Synergistic Drug Combinations Research Articles

Cyclin-Dependent Kinases (CDKs) and the Human Cytomegalovirus-Encoded CDK Ortholog pUL97 Represent Highly Attractive Targets for Synergistic Drug Combinations.

Human cytomegalovirus (HCMV) is a pathogenic human herpesvirus associated with serious, potentially life-threatening symptoms in the immunocompromised or immunonaïve host. The limitations encountered by antiviral therapy options currently available include a narrow panel of accessible targets, the induction of viral drug resistance as well as severe drug dosage-mediated side-effects. Improved drug-targeting strategies to resolve these issues are the focus of our investigations. In particular, pharmaceutical kinase inhibitors (PKIs), either directed to host kinases or directed to the viral protein kinase pUL97, have been considered to overcome these restrictions. Recently, we reported the identification of a synergistic combination of two PKIs directed to host cyclin-dependent kinase 7 (CDK7) and viral CDK ortholog pUL97. Here, we substantiate these findings with the following results: (i) true drug synergy was exhibited by various chemical classes of PKI pairs directed to pUL97 and CDK7; (ii) no putative amplification of cytotoxicity by these drug combinations was observed; (iii) a reduction in drug dosage levels for synergistic combinations was defined on a quantitative basis and compared to monotreatments; (iv) the quantities of target proteins CDK7 and pUL97 expressed in HCMV-infected cells were assessed by confocal imaging, indicating a strong down-modulation of CDK7 levels as a result of synergistic drug treatment; (v) the functional importance of these target kinases, both binding to cyclin H, was illustrated by assessing HCMV replication under the viral genomic deletion of ORF-UL97 or cellular cyclin knock-out; (vi) new combinations of HCMV-specific drug synergy were demonstrated for solely host-directed treatments using PKIs against CDK2, CDK7, CDK8 and/or CDK9 and (vii) a triple PKI combination provided further support for the synergy approach. With these combined findings, this study highlights the potential of therapeutic drug combinations of approved, developmental and preclinical PKIs for expanding future options for anti-HCMV therapy.

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.

SHetA2 Attack on Mortalin and Colleagues in Cancer Therapy and Prevention.

Heat Shock Proteins of the 70-kDa family (HSP70s) do not cause cancer by themselves, but instead protect cells as they transform into cancer. These molecular chaperones bind numerous client proteins and utilize ATP hydrolysis to facilitate proper protein folding, formation of functional complexes and cellular localizations, or degradation of irreparably damaged proteins. Their transient upregulation by stressful situations avoids induction of programmed cell death. Continued upregulation of the mortalin, heat shock cognate (hsc70) and glucose regulated protein 78 (Grp78) support cancer development and progression by supporting pro-proliferative and metabolic functions and repressing pro-death functions of oncoproteins and tumor suppressor proteins. This review describes the discovery and development of a lead anti-cancer compound, sulfur heteroarotinoid A2 (SHetA2, NSC726189), which was originally developed to bind retinoic acid receptors, but was subsequently found to work independently of these receptors. The discovery and validation of mortalin, hsc70 and Grp78 as SHetA2 target proteins is summarized. The documented and hypothesized roles of these HSP70 proteins and their clients in the mechanism of SHetA2 inhibition of cancer without toxicity are discussed. Use of this mechanistic data to evaluate drug action in a cancer clinical trial and develop synergistic drug combinations is explained. Knowledge needed to optimize SHetA2 analogs for use in cancer therapy and prevention is proposed as future directions.

Abstract P5-04-03: Prolactin drives a dynamic STAT5A/HDAC6/HMGN2 cis-regulatory landscape exploitable in ER+ breast cancer

