the reliability of CMC, we performed a validation of criteria used to build CMC, comparison of CMC with various cancer data (publications and databases), and enrichment analyses of biological pathways and processes such as Gene Ontology or DisGeNET. All validation steps showed a strong association of CMC with cancer/cancer-related processes confirming its usefulness as a reference list of miRNA genes associated with cancer.

Conumee 2.0: enhanced copy-number variation analysis from DNA methylation arrays for humans and mice.

Copy-number variations (CNVs) are common genetic alterations in cancer and their detection may impact tumor classification and therapeutic decisions. However, detection of clinically relevant large and focal CNVs remains challenging when sample material or resources are limited. This has motivated us to create a software tool to infer CNVs from DNA methylation arrays which are often generated as part of clinical routines and in research settings. We present our R package, conumee 2.0, that combines tangent normalization, an adjustable genomic binning heuristic, and weighted circular binary segmentation to utilize DNA methylation arrays for CNV analysis and mitigate technical biases and batch effects. Segmentation results were validated in a lung squamous cell carcinoma dataset from TCGA (n = 367 samples) by comparison to segmentations derived from genotyping arrays (Pearson's correlation coefficient of 0.91). We further introduce a segmented block bootstrapping approach to detect focal alternations that achieved 60.9% sensitivity and 98.6% specificity for deletions affecting CDKN2A/B (60.0% and 96.9% for RB1, respectively) in a low-grade glioma cohort from TCGA (n = 239 samples). Finally, our tool provides functionality to detect and summarize CNVs across large sample cohorts. Conumee 2.0 is available under open-source license at: https://github.com/hovestadtlab/conumee2.

Abstract B098: Unbiased in vitro and in vivo drug anchor screens identify mechanisms of resistance and sensitization for MTA-cooperative PRMT5 inhibitors in MTAP-deleted cancer models

Abstract Homozygous deletion of the MTAP gene is one of the most common genetic alterations in cancer, affecting 10-15% of all human cancer. Clinical MTA-cooperative PRMT5 inhibitors, including TNG908 and TNG462, were developed to leverage the well-characterized synthetic lethal interaction between PRMT5 inhibition and MTAP deletion. Indeed, the first disclosed clinical data for an MTA-cooperative PRMT5 inhibitor demonstrated the ability of TNG908 to selectively inhibit PRMT5 in MTAP-deleted tumors while sparing adjacent normal, MTAP-proficient cells in patients. To further explore the mechanism of action of MTA-cooperative PRMT5 inhibitors in MTAP-deleted cells, we employed multiple CRISPR-based editing platforms (CRISPRn, CRISPRi and CRISPRa) to conduct unbiased MTA-cooperative PRMT5 inhibitor anchor screens in vitro and in vivo. A panel of cancer models representing multiple histologies revealed potential PRMT5 inhibitor sensitization and resistance pathways that also include mechanisms that are specific to MTA-cooperative inhibitors. Specifically, the results of these screens identify 1) the effect of MTA/SAM metabolism on MTA-cooperative PRMT5 inhibitors, and 2) identification of potential sub-stratification strategies for patients with MTAP-deleted cancer, and 3) the activity of PRMT5 in anti-apoptotic pathways and synergy between PRMT5 and BCLxL inhibition. Citation Format: Steven Lombardo, Matthew R Tonini, Lauren Grove, Silvia Fenoglio, James Tepper, Binzhang Shen, Hannah Stowe, Shangtao Liu, Samuel R Meier, Ashley H Choi, Tenzing Khendu, Yi Yu, Kevin M Cottrell, John P Maxwell, Jannik N Andersen, Alan Huang, Kimberly J Briggs, Teng Teng. Unbiased in vitro and in vivo drug anchor screens identify mechanisms of resistance and sensitization for MTA-cooperative PRMT5 inhibitors in MTAP-deleted cancer models [abstract]. In: Proceedings of the AACR-NCI-EORTC Virtual International Conference on Molecular Targets and Cancer Therapeutics; 2023 Oct 11-15; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2023;22(12 Suppl):Abstract nr B098.

Predicting therapeutic responses in head and neck squamous cell carcinoma from TP53 mutation detected by cell-free DNA.

