Search for: All records

Plexiform neurofibromas (PNF) are peripheral nerve sheath tumors that cause significant morbidity in persons with neurofibromatosis type 1 (NF1), yet treatment options remain limited. To identify novel therapeutic targets for PNF, we applied an integrated multi-omic approach to quantitatively profile kinome enrichment in a mouse model that has predicted therapeutic responses in clinical trials for NF1-associated PNF with high fidelity. Experimental Design:Utilizing RNA sequencing combined with chemical proteomic profiling of the functionally enriched kinome using multiplexed inhibitor beads coupled with mass spectrometry, we identified molecular signatures predictive of response to CDK4/6 and RAS/MAPK pathway inhibition in PNF. Informed by these results, we evaluated the efficacy of the CDK4/6 inhibitor, abemaciclib, and the ERK1/2 inhibitor, LY3214996, alone and in combination in reducing PNF tumor burden in Nf1flox/flox;PostnCre mice. Results:Converging signatures of CDK4/6 and RAS/MAPK pathway activation were identified within the transcriptome and kinome that were conserved in both murine and human PNF. We observed robust additivity of the CDK4/6 inhibitor, abemaciclib, in combination with the ERK1/2 inhibitor, LY3214996, in murine and human NF1(Nf1) mutant Schwann cells. Consistent with these findings, the combination of abemaciclib (CDK4/6i) and LY3214996 (ERK1/2i) synergized to suppress molecular signatures of MAPK activation and exhibited enhanced antitumor activity in Nf1flox/flox;PostnCre mice in vivo. Conclusions:These findings provide rationale for the clinical translation of CDK4/6 inhibitors alone and in combination with therapies targeting the RAS/MAPK pathway for the treatment of PNF and other peripheral nerve sheath tumors in persons with NF1. 

Abstract Colorectal cancer (CRC) cells display remarkable adaptability, orchestrating metabolic changes that confer growth advantages, pro‐tumor microenvironment, and therapeutic resistance. One such metabolic change occurs in glutamine metabolism. Colorectal tumors with high glutaminase (GLS) expression exhibited reduced T cell infiltration and cytotoxicity, leading to poor clinical outcomes. However, depletion of GLS in CRC cells has minimal effect on tumor growth in immunocompromised mice. By contrast, remarkable inhibition of tumor growth is observed in immunocompetent mice when GLS is knocked down. It is found that GLS knockdown in CRC cells enhanced the cytotoxicity of tumor‐specific T cells. Furthermore, the single‐cell flux estimation analysis (scFEA) of glutamine metabolism revealed that glutamate‐to‐glutathione (Glu‐GSH) flux, downstream of GLS, rather than Glu‐to‐2‐oxoglutarate flux plays a key role in regulating the immune response of CRC cells in the tumor. Mechanistically, inhibition of the Glu‐GSH flux activated reactive oxygen species (ROS)‐related signaling pathways in tumor cells, thereby increasing the tumor immunogenicity by promoting the activity of the immunoproteasome. The combinatorial therapy of Glu‐GSH flux inhibitor and anti‐PD‐1 antibody exhibited a superior tumor growth inhibitory effect compared to either monotherapy. Taken together, the study provides the first evidence pointing to Glu‐GSH flux as a potential therapeutic target for CRC immunotherapy. 

Abstract Quantitative assessment of single cell fluxome is critical for understanding the metabolic heterogeneity in diseases. Unfortunately, laboratory-based single cell fluxomics is currently impractical, and the current computational tools for flux estimation are not designed for single cell-level prediction. Given the well-established link between transcriptomic and metabolomic profiles, leveraging single cell transcriptomics data to predict single cell fluxome is not only feasible but also an urgent task. In this study, we present FLUXestimator, an online platform for predicting metabolic fluxome and variations using single cell or general transcriptomics data of large sample-size. The FLUXestimator webserver implements a recently developed unsupervised approach called single cell flux estimation analysis (scFEA), which uses a new neural network architecture to estimate reaction rates from transcriptomics data. To the best of our knowledge, FLUXestimator is the first web-based tool dedicated to predicting cell-/sample-wise metabolic flux and metabolite variations using transcriptomics data of human, mouse and 15 other common experimental organisms. The FLUXestimator webserver is available at http://scFLUX.org/, and stand-alone tools for local use are available at https://github.com/changwn/scFEA. Our tool provides a new avenue for studying metabolic heterogeneity in diseases and has the potential to facilitate the development of new therapeutic strategies. 