Abstract The hormone, prolactin, has been implicated in breast cancer pathogenesis and regulates chromatin engagement by the transcription factor, STAT5A. STAT5A is known to inducibly bind promoters and cis-regulatory elements genome wide, though the mechanisms by which it exerts specificity and regulation of target gene expression remain enigmatic. We previously identified HDAC6 and HMGN2 as cofactors that facilitate prolactin induced, STAT5A mediated gene expression. Here, multi-condition STAT5A, HDAC6, and HMGN2 ChIP-seq with parallel condition RNA-seq are utilized to reveal the cis-regulatory landscape and cofactor dynamics underlying prolactin stimulated gene expression in breast cancer. We find that prolactin regulated genes are significantly enriched for cis-regulatory elements bound by HDAC6 and HMGN2, and that inducible STAT5A binding at enhancers, rather than promoters, conveys specificity for prolactin regulated genes. The selective HDAC6 inhibitor, ACY-241, blocks prolactin induced STAT5A chromatin engagement at cis-regulatory elements as well as a significant proportion of prolactin stimulated gene expression. We identify functional pathways known to contribute to the development and/or progression of breast cancer that are activated by prolactin and inhibited by ACY-241. Additionally, we find that the DNA sequences underlying shared STAT5A and HDAC6 binding-sites at enhancers are differentially enriched for estrogen response elements (ESR1 and ESR2 motifs) relative to enhancers bound by STAT5A alone. Gene set enrichment analysis identifies significant overlap of ERα regulated genes with genes regulated by prolactin, particularly prolactin regulated genes with promoters or enhancers co-occupied by both STAT5A and HDAC6. Lastly, the therapeutic efficacy of ACY-241 is demonstrated in in vitro and in vivo breast cancer models, where we identify synergistic ACY-241 drug combinations and observe differential sensitivity of ER+ models relative to ER- models. Citation Format: Charles V Clevenger, Tia Turner, Chuck Harrell, Justin Craig. Prolactin drives a dynamic STAT5A/HDAC6/HMGN2 cis-regulatory landscape exploitable in ER+ breast cancer [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P5-04-03.

An enhanced cascade-based deep forest model for drug combination prediction.

Combination therapy has shown an obvious curative effect on complex diseases, whereas the search space of drug combinations is too large to be validated experimentally even with high-throughput screens. With the increase of the number of drugs, artificial intelligence techniques, especially machine learning methods, have become applicable for the discovery of synergistic drug combinations to significantly reduce the experimental workload. In this study, in order to predict novel synergistic drug combinations in various cancer cell lines, the cell line-specific drug-induced gene expression profile (GP) is added as a new feature type to capture the cellular response of drugs and reveal the biological mechanism of synergistic effect. Then, an enhanced cascade-based deep forest regressor (EC-DFR) is innovatively presented to apply the new small-scale drug combination dataset involving chemical, physical and biological (GP) properties of drugs and cells. Verified by the dataset, EC-DFR outperforms two state-of-the-art deep neural network-based methods and several advanced classical machine learning algorithms. Biological experimental validation performed subsequently on a set of previously untested drug combinations further confirms the performance of EC-DFR. What is more prominent is that EC-DFR can distinguish the most important features, making it more interpretable. By evaluating the contribution of each feature type, GP feature contributes 82.40%, showing the cellular responses of drugs may play crucial roles in synergism prediction. The analysis based on the top contributing genes in GP further demonstrates some potential relationships between the transcriptomic levels of key genes under drug regulation and the synergism of drug combinations.

PRODeepSyn: predicting anticancer synergistic drug combinations by embedding cell lines with protein-protein interaction network.

Although drug combinations in cancer treatment appear to be a promising therapeutic strategy with respect to monotherapy, it is arduous to discover new synergistic drug combinations due to the combinatorial explosion. Deep learning technology holds immense promise for better prediction of in vitro synergistic drug combinations for certain cell lines. In methods applying such technology, omics data are widely adopted to construct cell line features. However, biological network data are rarely considered yet, which is worthy of in-depth study. In this study, we propose a novel deep learning method, termed PRODeepSyn, for predicting anticancer synergistic drug combinations. By leveraging the Graph Convolutional Network, PRODeepSyn integrates the protein–protein interaction (PPI) network with omics data to construct low-dimensional dense embeddings for cell lines. PRODeepSyn then builds a deep neural network with the Batch Normalization mechanism to predict synergy scores using the cell line embeddings and drug features. PRODeepSyn achieves the lowest root mean square error of 15.08 and the highest Pearson correlation coefficient of 0.75, outperforming two deep learning methods and four machine learning methods. On the classification task, PRODeepSyn achieves an area under the receiver operator characteristics curve of 0.90, an area under the precision–recall curve of 0.63 and a Cohen’s Kappa of 0.53. In the ablation study, we find that using the multi-omics data and the integrated PPI network’s information both can improve the prediction results. Additionally, the case study demonstrates the consistency between PRODeepSyn and previous studies.