Head and neck squamous cell carcinoma (HNSCC) is an epithelial malignant tumor originating from the oral cavity, oropharynx, nasal cavity, sinuses, nasopharynx, hypopharynx, or larynx. Mutations in TP53 are the most common of all somatic genomic changes in HNSCC, and TP53 mutations are associated with the response to immunotherapy and chemotherapy. Tumor-derived circulating cell-free DNA (cfDNA) is a minimally invasive method to determining genetic alterations in cancer. This study aimed to explore the therapeutic responses of patients with HNSCC with TP53 mutation and the accuracy of cfDNA for detecting TP53 mutation. Information on TP53 mutations, patient survival time, and clinical data in HNSCC were downloaded from The Cancer Genome Atlas database. The difference in immune infiltration between the TP53-mutant group and the wild-type group was compared. We applied the single-sample gene set enrichment analysis method on the transcriptome of HNSCC samples to assess the distribution of immune cell types between the two groups. The chemotherapy response was constructed using the R software package, "pRRophetic". Gene set enrichment analysis was performed based on the TP53 mutation. The next-generation sequencing was executed on cfDNA from nine patients with HNSCC to detect genetic alterations. Tumor biopsy (n=9) was sequenced using the same technique. TP53 was the most frequently mutated gene in HNSCC. The TP53 mutation was related to immune cells and the expression of immune-associated genes. The TP53 mutation group showed lower response to immunotherapy but high sensitivity to some chemotherapies compared with the wild-type group. TP53 was the most frequently mutated gene (6/9; 66.67%) in cfDNA. Only 27.27% of TP53 mutations in tumor tissue were detected outside of cfDNA. TP53 mutation could be used as a specific predictor of treatment response in patients with HNSCC. Using cfDNA to detect the TP53 mutations in patients with HSNCC is a feasible method. The results suggested that the therapeutic response in patients could be predicted by detecting TP53 mutations in cfDNA, and large-scale and prospective studies are needed to validate this hypothesis.

Large-scale pathogenicity prediction analysis of cancer-associated kinase mutations reveals variability in sensitivity and specificity of computational methods.

Mutations in kinases are the most frequent genetic alterations in cancer; however, experimental evidence establishing their cancerous nature is available only for a small fraction of these mutants. Predicition analysis of kinome mutations is the primary aim of this study. Further objective is to compare the performance of various softwares in pathogenicity prediction of kinase mutations. We employed a set of computational tools to predict the pathogenicity of over forty-two thousand mutations and deposited the kinase-wise data in Mendeley database (Estimated Pathogenicity of Kinase Mutants [EPKiMu]). Mutations are more likely to be drivers when being present in the kinase domain (vs. non-kinase domain) and belonging to hotspot residues (vs. non-hotspot residues). We identified that, while predictive tools have low specificity in general, PolyPhen-2 had the best accuracy. Further efforts to combine all four tools by consensus, voting, or other simple methods did not significantly improve accuracy. The study provides a large dataset of kinase mutations along with their predicted pathogenicity that can be used as a training set for future studies. Furthermore, a comparative sensitivity and selectivity of commonly used computational tools is presented. Primary-structure-based in silico tools identified more cancerous/deleterious mutations in the kinase domains and at the hot spot residues while having higher sensitivity than specificity in detecting deleterious mutations.

Prostate cancer genetic alterations in Hispanic men.

Differences in DNA alterations in prostate cancer among White, Black, and Asian men have been widely described. This is the first description of the frequency of DNA alterations in primary and metastatic prostate cancer samples of self-reported Hispanic men. We utilized targeted next-generation sequencing tumor genomic profiles from prostate cancer tissues that underwent clinical sequencing at academic centers (GENIE 11th). We decided to restrict our analysis to the samples from Memorial Sloan Kettering Cancer Center as it was by far the main contributor of Hispanic samples. The numbers of men by self-reported ethnicity and racial categories were analyzed via Fisher's exact test between Hispanic-White versus non-Hispanic White. Our cohort consisted of 1412 primary and 818 metastatic adenocarcinomas. In primary adenocarcinomas, TMPRSS2 and ERG gene alterations were less common in non-Hispanic White men than Hispanic White (31.86% vs. 51.28%, p = 0.0007, odds ratio [OR] = 0.44 [0.27-0.72] and 25.34% vs. 42.31%, p = 0.002, OR = 0.46 [0.28-0.76]). In metastatic tumors, KRAS and CCNE1 alterations were less prevalent in non-Hispanic White men (1.03% vs. 7.50%, p = 0.014, OR = 0.13 [0.03, 0.78] and 1.29% vs. 10.00%, p = 0.003, OR = 0.12 [0.03, 0.54]). No significant differences were found in actionable alterations and androgen receptor mutations between the groups. Due to the lack of clinical characteristics and genetic ancestry in this dataset, correlation with these could not be explored. DNA alteration frequencies in primary and metastatic prostate cancer tumors differ among Hispanic-White and non-Hispanic White men. Notably, we found no significant differences in the prevalence of actionable genetic alterations between the groups, suggesting that a significant number of Hispanic men could benefit from the development of targeted therapies.

