Prediction of drugs having opposite effects on disease genes in a directed network

Background

Developing novel uses of approved drugs, called drug repositioning, can reduce costs and times in traditional drug development. Network-based approaches have presented promising results in this field. However, even though various types of interactions such as activation or inhibition exist in drug-target interactions and molecular pathways, most of previous network-based studies disregarded this information.

Methods

We developed a novel computational method, Prediction of Drugs having Opposite effects on Disease genes (PDOD), for identifying drugs having opposite effects on altered states of disease genes. PDOD utilized drug-drug target interactions with ‘effect type’, an integrated directed molecular network with ‘effect type’ and ‘effect direction’, and disease genes with regulated states in disease patients. With this information, we proposed a scoring function to discover drugs likely to restore altered states of disease genes using the path from a drug to a disease through the drug-drug target interactions, shortest paths from drug targets to disease genes in molecular pathways, and disease gene-disease associations.

Results

We collected drug-drug target interactions, molecular pathways, and disease genes with their regulated states in the diseases. PDOD is applied to 898 drugs with known drug-drug target interactions and nine diseases. We compared performance of PDOD for predicting known therapeutic drug-disease associations with the previous methods. PDOD outperformed other previous approaches which do not exploit directional information in molecular network. In addition, we provide a simple web service that researchers can submit genes of interest with their altered states and will obtain drugs seeming to have opposite effects on altered states of input genes at http://gto.kaist.ac.kr/pdod/index.php/main.

Conclusions

Our results showed that ‘effect type’ and ‘effect direction’ information in the network based approaches can be utilized to identify drugs having opposite effects on diseases. Our study can offer a novel insight into the field of network-based drug repositioning.

     

Drug Cocktail Optimization in Chemotherapy of Cancer

Background

In general, drug metabolism has to be considered to avoid adverse effects and ineffective therapy. In particular, chemotherapeutic drug cocktails strain drug metabolizing enzymes especially the cytochrome P450 family (CYP). Furthermore, a number of important chemotherapeutic drugs such as cyclophosphamide, ifosfamide, tamoxifen or procarbazine are administered as prodrugs and have to be activated by CYP. Therefore, the genetic variability of these enzymes should be taken into account to design appropriate therapeutic regimens to avoid inadequate drug administration, toxicity and inefficiency.

Objective

The aim of this work was to find drug interactions and to avoid side effects or ineffective therapy in chemotherapy.

Data sources and methods

Information on drug administration in the therapy of leukemia and their drug metabolism was collected from scientific literature and various web resources. We carried out an automated textmining approach. Abstracts of PubMed were filtered for relevant articles using specific keywords. Abstracts were automatically screened for antineoplastic drugs and their synonyms in combination with a set of human CYPs in title or abstract.

Results

We present a comprehensive analysis of over 100 common cancer treatment regimens regarding drug-drug interactions and present alternatives avoiding CYP overload. Typical concomitant medication, e.g. antiemetics or antibiotics is a preferred subject to improvement. A webtool, which allows drug cocktail optimization was developed and is publicly available on http://bioinformatics.charite.de/chemotherapy.     

Are We Heeding the Warning Signs? Examining Providers’ Overrides of Computerized Drug-Drug Interaction Alerts in Primary Care

Background

Health IT can play a major role in improving patient safety. Computerized physician order entry with decision support can alert providers to potential prescribing errors. However, too many alerts can result in providers ignoring and overriding clinically important ones.

Objective

To evaluate the appropriateness of providers’ drug-drug interaction (DDI) alert overrides, the reasons why they chose to override these alerts, and what actions they took as a consequence of the alert.

Design

A cross-sectional, observational study of DDI alerts generated over a three-year period between January 1st, 2009, and December 31st, 2011.

Setting

Primary care practices affiliated with two Harvard teaching hospitals. The DDI alerts were screened to minimize the number of clinically unimportant warnings.

Participants

A total of 24,849 DDI alerts were generated in the study period, with 40% accepted. The top 62 providers with the highest override rate were identified and eight overrides randomly selected for each (a total of 496 alert overrides for 438 patients, 3.3% of the sample).

Results

Overall, 68.2% (338/496) of the DDI alert overrides were considered appropriate. Among inappropriate overrides, the therapeutic combinations put patients at increased risk of several specific conditions including: serotonin syndrome (21.5%, n=34), cardiotoxicity (16.5%, n=26), or sharp falls in blood pressure or significant hypotension (28.5%, n=45). A small number of drugs and DDIs accounted for a disproportionate share of alert overrides. Of the 121 appropriate alert overrides where the provider indicated they would “monitor as recommended”, a detailed chart review revealed that only 35.5% (n=43) actually did. Providers sometimes reported that patients had already taken interacting medications together (15.7%, n=78), despite no evidence to confirm this.