Drug response profiling identifies sensitive synergistic drug combinations for patients with refractory/relapsed acute lymphoblastic leukemia

Drug response profiling identifies sensitive synergistic drug combinations for patients with refractory/relapsed acute lymphoblastic leukemia

Selected Articles from This Issue

Dysregulation of the protein arginine methyltransferase 5 (PRMT5) is associated with lymphomas, lung cancers, and breast cancers. In this First Disclosure, Brehmer and colleagues outline JNJ-64619178, an inhibitor of PRMT5 with potent anti-proliferative activity. JNJ-64619178 occupies the SAM and substrate pockets of PRMT5, trapping it in a catalytically inactive state. PMRT5 inhibition by JNJ-64619178 increased alternatively spliced events and resulted in the killing of primary AML samples ex vivo. Taken together, the results support the continued clinical study of JNJ-64619178 (NCT03573310).The approval of Trastuzumab deruxtecan demonstrates the potency of the DNA topoisomerase I inhibitor DXd in antibody-drug conjugates (ADCs). DXd-ADCs exhibit a short systemic half-life and ability to optimize the drug-antibody ratio. In this First Disclosure, Okajima and colleagues outline the design of another DXd-containing ADC targeting Trophoblast cell surface antigen 2 (TROP2). Datopotamab deruxtecan is designed to have a shorter half-life than sacituzumab govitecan, a TROP2-directed ADC approved by the FDA for TNBC. Dato-DXd showed potent anti-tumor activity and was absent of severe gastrointestinal toxicity. The results support the first-in-human phase 1 study for patients with advanced solid tumors (NCT03401385).Chromosome region maintenance 1 is overexpressed in ovarian cancer and its inhibition has been shown to achieve a 30% disease control rate in recurrent disease. Here, Handley and colleagues seek to enhance the effects of selinexor through synergistic drug combinations. Their screen found the PARP inhibitor olaparib as the most likely candidate. Selinexor decreased the expression of DNA damage repair proteins, generating an increased susceptibility to PARP inhibitors. The combination reduced tumor nodules and weight in vivo in a mouse model. Their results outline an actionable combination for advanced ovarian cancer patients.The advent of KRAS inhibitors, as well as the disagreement between retrospective studies in immunotherapy and KEYNOTE-189, demonstrate the need to characterize the KRAS mutational landscape of non-small cell lung cancer (NSCLC). Therefore, Judd and colleagues sequenced 592 genes from 17,095 NSCLC specimens to assess co-occuring genomic alterations, tumor mutational burden (TMB), and PD-L1 expression. They demonstrate the rate of co-mutation depends on the KRAS mutation type. Mutated KRAS was significantly increased in PD-L1 expression compared to wild type, with KRAS G12C the most likely to be PD-L1 positive. These studies inform future therapeutic intervention for KRAS-mutated NSCLC.

DDRE-28. IDENTIFICATION OF SYNERGISTIC DRUG COMBINATIONS FOR THE THERAPY OF IDHmut GLIOMA

Abstract OBJECTIVE Isocitrate Dehydrogenase (IDH) mutation in glioma results in a multitude of biological differences with consequences for survival and therapy response. Therefore, IDH mutated (IDHmut) and wildtype (IDHwt) tumors are regarded as separate entities with the need for adjusted therapy like the combination of procarbazine, CCNU and vincristine (PCV). However, as vincristine has often severe side effects like neuropathy new effective therapy options are required. Therefore, we searched for combinations of FDA-approved drugs which effectively inhibit the growth of IDHmut cells in vitro. METHODS We tested different drug combinations of a drug library consisting of 146 FDA-approved drugs on two established IDHmut GSC lines. Based on a previous single agent drug screen, six drugs were selected (Idarubicin, Ixazumib, Ponatinib, Neratinib, Romidepsin) to be combined with all 146 drugs of the library. Cell viability was assessed by the CellTiterGlo 3D assay (Promega) in 96 well plates, while Caspase-Glo 3/7 3D assay was used to measure induction of apoptosis. RESULTS Out of 1460 drug combinations tested altogether 21 synergistic drug combinations could be identified and validated. The combination with the highest blood-brain-barrier permeability score was further investigated. Finally, drug-concentrations elucidating the highest synergistic effect on proliferation was further studied in a 8-point dose-response matrix followed by validation in additional four IDHmut GSC lines. CONCLUSION This work can lay the foundation for future improvements of the therapy of patients suffering from LGGs.