Abstract 1707: Establishment of a targeted exosome drug delivery system containing natural bacterial toxins and antitoxins

Abstract Background: The inactivation of p53, a tumor suppressor, and the activation of the Ras oncogene are the most frequent genetic alterations in cancer. We have shown that a unique E. coli MazF-MazE toxin-antitoxin (TA) system, controlled by Ras and p53, delivered via adenoviruses can be used to selectively and effectively eradicate Ras mutated cancer cells (Shapira et al., oncogene, 2021). Because of their low immunogenicity and known advantage as ideal cell transporters, exosomes offer great promise as cancer immunotherapy agents. CD24 is a GPI-anchored protein that highly expressed in most of human malignancies and not normal cells. Aim: Development of a drug delivery platform for effective and selective eradication of cancer cells while sparing normal cells based on the Trojan horse strategy and genetic signature. Methods: Two delivery systems have been designed and developed; targeted lentivirus and exosomes. Trex293 cells were used to establish stable scFv-CD24-expressing clones, which were adapted to suspension culture in a chemically-defined media. Exosomes were purified by polymer precipitation method and characterized by several methods including nano-particle tracking analysis, Western blot and EXO-ELISA. CD24 stable transfected HCT116-/- (P53 null) and HCT116+/+ (WTP53) clones were established, and evaluated for CD24 and P53 expression. HCT116+/+-CD24 cells were treated with exosomes loaded with GFP or MazF expressing plasmids and analyzed for GFP expression and cell viability. Results: Targeted lentiviruses carrying the TA system effiently shrinked CRC-derived xenografts tumors in mice. Then, a new genetically encoded pseudotyping platform, based on exosomes, was developed. A homogenic and pure population of exosomes was purified from 3D culture of the highest expressing clone (TREx-scFv). Stable and high expression of CD24 were shown, by Western blot and flow cytometry, in both, HCT116-/- and HCT116+/+ cells. HCT116+/+-CD24 cells treated with 5FU showed active P53. The ability of loading DNA into the purified exosomes and deliver them to the target cells were evaluated and showed high and intensive GFP expression. mazF was successfully loaded into the scFv targeted-exosomes and induced a massive cell death, in a dose-dependent manner. Conclusions: By exploiting the P53 genetic status as well as the Ras pathway, an effective and safe treatment could be developed that targets tumor cells specifically while sparing normal tissues. The effectiveness of this out-of-the-box approach holds great promise for the treatment and management of cancer. Citation Format: Shiran Shapira, Daniela Eisurovich, Dina Kazanov, Nadir Arber. Establishment of a targeted exosome drug delivery system containing natural bacterial toxins and antitoxins [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 1707.

Epigenetic liquid biopsies for minimal residual disease, what's around the corner?

Liquid biopsy assays for minimal residual disease (MRD) are used to monitor and inform oncological treatment and predict the risk of relapse in cancer patients. To-date, most MRD assay development has focused on targeting somatic mutations. However, epigenetic changes are more frequent and universal than genetic alterations in cancer and circulating tumor DNA (ctDNA) retains much of these changes. Here, we review the epigenetic signals that can be used to detect MRD, including DNA methylation alterations and fragmentation patterns that differentiate ctDNA from noncancerous circulating cell-free DNA (ccfDNA). We then summarize the current state of MRD monitoring; highlight the advantages of epigenetics over genetics-based approaches; and discuss the emerging paradigm of assaying both genetic and epigenetic targets to monitor treatment response, detect disease recurrence, and inform adjuvant therapy.

6MO Pan-cancer characterization of receptor tyrosine kinases alterations to sort targetable drivers from passengers

Receptor tyrosine kinases (RTK) or ligands have frequent genetic alterations in cancers. Either amplifications, missense mutations or fusions of RTK are biomarkers for some targeted therapies in a fraction of cancers, with heterogeneous responses to tyrosine kinase inhibitors (TKI), monoclonal antibodies (mabs) or andibody-drug conjugates (ADC).

IKZF3 amplification frequently occurs in HER2-positive breast cancer and is a potential therapeutic target.