Boolean matrix factorization (BMF) has been widely utilized in fields such as recommendation systems, graph learning, text mining, and -omics data analysis. Traditional BMF methods decompose a binary matrix into the Boolean product of two lower-rank Boolean matrices plus homoscedastic random errors. However, real-world binary data typically involves biases arising from heterogeneous row- and column-wise signal distributions. Such biases can lead to suboptimal fitting and unexplainable predictions if not accounted for. In this study, we reconceptualize the binary data generation as the Boolean sum of three components: a binary pattern matrix, a background bias matrix influenced by heterogeneous row or column distributions, and random flipping errors. We introduce a novel Disentangled Representation Learning for Binary matrices (DRLB) method, which employs a dual auto-encoder network to reveal the true patterns. DRLB can be seamlessly integrated with existing BMF techniques to facilitate bias-aware BMF. Our experiments with both synthetic and real-world datasets show that DRLB significantly enhances the precision of traditional BMF methods while offering high scalability. Moreover, the bias matrix detected by DRLB accurately reflects the inherent biases in synthetic data, and the patterns identified in the bias-corrected real-world data exhibit enhanced interpretability. 

Matrix low rank approximation is an effective method to reduce or eliminate the statistical redundancy of its components. Compared with the traditional global low rank methods such as singular value decomposition (SVD), local low rank approximation methods are more advantageous to uncover interpretable data structures when clear duality exists between the rows and columns of the matrix. Local low rank approximation is equivalent to low rank submatrix detection. Unfortunately,existing local low rank approximation methods can detect only submatrices of specific mean structure, which may miss a substantial amount of true and interesting patterns. In this work, we develop a novel matrix computational framework called RPSP (Random Probing based submatrix Propagation) that provides an effective solution for the general matrix local low rank representation problem. RPSP detects local low rank patterns that grow from small submatrices of low rank property, which are determined by a random projection approach. RPSP is supported by theories of random projection. Experiments on synthetic data demonstrate that RPSP outperforms all state-of-the-art methods, with the capacity to robustly and correctly identify the low rank matrices when the pattern has a similar mean as the background, background noise is heteroscedastic and multiple patterns present in the data. On real-world datasets, RPSP also demonstrates its effectiveness in identifying interpretable local low rank matrices. 

Characterized by the accumulation of somatic mutations in blood cell lineages, clonal hematopoiesis of indeterminate potential (CHIP) is frequent in aging and involves the expansion of mutated hematopoietic stem and progenitor cells (HSC/Ps) that leads to an increased risk of hematologic malignancy. However, the risk factors that contribute to CHIPassociated clonal hematopoiesis (CH) are poorly understood. Obesity induces a proinflammatory state and fatty bone marrow (FBM), which may influence CHIP-associated pathologies. We analyzed exome sequencing and clinical data for 47,466 individuals with validated CHIP in the UK Biobank. CHIP was present in 5.8% of the study population and was associated with a significant increase in the waist-to-hip ratio (WHR). Mouse models of obesity and CHIP driven by heterozygosity of Tet2, Dnmt3a, Asxl1, and Jak2 resulted in exacerbated expansion of mutant HSC/Ps due in part to excessive inflammation. Our results show that obesity is highly associated with CHIP and that a proinflammatory state could potentiate the progression of CHIP to more significant hematologic neoplasia. The calcium channel blockers nifedipine and SKF-96365, either alone or in combination with metformin, MCC950, or anakinra (IL-1 receptor antagonist), suppressed the growth of mutant CHIP cells and partially restored normal hematopoiesis. Targeting CHIP-mutant cells with these drugs could be a potential therapeutic approach to treat CH and its associated abnormalities in individuals with obesity. 