Selective Inhibition of Class II HDAC Isoforms 4 and 5 Provides a Promising Therapeutic Intervention for MLL-Rearranged Acute Lymphoblastic Leukemia in Infants

▪MLL-rearranged acute lymphoblastic leukemia (ALL) is characterized by deregulation of the epigenome and shows susceptibility towards epigenetic perturbators such as histone deacetylase (HDAC) inhibitors. Hence, HDACs represent attractive drug targets and a variety of small molecule HDAC inhibitors have been developed and evaluated for the treatment of hematological malignancies. However, most broad-spectrum inhibitors, which simultaneously target the majority of human HDAC isoforms, often induce toxicity, especially in combination with other therapeutic agents. Therefore, selective inhibition of only one or two HDAC isoforms may represent a better alternative, provided that disease-specific dependency on specific HDACs has been identified.We examined the effects of shRNA-mediated knock-down of the class II HDACs (i.e. HDAC4, HDAC5, HDAC6, HDAC7 and HDAC9) in the MLL-rearranged ALL cell lines SEM and ALL-PO. Except for HDAC9, loss of expression (both on the mRNA and protein level) of all HDACs led to strong reductions in viable cells (0.70 to 0.19-fold; p=0.02-0.0016) in both models due to apoptosis, cell cycle arrest, or a combination thereof.Next, we evaluated the in vitro efficacy of a variety of class II HDAC-specific inhibitors on a panel of MLL-rearranged ALL (n=5) using 4-day viability MTT assays. This revealed that the selective HDAC4/5 inhibitor LMK-235 was able to recapitulate the loss-of-function phenotype of HDAC4 and HDAC5. Dose response curves showed complete growth inhibition in MLL-rearranged ALL cell lines (n=5), as well as in primary MLL-rearranged infant ALL patient samples (n=4), with IC 50 values of ~100 nM and 40-100 nM, respectively. Importantly, at these concentrations, LMK-235 hardly affected whole bone marrow samples derived from healthy individuals (n=2), for which IC 50 values were ~1 µM.To further explore the potential of class II HDAC inhibitor-based therapeutic strategies, we performed a combinatorial drug screen to identify compounds that synergize with LMK-235. For this, a compound library (comprising >200 unique agents) was screened in the absence and presence of varying concentrations of LMK235 in the MLL-rearranged cell line models SEM and ALL-PO. This, and subsequent validation experiments in additional cell line models, revealed that Venetoclax (BCL2 inhibitor), Trametinib (MEK/ERK inhibitor), Ponatinib (multi-tyrosine kinase inhibitor) and Omipalisib (a PI3K/mTOR inhibitor) strongly synergized with LMK-235. Average ZIP synergy scores ranged from 10-30, with peak ZIP scores up to 40. Importantly, synergistic effects were consistent over all concentration combinations tested. The addition of 50-100 nM LMK-235 strongly reduced IC 50 values for Omipalisib, Ponatinib and Venetoclax (0.27-fold p=0.003, 0.11-fold p=0.0005, 0.75-fold p=0.0004, respectively) in both models.In preparation to assess the in vivo efficacy of LMK-235 in patient-derived xenograft (PDX) mouse models of MLL-rearranged infant ALL, pharmacokinetic/pharmacodynamic (PK/PD) analysis was performed in immunodeficient NSG mice (n=5). For this, mice were treated with 20 mg/kg of LMK-235, daily administered via intraperitoneal injections for a total of 29 days. While none of the mice showed signs of toxicity or weight loss, LMK-235 plasma levels were stably maintained at concentrations that are highly effective against MLL-rearranged ALL cells in vitro.Taken together, these data demonstrate that various class II HDAC isoforms are targetable vulnerabilities in MLL-rearranged ALL and that pharmaceutical inhibition of HDAC4/5 by LMK-235 represents an attractive therapeutic option. Moreover, high levels of synergy observed between this HDAC inhibitor and various agents belonging to drug classes already reported to be effective against MLL-rearranged ALL, warrants pre-clinical evaluation in vivo. Currently, the assessment of the in vivo efficacy of LMK-235 monotherapy in MLL-rearranged infant ALL PDX models is in progress, after which promising synergistic HDAC inhibitor-based drug combinations will be evaluated. To determine the additional therapeutic value, the efficacy of LMK-235 and promising synergistic combinations will be evaluated in the background of conventional combination chemotherapy, where PDX models will receive a mouse-adapted version of induction therapy currently applied for treatment of MLL-rearranged infant ALL patients. DisclosuresNo relevant conflicts of interest to declare.