Conclusions and Relevance

We found that providers continue to override important and useful alerts that are likely to cause serious patient injuries, even when relatively few false positive alerts are displayed.     

Predicting Pharmacodynamic Drug-Drug Interactions through Signaling Propagation Interference on Protein-Protein Interaction Networks

As pharmacodynamic drug-drug interactions (PD DDIs) could lead to severe adverse effects in patients, it is important to identify potential PD DDIs in drug development. The signaling starting from drug targets is propagated through protein-protein interaction (PPI) networks. PD DDIs could occur by close interference on the same targets or within the same pathways as well as distant interference through cross-talking pathways. However, most of the previous approaches have considered only close interference by measuring distances between drug targets or comparing target neighbors. We have applied a random walk with restart algorithm to simulate signaling propagation from drug targets in order to capture the possibility of their distant interference. Cross validation with DrugBank and Kyoto Encyclopedia of Genes and Genomes DRUG shows that the proposed method outperforms the previous methods significantly. We also provide a web service with which PD DDIs for drug pairs can be analyzed at http://biosoft.kaist.ac.kr/targetrw.     

Randomised Trial of Text Messaging on Adherence to Cardiovascular Preventive Treatment (INTERACT Trial)

Background

About one third of patients prescribed blood pressure or lipid-lowering drugs for the prevention of coronary heart disease and stroke do not take their medication as prescribed. We conducted a randomized trial to evaluate text messaging as a means of improving adherence to cardiovascular disease preventive treatment.

Methods

303 patients taking blood pressure and/or lipid-lowering medications were randomly assigned to being sent text messages (Text group, 151) or not being sent them (No text group, 152). Texts were sent daily for 2 weeks, alternate days for 2 weeks and weekly thereafter for 22 weeks (6 months overall), using an automated computer programme. Patients were asked to respond on whether they had taken their medication, whether the text reminded them to do so if they had forgotten, and if they had not taken their medication to determine if there was a reason for not doing so. At 6 months, use of medication was assessed.

Results

Two patients were lost to follow-up, providing data on 301 for analysis. In the No text group 38/151 (25%) took less than 80% of the prescribed regimen (ie. stopped medication completely or took it on fewer than 22 of the last 28 days of follow-up) compared to 14/150 patients (9%) in the Text group – an improvement in adherence affecting 16 per 100 patients (95% CI 7 to 24), p<0.001. The texts reminded 98/151 patients (65%) to take medication on at least one occasion and lead to 20/151 (13%) who stopped taking medication because of concern over efficacy or side-effects, resuming treatment.

Conclusions

In patients taking blood pressure or lipid-lowering treatment for the prevention of cardiovascular disease, text messaging improved medication adherence compared with no text messaging.

Trial Registration

Controlled-Trials.com ISRCTN74757601     

A Cluster Randomised Controlled Trial of a Pharmacist-Led Collaborative Intervention to Improve Statin Prescribing and Attainment of Cholesterol Targets in Primary Care

Background

Small trials with short term follow up suggest pharmacists’ interventions targeted at healthcare professionals can improve prescribing. In comparison with clinical guidance, contemporary statin prescribing is sub-optimal and achievement of cholesterol targets falls short of accepted standards, for patients with atherosclerotic vascular disease who are at highest absolute risk and who stand to obtain greatest benefit. We hypothesised that a pharmacist-led complex intervention delivered to doctors and nurses in primary care, would improve statin prescribing and achievement of cholesterol targets for incident and prevalent patients with vascular disease, beyond one year.

Methods

We allocated general practices to a 12-month Statin Outreach Support (SOS) intervention or usual care. SOS was delivered by one of 11 pharmacists who had received additional training. SOS comprised academic detailing and practical support to identify patients with vascular disease who were not prescribed a statin at optimal dose or did not have cholesterol at target, followed by individualised recommendations for changes to management. The primary outcome was the proportion of patients achieving cholesterol targets. Secondary outcomes were: the proportion of patients prescribed simvastatin 40 mg with target cholesterol achieved; cholesterol levels; prescribing of simvastatin 40 mg; prescribing of any statin and the proportion of patients with cholesterol tested. Outcomes were assessed after an average of 1.7 years (range 1.4–2.2 years), and practice level simvastatin 40 mg prescribing was assessed after 10 years.