Breast cancer is one of the leading causes of cancer death in women, and although treatment outcome has substantially improved in the past decades, advanced or metastatic breast cancers still carry a poor prognosis. Gene amplification is one of the frequent genetic alterations in cancer, and oncogene amplification may be associated with cancer aggressiveness and oncogenicity. Targeting amplified genes such as HER2 has vastly improved disease outcome and survival, and anti-HER2 therapeutics have revolutionized the standard of care in HER2 breast cancer. Besides currently known druggable gene amplifications including ERBB2 and FGFR2, other frequently amplified genes are relatively less well known for function and clinical significance. By querying four large databases from TCGA and AACR-Genie, from a total of 11,890 patients with invasive ductal breast carcinoma, we discover IKZF3, CCND1, ERBB2 to be consistently amplified across different cohorts. We further identify IKZF3 as a frequently amplified gene in breast cancer with a prevalence of 12-15% amplification rate. Interestingly, IKZF3 amplification is frequently co-amplified with ERBB2/HER2, and is also associated with worse prognosis compared to IKZF3 non-amplified cancers. Analysis of HER2 breast cancer patients treated with trastuzumab revealed decrease in both ERBB2/HER2 and IKZF3 expression. Further investigation using the DepMap for gene dependency by genome-wide CRISPR screening revealed dependence on IKZF3 in HER2 breast cancer cell lines. Our study utilized an integrative analysis of large-scale patient genomics, transcriptomics and clinical data to reveal IKZF3 as a frequently amplified gene, and suggest a potential role of IKZF3 as a druggable target for HER2 breast cancer.

Analysis of genomic alterations in cancer associated human pancreatic stellate cells

Pancreatic stellate cells (PSCs) constitute important cells of the pancreatic microenvironment and their close interaction with cancer cells is important in pancreatic cancer. It is currently not known whether PSCs accumulate genetic alterations that contribute to tumor biology. Our aim was to analyze genetic alterations in cancer associated PSCs. PSC DNA was matched to DNA isolated from pancreatic cancer patients’ blood (n = 5) and analyzed by Next-Generation Sequencing (NGS). Bioinformatic analysis was performed using the GATK software and pathogenicity prediction scores. Sanger sequencing was carried out to verify specific genetic alterations in a larger panel of PSCs (n = 50). NGS and GATK analysis identified on average 26 single nucleotide variants in PSC DNA as compared to the matched blood DNA that could be visualized with the Integrative Genomics Viewer. The absence of PDAC driver mutations (KRAS, p53, p16/INK4a, SMAD4) confirmed that PSC isolations were not contaminated with cancer cells. After filtering the variants, using different pathogenicity scores, ten genes were identified (SERPINB2, CNTNAP4, DENND4B, DPP4, FGFBP2, MIGA2, POLE, SNRNP40, TOP2B, and ZDHHC18) in single samples and confirmed by Sanger sequencing. As a proof of concept, functional analysis using control and SERPINB2 knock-out fibroblasts revealed functional effects on growth, migration, and collagen contraction. In conclusion, PSC DNA exhibit a substantial amount of single nucleotide variants that might have functional effects potentially contributing to tumor aggressiveness.

Identification of gene signatures for COAD using feature selection and Bayesian network approaches

The combination of TCGA and GTEx databases will provide more comprehensive information for characterizing the human genome in health and disease, especially for underlying the cancer genetic alterations. Here we analyzed the gene expression profile of COAD in both tumor samples from TCGA and normal colon tissues from GTEx. Using the SNR-PPFS feature selection algorithms, we discovered a 38 gene signatures that performed well in distinguishing COAD tumors from normal samples. Bayesian network of the 38 genes revealed that DEGs with similar expression patterns or functions interacted more closely. We identified 14 up-DEGs that were significantly correlated with tumor stages. Cox regression analysis demonstrated that tumor stage, STMN4 and FAM135B dysregulation were independent prognostic factors for COAD survival outcomes. Overall, this study indicates that using feature selection approaches to select key gene signatures from high-dimensional datasets can be an effective way for studying cancer genomic characteristics.

Cancer genetic alterations and risk of venous thromboembolism

Cancer has long been known to incur an increased risk of venous thromboembolism (VTE). Multiple risk factors for cancer-associated thrombosis (CAT) have been identified, and several pathophysiological mechanisms elucidated. However, until recently there was scant data available about the influence of cancer-specific somatic genetic alterations on the risk of venous thromboembolism. In the last few years, several gene loci were found to modulate the risk of CAT, usually causing an increase in risk but sometimes found to have a protective effect. Notably, cancer-specific somatic genetic alterations in KRAS, IDH1, ALK and ROS1 have been found to alter the risk of CAT by independent groups. Work in this field is limited by the high-dimensionality and often sparse nature of genomic datasets. Also, early data suggest that for certain genes the effect on VTE risk can be tumor type-specific, which suggests that predictive models must factor such interactions in order to optimally leverage genetic information. Notably, individual gene effects appear to be often small and no one gene explains most of the variability of CAT risk. Ultimately, improved knowledge of the genetic determinants of CAT will help ameliorate risk stratification for this complication and hopefully provide mechanistic insights. Better risk stratification could lead to enhanced pharmacological VTE prophylaxis, while advancements in the understanding of the biology of CAT could conceivably lead to non-anticoagulant targeted therapies for thrombosis.