Microfluidic Multi-Drug and Multi-Dose Functional Profiling Towards Personalized Clinical Predictions in Pediatric Leukemia

Treatment of pediatric acute lymphoblastic leukemia (ALL) has seen dramatic improvements over the last several decades, leading to survival rates of over 90% for specific groups. However, for patients with relapsed/refractory disease, achieving complete remission remains a significant challenge. To further improve treatment outcomes for these patients, improved functional screens are needed to better match high risk leukemia patients with novel therapeutic strategies.Current functional screens are limited by the large patient samples required, particularly for combination drug screening and evaluation of higher order drug interactions. To that end, this work aims to develop a microfluidic multi-drug and multi-dose chemosensitivity assay to directly test candidate combinations of therapeutics in patient derived ALL cells. This will allow for analysis of response to combinations of 3 small molecule drugs while minimizing required patient sample and reducing experimental workload. Once validated, this assay can be used to identify biomarkers of response to novel combination regimens in pediatric ALL, and ultimately, could be used to prospectively guide therapy in newly diagnosed patients.Specifically, this is accomplished by generating stable, overlapping concentration gradients of small molecule drugs across 3D cultures of leukemia cells with and without human bone marrow derived mesenchymal stem cells (Fig 1A). We have demonstrated the ability to generate and maintain these gradients over 48 hours, and have shown that experimentally generated concentration gradients closely match steady state computational fluid dynamics (CFD) model predictions (Fig 1B). Moreover, exposing a T-ALL cell line (Jurkats) to a single gradient of cytarabine results in a gradient in cell viability. In drug treated devices, increasing drug dose led to decreasing averaged viability, while no appreciable difference was seen in vehicle control devices (Fig 1C).A commonly used 3 drug combination, Daunorubicin, Vincristine, and Prednisolone, was then used to validate the utility of the device for evaluating multi-dose drug combinations. Here, Jurkats were exposed to superimposed concentration gradients of Daunorubicin, Vincristine, and Prednisolone for 48 hours, after which cell viability as a function of concentration was examined. Specifically, a ternary contour map of the average viability across the device was generated, clearly depicting the range of response across the device (Fig 2B).Previous results have demonstrated Jurkats' sensitivity to Daunorubicin as well as Vincristine and relative resistance to Prednisolone, and these results are closely recapitulated in the device. Previous reports in literature have also demonstrated potential for antagonism between anthracyclines and vinca alkaloids, specifically between Doxorubicin and Vincristine in Jurkats (Ehrhardt, 2011). Our results also suggest that some amount of antagonism may exist between Daunorubicin and Vincristine in Jurkats at these concentration ranges tested, evidenced by the high average viability in regions of high vincristine concentrations and mid-range daunorubicin concentrations.Finally, we then used this system to evaluate response to drug combinations in a test patient sample. Specifically, combinations of Vincristine, Prednisolone, and Daunorubicin, as well as combinations of Nelarabine, Etoposide, and Cytoxan were evaluated in a standard risk B-ALL bone marrow aspirate sample. Drug treated devices resulted in significantly reduced viability relative to the vehicle control, with a differential in response observed between the 2 combinations tested (Fig 3A). Moreover, at these drug concentration ranges, it appears that combinations of Etoposide and Ara-G are particularly ineffective relative to other combinations in this sample (Fig 3B). Future studies will work to improve the baseline viability of banked samples in the device, and include testing of fresh leukemia samples as well.We are also working to develop methods to use data collected from these devices to identify synergistic or antagonistic drug combinations, and in the future, can correlate this with clinical outcome metrics as well as biomarkers of response.Tissue samples were provided by the Children's Healthcare of Atlanta Pediatric Bio-Repository. Other investigators may have received specimens from the same subjects. [Display omitted] DisclosuresLam: Sanguina, Inc.: Current holder of individual stocks in a privately-held company.