Findings

We randomised 31 practices (72 General Practitioners (GPs), 40 nurses). Prior to randomisation a subset of eligible patients were identified to characterise practices; 40% had cholesterol levels below the target threshold. Improvements in data collection procedures allowed identification of all eligible patients (n = 7586) at follow up. Patients in practices allocated to SOS were significantly more likely to have cholesterol at target (69.5% vs 63.5%; OR 1.11, CI 1.00–1.23; p = 0.043) as a result of improved simvastatin prescribing. Subgroup analysis showed the primary outcome was achieved by prevalent but not incident patients. Statistically significant improvements occurred in all secondary outcomes for prevalent patients and all but one secondary outcome (the proportion of patients with cholesterol tested) for incident patients. SOS practices prescribed more simvastatin 40 mg than usual care practices, up to 10 years later.

Interpretation

Through a combination of educational and organisational support, a general practice based pharmacist led collaborative intervention can improve statin prescribing and achievement of cholesterol targets in a high-risk primary care based population.

Trial Registration

International Standard Randomised Controlled Trials Register ISRCTN61233866     

Benchmarking Human Protein Complexes to Investigate Drug-Related Systems and Evaluate Predicted Protein Complexes

Protein complexes are key entities to perform cellular functions. Human diseases are also revealed to associate with some specific human protein complexes. In fact, human protein complexes are widely used for protein function annotation, inference of human protein interactome, disease gene prediction, and so on. Therefore, it is highly desired to build an up-to-date catalogue of human complexes to support the research in these applications. Protein complexes from different databases are as expected to be highly redundant. In this paper, we designed a set of concise operations to compile these redundant human complexes and built a comprehensive catalogue called CHPC2012 (Catalogue of Human Protein Complexes). CHPC2012 achieves a higher coverage for proteins and protein complexes than those individual databases. It is also verified to be a set of complexes with high quality as its co-complex protein associations have a high overlap with protein-protein interactions (PPI) in various existing PPI databases. We demonstrated two distinct applications of CHPC2012, that is, investigating the relationship between protein complexes and drug-related systems and evaluating the quality of predicted protein complexes. In particular, CHPC2012 provides more insights into drug development. For instance, proteins involved in multiple complexes (the overlapping proteins) are potential drug targets; the drug-complex network is utilized to investigate multi-target drugs and drug-drug interactions; and the disease-specific complex-drug networks will provide new clues for drug repositioning. With this up-to-date reference set of human protein complexes, we believe that the CHPC2012 catalogue is able to enhance the studies for protein interactions, protein functions, human diseases, drugs, and related fields of research. CHPC2012 complexes can be downloaded from http://www1.i2r.a-star.edu.sg/xlli/CHPC2012/CHPC2012.htm.     

A Preliminary Investigation of Individual Differences in Subjective Responses to D-Amphetamine,Alcohol, and Delta-9-Tetrahydrocannabinol Using a Within-Subjects Randomized Trial

Polydrug use is common, and might occur because certain individuals experience positive effects from several different drugs during early stages of use. This study examined individual differences in subjective responses to single oral doses of d-amphetamine, alcohol, and delta-9-tetrahydrocannabinol (THC) in healthy social drinkers. Each of these drugs produces feelings of well-being in at least some individuals, and we hypothesized that subjective responses to these drugs would be positively correlated. We also examined participants’ drug responses in relation to personality traits associated with drug use. In this initial, exploratory study, 24 healthy, light drug users (12 male, 12 female), aged 21–31 years, participated in a fully within-subject, randomized, counterbalanced design with six 5.5-hour sessions in which they received d-amphetamine (20mg), alcohol (0.8 g/kg), or THC (7.5 mg), each paired with a placebo session. Participants rated the drugs’ effects on both global measures (e.g. feeling a drug effect at all) and drug-specific measures. In general, participants’ responses to the three drugs were unrelated. Unexpectedly, “wanting more” alcohol was inversely correlated with “wanting more” THC. Additionally, in women, but not in men, “disliking” alcohol was negatively correlated with “disliking” THC. Positive alcohol and amphetamine responses were related, but only in individuals who experienced a stimulant effect of alcohol. Finally, high trait constraint (or lack of impulsivity) was associated with lower reports of liking alcohol. No personality traits predicted responses across multiple drug types. Generally, these findings do not support the idea that certain individuals experience greater positive effects across multiple drug classes, but instead provide some evidence for a “drug of choice” model, in which individuals respond positively to certain classes of drugs that share similar subjective effects, and dislike other types of drugs.