Pancreatic cancer or pancreatic ductal adenocarcinoma (PDAC) is characterized by a profound inflammatory tumor microenvironment (TME) with high heterogeneity, metastatic propensity, and extreme hypoxia. The integrated stress response (ISR) pathway features a family of protein kinases that phosphorylate eukaryotic initiation factor 2 (eIF2) and regulate translation in response to diverse stress conditions, including hypoxia. We previously demonstrated that eIF2 signaling pathways were profoundly affected in response to Redox factor-1 (Ref-1) knockdown in human PDAC cells. Ref-1 is a dual function enzyme with activities of DNA repair and redox signaling, responds to cellular stress, and regulates survival pathways. The redox function of Ref-1 directly regulates multiple transcription factors including HIF-1α, STAT3, and NF-κB, which are highly active in the PDAC TME. However, the mechanistic details of the crosstalk between Ref-1 redox signaling and activation of ISR pathways are unclear. Following Ref-1 knockdown, induction of ISR was observed under normoxic conditions, while hypoxic conditions were sufficient to activate ISR irrespective of Ref-1 levels. Inhibition of Ref-1 redox activity increased expression of p-eIF2 and ATF4 transcriptional activity in a concentration-dependent manner in multiple human PDAC cell lines, and the effect on eIF2 phosphorylation was PERK-dependent. Treatment with PERK inhibitor, AMG-44 at high concentrations resulted in activation of the alternative ISR kinase, GCN2 and induced levels of p-eIF2 and ATF4 in both tumor cells and cancer-associated fibroblasts (CAFs). Combination treatment with inhibitors of Ref-1 and PERK enhanced cell killing effects in both human pancreatic cancer lines and CAFs in 3D co-culture, but only at high doses of PERK inhibitors. This effect was completely abrogated when Ref-1 inhibitors were used in combination with GCN2 inhibitor, GCN2iB. We demonstrate that targeting of Ref-1 redox signaling activates the ISR in multiple PDAC lines and that this activation of ISR is critical for inhibition of the growth of co-culture spheroids. Combination effects were only observed in physiologically relevant 3D co-cultures, suggesting that the model system utilized can greatly affect the outcome of these targeted agents. Inhibition of Ref-1 signaling induces cell death through ISR signaling pathways, and combination of Ref-1 redox signaling blockade with ISR activation could be a novel therapeutic strategy for PDAC treatment. 

Metastatic cancer accounts for over 90% of all cancer deaths, and evaluations of metastasis potential are vital for minimizing the metastasis-associated mortality and achieving optimal clinical decision-making. Computational assessment of metastasis potential based on large-scale transcriptomic cancer data is challenging because metastasis events are not always clinically detectable. The under-diagnosis of metastasis events results in biased classification labels, and classification tools using biased labels may lead to inaccurate estimations of metastasis potential. This issue is further complicated by the unknown metastasis prevalence at the population level, the small number of confirmed metastasis cases, and the high dimensionality of the candidate molecular features. Our proposed algorithm, called P ositive and unlabeled L earning from U nbalanced cases and S parse structures ( PLUS ), is the first to use a positive and unlabeled learning framework to account for the under-detection of metastasis events in building a classifier. PLUS is specifically tailored for studying metastasis that deals with the unbalanced instance allocation as well as unknown metastasis prevalence, which are not considered by other methods. PLUS achieves superior performance on synthetic datasets compared with other state-of-the-art methods. Application of PLUS to The Cancer Genome Atlas Pan-Cancer gene expression data generated metastasis potential predictions that show good agreement with the clinical follow-up data, in addition to predictive genes that have been validated by independent single-cell RNA-sequencing datasets. 

An immunotherapy trial often uses the phase I/II design to identify the optimal biological dose, which monitors the efficacy and toxicity outcomes simultaneously in a single trial. The progression-free survival rate is often used as the efficacy outcome in phase I/II immunotherapy trials. As a result, patients developing disease progression in phase I/II immunotherapy trials are generally seriously ill and are often treated off the trial for ethical consideration. Consequently, the happening of disease progression will terminate the toxicity event but not vice versa, so the issue of the semi-competing risks arises. Moreover, this issue can become more intractable with the late-onset outcomes, which happens when a relatively long follow-up time is required to ascertain progression-free survival. This paper proposes a novel Bayesian adaptive phase I/II design accounting for semi-competing risks outcomes for immunotherapy trials, referred to as the dose-finding design accounting for semi-competing risks outcomes for immunotherapy trials (SCI) design. To tackle the issue of the semi-competing risks in the presence of late-onset outcomes, we re-construct the likelihood function based on each patient's actual follow-up time and develop a data augmentation method to efficiently draw posterior samples from a series of Beta-binomial distributions. We propose a concise curve-free dose-finding algorithm to adaptively identify the optimal biological dose using accumulated data without making any parametric dose–response assumptions. Numerical studies show that the proposed SCI design yields good operating characteristics in dose selection, patient allocation, and trial duration.