Prediction of synergistic drug combinations using PCA-initialized deep learning

BackgroundCancer is one of the main causes of death worldwide. Combination drug therapy has been a mainstay of cancer treatment for decades and has been shown to reduce host toxicity and prevent the development of acquired drug resistance. However, the immense number of possible drug combinations and large synergistic space makes it infeasible to screen all effective drug pairs experimentally. Therefore, it is crucial to develop computational approaches to predict drug synergy and guide experimental design for the discovery of rational combinations for therapy.ResultsWe present a new deep learning approach to predict synergistic drug combinations by integrating gene expression profiles from cell lines and chemical structure data. Specifically, we use principal component analysis (PCA) to reduce the dimensionality of the chemical descriptor data and gene expression data. We then propagate the low-dimensional data through a neural network to predict drug synergy values. We apply our method to O’Neil’s high-throughput drug combination screening data as well as a dataset from the AstraZeneca-Sanger Drug Combination Prediction DREAM Challenge. We compare the neural network approach with and without dimension reduction. Additionally, we demonstrate the effectiveness of our deep learning approach and compare its performance with three state-of-the-art machine learning methods: Random Forests, XGBoost, and elastic net, with and without PCA-based dimensionality reduction.ConclusionsOur developed approach outperforms other machine learning methods, and the use of dimension reduction dramatically decreases the computation time without sacrificing accuracy.

ParsVNN: parsimony visible neural networks for uncovering cancer-specific and drug-sensitive genes and pathways.

Prediction of cancer-specific drug responses as well as identification of the corresponding drug-sensitive genes and pathways remains a major biological and clinical challenge. Deep learning models hold immense promise for better drug response predictions, but most of them cannot provide biological and clinical interpretability. Visible neural network (VNN) models have emerged to solve the problem by giving neurons biological meanings and directly casting biological networks into the models. However, the biological networks used in VNNs are often redundant and contain components that are irrelevant to the downstream predictions. Therefore, the VNNs using these redundant biological networks are overparameterized, which significantly limits VNNs’ predictive and explanatory power. To overcome the problem, we treat the edges and nodes in biological networks used in VNNs as features and develop a sparse learning framework ParsVNN to learn parsimony VNNs with only edges and nodes that contribute the most to the prediction task. We applied ParsVNN to build cancer-specific VNN models to predict drug response for five different cancer types. We demonstrated that the parsimony VNNs built by ParsVNN are superior to other state-of-the-art methods in terms of prediction performance and identification of cancer driver genes. Furthermore, we found that the pathways selected by ParsVNN have great potential to predict clinical outcomes as well as recommend synergistic drug combinations.

In This Issue

In This Issue

CLINICAL OUTCOMES ASSOCIATED WITH IVERMECTIN USE IN COVID-19: AN UPDATED SYSTEMATIC REVIEW AND META-ANALYSIS

TOPIC: Chest Infections TYPE: Original Investigations PURPOSE: Despite exhaustive efforts by the global healthcare to treat COVID-19, the death toll continues to rise. Ivermectin, a known anti-parasitic agent, when re-purposed for treating COVID-19, has demonstrated positive results in several studies. We aim to evaluate the benefit and risk of Ivermectin therapy in COVID-19 patients. METHODS: We conducted a systematic search for full-text manuscripts published from February 1, 2020, to August 15, 2021, that focused on the efficacy and safety of Ivermectin use in COVID-19. Overall mortality and need for intensive care unit (ICU) admission were primary outcomes;secondary outcomes were adverse effects, need for mechanical ventilation, viral clearance, time to viral clearance, need for hospitalization, and length of hospital stay. Random-effects models were used for the quantitative and qualitative analyses. RESULTS: A total of 52 studies (n= 17561) were included in the qualitative analysis and out of these, 44 studies (n=14019) were included in the quantitative analysis. In the qualitative analysis of Ivermectin treatment, a mortality rate of 3.6%, ICU admission rate of 5.4%, mechanical ventilation rate of 3.9%, and adverse event rate of 6.6% were observed. In the overall mortality meta-analysis, odds of death were lower in the Ivermectin-arm compared to the non-Ivermectin arm (OR 0.54, p=0.009). However, in the subgroup analysis of 15 randomized controlled trials, we observed lower odds of mortality in the mild/moderate sub-group (OR 0.31, p=0.06) but without statistical significance. In the severe/critical sub-group, odds were only marginally lowered and were not statistically significant.(OR 0.86, p=0.56). The benefit with Ivermectin was not statistically significant in the meta-analysis of the need for ICU admission (OR 0.48, p=0.17), mechanical ventilation (OR=0.75, p=0.12) and duration of hospitalization (MD -1.80, p= 0.06). The meta-analysis of adverse effects was inconclusive (OR 0.87, p=0.30). Ivermectin, frequently used as adjunctive treatment, was linked with higher odds of achieving viral clearance (OR 3.52, p=0.0002), in a shorter duration (MD -4.12, p=0.02) as well as reduction in the need for hospitalization (OR 0.34, p=0.008). CONCLUSIONS: In the updated quantitative analysis, we found that mortality benefit with Ivermectin treatment is uncertain. Trends of decreased need for ICU admissions and mechanical ventilation, and duration of hospitalization were observed but were not significant. Mostly as an adjuvant treatment, Ivermectin may help accelerate viral clearance as well as reduce the need for hospitalization. Meta-analysis for adverse events suggested that there may be no difference in the adverse event rate with Ivermectin and other treatments, but this cannot be concluded with confidence. The qualitative analysis showed that Ivermectin led to better clinical outcomes in COVID-19 patients and a lower incidence of adverse events. CLINICAL IMPLICATIONS: Well-designed larger observational studies and clinical trials are needed to confirm Ivermectin's mortality benefit in COVID-19 and to further investigate its ideal dosage and timing in the disease course, drug interactions, and possible synergistic drug combinations to achieve maximum benefit. We also need to evaluate the impact of Ivermectin in patients infected with the newly emerging strains and its role in vaccinated patients. Finally, we recommend physicians to exercise caution while prescribing Ivermectin for COVID-19 while we await evidence-based guidelines. DISCLOSURES: No relevant relationships by Vikas Bansal, source=Web Response No relevant relationships by Abhishek Bhurwal, source=Web Response No relevant relationships by Shree Spandana Ghanta, source=Web Response No relevant relationships by Smruti Karale, source=Web Response No relevant relationships by Rahul Kashyap, source=Web Response No relevant relationships by Hira Khan, source=Web Response No relevant relationships by Janaki Makadia, source=Web Response No relevant relationsh ps by Ishita Mehra, source=Web Response No relevant relationships by HEMANT MUTNEJA, source=Web Response No relevant relationships by ROMIL SINGH, source=Web Response No relevant relationships by Muhammad Tayyeb, source=Web Response No relevant relationships by Aysun Tekin, source=Web Response