Trial Registration

ClinicalTrials.gov {"type":"clinical-trial","attrs":{"text":"NCT02485158","term_id":"NCT02485158"}}NCT02485158     

A Network-Based Target Overlap Score for Characterizing Drug Combinations: High Correlation with Cancer Clinical Trial Results

Drug combinations are highly efficient in systemic treatment of complex multigene diseases such as cancer, diabetes, arthritis and hypertension. Most currently used combinations were found in empirical ways, which limits the speed of discovery for new and more effective combinations. Therefore, there is a substantial need for efficient and fast computational methods. Here, we present a principle that is based on the assumption that perturbations generated by multiple pharmaceutical agents propagate through an interaction network and can cause unexpected amplification at targets not immediately affected by the original drugs. In order to capture this phenomenon, we introduce a novel Target Overlap Score (TOS) that is defined for two pharmaceutical agents as the number of jointly perturbed targets divided by the number of all targets potentially affected by the two agents. We show that this measure is correlated with the known effects of beneficial and deleterious drug combinations taken from the DCDB, TTD and Drugs.com databases. We demonstrate the utility of TOS by correlating the score to the outcome of recent clinical trials evaluating trastuzumab, an effective anticancer agent utilized in combination with anthracycline- and taxane- based systemic chemotherapy in HER2-receptor (erb-b2 receptor tyrosine kinase 2) positive breast cancer.     

Genome-Scale Screening of Drug-Target Associations Relevant to Ki Using a Chemogenomics Approach

The identification of interactions between drugs and target proteins plays a key role in genomic drug discovery. In the present study, the quantitative binding affinities of drug-target pairs are differentiated as a measurement to define whether a drug interacts with a protein or not, and then a chemogenomics framework using an unbiased set of general integrated features and random forest (RF) is employed to construct a predictive model which can accurately classify drug-target pairs. The predictability of the model is further investigated and validated by several independent validation sets. The built model is used to predict drug-target associations, some of which were confirmed by comparing experimental data from public biological resources. A drug-target interaction network with high confidence drug-target pairs was also reconstructed. This network provides further insight for the action of drugs and targets. Finally, a web-based server called PreDPI-K_i was developed to predict drug-target interactions for drug discovery. In addition to providing a high-confidence list of drug-target associations for subsequent experimental investigation guidance, these results also contribute to the understanding of drug-target interactions. We can also see that quantitative information of drug-target associations could greatly promote the development of more accurate models. The PreDPI-K_i server is freely available via: http://sdd.whu.edu.cn/dpiki.     

Network-assisted investigation of antipsychotic drugs and their targets

Antipsychotic drugs are tranquilizing psychiatric medications primarily used in the treatment of schizophrenia and similar severe mental disorders. So far, most of these drugs have been discovered without knowing much on the molecular mechanisms of their actions. The available large amount of pharmacogenetics, pharmacometabolomics, and pharmacoproteomics data for many drugs makes it possible to systematically explore the molecular mechanisms underlying drug actions. In this study, we applied a unique network-based approach to investigate antipsychotic drugs and their targets. We first retrieved 43 antipsychotic drugs, 42 unique target genes, and 46 adverse drug interactions from the DrugBank database and then generated a drug-gene network and a drug-drug interaction network. Through drug-gene network analysis, we found that seven atypical antipsychotic drugs tended to form two clusters that could be defined by drugs with different target receptor profiles. In the drug-drug interaction network, we found that three drugs (zuclopenthixol, ziprasidone, and thiothixene) tended to have more adverse drug interactions than others, while clozapine had fewer adverse drug interactions. This investigation indicated that these antipsychotics might have different molecular mechanisms underlying the drug actions. This pilot network-assisted investigation of antipsychotics demonstrates that network-based analysis is useful for uncovering the molecular actions of antipsychotics.     