An integrated network representation of multiple cancer-specific data for graph-based machine learning

Genomic profiles of cancer cells provide valuable information on genetic alterations in cancer. Several recent studies employed these data to predict the response of cancer cell lines to drug treatment. Nonetheless, due to the multifactorial phenotypes and intricate mechanisms of cancer, the accurate prediction of the effect of pharmacotherapy on a specific cell line based on the genetic information alone is problematic. Emphasizing on the system-level complexity of cancer, we devised a procedure to integrate multiple heterogeneous data, including biological networks, genomics, inhibitor profiling, and gene-disease associations, into a unified graph structure. In order to construct compact, yet information-rich cancer-specific networks, we developed a novel graph reduction algorithm. Driven by not only the topological information, but also the biological knowledge, the graph reduction increases the feature-only entropy while preserving the valuable graph-feature information. Subsequent comparative benchmarking simulations employing a tissue level cross-validation protocol demonstrate that the accuracy of a graph-based predictor of the drug efficacy is 0.68, which is notably higher than those measured for more traditional, matrix-based techniques on the same data. Overall, the non-Euclidean representation of the cancer-specific data improves the performance of machine learning to predict the response of cancer to pharmacotherapy. The generated data are freely available to the academic community at https://osf.io/dzx7b/.

Chimeric RNA Design Principles for RNA-Mediated Gene Fusion

One common genetic alteration in cancer is gene fusion resulting from chromosomal translocations. The mechanisms that create such oncogenic fusion genes are not well understood. Previously, we provided the direct evidence that expression of a designed chimeric RNA can drive the formation of TMPRSS2-ERG gene fusion. Central to this RNA-mediated gene fusion mechanism is a proposed three-way junction formed by RNA/DNA hybrid and the intergenic DNA stem formed by target genes. In this study, we determined the important parameters for chimeric RNA-mediated gene fusion using TMPRSS2-ERG fusion gene as the model. Our results indicate that both the chimeric RNA lengths and the sizes of unpaired bulges play important roles in inducing TMPRSS2-ERG gene fusion. The optimal length of unpaired bulges was about 35 nt, while the optimal chimeric RNA length was about 50 nt for targeting. These observations were consistent regardless of the target locations within TMPRSS2 and ERG genes. These empirically determined parameters provide important insight for searching cellular RNAs that may initiate oncogenic fusion genes. The knowledge could also facilitate the development of useful genomic technology for manipulating mammalian genomes.

Liquid biopsy detection of genomic alterations in pediatric brain tumors from cell-free DNA in peripheral blood, CSF, and urine.

The ability to identify genetic alterations in cancers is essential for precision medicine; however, surgical approaches to obtain brain tumor tissue are invasive. Profiling circulating tumor DNA (ctDNA) in liquid biopsies has emerged as a promising approach to avoid invasive procedures. Here, we systematically evaluated the feasibility of profiling pediatric brain tumors using ctDNA obtained from plasma, cerebrospinal fluid (CSF), and urine. We prospectively collected 564 specimens (257 blood, 240 urine, and 67 CSF samples) from 258 patients across all histopathologies. We performed ultra-low-pass whole-genome sequencing (ULP-WGS) to assess copy number variations and estimate tumor fraction and developed a pediatric CNS tumor hybrid capture panel for deep sequencing of specific mutations and fusions. ULP-WGS detected copy number alterations in 9/46 (20%) CSF, 3/230 (1.3%) plasma, and 0/153 urine samples. Sequencing detected alterations in 3/10 (30%) CSF, 2/74 (2.7%) plasma, and 0/2 urine samples. The only positive results were in high-grade tumors. However, most samples had insufficient somatic mutations (median 1, range 0-39) discoverable by the sequencing panel to provide sufficient power to detect tumor fractions of greater than 0.1%. Children with brain tumors harbor very low levels of ctDNA in blood, CSF, and urine, with CSF having the most DNA detectable. Molecular profiling is feasible in a small subset of high-grade tumors. The level of clonal aberrations per genome is low in most of the tumors, posing a challenge for detection using whole-genome or even targeted sequencing methods. Substantial challenges therefore remain to genetically characterize pediatric brain tumors from liquid biopsies.