A novel design of microfluidic platform for metronomic combinatorial chemotherapy drug screening based on 3D tumor spheroid model.

For treating cancer at various stages, chemotherapy drugs administered in combination provide better treatment results with lower side effects compared to single-drug therapy. However, finding the potential drug combinations has been challenging due to the large numbers of possible combinations from approved drugs and the failure of in vitro 2D well plate-based cancer models. 3D spheroid-based high-throughput microfluidic platforms recapitulate some of the important features of native tumor tissue and offer a promising alternative to evaluate the combinatory effects of the drugs. This study develops a novel polydimethylsiloxane (PDMS) based microfluidic design with a dynamic environment and strategically placed U-shaped wells for testing all seven possible combinations (three single-drug treatments, three pairwise combinations, treatment with all three drugs) of three chemotherapy drugs (Paclitaxel, Vinorelbine, and Etoposide) on lung tumor spheroids. The design of U-shaped wells has been validated with computational results. Firstly, we test all combinations of drugs on the conventional well plate in static conditions with 3D tumor spheroids. Based on static drug testing results, we show a proof-of-concept by testing the most effective drug combination on the microfluidic device in a dynamic environment. The concentration of the drugs used in combination falls below the maximum tolerated dose (MTD) of the individual drugs, towards low dose metronomic (LDM) chemotherapy. LDM combinatorial chemotherapy identified in this study can potentially lower toxicity and provide better treatment results in cancer patients. The device can be further used to culture patient-specific tumor spheroids and identify synergistic drug combinations for personalized medicine.

DeepDDS: deep graph neural network with attention mechanism to predict synergistic drug combinations.

Drug combination therapy has become an increasingly promising method in the treatment of cancer. However, the number of possible drug combinations is so huge that it is hard to screen synergistic drug combinations through wet-lab experiments. Therefore, computational screening has become an important way to prioritize drug combinations. Graph neural network has recently shown remarkable performance in the prediction of compound-protein interactions, but it has not been applied to the screening of drug combinations. In this paper, we proposed a deep learning model based on graph neural network and attention mechanism to identify drug combinations that can effectively inhibit the viability of specific cancer cells. The feature embeddings of drug molecule structure and gene expression profiles were taken as input to multilayer feedforward neural network to identify the synergistic drug combinations. We compared DeepDDS (Deep Learning for Drug-Drug Synergy prediction) with classical machine learning methods and other deep learning-based methods on benchmark data set, and the leave-one-out experimental results showed that DeepDDS achieved better performance than competitive methods. Also, on an independent test set released by well-known pharmaceutical enterprise AstraZeneca, DeepDDS was superior to competitive methods by more than 16% predictive precision. Furthermore, we explored the interpretability of the graph attention network and found the correlation matrix of atomic features revealed important chemical substructures of drugs. We believed that DeepDDS is an effective tool that prioritized synergistic drug combinations for further wet-lab experiment validation. Source code and data are available at https://github.com/Sinwang404/DeepDDS/tree/master.