Machine learning liver-injuring drug interactions with non-steroidal anti-inflammatory drugs (NSAIDs) from a retrospective electronic health record (EHR) cohort

Drug-drug interactions account for up to 30% of adverse drug reactions. Increasing prevalence of electronic health records (EHRs) offers a unique opportunity to build machine learning algorithms to identify drug-drug interactions that drive adverse events. In this study, we investigated hospitalizations’ data to study drug interactions with non-steroidal anti-inflammatory drugs (NSAIDS) that result in drug-induced liver injury (DILI). We propose a logistic regression based machine learning algorithm that unearths several known interactions from an EHR dataset of about 400,000 hospitalization. Our proposed modeling framework is successful in detecting 87.5% of the positive controls, which are defined by drugs known to interact with diclofenac causing an increased risk of DILI, and correctly ranks aggregate risk of DILI for eight commonly prescribed NSAIDs. We found that our modeling framework is particularly successful in inferring associations of drug-drug interactions from relatively small EHR datasets. Furthermore, we have identified a novel and potentially hepatotoxic interaction that might occur during concomitant use of meloxicam and esomeprazole, which are commonly prescribed together to allay NSAID-induced gastrointestinal (GI) bleeding. Empirically, we validate our approach against prior methods for signal detection on EHR datasets, in which our proposed approach outperforms all the compared methods across most metrics, such as area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve (AUPRC).     

Distinctive Behaviors of Druggable Proteins in Cellular Networks

The interaction environment of a protein in a cellular network is important in defining the role that the protein plays in the system as a whole, and thus its potential suitability as a drug target. Despite the importance of the network environment, it is neglected during target selection for drug discovery. Here, we present the first systematic, comprehensive computational analysis of topological, community and graphical network parameters of the human interactome and identify discriminatory network patterns that strongly distinguish drug targets from the interactome as a whole. Importantly, we identify striking differences in the network behavior of targets of cancer drugs versus targets from other therapeutic areas and explore how they may relate to successful drug combinations to overcome acquired resistance to cancer drugs. We develop, computationally validate and provide the first public domain predictive algorithm for identifying druggable neighborhoods based on network parameters. We also make available full predictions for 13,345 proteins to aid target selection for drug discovery. All target predictions are available through canSAR.icr.ac.uk. Underlying data and tools are available at https://cansar.icr.ac.uk/cansar/publications/druggable_network_neighbourhoods/.     

Indication Alerts Intercept Drug Name Confusion Errors during Computerized Entry of Medication Orders

Background

Confusion between similar drug names is a common cause of potentially harmful medication errors. Interventions to prevent these errors at the point of prescribing have had limited success. The purpose of this study is to measure whether indication alerts at the time of computerized physician order entry (CPOE) can intercept drug name confusion errors.

Methods and Findings

A retrospective observational study of alerts provided to prescribers in a public, tertiary hospital and ambulatory practice with medication orders placed using CPOE. Consecutive patients seen from April 2006 through February 2012 were eligible if a clinician received an indication alert during ordering. A total of 54,499 unique patients were included. The computerized decision support system prompted prescribers to enter indications when certain medications were ordered without a coded indication in the electronic problem list. Alerts required prescribers either to ignore them by clicking OK, to place a problem in the problem list, or to cancel the order. Main outcome was the proportion of indication alerts resulting in the interception of drug name confusion errors. Error interception was determined using an algorithm to identify instances in which an alert triggered, the initial medication order was not completed, and the same prescriber ordered a similar-sounding medication on the same patient within 5 minutes. Similarity was defined using standard text similarity measures. Two clinicians performed chart review of all cases to determine whether the first, non-completed medication order had a documented or non-documented, plausible indication for use. If either reviewer found a plausible indication, the case was not considered an error. We analyzed 127,458 alerts and identified 176 intercepted drug name confusion errors, an interception rate of 0.14±.01%.

Conclusions

Indication alerts intercepted 1.4 drug name confusion errors per 1000 alerts. Institutions with CPOE should consider using indication prompts to intercept drug name confusion errors.     

ASDCD: Antifungal Synergistic Drug Combination Database

Finding effective drugs to treat fungal infections has important clinical significance based on high mortality rates, especially in an immunodeficient population. Traditional antifungal drugs with single targets have been reported to cause serious side effects and drug resistance. Nowadays, however, drug combinations, particularly with respect to synergistic interaction, have attracted the attention of researchers. In fact, synergistic drug combinations could simultaneously affect multiple subpopulations, targets, and diseases. Therefore, a strategy that employs synergistic antifungal drug combinations could eliminate the limitations noted above and offer the opportunity to explore this emerging bioactive chemical space. However, it is first necessary to build a powerful database in order to facilitate the analysis of drug combinations. To address this gap in our knowledge, we have built the first Antifungal Synergistic Drug Combination Database (ASDCD), including previously published synergistic antifungal drug combinations, chemical structures, targets, target-related signaling pathways, indications, and other pertinent data. Its current version includes 210 antifungal synergistic drug combinations and 1225 drug-target interactions, involving 105 individual drugs from more than 12,000 references. ASDCD is freely available at http://ASDCD.amss.ac.cn.