Synergistic Activity of Capsaicin and Colistin Against Colistin-Resistant Acinetobacter baumannii: In Vitro/Vivo Efficacy and Mode of Action.

Acinetobacter baumannii is an opportunistic pathogen predominantly associated with nosocomial infections. With emerging resistance against polymyxins, synergistic combinations of drugs are being investigated as a new therapeutic approach. Capsaicin is a common constituent of the human diet and is widely used in traditional alternative medicines. The present study evaluated the antibacterial activities of capsaicin in combination with colistin against three unrelated colistin-resistant Acinetobacter baumannii strains in vitro and in vivo, and then further studied their synergistic mechanisms. Using the checkerboard technique and time-kill assays, capsaicin and colistin showed a synergistic effect on colistin-resistant A. baumannii. A mouse bacteremia model confirmed the in vivo effects of capsaicin and colistin. Mechanistic studies shown that capsaicin can inhibit the biofilm formation of both colistin-resistant and non-resistant A. baumannii. In addition, capsaicin decreased the production of intracellular ATP and disrupted the outer membrane of A. baumannii. In summary, the synergy between these drugs may enable a lower concentration of colistin to be used to treat A. baumannii infection, thereby reducing the dose-dependent side effects. Hence, capsaicin–colistin combination therapy may offer a new treatment option for the control of A. baumannii infection.

Deep learning identifies synergistic drug combinations for treating COVID-19.

Effective treatments for COVID-19 are urgently needed. However, discovering single-agent therapies with activity against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has been challenging. Combination therapies play an important role in antiviral therapies, due to their improved efficacy and reduced toxicity. Recent approaches have applied deep learning to identify synergistic drug combinations for diseases with vast preexisting datasets, but these are not applicable to new diseases with limited combination data, such as COVID-19. Given that drug synergy often occurs through inhibition of discrete biological targets, here we propose a neural network architecture that jointly learns drug-target interaction and drug-drug synergy. The model consists of two parts: a drug-target interaction module and a target-disease association module. This design enables the model to utilize drug-target interaction data and single-agent antiviral activity data, in addition to available drug-drug combination datasets, which may be small in nature. By incorporating additional biological information, our model performs significantly better in synergy prediction accuracy than previous methods with limited drug combination training data. We empirically validated our model predictions and discovered two drug combinations, remdesivir and reserpine as well as remdesivir and IQ-1S, which display strong antiviral SARS-CoV-2 synergy in vitro. Our approach, which was applied here to address the urgent threat of COVID-19, can be readily extended to other diseases for which a dearth of chemical-chemical combination data exists.

GraphSynergy: a network-inspired deep learning model for anticancer drug combination prediction.

To develop an end-to-end deep learning framework based on a protein-protein interaction (PPI) network to make synergistic anticancer drug combination predictions. We propose a deep learning framework named Graph Convolutional Network for Drug Synergy (GraphSynergy). GraphSynergy adapts a spatial-based Graph Convolutional Network component to encode the high-order topological relationships in the PPI network of protein modules targeted by a pair of drugs, as well as the protein modules associated with a specific cancer cell line. The pharmacological effects of drug combinations are explicitly evaluated by their therapy and toxicity scores. An attention component is also introduced in GraphSynergy, which aims to capture the pivotal proteins that play a part in both PPI network and biomolecular interactions between drug combinations and cancer cell lines. GraphSynergy outperforms the classic and state-of-the-art models in predicting synergistic drug combinations on the 2 latest drug combination datasets. Specifically, GraphSynergy achieves accuracy values of 0.7553 (11.94% improvement compared to DeepSynergy, the latest published drug combination prediction algorithm) and 0.7557 (10.95% improvement compared to DeepSynergy) on DrugCombDB and Oncology-Screen datasets, respectively. Furthermore, the proteins allocated with high contribution weights during the training of GraphSynergy are proved to play a role in view of molecular functions and biological processes, such as transcription and transcription regulation. The introduction of topological relations between drug combination and cell line within the PPI network can significantly improve the capability of synergistic drug combination identification.